EDIT (17/08/2022): I have answered this question with an evolution of the argument. See accepted answer below.

There are a number of arguments which aim to prove the impossibility of free will.

The Standard Argument (incorporating the Determinism Objection and the Randomness Objection) is well known and powerful, although subject to a variety of criticisms.

I seek here to provide an argument immediately testable via personal investigation.

Is the following argument sound?

Note: here, an 'act' is defined as 'a thing done', as per Oxford Languages definition #2.

  1. In order for an act to be voluntary, a person must decide to perform it.

  2. A decision is an act. Therefore, in order for a decision to be voluntary, a person must decide to decide it.

  3. This leads to an infinite regress of prior decisions, in which any voluntary decision requires an infinite chain of prior decisions.

  4. Insofar as free will requires the ability to make voluntary decisions, free will is impossible.

Related reading:

  • 1
    Comments are not for extended discussion; this conversation has been moved to chat.
    – Philip Klöcking
    Commented Aug 4, 2021 at 14:55
  • 1
    @GuyInchbald Sorry, you have been unpersoned. You must use SE chat or face deletion. There is no alternative, that's how this works. Commented Aug 4, 2021 at 16:10
  • 1
    @user253751 Sorry I haven't a clue what is going on here. I was clarifying the original question, which is what comments are for. The Help centre has no hits for "unperson", philosophy.stackexchange.com/help/search?q=unperson and my personal page carries no hint that I can see. Would you mind pointing me to where all this is explained, so we can get off this thread? Commented Aug 4, 2021 at 17:00
  • 1
    On your recent edit comment, while I accept that no argument below has been convincing enough to qualify as "the right answer", I'm not sure that all of them lack detail.
    – Paul Ross
    Commented Dec 7, 2021 at 18:12
  • 3
    It's the old "you can do what you want, but how do you want what you want?" argument. It's a valid one. It can be countered by stating that "wants" are uncaused: we wouldn't "want to want to do something", just directly "want to do something". Alas, this counter (1) does not prove anything, just asserts an unfalsifiable ad hoc hypothesis, (2) does not explain how this uncaused "want" triggers a chain of physical events that results in performing an action.
    – armand
    Commented Jul 22, 2022 at 7:41

15 Answers 15


Seems like no one brought up Frankfurt and hierarchical compatabilism.

  • First-order desires: desires that are directed to objects or states of affairs.
    We desire things like being healthy, being well-informed, and being paid.

  • Second-order desires: self-conscious beings are not only aware of the first-order desires, but can have desires about those desires.
    A smoker can have the desire to not desire to smoke.

  • Second-order volitions: second-order desires we want to act on.
    There are second-order desires we don't wish to act on, like a priest desiring to know what it's like to be married.

To act freely, says Frankfurt, is to act on a second-order volition. If you do not formulate second-order volitions, or do not act on the ones you do form, your actions are not free - you are a slave to your first-order desires.

According to Frankfurt, free actions are caused by second-order volitions that one decisively identifies with. "Decisively identifies with" is a necessary condition to forestall an infinite regress. We may formulate a third-order desire and so on, but a second-order desire that we decisively identify with, claims Frankfurt, "resounds" throughout the potentially endless array of higher orders, halts the regress, and brings coherence to our preference structure.

Note that traditional compatabilism maintains that you cannot act freely if your actions are externally constrained. Frankfurt denies this. As long as you act on your second-order volitions, you are responsible for you actions, whether or not you could do otherwise.

From Doing Philosophy 4th ed.. Theodore Shick Jr, Lewis Vaugh. 2010. Pp. 220-225

  • 1
    Makes me think of Sartre's claim that a prisoner is as free as anyone else
    – CriglCragl
    Commented Dec 5, 2021 at 23:57
  • 1
    @CriglCragl. Thanks. After reading your comment I found this article, which goes into a bit more detail (See paragraph 10 for the prisoner ref.). Commented Dec 7, 2021 at 11:57
  • 1
    @Dcleve We clearly never experience endless recursive decisions though. I take hierarchical compatibilism as saying call that final volition before a free action second order, even if there are more layers of decisions than two. As long as we are not slaves to our desires, we are free. We clearly are not slaves to desire, and we decisively identify with choices and act in finite time. The cut off is not arbitrary. We really do make decisive choices and call them free. This is compatibilsm after all, we experience free choice even if there is causal determinism.
    – J Kusin
    Commented Dec 7, 2021 at 22:47
  • 1
    @JKusin -- But if you have abandoned the necessity of applying causal chains of logical justification at the 2nd step, one can much more justifiably abandon it at the first step. This removes any justification for assuming causal determinism. I point this out in a newly posted answer, referencing empirical pragmatism as an alternate truth reference. :-)
    – Dcleve
    Commented Dec 7, 2021 at 23:06
  • 1
    @Futilitarian it was for an intro class And I found it good for that purpose. Wide breadth of concepts
    – J Kusin
    Commented Dec 8, 2021 at 15:08

