The problem of free will is one of the more vexing problems in philosophy, as well as one that seems to have important practical ramifications given its bearing upon great questions relating to ethics, theology, interpersonal relationships, political theory, criminal justice, and more. It is a problem that has considerable personal significance to me, as the vexation it caused me during my undergraduate days propelled me into graduate school (for better or worse). In the light of this, I have many - probably too many - thoughts about it, so this will undoubtedly be but the first of many posts I write on the subject. All I aim to do here is set out the problem, which I will do by presenting two arguments for the conclusion that free will is impossible. The point is not to convince you that free will is, in fact, impossible (in fact, I rather fear convincing anyone of that) - but, instead, to make vivid why that conclusion is surprisingly difficult to resist, as well as what sort of moves one would have to make in order to resist it.
I should begin by specifying what I mean by “free will”. Free will, as I understand it, is a capacity possessed by some agent, S, if and only if:
(a) S makes choices,
(b) when S makes choices, there are actual alternative possibilities open to S, and
(c) S is responsible for at least some of S’s choices in such a way that S can deserve praise and blame (and, potentially, reward and punishment) in virtue of these choices (thereby making it the case that S is morally responsible1 for these choices).
This is, I think, roughly the commonsense notion of free will. Some philosophers would insist that I am wrong in this judgment; that, in fact, ordinary people (the “folk”, as philosophers once liked to say…) do not believe either that they can do otherwise than they do, or that there is any essential connection between free will and moral responsibility. And some of these philosophers believe that they have some empirical evidence from experimental philosophy (“x-phi”, as we call it) on their side. As I will perhaps explain in some future post, I am unconvinced that the findings of any studies actually do support the conclusion that the commonsense view of free will is significantly less robust than I have supposed. But for now, I will set that matter aside and trust that the notion of free will under discussion is a perfectly natural one.
The problem of free will is often cast in terms of determinism. Determinism is the view according to which, at any given moment in time, there is only one possible future, given the past and the laws of nature.2 If determinism is true, then there are never actual (as opposed to merely hypothetical) alternative possibilities, and we are not morally responsible for any of our choices, given that the way in which we are when making them is causally fixed by factors far beyond our control. So, read straightforwardly, conditions (b) and (c) above cannot be satisfied on determinism.
It is widely thought that if determinism is false, then there is room for free will. It is certainly true that if determinism is false, then this opens the possibility that at least some of our choices are not causally necessitated. The trouble is that undetermined choices are not obviously choices at all. For it seems constitutive of something’s counting as a choice made by someone, S, that it must have been brought about by S. But, arguably, a choice that is undetermined cannot properly be said to have been brought about by the agent who made it. Indeed, it arguably cannot be said to have been brought about by anything at all, so that an “undetermined choice” would not be under the control of the agent who made it (given that it would not flow from any of that agent’s psychological states but would, instead, simply happen to the agent).
Given that, necessarily, determinism or its negation must be true (one need not exactly be a formally trained logician to get that), the free will skeptic can construct The Dilemma of Determinism as follows.
(1) Either: determinism is true, or it is false.
(2) If determinism is true, then free will is impossible (given that if determinism is true, then we never choose between actual alternatives and are not responsible for our choices).
(3) If indeterminism is true, then free will is impossible (given that we cannot be responsible for undetermined choices, since such choices cannot be under our control).
(4) Therefore, free will is impossible.*
This argument is certainly formally valid (i.e., its premises entail it conclusion), and (1) is logically necessary. Compatibilists insist that (2) is false, because they redefine free will in such a manner as to make it compatible with determinism. Compatibilists have produced ingenious work identifying the various varieties of freedom that are possible within a causally determined universe, but whether any of these varieties of count as free will is another question. Metaphysical libertarians (not to be confused with political libertarians!) deny (3), insisting that undetermined events are not necessarily random. Personally, I have always struggled to see how this could be so, but given that minds far greater than mine have endorsed metaphysical libertarianism, I allow for the possibility that I am simply missing something.
There is a route to free will skepticism that bypasses talk of determinism and negation by way of an old argument that has been developed most clearly and forcefully in recent years by Galen Strawson (who, full disclosure, was my dissertation supervisor). While Strawson’s full argument, which can be found in “The Impossibility of Moral Responsibility” and Freedom and Belief (see below for the full references) is, of necessity, long and technical, the basic insight driving it is quite straightforward.
Strawson notes, first, that to be morally responsible for what we do, we must be responsible for the way that we are, mentally, in certain fundamental respects, so that we are in this way ultimately responsible for what we do (i.e., Strawson thinks that moral responsibility is secured by ultimate responsibility).
Second, he points out that no one is ultimately responsible for the way in which they are in any respect, given that the way in which we are at a given time t is, necessarily, fixed by factors over which we have no control at t. We could be ultimately responsible for the way that we are only if we could create ourselves – if we could be causa sui. But it seems that nothing could be a causa sui, for such an entity would, by definition, have to exist and not exist at the same time.
Strawson states an abbreviated version of his Basic Argument in an article for The Stone as follows.
“(1) You do what you do – in the circumstances in which you find yourself – because of the way you then are.
(2) So if you’re going to be ultimately responsible for what you do, you’re going to have to be ultimately responsible for the way you are – at least in certain mental respects.
(3) But you can’t be ultimately responsible for the way you are in any respect at all.
