Free will and the limits of reason

When we say that a person has free will, we generally mean that he chooses his own actions and that, for every choice he makes, he could have chosen otherwise.

That “could have” is a little slippery. Usually we mean it in a counterfactual sense. When a falling stone just misses my head and I say that it could have hit me, I am not claiming that the motion of the stone was not deterministic; I mean simply that it would have hit me if conditions (my location, the direction of the wind, etc.) had been slightly different. When we say that a person could have chosen otherwise, it could likewise be interpreted to mean that would have chosen differently if certain conditions (such as his character, motives, or mood) had been slightly different — but proponents of free will generally mean something much stronger: that exactly the same person in exactly the same situation and the same mental state could have chosen otherwise — that precisely the same set of causes could just as easily have resulted in a different effect. The actions of a person who has free will are supposed to be impossible to predict even in principle, even by a psychic Laplace’s demon with complete information about the person’s character and mental state. But — and this is where the contradiction comes in — the actions are still seen as being decided by the person.

Basically, the claim that I have free will boils down to an assertion that both of the following propositions are true:

  1. I determine my actions.
  2. My actions are not determined.

Thus, the claim that I have free will is logically false. Free will is not just something which we happen not to have because the universe is deterministic; it is something fundamentally incoherent, which could not possibly exist in any conceivable universe.


When I made a similar argument in my post Free will: a problem for everyone, Bruce Charlton left the following comment:

Reason reasons-about only that which follows reason. Therefore if there was anything which did not follow reason, such a phenomenon would necessarily be invisible to reason.

So, the argument you make excludes the possibility of free will, from the initial assumptions. . . . reasoning excludes free will a priori – reason does not disprove free will.

Dr. Charlton goes on to compare reason’s exclusion of free will to the way science excludes miracles and the supernatural — not because it can disprove them, but precisely because it cannot disprove them, because they are deemed untestable and thus outside the domain of science.

Like science, reason supports some things, disproves others, and excludes still others from its domain. Simple propositions, for example, are excluded. Reason can’t tell you whether all men or mortal or whether Socrates is a man; it can only tell you that if those two propositions are true then it follows that Socrates is mortal. All our first principles and empirical facts come from outside reason and cannot be disproved by reason. What reason can disprove is a related set of propositions, and it does this by showing that the set is self-contradictory — but this is precisely what reason does with regard to free will, as I have shown above. If reducing something to a contradiction doesn’t count as disproving it, it’s hard to imagine what would qualify.


So, given that reason disproves free will, is there any respectable way to go on believing in free will anyway? Usually when you run into a contradiction, you can question the premises, but in this case the only two premises are part of the definition of free will; to reject either or both of the premises is to reject free will. (You can still use the term “free will” for whatever is left — that’s what most philosophers do — but you’d still be rejecting free will in the sense that most non-philosophers intuitively understand and believe in it.)

Since ditching the premises is out, the next option is to remember that deductive reason is infallible only in theory. Not everything that is logically true is obvious (that’s why math tests are hard), and an argument which looks airtight could still be fallacious in ways that escape our notice. Most people justifiably reject Zeno’s proof that motion is impossible even if they can’t find any logical flaw in his reasoning; the overwhelming empirical evidence that fleet-footed warriors can in fact catch tortoises overrides the claims of reason. In a similar way, and on similar empirical grounds, most scientists continue to treat both relativity and quantum theory as true even though they are known to be logically incompatible.

So, if there were strong empirical evidence that we do in fact have free will, it would be reasonable to dismiss the above disproof of it as a curious paradox which, like Zeno’s, must somehow be wrong even if we can’t understand how — if there were evidence. But in fact the only evidence we have for free will is summed up well by Spinoza (Ethics II, Prop. 35, Schol.):

Men are deceived in that they think themselves free, i.e., they think that, of their own free will, they can either do a thing or forbear doing it, an opinion which consists only in this, that they are conscious of their actions and ignorant of the causes by which they are determined. This, then, is their idea of freedom — that they do not know any cause of their actions.


Filed under Philosophy

8 responses to “Free will and the limits of reason

  1. Bruce G Charlton

    Given that you will be aware that using reason to deny free will is itself paradoxical ,I am guessing that you are trying to find the flaw in your argument?

    One general point is to ask whether the great logicians who did believe in free will (e.g. everybody up to the 17th century) were unaware of the simple argument you outline. Clearly they were not. So either they are dishonest or grossly incompetent; or else there is more to be said.

    So far as I can see, the argument you make is that this is a wholly deterministic universe (in which everything is precisely caused by something else) therefore there is no such thing as free will. (and no such things as randomness either).

    Inter alia, this argument leads to one of Aristotle’s/ Aquinas’s proofs of God (the ‘second way’, I think) – since otherwise there is an infinite regress of causes. There must be an uncaused cause which keeps everything going, moment to moment. (Of course this God is only a small part of the Christian God. The Christian God is not wholly deducible from reason – only partly so.)

    I personally find this argument very obvious and compelling. And if it is valid, then the existence of God then adds another factor to the mix.