This argument constructs a paradox of the type popularized by Zeno, i.e.:

  • one cannot do x until one has done x'
  • one cannot do x' until one has done x''
  • one cannot do x'' until one has done x'''...
  • therefore one can never do x

This is how Zeno argued that Achilles could never finish a race because — logically speaking — Achilles could never begin a race. There is always some minuscule distance he must traverse before he can traverse any distance at all.

Of course, this logical argument never actually stopped anyone from finishing a race, much less Achilles, which is why it's normally referred to as a paradox. It doesn't point out a flaw in the universe; it points out a weakness in our system of logic. Nor does the given argument prevent anyone from having the experience of making a free choice. If I choose to run a footrace, I will (most likely) cross the finish line; all the arguments to the contrary won't change either the choice or the outcome.

  • We can demonstrate the contradiction between Zeno's race and reality. Demonstrating how to escape the infinite regress of required decisions seems to be far more difficult. As for the experience of free choice, arguments to the contrary may not change the outcome, but they have huge ramifications for the experience. The possibility free will is illusory; that we may all be merely the embodiments of unique sets of circumstances - some of us far luckier than others - moves me to be far more compassionate towards, and non-judgemental of, others and myself (Perhaps therein lies the attraction). Commented Dec 8, 2021 at 5:26
  • 3
    @Futilitarian: The essence of scientific reasoning is that observation trumps theory; if we have an observation and a theory that are mutually incompatible, we retain the observation and discard (or modify) the theory. We all observe ourselves making decisions. There are obvious constraints and limitations, but within those boundaries we don't see a 'machine'; we see a choice. Further observation might eventually contradict that impression (the way astrophysics contradicts the impression that the sun rises and sets). But we currently have no reason to doubt the existence of will. Commented Dec 8, 2021 at 5:41
  • @Futilitarian: The infinite regress you mention is a red herring; will can be illusory without this kind of infinite regress, and invoking it doesn't prove anything. Besides, what you're really attacking (philosophically speaking) isn't 'will' at all; the argument attacks morality. The only purpose of 'choice' is to select better or worse outcomes. If there's no choice, there's no better or worse, but merely 'what is'. In that context serial killers and saints have the same moral standing — they both do exactly what they must do — and every act is as justified as every other. Commented Dec 8, 2021 at 5:48
  • @Futilitarian: Even absurdism gives us the choice of constructing meaning out of an otherwise meaningless reality. If you take that away, what are we? Commented Dec 8, 2021 at 5:49
  • Observation trumping theory makes perfect sense until we consider theory which questions the reliability of observation (in this case; the 'observation' that we have free will). We have (almost) no reason to doubt the existence of will, but we do have good reason (via Strawson for example) to doubt that it is free. Infinite regresses seem quite capable of proving impossibility. The impossibility of moral responsibility and traditional 'meaning' are but two consequences of free will's absence; they are not the driver of relevant theory, and a theory's repugnance is no reason to reject it. Commented Dec 8, 2021 at 7:13

You are always free to decide what you wanna decide to decide what you decide about what you wanna decide about what you wanna decide about to decide your final decision what to decide how you decide the decision of making a free decision about what you wanna decide so the decision will not be decided by others who wanna decide for you what you are supposed to decide if you have to make a decision what to decide if you wanna decide what the others have to decide if they wanna make you decide what decision to make about a decision how to decide the decision they want you to make about you deciding how they should decide deciding your decision to make a decision how every one should make decisions about decisions that in general decide how to make decisions for them for deciding a decision.

It's your decision.


The infinite regress argument does not work because it stops after two steps. "They must be able to decide what to decide what to decide" is not valid English grammar for a reason.

You have said that they must decide what to decide, and that's the end of it.

Task: Decide what to decide.
Result: What to eat.
Task: Decide what to eat.
Result: Pizza.

