A very short story on (the absence of) free will

April 26, 2020 • 9:45 am

This story was published in Nature some years ago and called to my attention by reader David. The author is the well known science fiction writer Ted Chiang, and though the story is 15 years old, it’s not only still relevant but, like the best sci-fi, makes you think hard. The story was published two years after Libet’s experiments showing that one could, using brain scans, predict with some accuracy when a subject was going to press a button before the subject became conscious of having made that decision. Since then, a host of other studies have shown that you can predict things that don’t involve physical actions, like deciding whether to add or subtract two numbers, up to ten seconds before the subject records having made a decision.

All the experiments show that brain scans can predict—not perfectly, mind you—what a subject will do before she has become conscious of that decision. And they cast severe doubt on the notion of libertarian free will: that at a given moment we could have decided to take any number of alternative decisions. People hate these results, and try to impugn them, for they don’t like the idea that decisions are determined by the brain before we think we’ve made them. People like Dan Dennett argue that if we fully grasp determinism, society will fall apart because we’ll become apathetic and refuse to get out of bed. This, of course, is not true. I got out of bed this morning and am busy writing this.

But to someone who’s science minded, determinism is the only game in town. Setting aside pure indeterminism—which would obtain if quantum processes affected our decisions—our choices and behaviors are the results of the laws of physics, and at any one time (leaving aside quantum factors) we could have made only one decision. If you doubt that, then you’ve bought into either the numinous, the supernatural, and mind/body dualism. Religious people, of course, are the most ardent believers in libertarian free will, because, at least in the Abrahamic faiths, you have a free choice about embracing God, Jesus, or Allah, and if you don’t you’re doomed.

I’ve harped about the hegemony of naturalism and determinism before, and emphasized its importance in structuring society and the judicial system. I won’t do it again; you can read what I’ve written by entering “free will” in the site’s search box.

I’ve also dismissed compatibilism as a semantic game constructed to ensure that people don’t become nihilisitic or depressed if they realize that they don’t have free will in the classical you-could-have-done-otherwise sense. The story below incorporates both determinism, a device based on the Libet experiment, and the supposed consequences of realizing that your behaviors are in fact determined. It’s a really nice and very short (1.5 page) science “fiction” story. Click on the screenshot.

The premise: people are given a Libet-ian device that predicts with a red light when they’re going to press a button on the device. They can’t outwit it because it’s apparently been programmed with every bit of information in the Universe. The consequences: many people go nuts.

Some excerpts (but just read the story):

By now you’ve probably seen a Predictor; millions of them have been sold by the time you’re reading this. For those who haven’t seen one, it’s a small device, like a remote for opening your car door. Its only features are a button and a big green LED. The light flashes if you press the button. Specifically, the light flashes one second before you press the button.

Most people say that when they first try it, it feels like they’re playing a strange game, one where the goal is to press the button after seeing the flash, and it’s easy to play. But when you try to break the rules, you find that you can’t. If you try to press the button without having seen a flash, the flash immediately appears, and no matter how fast you move, you never push the button until a second has elapsed. If you wait for the flash, intending to keep from pressing the button afterwards, the flash never appears. No matter what you do, the light always precedes the button press. There’s no way to fool a Predictor.

The heart of each Predictor is a circuit with a negative time delay — it sends a signal back in time. The full implications of the technology will become apparent later, when negative delays of greater than a second are achieved, but that’s not what this warning is about. The immediate problem is that Predictors demonstrate that there’s no such thing as free will.

There have always been arguments showing that free will is an illusion, some based on hard physics, others based on pure logic. Most people agree these arguments are irrefutable, but no one ever really accepts the conclusion. The experience of having free will is too powerful for an argument to overrule. What it takes is a demonstration, and that’s what a Predictor provides.

Typically, a person plays with a Predictor compulsively for several days, showing it to friends, trying various schemes to outwit the device. The person may appear to lose interest in it, but no one can forget what it means — over the following weeks, the implications of an immutable future sink in. Some people, realizing that their choices don’t matter, refuse to make any choices at all. Like a legion of Bartleby the Scriveners, they no longer engage in spontaneous action. Eventually, a third of those who play with a Predictor must be hospitalized because they won’t feed themselves. The end state is akinetic mutism, a kind of waking coma. They’ll track motion with their eyes, and change position occasionally, but nothing more. The ability to move remains, but the motivation is gone.

The reason that people go nuts in this scenario, which doesn’t occur with determinists like me, is apparently that they get a repeated, irrefutable, and tangible demonsration that they can’t choose otherwise. I guess this demonstration hits closer to home than simply the mental realization that our choices are determined before we’re conscious of having made them. But why would people buy such a device if the chances are high it would drive them nuts? (They have no choice, of course!)

Setting aside the fact that it would be impossible to build a Predictor, for it has to be in possession of all the information in the Universe, including things that happened up to the last second, it would be a useful way of demonstrating determinism. There are other problems with the description as well, but I’ll let you suss them out.

The end of the story? The narrator sends a warning back to the past about the dangers of the device:

And yet I know that, because free will is an illusion, it’s all predetermined who will descend into akinetic mutism and who won’t. There’s nothing anyone can do about it — you can’t choose the effect the Predictor has on you. Some of you will succumb and some of you won’t, and my sending this warning won’t alter those proportions. So why did I do it?