    Another aspect is that free will is an attribute of the soul – and the soul is conceptualized differently from other things in the universe. Specifically, it is non-material. We don’t have any strong intuitions about whether determinism applies to the non-material.

    More specifically, for Christians (and some others) free will is specifically attributed to the human soul, and not to (many) other things. So (roughly) the Christian metaphysician would say – yes – you are right that the universe is deterministic – except for human souls, which have free will, because it was given by God.

    Indeed Christianity is nonsense without free will; or, free will is an axiom of salvation.


    In sum, I think that the argument you make is merely a reductio ad absurdum of secular materialism.

    A secular materialism which regards the whole universe as rational and deterministic with no exceptions is simply nonsense. It is self-refuting.

    Secular materialism has no grounds for believing in the validity of reason or the deterministic nature of the universe, since although these seem obviously valid to us, there is no reason that their obviousness to a contingent and purposeless temporary organism such as the human being is (to secular materialism) means that they are true.

    Such beliefs are more likely a delusion, or random nonsense of some kind – after all why should a contingent organism have access to any truths of any kind? No reason that I can see.

    So there are no grounds from which reason can destroy free will, because the grounds used to validate reason are also grounds for validating free will.

  2. No, I am most certainly NOT arguing that this is a deterministic universe and that therefore there is no such thing as free will! I can see I have completely failed to make myself clear.

  3. Bruce G Charlton

    Sorry – apparently I was attacking a straw man of my own imagination!

    Would you care to try again in explaining to me?

  4. Here’s another stab at making my point:

    If the universe is deterministic, there is no free will — because we cannot act otherwise than we do, and because all our actions are ultimately determined by events that happened before we were born.

    If the universe is not deterministic, there is also no free will — because if my actions are not determined, it follows that they are not determined (i.e., decided, chosen) by me.

    Therefore, regardless of whether or not the universe is deterministic, free will is impossible — not impossible like miracles are impossible, but impossible like square circles are impossible. No matter what the facts may happen to be with regard to determinism, God, the soul, etc., there can be no free will (as that term is popularly understood) because it is a logically incoherent concept.

    Is that any clearer? I feel like I’m just repeating the same things and not getting through.

    I’m glad you asked for an explanation, though, because when I sat down to try to spell everything out as explicitly as possible, Spinoza-style, I discovered that I wasn’t able to do so, that my own thinking on this topic had been much less rigorous than I had previously imagined — that, for example, I was unable even to give satisfactory definitions of such basic terms as “determinism” and “possible.” Another problem is that my opinions about free will were formed before I had really come to understand relativity and the nature of time, and they are sorely in need of an eternalist/antisolipsist update. I’ve been thinking these and related topics over in recent weeks and have hit upon some potentially fruitful ideas, which I will post here whenever I find the time.

  5. Bruce G Charlton

    I think I see your line of argument – but surely it is not coherent?

    Surely an algebraic formulation cannot say anything about the real universe?

    Otherwise you could substitute any imprecise and undefined word into your formulation and claim to have discovered something about reality.

    Here you make determinism your assumption (and seem to discover no place for free will), but surely you could equally well make free will your assumption and ‘discover’ no place for determinism (in the sense you seem to mean)?; and logical analysis does not allow you to choose one of these terms over another.

  6. Bruce, my rejection of free will does not depend on assuming determinism, nor does logic allow me to choose whichever of the two I prefer and reject the other.

    Where D means “the universe is deterministic” and F means “we have free will,” you seem to think I’m saying ~(D&F) and that therefore D&~F and F&~D are both equally logical. But in fact my argument is that (~(D&F))&(~(~D&F)), so ~F is the the only possible conclusion.

  7. O… K…

    I’m afraid that means nothing to me (I am one of those who translate algebra into words rather than the reverse).

    But maybe the key point is to ask – do you regard the material physical universe as everything? or do you concur with the vast majority of historical humankind that there are also immaterial realities (such as the soul, spirits etc) – which would have properties such as being undetectable to physical instrumentation?

    I suspect that the denial of free will is an aspect of materialism – an aspect of the denial of the soul, supernatural etc. There is not really a problem with understanding the possibility of free will if the soul is regarded as real and (for example) angels, spirits, demons etc. – which were usually regarded as having a non-material nature.

    The point is (again!) that when free will is excluded from the metaphysical system , of course analysis of the system makes free will seems impossible. When free will is built into the metaphysical system then there is not this problem.

    Humans have a natural spontaneous metaphysical system which includes the immateruial, and free will. The main question (for me) is whether there was *ever* any reason for the evolution of thought to reject the immaterial (and reject free will) and insist that only the material is real – as Western Intellectuals have believed for several generations.

    There can never (in principle) be proof (ie scientific type proof) of the validity of a metaphysical system – and there don’t seem to be any compelling reasons for rejecting the spontaneous metaphysical system of humans – only ‘reasons’ related to the specialization and professional evolution of philosophy and modern intellectual life.

  8. Pingback: The necessity of agency | Bugs to fearen babes withall

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s