  • 2
    The fact that the expression "They must be able to decide what to decide to decide...", becomes almost incoherent is precisely the point of the argument. It is also a perfectly legitimate example of English grammar. Your claim that "that's the end of it" doesn't make much sense to me. If you provide more information, I'm happy to clarify. In the meantime: The final 'decision', D, requires a prior decision, PD, as to whether D is desirable. Without this arbitration, D is made involuntarily. It is an illusory decision. But if D illusory, so is PD (without PD2), and so on, ad infinitum... Commented Aug 4, 2021 at 14:59
  • 2
    It sounds like you have a problem with how he worded it, rather than the actual concept he's communicating @user253751
    – TKoL
    Commented Aug 4, 2021 at 15:05
  • @Futilitarian What is PD2? Can you express it? Commented Aug 4, 2021 at 15:15
  • @User. Prior Decision 2. This might not be the best term. Come up with anything you like. It's simply a description of the next link in the ridiculous 'decision' chain that arises out of an attempt to get to a true, non-illusory decision, a phenomenon I'm claiming to be impossible. Commented Aug 4, 2021 at 15:29
  • @Futilitarian but what is Prior Decision 2? What is the question whose answer needs to be decided? Commented Aug 4, 2021 at 15:37

Apologies, our discussion of what you mean has been cut short. So here is my best shot.

You write:

In order for a person to be in conscious control over their decisions, they must be able to decide what to decide (otherwise, the decision becomes involuntary...

But you said in the deleted discussion that you are not proposing a chain of meta-decisions distinct from the causal chain. This appears to contradict your notion to "decide what to decide", as there is no such prior decision.

It seems to me that you are conflating "decisions about decisions" with prior causes.

There is no prior decision event to require free will; the regress of decisions never gets started.

There is however a chain of prior causes, and this "regress" of causes is of course valid.

All this sheds no light on whether the one actual decision is free or not.

  • 1
    Again, I didn't say that. Again, I stated explicitly, "I have not at any time argued for any "prior meta-decision as in any way distinct from the causal chain". I can't be any clearer. "Decisions about decisions" (should they exist) would be prior causes. I've addressed your next point about 'prior decision event' already, succinctly. I'm happy to continue in chat, but this is the third time you've made an inaccurate claim about my comments, and we're simply getting nowhere. Thanks for the chat. All the best. Bye. Commented Aug 4, 2021 at 15:25
  • @Futilitarian The point is, your statement that "they must be able to decide what to decide (otherwise, the decision becomes involuntary." is wholly unjustified. The condition you set up for your straw free will cannot in fact be set up. If I claim that I am perfectly capable of freely deciding what to do, without any prior decisions-that-aren't-decisions or whatever, on what basis do you refute that? Commented Aug 4, 2021 at 15:31

The Stoic position would be to reject premise 2. Someone may be in conscious control of their actions as the determining factors of their decisions are "internal rather than external", even if their own internal range of choices is narrowly conceived. The causal physical/chemical workings of my brain and that relationship to the biological system that is my body may reasonably have established, predictable and even exploitable input-output relations, but they are nonetheless my behavioural functions, and their operation is my responsibility and my means of positive liberty.

The reason this is conceptually useful is in attributations of moral responsibility. We might reasonably say I was coerced, deprived of my free will to act, if someone else outside the boundaries of my person was holding me hostage, and this is the difference that makes a difference when it comes to understanding my freedom to choose whether to do something that would otherwise be deemed morally judgeworthy. However, that same justification would not be available to someone who, independently of outside coercion, simply did that morally wrong thing and held that physical causal determinism deprived them of moral agency.

  • Thanks. Am I right in saying that is a compatibilist perspective? I can certainly see the considerable practical appeal and utility of such an outlook. Most of us, I suspect - even hard determinists - resort to similar assumptions in day-to-day life. It's not a reality to which I intellectually subscribe, yet is simultaneously a 'reality' with which I typically operate. It's a weird feeling sometimes. Commented Aug 4, 2021 at 16:15
  • 1
    @Futilitarian, yes, I think it's fair to call this compatibilist. A Stoic would accept that reality at the atomic level might be properly deterministic (though not committed to it!); either way, this doesn't change the individuation of the personal agent as capable of making choices and responsible for understanding the choices they make and their consequences, and if needs' be refining their personal decision processes as they grow. Commented Aug 4, 2021 at 18:19

Here is a Libertarian answer, referencing empirical pragmatism vs analyticity.

This sort of logic argument has serious flaws when applied to claims about our world. For example "we can't have an infinite regress of causes" is shared between this argument and Munchausen's Trilemma. The resolution of this can be achieved the same way Munchausen's Trilemma can be resolved pragmatically -- by admitting that logical soundness per classical logic is not sufficient to establish reality in our universe, as the universe does not always follow classical logic.