Because I had no choice.

A nice story!


73 thoughts on “A very short story on (the absence of) free will

  1. One of the best short science fiction stories I have ever read. As I read, another part of my mind had doubts and fears. It was wonderful.

  2. Since you are stuck at home, download and watch (or stream)

    The show, which recently came out, is an exploration of exactly the posted issues.

  3. I love Ted Chiang, and would give a special recommendation to “The Merchant and the Alchemist’s Gate.” Related themes, but quite different.

    The prediction device is physically impossible. Short explanation: In order to gather all the physically relevant information one second early, the device needs to access the space out to one light-second (300,000 km) from your action. But to collect that information into the device takes another second. So now it needs information up to two light-seconds away… The device can never catch up.

    Philosopher Jenann Ismael gives a fuller explanation here.

    As some of us compatibilists keep pointing out, to evaluate the implications of scientific determinism, you need to be careful about what physics actually says.

    1. Might it not be that complete information about the person’s brain is sufficient to predict what they will do in the next second? And not down to the elementary particle level either – a description of the gross properties of each neuron and its connections, and its present state, might be enough.

      1. I was thinking that since a main part of the discovery of our lack of free will were brain scans that predicted what a subject would do before they were able to be conscious of it. So I don’t see why the device needs all the info from the universe.

        1. Right, Mark. No need to wait for time travel to send a signal back in time. There are already experiments where the subject’s choice is predicted several seconds in advance of conscious awareness. But this does not affect free will. After all, it is still the person’s own brain that is making the decision (and then making him aware of the decision a few seconds later).

          Free will is when a person decides for themselves what they will do, free of coercion and other forms of undue influence.

      2. Sure, usually that would be true. However – and Ismael talks about this a lot, and I should have mentioned it – a flashing light on a device would also influence many people. It would influence them to do the opposite. So the device would simply defeat its own predictions, given physics plus human nature as we know it.

        Not me though. I’d use the device to make a killing on Wall Street. Go to the big boys and tell them I can buy any single stock 0.9 seconds before it makes any steep rise (of a given size). I think you can figure out the rest…

        1. That’s pretty much the scam high-frequency traders pulled for years by using high-tech equipment to front-run the stock and commodities markets by mere milliseconds. Michael Lewis wrote a book about it, Flash Boys.

    2. I think it’s probably untrue that such a device needs to access all data within a few light-seconds to be accurate. This a very simple yes/no decision (push the button), and AIUI psychologists and physiologists can already predict such decisions a few seconds in advance just by getting the appropriate real-time scan of the brain.

      What *I* would say about the difficulty of such a device is that even with a deterministic organism, the red light is input into the system which will then impact the organism’s decision. So you might get all sorts of recursive loops and such going that makes it impossible to analytically solve for the outcome, the same way it is analytically impossible to solve a three-body problem even if we all agree it’s a deterministic system.

      1. Your point was what I was trying to get at when I said in another comment that the device would defeat its own predictions by signaling the human being. But your way of putting it is more elegant. Beyond the three-body problem, an even closer comparison would be a computer solving its own halting problem.

        Anyway, your point gets to the heart of why this Laplacean Demon device is *not* a problem for free will. Real human beings always get the last word when someone merely predicts (without brute force to coerce the person) what they will do.

      2. I think you can tie in an awful lot of external things. What if I say I will push the button when my cell phone rings? I can uphold my end of that bargain pretty easily, if I have the two devices at hand. Can the Predictor predict when my phone will ring? And if it will be my sister calling? Can it predict when I will next see a lightening flash? These are all conditions I can choose for pushing the button, and they would be difficult to monitor.

    3. Why not go with a Feynman diagram for how it works? Particles can go backward in time: an electron moving backwards in time is just a positron, etc.
      So in principle the light can actually be triggered by the button push, with the signal going back in time to turn the light on first. No information gathering required.

      1. Yes, in principle. You would need to get extraordinarily lucky with respect to entropy though, in order to make one macroscopic event depend on an earlier macroscopic event. See Sean Carroll’s discussion of probability and entropy.

        If the “prediction” was *consistently* correlated with the “later” button press, it would mean that entropy was lower “later”, But then the arrow of time goes the opposite way from what we were supposing. People in that universe would experience pushing a button, then seeing a light go on one second later.

    4. I agree 100% with your assessment of “The Merchant and The Alchemist’s Gate.” It has the feel of a classic tale.

  4. Some people, realizing that their choices don’t matter, refuse to make any choices at all.

    But that is still choosing how to behave, so is not “refusing to make any choices”.

    My prediction: people would quickly get bored playing with the Predictor, and it would make no difference to what they do after that.

    That’s because people don’t actually need the concept of libertarian, contra-causal free will in order to just get on with life as human beings.

    1. “But that is still choosing how to behave, so is not “refusing to make any choices”.”

      No, those are new circumstances so the reactions are different but there still is no choice.

      1. “Choice” does not mean “dualistic, contra-causal choose”, it means “choose” in the same way that a chess-playing computer chooses a move. So yes, there are choices.

      1. Trump voters determined that their own votes were for Trump, obviously. But do you claim that their votes were predetermined since the Big Bang?

    1. Presumably not, or PCC(E)’s $100 bet would be a foregone conclusion and he wouldn’t have made it. Assuming he had a choice… I freely admit I don’t understand this issue.