(4) So you can’t be ultimately responsible for what you do.”
In defense of (3), Strawson offers the following argument.
“(a) It’s undeniable that the way you are initially is a result of your genetic inheritance and early experience.
(b) It’s undeniable that these are things for which you can’t be held to be in any way responsible (morally or otherwise).
(c) But you can’t at any later stage of life hope to acquire true or ultimate moral responsibility for the way you are by trying to change the way you already are as a result of genetic inheritance and previous experience.
(d) Why not? Because both the particular ways in which you try to change yourself, and the amount of success you have when trying to change
yourself, will be determined by how you already are as a result of your genetic inheritance and previous experience.
(e) And any further changes that you may become able to bring about after you have brought about certain initial changes will in turn be determined, via the initial changes, by your genetic inheritance and previous experience.”
Now, it might be thought, given Strawson’s invocation of early experiences and genetic inheritance, that his argument is a problem only if naturalism is true. On a common way of cashing out the notoriously hard-to-define view, naturalism holds that reality consists only of those entities and properties posited by the natural sciences.
However, Strawson’s basic point is that it is impossible to “get behind ourselves”, as it were, in order to create ourselves; this is a conceptual point that holds even if naturalism is false and we are not material beings, but immaterial minds, or souls. Just as we are unable to bring about the natural facts that make us who we are if naturalism is true, so we are unable to bring it about that we possess the immaterial mind or soul that we do if some form of non-naturalism, or supernaturalism, is true.
Strawson’s argument does not entail that compatibilists (and libertarians, if libertarian freedom is in fact coherent) do not describe real and potentially significant varieties of freedom; the Basic Argument entails only that such accounts fail to secure ultimate responsibility. And insofar as the intuitive, or commonsense, variety of free will is bound to the concept of ultimate moral responsibility, the Basic Argument entails that free will, as experienced and (at least implicitly construed) by the average person is impossible.
It is almost impossible to deny that The Basic Argument proves the impossibility of something. Whether you think it proves the impossibility of free will will depend upon whether you agree with Strawson that free will is bound to ultimate moral responsibility, as he conceives it (i.e., as requiring self-creation). A common way of resisting the argument is by insisting that free will is tied to a more limited form of responsibility.
The conclusion of both The Dilemma of Determinism and the Basic Argument is the same: free will, conceived as a capacity, the possession of which confers ultimate responsibility for at least some of one’s choices, is impossible. This conclusion, if true, would have important practical and moral ramifications and, therefore, matters a great deal. Most significantly, if ultimate moral responsibility is impossible, then it follows that our moral, legal, political, and interpersonal practices need to be reformed insofar as they contain retributivist elements that cannot be justified solely in terms of their instrumental value. For if no one is ultimately responsible for what they do, then (contra the retributivists) there is nothing intrinsically valuable in anyone’s suffering. I will devote some future posts to exploring what I take to be the consequences of free will skepticism.
It should mention two things that clearly do not follow from free will skepticism. First, it does not follow from free will skepticism that ethical nihilism – the view according to which nothing matters objectively, and there are no moral facts – is true. It would, to my mind, be a bizarre view of moral facts indeed that understood them to hold only if free will is possible. Inflicting suffering for fun is sure wrong regardless of the capacities we humans do, or do not, possess.
Moreover, it does not follow from free will skepticism that Fatalism, the view according to which the same future will unfold regardless of what we do, is true. Free will skeptics do not claim that our acts are causally inefficacious, or that the future would not unfold differently if we acted differently, but only that we are not free in the sense of being ultimately in control of whatever it is that we in fact end up doing.
That thought may at first seem bleak. Perhaps in one way it is. There is, regrettably, no law of nature dictating that the truth must be what we would wish.
However, it is, perhaps, open to one to affirm the reality of free will on the grounds that its reality is obvious and fundamental (perhaps a prerequisite to rational deliberation itself), even if an account of it is hopelessly elusive. Although I cannot myself quite endorse this mysterian position, it has been adopted by no lesser minds than Immanuel Kant,3 Thomas Nagel, and Peter Van Inwagen, and I will have more to say about it in the future.
References
Strawson, Galen. 1986/2010. Freedom and Belief. Oxford: New York.
Strawson, Galen. 1994. “The Impossibility of Moral Responsibility”. Philosophical Studies, An International Journal For Philosophy in the Analytic Tradition. Vol. 75, No. 1/2: pp. 5-24.
Moral responsibility just is that status in virtue of which praise and blame, punishment and reward can be deserved.
For my own part, I think it matters a great deal whether irreducible agents figure in the causal story, but I will leave this subtlety for another time.
I, at least, think it fair to describe Kant as a mysterian; however, as he was both a brilliant thinker and a famously obscure writer, interpretations vary.
>>>For if no one is ultimately responsible for what they do, then (contra the retributivists) there is nothing intrinsically valuable in anyone’s suffering.<<<
I would say that *even if* libertarian Free Will is true, and agents really can be ultimately morally responsible for their evil acts, that *even then* there would be nothing intrinsically valuable in their suffering. Even in that case, I think the only value to suffering is its ability to change behavior and deter evil actions.
BTW, I think the arguments against the coherency of LFW are conclusive... but I believe in it anyway 🤷🏻♂️--albeit in a very limited capacity. I guess I'd have to consider myself a mysterian about LFW.