With Munchausen, we DO have knowledge, so three sound arguments as to how we cannot are -- refuted in their applicability to this world by observation. For Free Will -- we experience, and observe, free will in operation every day. As with our acquisition of knowledge, this is primary data. The THEORY that classical logic somehow trumps primary data, is an effort to prioritize theory over empirical observation, and is readily rejected by any empiricist.

Note, the applicability of logics to this world is as much an empirical question as the applicability of any other science hypothesis, and the plethora of alternative logics means that none of them have any presumption of validity. https://www.cambridge.org/core/journals/think/article/guide-to-logical-pluralism-for-nonlogicians/EDFDFA1C9EB65DB71848DABD6B12D877

  • Interesting. Upon reading your answer however, my immediate response is that given logic seemingly provides the most reliable means by which we can arrive at truth, so I'm unsure as to why I would discard it in relation to something so important. I'm unfamiliar with Munchausen and will investigate him and your link. In chat @Conifold also directed me to this link re. problems with infinite regresses: [philosophy.stackexchange.com/questions/32189/…. I clearly don't understand them sufficiently yet. Commented Dec 8, 2021 at 5:11
  • 2
    @Futilitarian -- empiricists regularly have problem sets where none of their models are universally applicable. That classical logic is a highly useful model of how our world's event relations behaves is unquestionable. But in other circumstances when our models break down, a wise pragmatic empiricist does not stick with the absurdist predictions of a mere model. For a prior PSE discussion of Munchausen, see: philosophy.stackexchange.com/questions/64638/…
    – Dcleve
    Commented Dec 8, 2021 at 5:36
  • 1
    Re. wise pragmatic empiricists; sure, but the infinite regress I employ does not lead to an absurdist prediction. On the contrary, it describes the absurdity of a condition which seems necessary for free will to exist, and concludes that free will, if reliant upon that absurdity, cannot exist. Commented Dec 8, 2021 at 9:08
  • @Futilitarian -- My answer noted that there are infinite logics. That one of them provides a theoretic inference that there is no free will, how is that relevant at all? Munchausen's trilemma is far better known than Strawson's regress, and we pragmatists have not abandoned belief in knowledge because of it.
    – Dcleve
    Commented Dec 17, 2021 at 16:04
  • 1
    Yes, but whereas the infinite regress - in Achilles's race for example - is demonstrably flawed, the one employed by my argument makes perfect sense as a mechanism by which a decision becomes impossible. It is not articulating a paradox at all. On the contrary, the regress is plainly denied as a reasonable device upon which to rely. As for 'observation over theory'; this only makes sense until the theory questions the reliability of our observations, as my syllogism does. We shouldn't continue to rely so heavily upon that which we have very good reason to doubt. Commented Dec 18, 2021 at 23:59

I suggest that the proposal is unsound. While it may or may not be the case that one must decide to make a decision, the same applies to a non-free will decision and so the paradox in fact says nothing about free will. Given that we clearly can make decisions, whether or not by free will, it follows that the infinite recursion proposal must be incorrect.

  • How does the idea "the same applies to a non-free will decision" render the paradox irrelevant to free will? Your final sentence is also a non-sequitur. The fact that we make decisions is not up for debate. The question is whether or not those decisions are voluntary. The claim that the 'infinite recursion proposal is incorrect' does not detract from the argument. It supports it. Commented Jan 6, 2022 at 6:46
  • @Futilitarian which way are you arguing that ‘the fact that we make decisions is not up for debate’? On the one hand we appear to be making decisions and yet you’re proposing that Aristotle’s frog precludes us from doing so. Whether or not our decisions result from free will doesn’t appear to relate to this as either cause or effect. We may be at crossed purposes. But in any case I put it to you that a free will decision may result from a non free will decision. In a simple case I ask you for a number between 1 and 10 and you tell me one.
    – Frog
    Commented Jan 6, 2022 at 20:04
  • Yes. We're definitely at cross purposes, Frog. Respectfully, I suggest you read the argument again and tackle any specific attributes of its premises you find problematic. In regards to me being able to pick a number, my argument is that any number I decide upon (or any decision I make) is involuntary as, for it to be voluntary, I would be required to decide to decide upon it (or choose to choose it), which leads into the infinite regress. The argument is demonstrating that the experience of decision-making is illusory. Some call this attitude 'willusionism' : ) Commented Jan 6, 2022 at 22:11
  • @Futilitarian I find part 3 problematic. Deciding to decide requires infinite regression whether or not free will is involved. I propose, though, that a non-free will decision could be used to decide to make a decision with a free will-based outcome (i.e. you are compelled to pick a number but you pick the number). It might be more interesting though to consider that a decision is not made in a vacuum but your ongoing experience of free will receives the stimulus of me asking you to pick a number and your continuum of free will continues with your response…
    – Frog
    Commented Jan 7, 2022 at 1:24
  • The latter would require that your capacity for free will begins at some point, so perhaps your proposal does apply in regard to your first ever free will decision.
    – Frog
    Commented Jan 7, 2022 at 1:25