  5. “… because free will is an illusion …”, yes, and yet still real.
    I’m half way into Harari’s book ‘Sapiens’ (recommended, although not sure how strongly) and one of his main points is that our species is characterized by the formation and acceptance of common illusions, which allows the conformation of large alliances among people not based on genes. He makes the points that we believe in plenty of non existent entities: corporations, money, countries, gods, … Free-will may very well be included in this list I’d argue.
    Couple more things that need to be repeated:
    i) Determinism =/= {NOT free will}.
    ii) That many people has the concept wrong is not an argument to throw it out, otherwise we would have to dispose of concepts like atoms, species, evolution, etc.

    1. Harari is wrong to believe that all inventions are illusions. “Corporations”, “money” and “countries” are not illusions. Instead, the gods are fallacious inventions, and consequently they are illusions.

  6. There are other problems with the description as well, but I’ll let you suss them out.

    One might be that (pace the first sentence of the story’s penultimate paragraph) determinism does not necessarily entail predetermination. That is the type of determinism presupposed by LaPlace’s demon. Several arguments against LaPlace’s demon have been posited.

    The story’s last line has a nice O. Henry-style turn to it.

    1. The difference between determinism and predeterminism is subtle, or perhaps non-existent. To me, predeterminism has a whiff of theological determinism to it, although some argue determinism and predeterminism are simply synonyms, since determinism argues all events are determined by antecedent (pre-) events. LaPlace’s demon, on the other hand, is about whether there are limits to knowledge an entity can have about the universe. On that, Wolpert’s theorem—the last argument against LaPlace’s demon in the Wikipedia article you reference—basically rules such an entity including the Predictor if such an entity is part of the physical universe. Too bad. A working Predictor is all it would take for me to finally take a side in the free will debate.

  7. I’ve never understood why these experiments are supposed to have any relevance to the notion of free will. All they show is that sometimes we may make choices unconsciously. So what? Making them consciously doesn’t change the fact that everything is predetermined. It seems evident to me that the notion of free will can totally be discarded by thinking alone—no need at all for any experimental corroboration. It’s just intrinsically incoherent.

    “People like Dan Dennett argue that if we fully grasp determinism, society will fall apart because we’ll become apathetic and refuse to get out of bed. This, of course, is not true. I got out of bed this morning and am busy writing this.” I totally agree. Even if you reject the idea of free will, the minute you’re hungry (for example), you’ll rise from your philosophy-induced slumber (assuming such a slumber would take place—exceedingly unlikely!) and go look for something to eat. Not much, if anything, would change from our day-to-day life from a denial of free will. Our moral philosophy may well change, but most probably only for the good: we may still need to lock up criminals for safety and deterrence, buy hopefully any inclination towards cruelty, and any notions of revenge or that making someone suffer is their just deserts, would tend to diminish, since we’d recognize that they are illogical, and in fact immoral.

    And to adduce that quantum-mechanics randomness may open a window for free will is nonsense, as then our will would just act randomly—it would still not be “free” (https://medium.com/@cdelosada/free-will-af8e582688b8?source=friends_link&sk=8f97e41da38664188f1a2549e5e271eb).

    1. I think what the Libet experiments show is even more mundane: thinking is a computational process that takes time to occur. Even though we experience decision making as an instantaneous conscious event, that’s only an illusion. That we have such an illusion is also not surprising. Much of our body’s processing is not accessible to our conscious mind.

      1. Paul,

        Exactly. I read the story and it doesn’t seem to offer anything beyond a Libet-experiment type challenge. It’s full of question-begging assumptions of the type that have been hashed out before.

        1. I didn’t read the story but, as I said, I find the Libet experiments interesting but nothing about them really adds much to our understanding of the brain. They merely show experimentally what most assumed was the case.

  8. A nice story. But I don’t think that people would become apathetic if they fully come to grasp our lack of free will. Its as if we discover we are really in a simulation, controlled by super-intelligent beings. Or if we discover that one day, no matter what you do, you will die and that will be that as there is no afterlife of any kind.
    In all cases we are programmed to carry on our lives anyway. Soon enough we would check whats’ on PBS or NPR or open that book at its bookmark and have a good read.
    We have no choice b/c our actions and feelings come from our brains and our brains are programmed to think and do things as if we had free will, or were not part of a computer simulation, or was immortal.

    1. Yes. Most people wouldn’t even draw the conclusion that the predictor’s infallibility means they have no free will. Millions of US citizens are going to vote for Trump later this year, even though the conclusion is obvious from his actions that he’s totally unfit to continue to be president (not that he ever was in the first place). The same people would say: “Huh! Just can’t beat that predictor gizmo. I wonder if there’s any beer left in the fridge.”

  9. Without trying to argue for or against determinism, I think it is quite tautological that the observation that a decision was made comes after the decision.. That does not argue against free will.. The conscious is what the brain observes.. the brain observes itself but can only see so much.. what it doesn’t see is the subconscious.. So the brain at some point makes a decision, whether freely or determined, and afterwards observes that it made a decision.. the observation is the conscious part, but that does not make the subconscious decision unfree..

    1. To add to my comment, the brain makes a little shortcut in that it doesn’t say ‘oh I observe that I made a decision’.. no, it says ‘I decide such and so’ because that’s what it did after all.. it’s just that you only become conscious of your decision at the moment of observation, which obviously is after the decision was actually made..

      1. I mostly agree but there are occasions when one makes a decision and, at the same time, has some inkling as to what influences caused it. Of course, these are only observations made after the fact and are subject to illusions as well. They are unlikely to give us a complete or accurate picture of how we arrived at a decision. Similarly, we can’t completely examine all our brain and muscle activity that went into making a basketball shot.

    2. Yup!

      IF you fully accept the soundness of the experiments, the most you get is that we would have learned something more about the relationship of the conscious to the unconscious. It could dispel the idea that the conscious thoughts “come first” as it feels. But it’s easy to become sloppy in making inferences. So if someone infers that, because our decisions happen in a pre-conscious manner and then enter consciousness, that “therefor I am not REALLY in control of my decisions” then that would only make sense if you reject your brain’s reasoning as “being me.” A dualistic idea where your consciousness is “you” but the activities by which your brain actually reasons is “not you.” And that’s a very weird leap to make.

      As Dennett always says of this strange reductionist tendency, by making yourself smaller and smaller e.g. where you become only “what is conscious at any exact moment” you can externalize almost everything – including yourself from 3 seconds ago!.

      But…why? It doesn’t solve any problems or do much work.

      It makes more sense to look at “me” being a system of conscious and unconcious, and we can fruitfully still understand “the actions I can take, the things I can and can’t control, the options I have” etc. Just as we do everyday.

      1. That’s ridiculous. If “will” ad agency mean anything, it means that you make a decision CONSCIOUSLY.

        In fact, compatibilist free will does no work. And all this is just dancing around the issue of whether determinism is true and applies to our decisions. This claiming that you can call your “free will” the same thing as your personally unobserved firing of neurons and all the antecedents to them is just philosophical wheel-spinning. Frankly, if you want to call that “you”, that’s fine, but everybody in the whole world save compatibilists think that a decision is made consciously.

        As I said, the important question is whether your decisions are results of the laws of physics and whether you could have decided otherwise. The answer is “yes” and “no” respectively.

        Your own view does no work, not a bit, and has no social implications. There is, apparently, nothing that you cannot subsume under the idea of “free will”, even if your brain conditions a “decision” three day in advance. “I decided that three days ago without realizing it”? No problem!

        1. … but everybody in the whole world save compatibilists think that a decision is made consciously.

          Is that really true about incompatibilists? I’d have said that the consciousness is just the tip of the iceberg of our decision making, most of which is unconscious. (Do people generally think that they are consciously aware of all the machinery that goes into their decisions?) But then I plead guilty to being a compatibilist.

        2. I think by raising consciousness on this pedestal is that you complicate issues that are in se very simpel..

          The brain is a neural network that has been trained with inputs and rewards or costs on outcome behaviour. Nothing in this resulting ‘intelligence’ requires there to be consciousness. You can drive home, lost in though, and arrive without having made a single conscious decision on which way to turn or which buttons or handles to operate. Yet decisions you made.

          Consciousness is just the icing on the cake that gives us insight in our inner processes, but even so we can only consciously *observe* what is happening inside, but neither are these observations necessarily complete nor are they necessarily objectively true.. They are a subjective interpretation of incomplete observations.. We don’t have to means to get to objective introspective truths as scientifically one would do with double-blind measuring and peer review, since we are the our own sole tool-less observers.. All of that to say that we have no means to be even sure that our introspection shows us the objective truth. Consciousness is flawed and should not be out on a pedestal. It is a mere after-the fact subjective interpretation of what happened in our neural network that made the decision.

          I am not arguing that we have free will (even though I think so), only that after-the-fact subjective observation of our decision does not exclude free will.

          1. Before Libet and the other experiments, people would have thought that what you do is the result of a decision, and cannot be predicted with any accuracy until the moment you feel that you’ve made a CONSCIOUS decision. That is, you couldn’t predict something before a conscious “decision.” Now science shows that that is wrong, and now you say, “Well, we expected that result.” Sorry, but we didn’t, which is why the Libet experiments were so controversial. Now they’ve been buttressed again and again. The upshot: our brain is running programs and consciousness of making a “decision” is probably a confabulation of those evolved and environmentally-induced programs.

            You are making a virtue of necessity, saying that OF COURSE we knew that when, in fact, we didn’t.

          2. The Libet experiments were only controversial, if that’s the right word, to those that believed the illusion of their brains making a decision at the instant one is conscious of having made it. Libet showed that this belief is based on an illusion. Those of us that never believed such a thing could be possible in the first place, were not surprised at Libet’s results. Making a decision is not magic. It’s a computation of some complexity and, therefore, it takes time to process. We may experience it as instantaneous but it can’t possibly be.

          3. “… our brain is running programs and consciousness of making a “decision” is probably a confabulation of those evolved and environmentally-induced programs.”

            I would agree with that (and I thought pretty much that prior to Libet’s results, which didn’t surprise me).

        3. “That’s ridiculous. If “will” ad agency mean anything, it means that you make a decision CONSCIOUSLY.”

          Only if you beg the question against a contrary interpretation. About the Will, Wikipedia aptly describes it:

          “Will, generally, is the faculty of the mind that selects, at the moment of decision, a desire among the various desires present; it itself does not refer to any particular desire, but rather to the mechanism responsible for choosing from among one’s desires.”