I'd already attack you're premises. Like what does that mean:

Note: here, an 'act' is defined as 'a thing done', as per Oxford Languages definition #2.

And henceforth:

A decision is an act.

Like what thing have you done with a decision? Is it really an act? Isn't that already a circular logic in that it implies that a decision changes an internal state of the body and is therefore deterministic? So isn't that hiding the conclusion as a premise? Do we know sufficiently how decisions work and that this is actually the case? Isn't it, at least as a hypothesis, also possible that a decision is not an act but something else that doesn't need a deterministic chain of effects that causes it?

Also isn't this a definition of "free will" that requires god like powers, that would invalidate free will but also be a fallacy of equivocation when it comes to what people mean by "free will"?

Like most of your decisions aren't perfectly free. Your actions are limited by the physical reality and also your thoughts and world view are limited by your direct and indirect experiences and on top of that most of the decisions that you make are forced upon you by having to deal with sensory inputs that you don't control.

That all being said, you still experience consciousness, you still experience the ability to choose between options or even if there is no choice and only one path to take; to value your action as good or bad. So you still experience an agency to make decisions, that is heavily restricted by outside forces but not ultimately determined. With all these limitations you still have a sense of agency. You can't control the world but you can control how you interact with it and how you feel about it, even to the point of deliberate physical and mental self-destruction.

  • Can you provide a definition of decision which somehow shows it not to be an act? That a decision is 'a thing done', I would have thought is relatively uncontroversial. It certainly isn't a thing 'not done'. Also uncontroversial, I would have thought, is the notion that truly free will be truly voluntary, not merely illusorily so. I'm unclear as to the equivocation. I acknowledge the 'experience' of free will exists but propose it is an illusion, which is the point of the argument. Commented Jul 20, 2022 at 10:22
  • Afaik it's unclear what exactly a decision is to begin with. I mean technically it has to be somewhere between the decision making process and the action caused by the decision. Because the valuation itself is important for a deliberate decision but it's not the decision itself, while the action is just the action. But a more obvious one would just be to assert that a decision does not require a decision and therefore is not an act, because those require a decision. Sure that just kicks the ball back into your field, but you're the one trying to prove that (no offense).
    – haxor789
    Commented Jul 20, 2022 at 10:43
  • Same for the illusion. A priori we do experience something that could be called free will, so that is already a fact. While the hypothesis of an illusion is just an assertion, so it would be up to you to prove that. And asserting things about the nature of decisions that lends itself to the conclusion would fail to do so or am I missing something?
    – haxor789
    Commented Jul 20, 2022 at 10:47
  • Thanks. No offence taken at all. It seems to me though that to claim "a decision does not require a decision and therefore is not an act, because those require a decision" is a far more onerous thing to demonstrate than that a decision is an act (if an act is defined as 'a thing done'). Again... if you a decision is not a 'thing done', what is it? I am not trying to define all the attributes of a decision, merely a single attribute required by the argument. Commented Jul 20, 2022 at 10:50
  • As for your later comment.. "so it would be up to you to prove that". Yes. Which is what the argument is attempting to do. If the assumption I've utilised, - namely that a decision is a thing done - is controversial, I would like to know why. If it is uncontroversial, or acceptable, then there is nothing unusual about a premise lending itself to a conclusion. That's how conclusions are reached. Via the premises. Commented Jul 20, 2022 at 10:54

The most glaring problem, here, in my eyes at least, is a "bad company" objection. Premise (3) is meant to be the kicker, in that (though this is not explicitly stated) an infinite regress of choices is supposedly impossible. However, if we accepted the idea, we might wonder:

  1. Suppose that every effect has a cause, and every cause is an effect of another cause, and so on. So, for there to be any effects, every effect must be preceded by infinitely many causes.
  2. In normal epistemic logic, for kS, k = "it is know that" and S = some sentence, it is also true that kS → kkS, and so on down the line.
  3. Not only the k-operator, but a proof operator, can face such a regress. Indeed, this is where rejections of foundationalism or advocacy of infinitism come into play. Another similar case would be for an understanding logic, such that uS → uuS.
  4. Modal logic can encode such a regress, though the prevailing assumption is that iterated modal operators are trivially reducible such that ◊◊S → ◊S, for instance.