          This concept does not decide the issue in favor of your claim; it’s all compatible with understanding our desires and reasoning emerge from unconscious processing in to consciousness. Whether our conscious self makes the choices or whether it is how we become conscious of our choices (or some mix): We are still reasoning agents, having desires, and reasoning about which actions will most probably fulfill those desires. We use this understanding because it is fruitful in both making sense of our and other people’s actions, predicting them, assessing probabilities, just making sense.

          So how does your view about the “will” and consciousness change this? I don’t see how. Will you reason any differently, as if you yourself are not a rational being interacting with other rational beings (which would make your replies incoherent).

          ” Frankly, if you want to call that “you”, that’s fine, but everybody in the whole world save compatibilists think that a decision is made consciously. “

          That doesn’t actually seem to be true.
          If it were, there wouldn’t be a notion of the unconscious causation in our thinking, but there clearly is. Tons of articles on this that people don’t seem to just reject.
          And as this article says:


          “Indeed, designers and advertisers have known how to control our non-conscious decisions for centuries. By using subtle cues designed to bypass conscious awareness, they can “trick” us so that we drive more safely, navigate cities in ways we do not realise and even drink more alcohol at the bar.”

          Further, scientists acknowledge the issue of bias as you know, big list here:


          The main issue with bias being that people can be wrong in their conscious account of how they came to a decision, unaware of hidden influences in how their mind works.

          So it seems clear to me that not only do scientists don’t simply assume all decisions are made fully “consciously,” it seems to me most people are actually pretty well-acquainted with the idea that they can’t always account for or control their decisions, and that hidden-in-the-mind/brain factors influence decisions. (Tons of social science and self-help-type literature turn on this!).

          But even IF everyone really eschewed any belief in the influence of the unconscious processes, and believed all decisions were made at the very moment of conscious awareness, that doesn’t decide the matter.
          It would say people are under *some level* of misunderstanding about how our decision process operates, but it doesn’t immediately council us to throw out the idea of agency, will or rationality.

          No more than discovering that our sight does not work as we intuit it to work – e.g. we don’t intuit that our eyes see the world upside down, we don’t intuit our blind spot, we don’t intuit the types of subconscious, heuristic assumptions our brains make about expectations, shadowing etc that can lead to illusions, or that our field of view is in fact not sharp through the whole thing, it’s an illusion brought on by unconscious processes of scanning a tiny point in focus).

          But would the fact most people have some naive understanding, some misapprehension about how vision works entail throwing away the concept we have “vision?”
          No, we just gain a better understanding of how the process works.

          So I do not see that it follows directly from anything you are saying that having a fuller understanding of how conscious and unconscious interact entails we ditch the concept of agency, of a “will,” of a “you” doing deliberating.

          “Your own view does no work, not a bit, and has no social implications.”

          I suggest compatibilism allows for a better explanation for human behavior (particularly choice-making) than what I’ve seen from hard incompatibilists. It can provide a sounder basis for the “social implications” of choice-making that we’d agree on, while identifying incorrect claims of “social implications” in a hard incompatibilist view.

          Just as it can be salutary for people to discover morality doesn’t depend on a Deity, so it can be for discovering free will never depended on God, dualism or a soul. Better explanations are valuable.

          1. This is all a bunch of well-put obscurantism, but obscurantism still. I just wrote a long response but realized that, like you, I was going to exceed the length limit of comments (please don’t write such long ones).

            In the end I’ll just say that you’re relabeling a term that nearly everyone (you mention scientists, but what about the average person?) conceives of as libertarian free will. And I’m intrigued why you and other compatibilists are so bloody eager to hold onto a concept that, in its original form, science has rejected. You can relabel “god” as “the universe” when science disproves the supernatural, but that doesn’t help matters. Telling people, “yes, you havee free will” without emphasizing that their behavior is physically determined is, in my view, intellectually dishonest. There is no “free will” floating out there that we’ve suddenly discovered means something different from what we originally thought.

            I also note the reluctance that you and other compatibilists have in telling people that their behaviors are purely the result of physical processes, and they could not have done otherwise. Why not say, “your brain compelled this behavior according to the laws of physics, and your view that you made a ‘conscious decision’ is post facto confabulation. Your consciousness of having “reasoned” is likely an illusion.” And if that’s the case, which science is telling us, then that’s progress. Your view of free will seems remarkably resistant to any scientific advances in neuroscience.

            A lawyer recently told me that he was at a meeting of judges, and nearly all of them agreed that the American justice system rests on a libertarian view of free will. If people call for execution or retributive punishment because “someone made the wrong choice”, compatibilism has no response. Determinism has the correct response, “They could not have chosen otherwise.” Compatibilism has a lot of mushy verbosity.

            I think we’ll have to agree to disagree. Finito.

          2. “Why not say, “your brain compelled this behavior according to the laws of physics, and your view that you made a ‘conscious decision’ is post facto confabulation. Your consciousness of having “reasoned” is likely an illusion.” ”

            As a compatibilist, I would readily say that and agree with that.

            (I would be less likely to say it *here* because there are no dualists here; everyone is a determinist, either a compatibilist or an incompatibilist, and are thus agreed on this.)

          3. “what about the average person?) conceives of as libertarian free will.”

            The operational definition of free will is a choice we make that is free of coercion and undue influence. That’s the definition that everyone understands and correctly applies when assessing moral and legal responsibility.

            The paradoxical definition of free will is a choice that is free of causal necessity. Since causal necessity is logically implied by reliable cause and effect, and since every freedom we have to do anything at all subsumes a world of reliable causation, the so-called “philosophical” definition is a self-contradicting oxymoron.

            Nearly everyone would agree that we live in a world of reliable cause and effect (deterministic) and that we make most of our decisions free of coercion and undue influence (a freely chosen “I will”).

            I don’t see how any other position could possibly be defended.

          4. If any action I take is causally necessary, and inevitable from any prior point in eternity, as you say, then I could not have done otherwise. Where is my free will if this is true?

          5. What you “can” do and what you “will” do are not the same thing.

            When you face a problem or issue that requires you to make a choice, two things are always true by logical necessity:(1) one must have at least two real possibilities to choose from, and (2) one must be able to choose either one. They are logically necessary because if either of them is false, then choosing cannot proceed. But it does.

            And this is why it is normal to hear someone say, “I chose A, but I could have chosen B instead”. It was inevitable from any prior point that the person “would” choose A. But it is also true that at the beginning they “could” have chosen A and they “could” have chosen B.

            Why? Because if we roll back time to the beginning of the choosing operation, it will once again be true that “I can choose A” and “I can choose B”. And anything that I “can” do today (present tense), then tomorrow I “could” have done yesterday (past tense).

            Choosing is an essential operation that evolved in intelligent species. It gives them an adaptability which aids their survival.

            Confusing what we “will” do with what we “can” do breaks the operation. So, we should avoid doing that.

            Choosing is how we deal with an inevitable future which is as yet unknown to us, because it has yet to be decided by us.

        4. ‘…the important question is whether your decisions are results of the laws of physics and whether you could have decided otherwise. The answer is ‘yes’ and ‘no’ respectively.”

          Compatibilists agree about determinism and that therefore (any indeterminism aside) one couldn’t have done otherwise in an actual situation as it played out, including conscious decision-making. What compatibilists call free will obtains basically in virtue of behaving (deliberating, deciding, acting) without coercion and in one’s right mind. This is compatible with determinism, and constitutes the exercise of real agency, but of course to call it free will is contentious given the libertarian definition, which says that free will only obtains if one could have done otherwise in actual situations.

          Naturalists in the debate about what we should call free will might disagree about the causal role of consciousness, but will likely agree that the brain, suitably ensconced in a body, does the decision-making. For those who agree that determinism is more or less the case, real agency must therefore derive from the properly functioning brain and body, not from anything contra-causal, whatever the contribution of consciousness. So I think it’s fair for naturalist-determinists to say that as agents we make real, albeit fully caused, choices, even if consciousness doesn’t add to what the brain is deterministically doing. To say that we don’t really make choices if determinism is true would be to adopt a libertarian conception of agency.

          1. I don’t agree. Your opinion is our own, not that of “naturalist-determinist” in general (or have you done a survey)? I for one don’t think that “making choices” is the same thing as “doing one thing among many”. That is not what most people conceive of as “making choices”, which is the libertarian view.

            And I don’t understand at all your view that it’s adopting a “libertarian conception of agency” if you deny that we don’t really make “choices”, given that what most people think of as “making choices” is “they could have made a different choice”. This sounds like an obscurantist strategy to try to tell people, “Yes, of course you have free will”, when you know full well that most people’s concept of “free will”, much less agency, is a libertarian one.

          2. “I for one don’t think that ‘making choices’ is the same thing as ‘doing one thing among many’. That is not what most people conceive of as ‘making choices’, which is the libertarian view.”

            If you think we don’t really make choices under determinism, then that’s to adopt the libertarian view. But you’re a naturalist-determinist, so why adopt it just because most folks, who are not determinists, do? To do so denies the existence of a real kind of agency. The process of choice-making between neurally-simulated alternatives is a real deterministic process the brain does, just as real as the heart deterministically pumping blood. To agree to this is not to assert a position on what we should call free will, which is a purely semantic issue. As you say, the important issue is whether *contra-causal* agency is real, and we agree it isn’t. So, like you, I deny we have what libertarians call free will, and agree about all the important personal and social implications of accepting determinism. Many compatibilists, however, don’t agree with us about those implications (e.g., the rejection of retributivism), which is why I emphatically don’t count myself as a compatibilist even though what they call free will is real.

          3. Jerry,

            I’m still unclear about what you’d suggest we do with words like “choice.”

            If you see the word as hopelessly entangled in dualism and/or libertarian free will assumptions, then do you discard the word?
            If so, can you suggest an alternative?
            The word seems to do a heck of a lot of work in human discourse, so I’m wondering which word or phrase would be able to replace it.

            Or, if you think that the word “choice” and the concept of “having a choice” can be re-rehabilitated, kept in use, so long as it is winnowed of illusions of dualism, how would you do this?

            I just don’t know how you change the word “choice” to still be useful, or to supplant it, in a way where the replacement doesn’t end up meaning essentially the same thing to be coherent and useful.

            If I’m test driving a new car which has a cloth interior, and the salesmen tells me “you have a choice between cloth and leather interior” that seems to convey real information.