So let's consider (4), here. Strong free will is the kind that depends on alternative possibilities, so it would have to connect up with modal logic. If singular possibility encodes for infinite possibility, and this unobjectionably enough, why would strong free will not work similarly? You are effectively claiming a decision operator such that dS → ddS, etc. I don't see a problem with a reversed collapse as such, from infinitely many decisions to one decision.

Another problem, though, is the idea that if ddS is involuntary, then so is dS. Inasmuch as we do not make choices in a vacuum, it is safe enough to say that the very fact that we face a decision problem can be caused by the historical environment (though per Kant (see below) this claim that transcendental freedom's "presence" is discernible in the physical world, might be rejected). A fact with a conditional structure, whose metaphysical antecedent is one subfact, but whose consequent is a disjunction of possible future subfacts, seems to be the kind of fact "inside of which" free will (or choice, rather) would occur. In other words, ddS (or dddS, etc., for that matter) might well be strictly determined by what came before, and yet dS itself would still be disjoint in effect, hence "free."

It is a mistake to think that belief in the possibility of free choices, or even the actuality of a faculty with such power, is necessarily a belief that every act that an agent with said power does, is a result of free will/choice. Kant thought that, per our phenomenal character, it would be possible, with an advanced enough science, to know all the empirically strict causes of our tangible behavior; but for all that, he located transcendental freedom and its causal spontaneity, then, in the noumenal realm. Sometimes people think this means that Kant is suggesting that our noumenal choice involves manifesting, or at least entering into, specific possible worlds, as though our souls could survey these worlds "from on high," and the implicit freedom of our empirical actions is the freedom to choose which set of empirical actions to incarnate in. That's fairly exotic a claim, and Kant might have looked for other abstract descriptions of how this procedure is actually supposed to work; but as far as his bold (and paradoxically, but epistemologically, modest) claims in his known work go, he openly admitted that he did not know, and could not know, much less more deeply understand, how transcendental freedom selects from the empirical possibilities, such as to allow us to be noumenally independent of physical causation and yet, in the phenomenal realm, predictably subject to the same.

More pointedly, though, right now, I would ask: why believe that either all our actions are determined or all are freely willed (in the strong way)? And mightn't some of our actions be random, no less, too?

So I don't know that your argument is unsound in the normal sense of that term, but it doesn't seem sound, either. Abstract, and obscure, philosophical premises are hard to judge on this kind of score.

  • 1
    Probably the best answer to date. I'm particularly interested in the limitations of infinite regresses to deny possibility and I understood some of what you wrote re. this. Do you think you could structure a similar argument, but for free will, utilising the 'reversed collapse' you describe? Or am I off track here? Commented Jul 21, 2022 at 8:28
  • 1
    I think you're on the right track in the important philosophical sense that we can ask such a question, the question is not meaningless, it might not lead to something so conclusive either (I'd be surprised if any philosophical reasoning of this kind was conclusive!), but it would help refine our sense of what is "at stake." For example, we might find that your argument is stronger or clearer, in the end, than a counterpoint argument for free will, at least in some context (e.g. in-world freedom vs. Kantian off-world freedom). Commented Jul 21, 2022 at 10:02
  1. Conscious control over decisions seems to be a strange way to put it. I would recommend either of the following: "conscious control over one's actions" or "the ability to make decisions".
  2. I don't exactly see what you mean by "able to decide what you decide" or "involuntary non-free decision". We only decide our voluntary actions and all decisions are free by definition. Voluntary actions are decided by the agent alone.
  3. Decisions are not past causal, there is no infinite regress. Decisions cannot be caused, decisions are the first causes of new causal chains of events. There is only forward causality.
  4. You may have disproven your own strawman version of free will, but you have not disproven our ability to make decisions.
  • 2
    Thank-you, Pertti. 1) Why is conscious control over decisions a strange way to put it? 2) If you do not decide what to decide, a decision is involuntary, in that it has occurred spontaneously. My assertion is that what we experience as a decision is in fact not a decision over which we have any but an illusory agency, for this very reason. 3) Please provide proof of your claim that decisions are new causal chains. You may be eligible for a Nobel if you are able to do this. 4) How have I strawmanned free will? Commented Aug 3, 2021 at 13:17
  • 1
    1) It is not clear what you mean. 2) Decisions do not occur, spontaneously or otherwise. Decisions are made for a reason. Whatever you do, you must decide what you do. If someone else decides, your action is involuntary. Agency cannot be illusory, someone must decide anyway. If you don't decide, someone else has to, otherwise your action is random and pointless. 3) There is no proof or a Nobel prize. Just think about it, it is plainly obvious: What causes your voluntary actions? You do. If they were caused by an external force, they would be involuntary. 4) You used an invalid definition. Commented Aug 3, 2021 at 13:35
  • 2
    Thanks again, Pertti. I recommend the online Stanford Encyclopedia of Philosophy as a good starting point for developing an understanding of these concepts, particularly plato.stanford.edu/entries/determinism-causal, and plato.stanford.edu/entries/freewill. They clearly explain how our experience of free will (including our decisions) may well be entirely illusory, and go a long way to explaining why any demonstration of self-causation would be a remarkable achievement. Commented Aug 3, 2021 at 13:59
  • I'm afraid, they do not explain what you think. You must have misinterpreted those articles in your own way. Commented Aug 3, 2021 at 14:05
  • 1
    Decisions cannot be caused so they're just random? They arise spontaneously and with no relationship to existing things or past events? My decisions don't feel random. I feel like they do have a causal relationship with the state of myself and my history.
    – TKoL
    Commented Aug 4, 2021 at 13:23