            How else can he convey this information? Maybe “you COULD buy this car with a leather interior instead.” Or “we offer this with a leather interior as well” or this car is sold in either cloth or leather interior..

            We’ve dropped the word “choice” but the conceptual work seems to entail what “choice” entailed in order to still make any sense.
            All of them imply or assume “what is possible” – e.g. “it is possible to buy either the cloth or the leather version” is doing the same work as “you have the choice between…”

            The objection to “choice” is that it assumes alternative possibilities. And determinism means “only one result is ever possible.”

            But that would apply to pretty much any other replacement for the word or concept of “having a choice” (that could make any sense) that I can think of.

          4. “Here are the available alternatives.” I think that’s good enough. We can still use the word “choice” in that sense, but what we can’t do is let people think they could have selected something other than what they did. Compatibilists still seem to want the term freighted with illusions that when you pick one of the available alternatives, you could have picked a different one.

            I gave you my non-freighted alternative as you asked, so this discussion has reached its end.

          5. The statement “you could not have chosen otherwise” is semantically confusing. Does it mean “there were no other alternatives available to you”, which I think most people would assume it means, or does it mean “under identical conditions you would have made the same choice”? Clearly, determinists mean the latter. Would it not be simpler and less argumentative if we simply stated it that way?

          6. “under identical conditions you would have made the same choice”

            If we believe determinism to be true, we’re talking about a single event. No two separate events will ever have identical conditions as that would imply that the universe stood still for a finite length of time.

          7. Under identical proximate conditions then. All of experimental science is based on the principle that replicating an experiment (proximate conditions) results in the same outcomes.

          8. I agree. The “libertarian conception of agency”, as you call it, seems to say that the only valid way of thinking about agency is to follow all causes back to the Big Bang. I think is is what Daniel Dennett calls greedy reductionism. We don’t think of things that way in most other contexts so I don’t see why we have to do this when talking about agency and free will. If we want to relate agency to the laws of physics, we can call it emergence but even that seems going too far in most contexts. Human agents process perceptions, have brain states, make decisions, and have free will. My choice of coffee vs tea is ultimately deterministic (or not, I’ll let the physicists sort it out) but I don’t find it useful to contemplate determinism when considering my behavior and that of my fellow humans.

        5. The answers to the important questions are yes, your decisions are the results of the laws of physics and the physical conditions, including that physical being which is you. And yes, you could have done otherwise. On a macroscopic level, the past is effectively set in stone, just like we intuitively think it is, but (you probably remember from the Christian List argument) the macroscopic past doesn’t suffice to determine your decisions. On a microscopic level, the past isn’t set in stone. The fundamental laws of physics are Charge-Parity-Time Reversible, and at that level the past is not independent of the present, our physical intuitions be damned.

          Common sense says consciousness has to exert control for free will, but it doesn’t say consciousness has to micromanage the details. The subjects in the Libet experiment were told not to plan their actions, but wait for an urge to move their finger. The only roles for conscious decision was when they agreed to follow the instructions, and when they vetoed an urge (“free won’t” as Libet said).

  10. While I’m deciding if I’m going to click to read the story or not… 😉

    Readers of this website might be interested in reading Chiang’s short stories. He’s the very opposite of prolific, and every story is worked out completely and satisfyingly. Some are conveniently collected in Stories of Your Life and Others. Ones that may be of particular interest are:

    Tower of Babylon, a retelling that, shall we say, retells it.

    Hell is the Absence of God, in which not only are heaven and hell real, but they are visible to all people as they go about their daily lives.

    Story of Your Life, which won “best novella of the 20th century” in one poll. Although in my opinion it would have some very stiff competition, it was certainly not an unreasonable choice.

    Not in the collection, but worth checking out (it’s been anthologized several times) is the story Exhalation, which is probably the best SF short story I’ve read on the actual practice of science.

  11. I am pretty skeptical of the possibility of structuring society around hard determinism. How would you hold anyone accountable for anything? Take the student loan crisis for example. If those students did not really choose to take out their loans then should they be forgiven if they are having trouble paying for them? And contrast that with folks that worked hard and sacrificed and paid them off. Obviously this is very simplified and there are many hard working, decent people that are struggling and because of uncontrollable (determined) economic conditions, can not make ends meet. Let’s say though that there are two people in the same economic conditions and by that I mean the same opportunity to pay off a loan. If the one chooses to take the opportunity and work and pay it off, and the other goes on vacation, does not pay off the loan, and wants it forgiven, how are you going to structure society to determine what is just? So much of our life is based on all sorts of contracts that humans are assumed to have chosen. I am not even saying that hard determinism is wrong. I do however very highly doubt it is possible to build a functioning society around it.

    1. I don’t think you’ve read what I’ve written about accountability. Go back and read that on this site, as I don’t really want to repeat myself here. The short answer is that you can hold people accountable because doing so has salubrious consequences.

      I function fine in this society being a hard determinist, and the legal system could sorely use a dose of determinism.

  12. I have some trouble with Jerry’s free will argument eliding into what seems like predestination or fatalism as he often says “I had no choice” (i.e. no contracausal free will) but maybe I misunderstand. I think I understand the core arguments and am convinced that we live in a determined world and dualism is wrong. Let me take the example of the Predictor that rather than going back in time, is the perfect Libet device that with 100% accuracy predicts one’s answer to all binary choices one second before one “decides”. Jerry writes “why would people buy such a device if the chances are high it would drive them nuts? (They have no choice, of course!)”

    I want to add a caveat that seems important. Isn’t it true that ‘people would buy it because they have no choice’…unless someone convinced them otherwise? Let’s say that I was prepared to buy a Predictor on Amazon and unbeknownst to me, I am one of the unfortunate people destined to lose his mind from using it. Let’s say just before I click “buy”, a friend just happens to walk by and asks me what’s up and I tell him I’m buying a Predictor and he tells me all about the risks, especially the likelihood that I’ll go looney given my personality profile so I decline.

    Doesn’t the introduction of new information into someone’s brain always have the potential to “change their mind” and the potential outcome? While this doesn’t refute determinism since it’s still a causal chain of events, it does seem to resemble a “changed mind” and a facsimile of choice even though I would be helplessly, irrevocably convinced by my friend’s argument to the degree my brain found it convincing. But I could not and would not have changed my mind if he had never happened by which is a random event making the “choice” appear rather contingent.

    1. In the incompatibilist interpretation of a “could not have chosen otherwise” event, they are talking about a single event, not two different events where the second is made based on new information. When they ask someone who just had coffee for breakfast, “Could you have chosen tea?”, they’re really talking about that single decision only, not two decisions with the possibility of having a different brain state due to different input.

      IMHO, this is not how people interpret such questions in everyday conversation. Most people, when asked this question, answer something like “Of course, if I had wanted to.” As you suggest, they are thinking in terms of two separate events, one where they want coffee and another where they want tea. If queried further as to why they sometimes want tea and other times want coffee, they will talk about being in a different state of mind some days. Most of the time we can’t really explain why we want tea sometimes and coffee other times but we know it has something to do with having different inputs. Perhaps I’ve had tea every morning this week and now I want something different. Perhaps when I’m less hydrated, I tend to prefer tea.

      1. “Most people, when asked this question, answer something like “Of course, if I had wanted to.” As you suggest, they are thinking in terms of two separate events, one where they want coffee and another where they want tea.”

        Yes. The idea that our everyday assessment of what we can do in our choice-making is derived from metaphysically and physical impossibilities is very strange.

        If I come home from the supermarket with some ground beef for dinner and someone asks “could you have bought liver instead?” I’d answer: Yes (insofar as I knew liver was available at the store). But the reason I didn’t buy liver is because I didn’t want to eat it. I don’t like liver. But, yeah, I could have bought it instead – they had liver for sale as well.

        It’s just usually presumed, to even make sense, that this is “if you had WANTED to do that other action.” After all, why would I buy liver if I didn’t want to? I think about things I’m able to do, if I want to do them or choose to take the action. It’s not about the physical or metaphysically impossible where I think the REAL question I’m being asked is “could you have bought the liver instead given precisely the same state of the universe, the same desires, decision, etc.”

  13. Wow! That short story is an absolute beauty. Ranks up there with the best of Arthur C Clarke. (I mention him because the last line is so Clarke-ian).

    I love stories that make you go ‘Ummmm.’

    Incidentally, I don’t think the fact that your brain is in the process of deciding (and detectable 10 seconds before you become conscious of it) has any implications for ‘free will’ or determinism. Simply because it’s all still part of the same brain – yours. It’s your ‘will’ made by your brain however it decided it.

    I’m not arguing against determinism here – I think it’s logically inevitable. I just don’t think it makes any difference whether the conscious brain ‘decided’ or whether subconscious parts of it did.

    I also think that though determinism is inevitable, the story is quite correct that ‘pretending’ to have free will makes life so much easier.


  14. What a coincidence! I purchased just last week the collection this story is in and read that story. I thought about your arguments on free will as I was reading it. Well worth reading, in my opinion.

  15. Indeed. I always come back to the simple formulation of “how could you ever choose an option that never occurred to you at the time?” I rather like the degrees of freedom argument as profoundly constraining our available choices at any given moment. Using the grocery store example, one could ask oneself “what fruit am I going to buy now?” Given an hour, you could probably write down 100 different fruits. In the moment, you’re probably evaluating a potential list of 10, maybe 20. Worse, those 10-20 will only be those you’ve bought in the past or see or maybe reflect on what your palate is signaling to you at that moment. So you weren’t really free to choose the 80 possibilities that didn’t occur to you. Nor fruits in the store you’ve never heard of before. The “choice” evaluation always seems to be post facto and unfairly so given the constraints of time to make a decision. Lastly, we also don’t choose what happens to enter our consciousness…it just flows automatically.

  16. Since I am my brain, whatever my brain decides, I have decided.

    The fact that conscious awareness of a choice may sometimes follow shortly after unconscious areas complete the calculation, does not alter the fact that I am what made the choice.

    The logical fact of universal causal necessity/inevitability does not alter the fact that it was my brain that did the choosing, and no other object in the physical universe.

    In fact, it was causally necessary, and inevitable from any prior point in eternity, that I would be that which encountered the issue, weighed my options, and made the choice.

    And when it is the case that I choose for myself what I will do, while free of coercion and other forms of undue influence, then it is a choice of my own free will.

    Causation never causes anything. Determinism never determines anything. Only the actual objects and forces that make up the physical universe can cause events. And I happen to be one of those objects.

Leave a Comment

Your email address will not be published. Required fields are marked *