Phyics, which is in an extremely refined state today, has never found any phenomenon in the universe that is not ultimately the interaction of aggomerations of elementary particles strictly following the laws of physics. And people do not have access to the chemical and electrical phenomena that take place in their own minds.

Therefore it is not the least bit surprising that although we generally have the impression of being able to make our own decisions independent of everything else, in fact our decisions are as determined as would be a computer making a decision based on a complicated algorithm after some amount of data has been input. The computer likewise doesn't "know" what its decision will be until it has been made, but it was nevertheless determined.

  • Free will has nothing to do with physics. Decisions cannot be determined by any twist of logic. Decisions are not physical events. Computers don't make any decisions. People, the programmers and the users make all the decisions concerning the computer's operation. Commented Sep 3, 2021 at 17:22
  • Everything that occurs in the universe is governed by the laws of physics, as far as we know: Nothing has ever been shown to be otherwise. That includes the operation of the human brain. Commented Sep 3, 2021 at 17:35
  • @DanielAsimov. So Daniel, what's your answer to the question? Is the argument sound as presented? Commented Sep 4, 2021 at 3:38
  • @DanielAsimov Human brain follows the laws of physics, as it is a physical object. The laws of physics don't apply to the human mind, which is the brain's capacity to process immaterial information. Brain physiology and psychology are different branches of science. Commented Sep 4, 2021 at 7:00
  • 1
    haxor789: I said nothing about "what physics argues". What I said is that physics has not found any phenomenon in the universe that fails to follow its laws. Commented Jul 21, 2022 at 15:51

Suppose I make a cup of tea, and the question is whether I did so voluntarily. Before I made a cup of tea (act 2), I decided to make a cup of tea (act 1).

What is wrong with saying that act 1 was involuntary (I have no choice about making a decision - either I make a cup of tea or I don't. There is no third option.) but act 2 was voluntary?

You appear to be making an unstated assumption that being forced to make a choice is the same thing as the choice being forced. That a voluntary decision can only arise from voluntary precursors.

This sort of makes sense for 'first cause' arguments, which are clearly related to their precursors, but I don't think the same can be said for the 'voluntary' property. If a voluntary decision can only arise from voluntary precursors, you need to say why.

Perhaps the problem is that the phrase 'voluntary decision' is ambiguous as to its referent - is 'decision' referring to (act 1) or (act 2)?

  • Yes, 'voluntary decision' refers to act 1, the decision. I am interested in your point re. the assumption I am making that a 'voluntary decision can only arise from voluntary precursors'. Do you believe therefore, that there is a flaw in premise 1: 'In order for an act to be voluntary, a person must decide to perform it'? Or have I misunderstood you? Commented Jul 21, 2022 at 10:21
  • Re. your 2nd para: Let's say an involuntary decision to drink (act 1) is followed by a voluntary action to drink (act 2). This would imply that volition can emerge independently of a decision. But this article ties volition to decision. "Volition is the capacity of humans...to initiate actions based on internal decision". In other words, voluntary drinking (act 2) would require voluntary decision. Commented Jul 21, 2022 at 11:25
  • I think that the "act" in premise 1 is referring to act 2. Premise 2 then applies the same principle to act 1, but this is a distinct application. Act 1 is involuntary, so requires no prior decision. Act 2 (the particular choice made) is voluntary, the outcome of act 1. You haven't done anything to prevent act 2 being voluntary. All you've done is require each voluntary act to be preceded by an involuntary one. In your point about volition, is the required 'voluntary decision' this is referring to act 1 or act 2? Commented Jul 21, 2022 at 11:27
  • The "act" in Premise 1 is any act, including decisions (for they are 'a thing done'). Your definition of Act 2 as "the particular choice made" explicitly states - via the word 'choice' - that it too is a decision, so it can't escape the quality of being a 'thing done'. And, as the argument shows, if any 'thing done' is to be voluntary, an infinite regress occurs. The only way to protect act two's volition would be by demonstrating that volition can occur without decision. Commented Jul 21, 2022 at 12:02
  • Let’s take your view of Act 1 being involuntary, and Act 2 being voluntary. How does Act 2 escape P1, “In order for an act to be voluntary a person must decide to perform it?”. How does drinking retain volition if the decision required for drinking was involuntary? Commented Jul 21, 2022 at 12:02

What I follows is not meant to be a definitive answer, but to offer - after consideration of the answers and comments both here and to linked questions I have asked on this stack - an evolution of the proposed argument which addresses what I believe to be the most important challenge to date; namely that the claim that a 'decision is an act' is philosophically controversial.

As Strawson (2003, p. 244) points out in his response to an imagined question:

"What about choices and decisions? These are clearly mental actions?"

Some are, but the case is far from clear... Very often there is no action at all: none of the activation of relevant considerations...It simply happens, driven by the need to make a decision...There is no direct action in the actual issuing of new content, any more than there is in the growth of trees one has planted.

With this in mind, I have added three premises which attempt to accomodate this challenge, although these additions do contain another assumption which be addressed below the argument.

As previously, an 'act' here is defined as 'a thing done', as per Oxford Languages definition #2.

1. Decisions may be either voluntary or involuntary.

2. Insofar as free will requires the ability to make voluntary decisions, those decisions which are involuntary cannot contribute to free will.

3. Any voluntary decision would constitute an 'act'.

4. However, in order for an act to be voluntary, a person must decide to perform it.

5. Therefore, in order for a decision to be voluntary, a person must decide to decide it (perform it).

6. This leads to an infinite regress of prior decisions, in which any voluntary decision requires an infinite chain of prior decisions.

7. Insofar as free will requires the ability to make voluntary decisions, free will is impossible.

Clearly, this argument depends on an assumption that a voluntary decision is always an act; that a dichotomy exists between voluntary decisions-as-actions and involuntary decisions. Or can a decision be a non-act and retain volition?

I would like to post this reformulation as a new post. Feel free to comment as to whether or not this would be appropriate. In the absence of any reasonable objections, I will do so.

I have 'accepted' the answer provisionally, but I will un-accept it in the face of any critique which establishes that it constitutes no improvement upon the original.


Strawson, G. (2003). Mental Ballistics: the Involuntariness of Spontaneity. Proceedings of the Aristotelian Society, pp. 227-256. DOI: 10.1111/j.0066-7372.2003.00071.x


Your argument fails. The mistake in your logic is that you assume for there to be free will, every free decision x must be preceded by another free decision x-1 to take free decision x, leading to an infinite regression. Instead it is entirely possible that the triggers for making decisions are involuntary but the decisions themselves are free.

  • There is a revised version of this argument here. But regardless, can you illustrate with an example? I have trouble imagining how that might work. Cheers Commented Feb 16, 2023 at 2:20
  • @Futilitarian I had answered that version too, some days ago, forgetting in the meantime that I had done so. What are you having trouble imagining? Commented Feb 16, 2023 at 6:45
  • Let's say I am involuntarily triggered to make a decision to eat an apple. How can the decision to eat the apple then be free? Commented Feb 16, 2023 at 6:55
  • No, you are involuntarily triggered to make a decision about whether to eat the apple. Commented Feb 16, 2023 at 6:56
  • Okay. So from where do you imagine volition arises? What is the mechanism? Why is it freed somehow from prior circumstance? Commented Feb 16, 2023 at 7:10

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .