Plain talk about free will from a physicist: Stop saying you have it!

January 11, 2016 • 10:30 am

We’ve taken a break from the many discussions on this site about free will, but, cognizant of the risks, I want to bring it up again. I think nearly all of us agree that there’s no dualism involved in our decisions: they’re determined completely by the laws of physics. Even the pure indeterminism of quantum mechanics can’t give us free will, because that’s simple randomness, and not a result of our own “will.”

So although most of us are pure determinists about our behaviors, and reject the libertarian we-could-have-done-otherwise brand of free will embraced by most people and nearly all religionists, many of us are still compatibilists. By and large, compatibilists reject dualism and embrace determinism (and the randomness of quantum phenomena), but still say that humans have “free will”. (What is deemed “compatible” is determinism and some notion of free will.) To do that, they simply redefine the classical notion of dualistic free will so that it means something else: the lack of constraint by others, the evolved complexity of our brain that processes a variety of inputs before spitting out a “decision,” and so on.

Given that most people’s notion of free will is a dualistic ghost-in-the machine one, and that we know that’s false, it’s not clear why the classical definition of free will has been replaced by compatibilism rather than determinism. To those compatibilists who gladly embrace a new definition of “free will”, I ask these questions:

What is the point of redefining free will so that it’s compatible with determinism? And given that compatibilistic definitions are diverse and often conflicting, which one is right? Or does it even matter?

All too often, the point of compatibilism is not to create some philosophical advance, but merely an attempt to stave off the damage that rejecting dualistic free will is said to pose to society. Although compatibilists often cover that up, I think that many of their efforts are directed at keeping the Little People from seeing that determinism reigns. The motive is that people’s false belief that they really can make different decisions in exactly the same circumstances is essential to keep society running smoothly. In precisely the same way, religion-friendly atheists say that the Little People need their gods, because without them the fabric of society would unravel. (There are a lot of similarities between belief in free will and belief in God.)

Explicit statements that we need to retain some concept of free will for the good of society have been made by several people, including Dan Dennett, the most sophisticated purveyor of compatibilism (he has two books on it), philosopher Eddy Nahmias, and, in 2014, Azim Shariff and Kathleen Vohs, who warned of the dangers of rejecting free will in a Scientific American article (reference below, sadly not free). Vohs was co-author of a famous 2008 paper (reference below, pdf free) showing that reading a passage about determinism caused a short-term increase in students’ tendency to cheat in psychological tests. Their results, however, have not been replicated in two subsequent tests.

Nevertheless, in the Scientific American article Shariff and Vohs make an extended argument about the dangers of science telling us that we don’t have free will. Two snippets:

The less we believe in free will, it seems, the less strength we have to restrain ourselves from the urge to lie, cheat, steal and feed hot sauce to rude people.

Yes, in one of their studies people who read passages denying the existence of free will tended to put twice as much hot salsa on tortilla chips intended for someone else who’d acted like an ass. Can you imagine a society in which everyone tries to burn the palate of others? That would truly be a disaster!

After considering the potential effects that rejecting free will could have on society, Shariff and Vohs conclude this way:

In the 18th century Voltaire famously asserted that if God did not exist, we would need to invent him because the idea of God is so vital to keeping law and order in society. Given that a belief in free will restrains people from engaging in the kind of wrongdoing that could unravel an ordered society, the parallel is obvious. What will our society do if it finds itself without the concept of free will? It may well reinvent it.

And that’s precisely what they’ve done by inventing compatibilistic notions of free will!

Given the dubious claim that rejecting free will damages society, and the undoubted benefits to our judicial system of embracing determinism, I’m still baffled by why compatibilists continue to argue that we NEED some notion of free will. If you’re going to argue that in the comments, I’d appreciate your telling me why we have to have such a notion rather than just rejecting the idea and embracing determinism. And why is the notion you embrace better than the alternative forms of compatibilism? (As I said, compatibilism is a lot like religion.)

And if rejecting free will has bad effects on society, so what? Doesn’t the truth matter—the truth that neuroscience is telling us about the determinism of our actions? Those who are compatibilists for the good of society are no better than atheists who argue that while there’s probably no god, it’s crucial that the Little People still believe in one. My view is that we should simply find out what’s true, and then deal with it.

Finally, free will is important because it’s one of those issues where philosophy and science can really make a difference in peoples’ lives. Science tells us that our behavior is not under our conscious control, and philosophy can tell us how to apply that to issues like reward and punishment. There are not that many areas of academic philosophy that can actually affect the lives of the average person, but this is one. Why on Earth do we waste our time arguing about compatibilist definitions of free will, definitions that are too arcane to affect society or the average person?

I’ve gone on too long, but I wanted to call your attention to a good new critique of free will—both the dualistic and compatibilist verisons—by Sabine Hossenfelder, a physicist at the Nordic Institute for Theoretical Physics. It’s on the BackReAction website, and is called “Free will is dead, let’s bury it“. Hossenfelder doesn’t pull any punches, and she writes very well. I’ll give a couple of excerpts:

There are only two types of fundamental laws that appear in contemporary theories. One type is deterministic, which means that the past entirely predicts the future. There is no free will in such a fundamental law because there is no freedom. The other type of law we know appears in quantum mechanics and has an indeterministic component which is random. This randomness cannot be influenced by anything, and in particular it cannot be influenced by you, whatever you think “you” are. There is no free will in such a fundamental law because there is no “will” – there is just some randomness sprinkled over the determinism.

In neither case do you have free will in any meaningful way.

These are the only two options, and all other elaborations on the matter are just verbose distractions. It doesn’t matter if you start talking about chaos (which is deterministic), top-down causation (which doesn’t exist), or insist that we don’t know how consciousness really works (true but irrelevant). It doesn’t change a thing about this very basic observation: there isn’t any known law of nature that lets you meaningfully speak of “free will”.

I consider compatibilism one of those “verbose distractions.” She then goes after the Chicken Little Compatibilists and cites the Shariff and Vohs article (that’s how I found it):

This conclusion that free will doesn’t exist is so obvious that I can’t help but wonder why it isn’t widely accepted. The reason, I am afraid, is not scientific but political. Denying free will is considered politically incorrect because of a wide-spread myth that free will skepticism erodes the foundation of human civilization.

For example, a 2014 article in Scientific American addressed the question “What Happens To A Society That Does not Believe in Free Will?” The piece is written by Azim F. Shariff, a Professor for Psychology, and Kathleen D. Vohs, a Professor of Excellence in Marketing (whatever that might mean).

In their essay, the authors argue that free will skepticism is dangerous: “[W]e see signs that a lack of belief in free will may end up tearing social organization apart,” they write. “[S]kepticism about free will erodes ethical behavior,” and “diminished belief in free will also seems to release urges to harm others.” And if that wasn’t scary enough already, they conclude that only the “belief in free will restrains people from engaging in the kind of wrongdoing that could unravel an ordered society.”

To begin with I find it highly problematic to suggest that the answers to some scientific questions should be taboo because they might be upsetting. They don’t explicitly say this, but the message the article send is pretty clear: If you do as much as suggest that free will doesn’t exist you are encouraging people to harm others. So please read on before you grab the axe.

Hossenfelder then goes on to criticize the Vohs and Schooler study for not showing what it claims to, and then dispels the canard that rejecting free will also denies people responsibility for what they do.

At the end she draws a connection between quantum mechanics and free will—a connection that eludes me. I know of the so-called “observer effect,” but didn’t realize that it, or Bell’s Theorem rejecting the existence of local hidden variables, had any connection to dualistic free will. So I invite readers to read what’s below and then enlighten me. And of course you are still welcome to defend compatibilism if you want, but do tell me why you think we have to retain a notion of free will. (All of us, of course, feel that we have free will, but that’s irrelevant.) At any rate, tell me what this means, especially the part I’ve bolded.

The reason I am hitting on the free will issue is not that I want to collapse civilization, but that I am afraid the politically correct belief in free will hinders progress on the foundations of physics. Free will of the experimentalist is a relevant ingredient in the interpretation of quantum mechanics. Without free will, Bell’s theorem doesn’t hold, and all we have learned from it goes out the window.

This option of giving up free will in quantum mechanics goes under the name “superdeterminism” and is exceedingly unpopular. There seem to be but three people on the planet who work on this, ‘t Hooft, me, and a third person of whom I only learned from George Musser’s recent book (and whose name I’ve since forgotten).

h/t: Hector


Vohs, K. D. and J. W. Schooler. 2008. The value of believing in free will: encouraging a belief in determinism increases cheating. Psychol. Sci. 19:49-54.

Shariff, A. F. and K. D. Vohs. 2014. The world without free will. Scientific American. Scientific American 310:76-79.

226 thoughts on “Plain talk about free will from a physicist: Stop saying you have it!

  1. These are the only two options, and all other elaborations on the matter are just verbose distractions. It doesn’t matter if you start talking about chaos (which is deterministic), top-down causation (which doesn’t exist), or insist that we don’t know how consciousness really works (true but irrelevant). It doesn’t change a thing about this very basic observation: there isn’t any known law of nature that lets you meaningfully speak of “free will”.

    It’s even worse than that. Any supernatural framework ever proposed suffers the same problem. Let’s say Christian theology holds. Does the soul have any sort of fundamental nature that characterizes it? If so, it is obeying some set of rules, whatever their origin, and is a slave to those rules and is not free. If not, then it has no anchor, flits about aimlessly, and is not willful.

    “Free will” is as coherent a notion as “married bachelor.”

    “Freedom” is a useful concept in certain limited contexts, as is the concept of the “will.” All else being equal, you can discuss something’s freedom on a particular axis of a range of options. Similarly, you can consider any particular individual’s conception of an idealized reality that the individual would prefer be actualized.

    But trying to merge the two into a single whole makes as much sense as waxing lyrical about how much bachelors enjoy their marriages.

    Whether or not your physics is a better match for reality as best understood today or for ancient fantasy.


    1. Ben, it’s even worse for those who believe in an all powerful deity. If it were true it would mean that at some point before She created the universe, as we know it, She would have to have known every word, thought, deed and event across the cosmos from then to the end of time. So in that case all is pre-ordained, no free will!

      1. And it is worse still for those who believe in the “sustainer” god of Catholicism and Islam: god is not only ultimately but *proximately* responsible for everything, if anything is, because ex hypothesi it concurs with *everything*. Leibnizian optimism is the only way out, but that can be answered with a laugh.

      2. friendlypig,


        The problem of Divine Omniscience for free will was recognized very early on, before Christianity (e.g. the Greeks) and early in Christianity.

        Issues like that split Christian thinkers into different camps concerning free will.
        Those who tried to reconcile accepting both that God knows our choices in advance with free will became essentially compatibilists.
        (This is, I believe, still the most dominant view).

        Others thought for free will to be truly free even God couldn’t predict our actions.
        (Concepts like God’s “middle knowledge” were brought in to dice this up).

        They are more like Libertarian. Others were more strict determinists/incompatibiists about our free will.

        This is why, as I have pointed out numerous times here, it’s simply false to think Free Will has always been one view that is only now trying to be “saved” by compatibilism.
        Or that dualism entails only one view of free will.

        Compatibilism has been around since humans started working through the implications of God(s) on our actions.

        1. I recommend reading the Catholic Encyclopaedia (you can Google it) on predestination for an expression of the knots intelligent people (for not everyone was a fool before the rise of modern science) tie themselves into when talking of determinism. Oddly, the Catholic position regarding punishment for ill actions does not seem so different from Jerry’s: God knew from the beginning of things (or before) that you or I would act in this or that less than satisfactory way (that’s to say, what you do is determined), but nevertheless you are responsible and therefore accountable for acting in that way. The theological doctrine of ‘free will’ comes across merely as a way of holding people accountable for what they do even though it has been determined from the beginning. In the end, though it is surely determinism that rules – the conclusion Calvin drew:

          ‘That predestination is indeed a sublime mystery appears not only from the fact that the depths of the eternal counsel cannot be fathomed, it is even externally visible in the inequality of the Divine choice. The unequal standard by which baptismal grace is distributed among infants and efficacious graces among adults is hidden from our view by an impenetrable veil. Could we gain a glimpse at the reasons of this inequality, we should at once hold the key to the solution of the mystery itself. Why is it that this child is baptized, but not the child of the neighbour? Why is it that Peter the Apostle rose again after his fall and persevered till his death, while Judas Iscariot, his fellow-Apostle, hanged himself and thus frustrated his salvation? Though correct, the answer that Judas went to perdition of his own free will, while Peter faithfully co-operated with the grace of conversion offered him, does not clear up the enigma. For the question recurs: Why did not God give to Judas the same efficacious, infallibly successful grace of conversion as to St. Peter, whose blasphemous denial of the Lord was a sin no less grievous than that of the traitor Judas? To all these and similar questions the only reasonable reply is the word of St. Augustine (loc. cit., 21): “Inscrutabilia sunt judicia Dei” (the judgments of God are inscrutable).’

          Notice that ‘the question recurs’… That is to say, as soon as one comes across a instance of free will, it is possible to introduce another factor that militates against the possibility of free will – but we want the notion of free will so that people may be held accountable, so the theologian throws in the sponge, gives up the chase, and declares with Augustine that God is inscrutable. Otherwise, there seems to be no end to such discussions, as Milton, speaking of the devils in Book 2 of ‘Paradise Lost’, suggests:

          ‘Others apart sat on a Hill retired,/ In thoughts more elevate, and reason’d high/
          Of Providence, Foreknowledge, Will and Fate,/ Fixt Fate, free will, foreknowledge absolute,/ And found no end, in wandering mazes lost…’

          1. To avoid misunderstanding, the quotation is not from some work of Calvin’s, but from the Catholic Encyclopaedia.

  2. I’m certainly not qualified to speak about Bell’s Theorem, except to say that I don’t understand it.

    In regard to this blog post generally, I’m reminded of Feynman’s advice, which I wish I could quote verbatim but must paraphrase: “[Unlike many in the so-called soft sciences] I [Feynman] know how difficult it is to prove something.” What I take from this, above all, is intellectual humility, the need to admit when we don’t know or understand something, the need to remain agnostic until conclusive proof is offered and understood.

    Thus, my take on free will: Based on the fact that everything that exist seems to emerge from the laws of physics, free will probably does not exist. Can we be sure of that, though?

    Hossenfelder glosses over what I think is the most important point: that we don’t really understand the nature of consciousness, how it is generated, how it works. Even Ed Witten has said that he believes that, even though we will probably map out every correlation between brain condition and conscious experience, we will never explain how the atoms that make up our brains actually generate the conscious experience.

    Why is our confusion about consciousness the most important point? Because free will, real or illusory, is ultimately a description of consciousness. If we don’t understand consciousness, how can we assert with such certainty that we understand it well enough to deny the existence of free will?

    Have you ever been struck, out of nowhere, with the thought, “You know, I know that the brain is a galaxy of atoms acting in unison, but isn’t it weird that atoms can really generate my consciousness? What is it that’s seeing right now? How can atoms see like this?”

    Imagine finding two rocks and arranging them in as many patterns as you like–spaced closely, then far apart, then held in your hands and spun around one another. Nothing seems to happen. Then you find a third rock and play around with them some more. Again, the rocks are as stupid and lifeless as before. But somehow, you continue like this until you manage to orchestrate the motions of a billion rocks, and suddenly they become a person, a consciousness. I know that the analogy breaks down eventually, as all analogies do, but isn’t this generally the state of affairs with the stupid, lifeless atoms that make up our brains, and isn’t this mind-bafflingly bizarre?

    So, to summarize, I’m no scientist, but it’s my best understanding that no one has ever demonstrated theoretically or experimentally how the strange quantum world can give rise to the classical world, or even if this in fact happens. It certainly seems to happen, it seems impossible that it doesn’t happen, but I understand that this demonstrated has never been achieved. And in a similar way, no one has demonstrated how consciousness, whatever it is, can arise from lifeless atoms. The level of complexity of the human brain seems to be beyond human comprehension, so how can any of us know what happens to matter arranged that complexly and that intricately, especially when we as a species are currently too stupid to understand what consciousness even is?

    1. Richard,

      Your problem with not understanding the why’s and how’s of consciousness are a non-sequitur regarding the assertion that humans operate via a libertarian free will. Yes I get it… it’s somewhat daunting to think of mindless atoms being able to generate such a thing as is human consciousness… but this is merely a distraction and has no bearing on whether free will exists (or is even if it is a coherent concept in the first place).


      1. That there is libertarian free will or there is no libertarian free will is an assertion about consciousness. Therefore, humanity’s inability to understand consciousness is relevant to this discussion and is not a non sequitur.

        1. No it is not an assertion about consciousness… It is like saying the T.V. in your living room is somehow in control of the programming that it displays. You seem to be equating consciousness with agency: in other words you are completely taken in by the illusion which we are saying free will is.

    2. Thus, my take on free will: Based on the fact that everything that exist seems to emerge from the laws of physics, free will probably does not exist. Can we be sure of that, though?


      Why is our confusion about consciousness the most important point? Because free will, real or illusory, is ultimately a description of consciousness. If we don’t understand consciousness, how can we assert with such certainty that we understand it well enough to deny the existence of free will?

      There will, of necessity, always be caveats about whether or not our understanding of anything is correct. It is always possible that any entity, no matter how allegedly perfectly knowledgeable, is suffering from some sort of delusion. However, lacking specific reason to suspect such is the case, indulging in such speculation is simply to lapse into conspiracy theory.

      That and related caveats aside…Sean Carroll has very aptly observed that the laws underlying the physics of everyday life are completely understood. Lots of questions remain, of course, about physics beyond everyday life, especially including high energy physics and physics at the smallest of scales…but we know with functionally-absolute certainty that whatever goes on in those realms, as fascinating as it will certainly be, resolves down to familiar Newtonian Mechanics at human scales (with nods to Einsteinian and Quantum Mechanics.

      The last remaining element…is that all that physics is Turin-computable, meaning that Church-Turing holds. Thus, cognition is computable. Period, full stop, end of story. We “just” have to figure out the computation. We might not have the intellectual capacity to figure it out, but we can be absolutely confident that an entity with such capacity could do so.

      This should be cause for wonder and humble amazement on our part, incidentally, and not despair…but the love of my life is arriving in a few minutes with some Korean pastries (she’s Japanese) and I’ve got to make some coffee (from Sumatra — and I’m half WASP and half Russian / Polish jew, so it’s about as international a brunch as things can get), so that’s all I have time for this morning, I’m afraid.




      1. Am I the only person to see strange typographical characters in Ben’s posts, mainly when he types an apostrophe or inverted comma? This started a couple of weeks ago (I’m still using the same browser, Firefox, and it’s only Ben’s posts that show these characters).

        Have you got a new computer Ben?

        BTW, I agree with your comment, and am envious of your new and happy situation 🙂

        1. I have a theory about the apostrophe (and it’s mine). The apostrophe is used to specify when something is missing. Ben is in love so, for him, nothing is missing. I believe he has a very sensitive keyboard.

      2. I recall that Sean Carroll, whose new book I have ordered, stated very clearly on this website that he does not agree with the incompatibilism espoused on this website… Perhaps his arguments might be addressed.

        1. Yes, thanks Gasper. It does seem to me odd that Jerry should not address the arguments made by such as Sean Carroll, or address the repeated criticisms of his position that have been made by Vaal, Coel, Gregory Kusnick and others, arguments and criticisms which are responsible and certainly bear no resemblance to the picture that is repeatedly painted, by both Jerry and Ben Goren, of that dreadful, craven creature: Peccator pusillanimis compatibilis.

          1. @ Tim

            I imagine Jerry just thinks he’s explained his own position so often (and probably heard/read yours, where “your” is plural, and includes pundits not on WEIT as well), so there’s not much to be accomplished by engaging at this point.

            Y’all have convinced me, however. 😉

            1. Thank you, Diane. The trouble is I don’t think the arguments and criticisms have really been engaged, but, rather, ignored. To be honest, I am very uninterested in the quarrel between ‘incompatibilism’ and ‘compatibilism’. What I am interested in an approach to our living that begins from the fact of determinism and at the same time does justice to the complexity of human life. That is all.

              1. Tim,

                Not surprisingly, I agree with you that the arguments don’t seem to have been addressed. It seems from our perspective (compatibilists) that we spend time correcting the same misrepresentation over and over again.

                I continue to find the very nature of disagreement fascinating. I think it would be very easy for someone on the “other side” of the debate to slip into the conclusion that Jerry was at this point being disingenuous for continuing to portray compatibilism as “re-defining” free will, despite having been shown numerous times that this is begging the question, that the very nature of free will is under debate, and that it has been conceived of in compatibilist terms (among others) for thousands of years. (This is one among various mischaracterizations compatibilists have been combating here).

                If we compatibilists were so inclined, it would be very tempting to think “Well, given he has been corrected on this enough times before, at this point it’s fair to judge his approach to the conversation as disingenuous.”

                But, and I emphasize this, I am NOT inclined to make such a judgement. Nor is the above some passive-aggressive dig. The point is in fact that I truly DON’T think Jerry is being disingenuous, I think he is honestly representing things as he sees them.

                But things can look like that from the other side of a disagreement, and Jerry has implied disingenuousness on the side of compatibilists because that is how it looks from his side. Something like: “The facts and logic for incompatibilism are too obvious – at this point there’s no excuse to give free will any pass, so compatibilists must have other motivations for supporting Free Will.”

                What’s interesting is that even among atheists who believe they are open to argument and wish to follow reason toward the truth, one gets the same sensation in a debate as dealing with “irrational” religious people.

                Jerry and the incompatibilists have expressed this comparison numerous times between compatibilists and theists.

                And from my perspective as a compatibilist arguing with incompatibilists feels *exactly the same* as arguing with theists. It’s like trying to argue someone out of an intuition they just won’t let go. Jerry/incompatibilists seem to have an intuition about what it would “really mean” to be “free,” and you can show all day long how useless it is, and how inconsistent it is with every other use of the word “free,” and how inconsistent the incompatibilist arguments are internally, and with how we accept truth in any other empirical realm….but nothing shakes this intuition. Logic and reason doesn’t seem to work.

                Again, that is only a description about “what it feels like” arguing the case here.
                It’s not a claim that compatibilism is actually the better theory, or that incompatibilists are in fact illogical or unreasonable.

                It’s just an observation about the nature of disagreement, and how easy it can be to mistakenly attribute ill-intention, of various sorts, to the other side.

              2. I could tell it was you from your nym so I figured “v” was just the more street version of “Vaal”. 😉

              3. Yes, it seems to me that one should choose one’s best adversaries – and Sean Carroll would certainly be one in this case. Then something might be advanced.

              4. What I am interested in an approach to our living that begins from the fact of determinism and at the same time does justice to the complexity of human life. That is all.

                That is also a lot. 🙂

                I think that discussion has been going on here amongst some of the rest of us, though, don’t you? (And by “us” I don’t mean “me”–I just sit back and listen to the thinkers.) I’m not the only one who appreciates all the thought that goes into posts by you, Vaal, Coel, etc.

                (I also appreciate the adamant opposition…ahem…)

              5. Jerry has addressed that repeatedly, mostly to the tune of explaining why accepting determinism should compel one away from retribution and punishment and rather towards compassion and rehabilitation.



              6. Yes indeed…. there is the desire to have a compassionate stance on punishment and there is the adoption of a particular scientific hypothesis. The question is which is the chicken and which is the egg…..

              7. Does it matter? The two are mutually supportive. Whether Newton’s conception of gravity was caused by an apple falling on his head or he dropped an apple to firm up his conception of gravity might be an interesting historical question, but it doesn’t change the fact that Newton’s language for describing ballistics is very useful.



      3. Good grief Ben, you draw out your Church Turing more often than Mattie Ross draws out her lawyer Daggett. Do you really think we are impressed with such technical term-dropping? You need be aware that it is totally irrelevant if a decisional process is either computable or non computable…. what is crucial is the sophistication of those algorithms being computed and more importantly the principal “authorship” of the decisional parts of the algorithms and of course the setting of associated data. In everyday terms it simply comes down to what Kane calls “ultimate responsibility” – WHO principally formed the decisional algorithm elements and the decisional data (which is in reality largely “self formed” over time. THAT is what really establishes what we call free will – Church Turing or no.

        1. ps: my good wishes on your recently finding love Ben…. a kindred spirit to share ones life with is indeed a wonderful thing, but if that someone goes out and brings back pastries as well -this is indeed the greatest of treasures

            1. That would depend on what definitions you’re working with.

              If we agree that a thermostat has no freedom, and that a pocket calculator has no freedom, and that a chess app running on your smartphone has no freedom, and that an airplane autopilot has no freedom, and so on, then it is absurd to conclude that humans or other biological computational systems have freedom.

              But, if we start with the assumption that humans have freedom, then it is the inevitable conclusion that thermostats also have the exact same type of freedom in at least some limited capacity — and, again, most people would consider that conclusion to be absurd.

              There’s the old conundrum of how many grains of sand constitutes a pile of sand. The mathematician solves the problem by declaring (depending on context) two grains, a single grain, or even no grains of sand to be a pile. Folk wisdom rarely if ever uses such a clear-cut and barebones distinction. The folk wisdom may well be what we all live our daily lives by, but it should be clear why it’s not actually all that wise in reality.

              That’s why I reject the entire conception of “free will” as incoherent as “married bachelor.” An essential part of the human experience is the projection of expected outcomes from possible actions we perceive available to us and basing our decisions on what to do on our analysis of those possibilities. When people say they’re exercising their free will, they’re inevitably pointing to that decision-making process. The process is real, but to therefore equate that process with “free will”…well, that’s as absurd as concluding that the Force is real because you saw Luke Skywalker use it in a movie theatre and you yourself once had a strong case of deja vu.




              1. “If we agree that a thermostat has no freedom, and that a pocket calculator has no freedom, and that a chess app running on your smartphone has no freedom, and that an airplane autopilot has no freedom, etc etc.”

                If we agree that a protozoa cannot talk, and that a sponge cannot talk, and that a lizard cannot talk ….then it is absurd to think that humans can talk.

              2. <sigh />

                You’re playing definition games. Speech and freedom are in no way analogous in the given context.

                Define the word, “free,” as you’re using it in this context. Dollars to donuts you’ve got a bait-and-switch going on analogous to the way that the religious use “faith” both to mean (e.g.) spousal fidelity and belief against evidence.



              3. You’re playing definition games. Speech and freedom are in no way analogous in the given context.”
                Of course they are analogous – they are functional properties that vary in degree. Yo have heard the expression “degrees of freedom” I assume. Speech is a property of communication which varies over species – ie from simple signalling in lizards to signalling in human speech. My point is that when such properties get more complex, and combine with other complex capabilities additional properties EMERGE. In the case of speech it is language. In degrees of freedom it is free-will

              4. Lacking a definition, there is no way to discuss the matter. Whilst “degrees of freedom” is certainly useful in certain contexts, in this context, it makes as much sense as, “a little bit pregnant.”

                So, again: your definition of, “freedom,” and why it’s the one we should be using here and now…?



              5. Just in case Ben, that you might have inferred by my absence from this thread that I thought you had possibly made some cogent point for which I could produce no response, I regret to tell you that the cause of my absence was merely a prolonged skiing holiday.
                The cognitive processes that one exhibits during skiing only reenforces ones assurance that one is exhibiting yet another aspect of Kane’s “self-formed” decisional capabilities.

    3. A great deal has been done by humans to understand the universe. Mostly we call this adventure, science.

      Turns out no one usually ever cares about free will, however, what science has done is put an enormous burden on those who think there the universe is not determined by the laws of physics.

      So if someone says today (2016) that there is free will, either they mean there are ghosts in the machine or they mean to say the universe appears to be consistent with the illusion of free will.

    4. Are you similarly perplexed by how computers work? After all, they’re just electrons moving around circuit boards somehow creating incredibly complex behaviors. They can remember things, they can predict things, they’re even slowly starting to be able to learn things. They can outplay our best chess players and perform computations far beyond the capability of the human mind to perform in fractions of the time.

      Human consciousness is more complex than any computer yet made, with very different parts that work under a heuristic programming language we still haven’t fully worked out… but I’ve never seen any reason to think that the brain is anything but an organic computer.

      Does the consciousness level of an insect cause you the same pause? In what way are we truly any different from them?

    1. Witten is smart, but I am perplexed why he does not parse our freewill on a more physical level. It is as if he thinks consciousness if something different than the atoms doing what atoms do.

      1. The problem is, though, that to an outside observer with our understanding of physics, there is no place for consciousness, and such an observer would assume that ‘meat computers’ had no capacity for consciousness. But we do.

        I see the two concepts as intimately connected, until consciousness is better understood, itneeds to remain an unanswered question.

        1. There are a great deal of big-thinkers who do not see consciousness as special. See here, EDGE – WHAT DO YOU THINK ABOUT MACHINES THAT THINK?:

          If anything science shows to an obverser that consciousness is the outcome of physical processes.

          Is consciousness important and not fully understood? Yes. Is life important and not fully understood? Yes. Is matter important and not fully understood? Yes.

          An 80kg plasma burning in the center of an arbitrary star is as complex a system as I am, contains about as much potential to store and compute information. I would like to think emergent phenomena, like consciousness are special, but the physics of us is no more special than the physics of other things.

        2. I truthfully do not understand the claim, or position, “to an outside observer with our understanding of physics, there is no place for consciousness, and such an observer would assume that ‘meat computers’ had no capacity for consciousness.”

          I agree that there seems to be many people that think that. As you seem to imply here (though I may not be interpreting you correctly) many of those people also seem to think that it is obviously so. That it follows logically, or necessarily. My response is, why do you (general you) think that? It does not seem obvious to me at all. Given our understanding of reality via the pursuit of science to date, what reasonable alternatives are there? Magic? Gods? Despite our significant lack of understanding of consciousness there is already strong evidence against those categories of hypotheses.

          It seems very clear to me that what is obvious is that there is no good reason to suppose that consciousness is some special category of mysterious. Yes, we don’t have it figured out but, so what? There are myriad things that we don’t have figured out. Every big mystery that has been figured out has had smart people of the day saying things like “it is some uniquely wondrous thing that we may never figure out.” Let’s wait and see. We have just begun to develop some tools to be able to start figuring out consciousness.

        3. You (and many others) assume that consciousness is different than the mere representational processes that go on in computers or meat computers. You assume this non-representational characteristic of consciousness based on self assessment of your own mental processes, and from all of the myriad of theories that have developed from similar self assessments.

          But you have no good reason to think consciousness possesses such stand alone qualities, that is, something beyond the kind of representational structures we would find in a self-driving car.

          Adult humans of course have many more representations of self and world (and self-world relationships) than any computer, and it is this which we call consciousness. But any one moment of the human brain does not necessary include anything beyond various representational structures.

          Human meat machines should fit into physics as much as any computer or amoeba or worm nervous system should.

          At least that’s my thinking, and there is a lot to be said about what a representation is.

  3. Vohs was co-author of a famous 2008 paper (reference below, pdf free) showing that reading a passage about determinism caused a short-term increase in students’ tendency to cheat in psychological tests. Their results, however, have not been replicated in two subsequent tests.

    It would be interesting to compare these results to similar tests that used passages about the existence or non-existence of some post-death judgment (either of the monotheistic heaven/hell type or even somethnig like karmic reincarnation). I bet we might observe a similar “local” lowering of social responsibility when people read about how no such judgment will occur.

    Given this hypothetical response, my thought is that given the stability and prosperity of atheist Europe, what seems very clear to me is that we can’t extrapolate social stability and prosperity from short-term local impacts on an individual’s sense of social responsibility. IOW, people cheating more after reading a passage doesn’t tell you what the murder rate per capita or income disparity would be if everyone read and believed that passage. The small test is simply not predictive of the population-wide long-term impact.

      1. If there was a direct inverse causal relationship between belief in God and per capita murder rate, we would at least have a realpolitik justification for promoting belief in God. My point is, empirically we don’t see any such relationship.

        1. I get your point. My point is that wanting a thing to be true or wanting a thing to be believed to be true can not override the real truth of that things existence or non-existence. What you are suggesting is truth be damned, let’s live with a lie(s) if it might make things better (better according to your perspective).

          1. Well I have the luxury of there being no truth (that I know of) so directly causing damage that we are justified in lying. All real life examples of empirical truths appear to be non-damaging or if damaging, they are so because of some weird human social or psychological construct. And in those cases I can just say, ‘change the construct’ is a solution both you and I agree on.

            But if you want to reductio my position you can certainly do it. Here, I’ll do it for you: hypothetically, if the survival of the human race depended on people thinking Bugs Bunny was a Disney character, yes, I would promote that belief. Because I’d rather the human race continue with that wrong belief than it die out. I’m not a philosophical absolutist about it the way you appear to be.

  4. Yes, Sabine Hossenfelder has also written at least 4 other posts on free will that are well worth reading. For those wishing to read them, and everyone should, just do a search on Backreaction and “free will”.

    The comments to her post are also worth reading, not for the silly attempts to bring conpatabilism back into the picture bot for her responses.

    But the part on Bell’s theorem and superdeterminism eluded me too. Hope someone can enlighten us. I did a search and read some other web sites on superdeterminism but it’s still a bit confusing. I’m a total determinist. Maybe I’m a superdeterminist too but I’m not sure as I don’t completely understand it. 🙂


    1. Well I googled it and then did a bit of thinking. I could be wrong, but here’s what I think she is saying:

      1. Bell’s theorem rules out local hidden variables. But it doesn’t rule out a completely determinate universe with non-local hidden variables.

      2. A completely determined universe with non-local hidden variables is not a popular view with physicists. Most believe instead that that universe really is quantum-indeterminate, and that our inability to find hidden variables behind QM is because there are none.

      3. Physicists have found out a lot about the QM-indeterminate interpretation of Bell’s theorem. If, instead, the superdeterminate interpretation turns out to be correct, they would have to throw all of this out and somewhat go back to the drawing board.

      Her use of the term ‘free will’ in her sentence is confusion. However, given the overall sense of her article, I think it is safe to assume that it’s use is just plain confusing language, and not some deep and opaque admission that free will is needed for some unusual variant of QM.

  5. I can’t help but feel I’m one of the “Little People,” not especially bright or educated, yet I doubt my acceptance of determinism has lead me to become more cruel or vindictive. Quite the opposite, in fact: I’m more compassionate when I remember people are not “morally responsible” for their actions.

    1. That has been Jerry’s argument in the past too. While I applaud your increase compassion, and I wish that our judicial system would focus more on fixing than punishment for its own sake, I don’t see how determinism really supports either side (the ‘be selfish’ side or the ‘be compassionate’ side).

      If we’re meat machines, we could easily be meat machines that alter our behavior more readily in response to punitive measures rather than therapy. Empiricism – not philosophy – will need to be used to figure out which sort of response to bad acts (treatment vs. punishment) is most likely to make a criminal’s behavior more socially acceptable in the future. Determinism doesn’t mean treatment will work better than punishment. It might – but it might not.

      1. “I don’t see how determinism really supports either side …”

        A belief in freewill justifies punishment and inequality beyond what is needed to deter crimes and motivate people. Freewill deniers lack these justifications.

        Freewill deniers are soft on crime.

      2. In a deterministic understanding of human behaviour, the revenge impulse that lies behind retributive and punitive justice has no gaps in which it can hide from critical scrutiny. And up close, this seemingly indispensable right to justice is basically a promise to cause harm, barely tolerable from an ethical standpoint only on the condition that it’s less unpleasant than the alternative. It’s a social tool evolved as a parochial and rough form of deterrence to aid genetic survival in small bands of pre-civilized humans, which hopefully by now we don’t automatically equate with a function that serves a coherent moral framework.

        It is, essentially, to stop a bad thing by doing a bad thing. I don’t doubt that deterrence is at least some of the time the only effective measure, but that doesn’t make it a good one.

        1. Well but your last paragraph is the key and why Jerry could be wrong. I don’t say is, I say could be. If you’re going to take a deterministic and rationalist approach to crime minimization, you should ultimately be dedicated to using what works. Liberals tend to interpret “use what works” to mean “eliminate any punitive measure that doesn’t work and replace it with rehabilitation that does.” However empirically, if it turns out that some punitive measure works better than a rehabilitative one, then the rationalist/deterministic approach would be to use the punitive measure. In some cases we probably have support for the liberal position; rehab works better than some forms of punishment. But it’s a two-edged sword, and I wouldn’t count on ‘what works’ always aligning with ‘what liberals view as an appropriate treatment for crime.’

          1. I think you’re missing the basic point here. Let’s say, hypothetically, that the empirically verified, most effective crime-prevention tactic is to punish past criminals with harsh penalties where everyone can see them. In fact, let’s go further and posit that it’s 100% effective, and every alternative flat-out doesn’t work or even makes the problem worse.

            I quote my previous comment:

            “barely tolerable from an ethical standpoint only on the condition that it’s less unpleasant than the alternative”

            “It is, essentially, to stop a bad thing by doing a bad thing.”

            Compromise is built into the retributive and punitive impulse by its paradoxical nature. Hitting somebody to stop them hitting somebody is basically a lose-lose, even if it were the most effective option in existence. It is a moral dilemma. The fact that it’s the least worst option doesn’t change the fact that it’s an ethically bad – or at least dubious – one.

            So the correct response if we discovered it was effective isn’t “Oh well, it’s OK then.” The correct response is “Well, doesn’t the universe just suck?”

            1. I agree with Tim; that is well said.

              But remember that Eric’s original point was simply that embracing a more humane rehabilitative approach and eschewing a harsher punitive approach is not a requisite consequence of espousing absolute determinism.

              1. Yes, Eric’s point is correct. I have repeatedly suggested in the past that it might just as readily follow from espousing determinism and being concerned with deterring bad acts that draconian punishments are not ruled out.

  6. Here is my two cents. With pure determinism only the past, not the future, can determine current events. But with information processing devices that can anticipate the future consequences of current events (our brains), potential future consequences can determine current events. This introduces a sort of retrocausality, which I think leads to the idea of compatible free will.

    Example. 55 million years ago, as Greenland was separating from Europe, enormous volcanic activity caused subsequent wide spread global warming. The consequence of the subsequent global warming could in no way feed back to influence the events that occurred 55 Mya. Past causes present causes future. Pure determinism.

    Today, human caused events, if continued, will cause substantial global warming in the future. However, we can anticipate these consequences, and take actions in the present to change them. In that sense, we have free will–possible future consequences (global warming in the example) determining What we do today. Similarly, anticipations of potential future consequences affect our every day present actions. Of course, it all unfolds in accordance with the laws of physics. No magic.

    1. with information processing devices that can anticipate the future consequences of current events (our brains), potential future consequences can determine current events

      No, this is not the case. If your brain obeys the laws of physics when it fashions future expectations, then those expectations are a result of determinism + QM indeterminism only, and this adds no new free will-style factor to the universe. If your brain does not obey the laws of physics, then we’re back to standard dualism. Which is not empirically supported.

        1. Which compatibilists say that?
          And how do they address the problem that quantum indeterminism /= agency and so /= will, free or otherwise?

          1. Lubos Motl
            He says the we have will and because it incorporates quantum effects it is free.
            Yes, our will has some degrees of freedom because of the quantum effects, but they don’t depend on us either! He says that they depend on us because they are part of us, but I disagree.

          1. In the realm of chess playing, I would say a chess playing computer like deep blue has a modicum of free will (of a very limited kind). Unlike human brains, deep blue does not fully know or understand the future consequences of its actions because it does not have consciousness. It simply maximizes a pre-specified value function. I can imagine a future AI machine that has “free will” to the same extent that we do.

            1. I’m really not grasping why you think there is a qualitative difference between deterministic computation that somehow uses only historic data (I’m not even sure what that means, what do you do with the data if you don’t have a model?), and deterministic computation that uses a model to predict possible future outcomes. Nor why you think that requires consciousness? Computer modeling and simulation has done this for decades, and does it far better than humans for almost anything requiring quantitative analysis. A computer weather model tells us rain is coming, so we take an umbrella.

              1. I do not know if there is a qualitative difference, and neither does anyone else. But there most certainly is a quantitative difference. That is why no one thinks AI has been achieved.

                Whether the modeling of how current events affects the future is done by biological brains or some artificial information processing system is, to me, irrelevant except that the latter are still very primitive. In my original post I said “information processing devices”. Until such devices existed, potential future outcomes could not affect current events. Now they can.

              2. My decision to take an umbrella because I have reason to expect it will rain and that I know an umbrella will keep me dry is an example of compatible free will. Where else in nature does such a thing (possible future event causing present event) occur?

              3. All I see here is that you’ve chosen to redefine “free will” to mean “deterministic computation that can make useful predictions about the future”.

                And, well, nobody else uses those to words to mean that.

              4. That is because to you “free will” must mean contra-causal dualism and nothing else is free will. By that definition, it is agreed that there is no free will.

  7. In order to Bell experiment have something to say about indeterminism, our decisions (because we are an active part of the experiment) have to be indeterminated too, we have to be real quantum objects. But that doesn’t mean that those decisions have to be determined only and exclusively by our will, our decisions could be very well determined by some quantum dice.

  8. Jerry,
    A reader “Jim” on the original site questioned the same paragraph you bolded. Below is Sabine’s response to Jim. I’m afraid it doesn’t entirely clear up what she meant (at least not to me), but maybe it helps:

    Regarding Bell’s theorem. In the derivation you have to assume that the detector settings are not correlated with the state you want to measure, which means that the experimenters must be able to make “free” decisions. Take away free will and this assumption is no longer reasonable. Superdeterminism is either non-local or has a backwards causation (in which case it allows for local hidden variables, which would violate Bell’s theorem if it was applicable). I don’t actually know why it’s called that way, sorry.

    1. So I should add that this explanation appears to me to say that the word “free” as used for detector settings is not necessarily the “free will” of philosophers. If, for example, the detector settings were set by a random number generator, they would not be correlated with the state you want to measure. So it seems to me this allows for a quantum indeterminate universe without philosophical free will. But again, I could be wrong.

  9. The common sense understanding of free will is not “ghost in the machine”. Everyone is aware that s/he has his/hers preferences. What’s more, those preferences – which are in fact constraints on a set of possible decisions – are what makes us individuals, is what makes us humans. Without preferences all decisions would be, in fact meaningless, they would be random choices.

    And we are aware that other have their preferences and that the better we know them, the better we can predict their decisions. Yet we still believe that we and them have the free will.

    Therefore, “free will” concept as discussed here is a phantom created by philosophers, not reflecting how this concept is understood by common people.

      1. If I only went by what I experience I would be a very devoted ‘dualist’. However the more that I pay attention to reality the more my evidence for the accuracy of my continuing sense of dualism goes from imaginary to non-existent. Understanding (or believing that I understand) the evidence that there is no such thing, and cannot be any such thing, as ‘free will’ in no way reduces my experience of dualism….and understanding that my ‘dualistic experience’ is an ex-post facto rationalization has a lot more explanatory power and potential for improvements than obsessing over the ‘nature of my immortal soul’ does.

    1. “The common sense understanding of free will is not “ghost in the machine”.”

      How ever did you come to this conclusion?

      And what does it have to do with whether free will exists?

      1. Hmmm…I think there must be a Sophisticated Philosopher equivalent to the Sophisticated Theologian. Szopeno’s claim seems to be an example of the “sophisticated” approach of claiming that the most common, mainstream understanding of something is actually a straw man version of it.

        1. I think all szopeno is pointing out is that many people are sloppy thinkers (and I include myself), and either don’t care it at unaware that they sometimes contradict themselves. Sure, lots of people may believe in contra-causal free will. They also often say things like “I couldn’t help myself”.

    2. “What’s more, those preferences – which are in fact constraints on a set of possible decisions […]”

      Our preferences, biases (conscious and unconscious), emotions, genes, brains, IQ levels, hormone levels, upbringing, social conditioning, received education… all translate to decisions which are not free in any sense of the word.

  10. It is a foreign argument to me to suggest the denial of free will erodes society.

    Only ignorant, fatalistic, people would start to ruin their lives with choices (lies, cheating, theft) motivated by a lack of free choice.

    Why don’t more people accept the fact that determinism is the law? 1) Because they want to live forever. 2) They think the deterministic universe we live in is consistent with one that has free will, simply because we are unable to predict the outcomes of determined events in any meaningful way that would make our lives seem determined.

  11. Jerry,
    Here is another response from Sabine to a commenter. This seems to clear it up. I have bolded what I think is the most relevant part.


    I think this is a misunderstanding. When I wrote that quantum mechanics needs free will, I was not referring to the wavefunction collapse or decoherence but to the detector settings. I think we all agree that free will or consciousness doesn’t play a role for the act of measurement.

    It is only if you believe that the experimentalists have the freedom to chose the detector settings that you can conclude the underlying law must have been non-classical, ie local hidden variables theories are ruled out. Superdeterminism merely means that they don’t have this freedom. It doesn’t a priori tell you anything more than that. This could be non-local influence or it could be a backward causation (indeed the two aren’t easily to distinguish). The main point is though that it is deterministic. Ie, giving up free will brings you back the option of having a deterministic time evolution.

    Now if you follow this line of thought a little further you might see why I think it’s interesting. First, for what I am concerned local hidden variables are the most obvious explanation for quantum effects, and once you give up on free will this is a possible explanation. Maybe more interestingly, it means that measurement outcomes are in principle predictable (though that might be in practice difficult). If they are predictable, you can use it for an effectively superluminal messaging. I say “effective” because there isn’t actually anything traveling superluminally, you’re just exploiting a correlation that was already present that you didn’t know of.

    I know it sounds wild, but give it a try, maybe you’ll come to like it 😉 Best,


    Now we are only left with the deep philosophical question of why someone named Sabine Hossenfelder choses to sign her work “B.” 🙂

    1. “It is only if you believe that the experimentalists have the freedom to chose the detector settings that you can conclude the underlying law must have been non-classical, ie local hidden variables theories are ruled out. Superdeterminism merely means that they don’t have this freedom.”

      I think this is disingenuous. The experimenter attempts to set the detector randomly. “Free will” isn’t really required, just statistical independence, randomness with respect to hypothesized Hidden Variables. But for superdeterminism to work, a whole lot more is needed. It’s not sufficient just to claim that the detector settings aren’t really “free” because of determinism. There needs to be some actual causal mechanism linking the experimenter’s actions to the Hidden Variable states of the particles such that the correlations come out in a highly specific way to give deceptive experimental results! So far as I can see, all we have so far from superdeterminism is the vague idea that everything in the universe shares a light cone back to the Big Bang, so that such a causal connection is possible in principle. But is it really plausible that there is some overarching deterministic process going on in the universe as a whole via an unknown mechanism that conspires to set up deceptive experimental results and fool us into thinking QM is non-deterministic?

      1. “Free will” isn’t really required, just statistical independence

        I agree, and I think that’s what Ms. Hossenfelder means to say, she just said it poorly. Her entire article has a very strong no-classical-free-will theme to it, so I think her usage of the term in this case may just be excessively confusing or a slip up. As far as I can tell, she’s not saying dualism is needed to set detectors independently.

  12. I had a shower thought, and I’m sure it’s been discussed by many before, but I haven’t seen it from this perspective.

    If for the history of the human species we have thought 1) there were god or gods that would punish our misbehavior and reward our good behavior or 1a) there were magical processes where we could influence the environment around us and 2) we had the free will to obey these gods/conduct the magic or not along with the ability to freely choose each choice we make, no matter how mundane or significant — from the perspective of evolutionary psychology will most humans function best if they still believe those things even though scientific methods have shown them not to be valid?

  13. Without first having a definition of ‘free will’ that everyone can agree on, it’s pointless to argue about whether we have it or not.

    What is undeniable, however, is that there is a very real psychological and neurological distinction between voluntary and involuntary behavior.

    It is in this sense that it is meaningful and not at all controversial to say that “I raised my arm up because I chose to do so (i.e., voluntarily)” versus “my arm moved up because of a twitch (involuntarily)”.

    1. If our consciousness is something that emerged from simpler processes (for me this is obviously the case) then voluntarily means “lots of involuntarily”. The border is arbitrary, hence the trouble.

  14. The interesting question is not free will vs determinism, by how people (and animals) learn. the social question is responsibility, and the legal question of responsibility hinges on the nature and potential of learning.

    1. Amplifying this a bit, I have felt like an evolutionist living among creationists most of my life, because most politicians and most writers and most journalists treat learning theory with the same degree of ignorant contempt that creationists treat evilution.

      Even AI researchers don’t accept the implications of evolution. Even writers of GAs.

  15. There’s always the issue of just what exactly is the “self” that is doing the willing. Is there a sense that the chemical processes in by brain actually are me, in a way that a gun to my head is not?

    Some Buddhists (in discussing the somewhat illusory nature of the self as has Sam Harris) talk about “thoughts without a thinker”. Can one have “will without a willer”?

    A saying widely but falsely attributed to Arthur Schopenhauer is “Man can do what he wills but he cannot will what he wills”.

    It’s certainly true that I have intentions of one kind or another and act on them. But it seems to me that the point here is to question “meta-intentions”, the existence of a second “higher” layer which allows my intentions to be self-modifying.

    If I understand Dan Dennett correctly (and I’m not sure I do), his compatibilism essentially boils down to the claim that just as some computer software is self-modifying code, just so are intentions are sufficiently complex as to have a self-modifying capacity and therefore one can call this free will.

    I’m curious as to whether Sam Harris and Dan Dennett have squared off on this.


    Above, JAC says that compatibilism is like religion. I wonder if a more precise formulation of what JAC wants to say is that compatibilism is like accomodationism.

    1. I am an ignoramus but here are my two cents.

      Some said: only stupid atoms. Not just only stupid atoms. You have to include what follows: atoms->molecules->compounds->tissues->organs->brains.

      You mix chloride and Sodium and you get salt. Salt have different properties than Chlorine and sodium. This is an emergence. Don’t forget emergences and don’t forget that chemistry is a magician: it can produce spontaneously complex molecules and curious process like catalysis. Don’t forget biochemistry and don’t forget evolution.

      OTOH, we don’t know how conscious emerge, but we do know from where it emerges. Kill the brain and you have killed consciousness. This is a proof by reductio ab absurdum.
      Maybe if we discard free will we will get hell. I doubt it. I think if all of us accept there is not free will, we will get, by deterministic reasons, to defend ourselves and society because our life and society will be in the line, and you know, survival is a deterministic tendency and in that situation we will be compelled to have whatever is needed for having a functional society. Deterministic bad behaviour will confront deterministic survival.

    2. Some Buddhists (in discussing the somewhat illusory nature of the self as has Sam Harris) talk about “thoughts without a thinker”.

      I consider that about as likely as a fart without a farter. The empirical evidence for each being biological processes is about as strong in both cases. But maybe your Buddhist would say that that lingering odor in the elevator could be the result of spiritual goings-on.

      1. You may be misinterpreting that line, or I might be of course. I think that what JonLynnHarvey may be referring to (though perhaps not Buddhists in general) is that much thinking going on in the brain may occur without being directed by whatever cognitive processes result in self awareness, or consciousness. I think that is very much not in contention. That there is lots of high level cognitive activity going on in our brains that we are not conscious of is not a new or unusual claim.

      2. Darrelle is sort of right. Buddhism (or more properly Buddhist philosophay) is pointing out the notion that the self is an artificial construct that really IS a set of processes in flux.

        To quote the book “The Three-Pound Enigma”
        Consciousness appears to be continuous and unified like a movie, it is actually spliced together from scraps of the brain.

        Or see the 2012 neuroscience book “The Self Illusion: How the Social Brain Creates Identity” subject of the 2011 Royal Institute Christmas lecture (the same series Dawkins did in the 90s.)

        I confess you have supplied the basic raw material for a round of Buddhist “Who farted?” jokes.

        Sam Harris discusses the issue here

    3. “Is there a sense that the chemical processes in by brain actually are me, in a way that a gun to my head is not?”

      This feels like a deepity. It takes something trivial and true, such as that external stimuli will affect how a system behaves, and then tries to attach some notion of universal oneness on top of it to explain it that is wholly unnecessary.

      The gun itself is not a part of your mental calculations. Your perception of the gun is. If you were blindfolded or otherwise had no knowledge of the gun’s presence, it would have no effect on you. That’s why your brain is you and the gun is not.

    4. In the FW&Determinism course I did 15 years ago at UBC, some of my colleagues and I came to the conclusion that in order to resolve the FW problem one had to solve the mind-body problem. As something of a combination of emergentist, functionalist and eliminative materialist, I think the problem is more or less solved, but I realize that I do not share this (admittedly somewhat optimistic viewpoint) with many. The topic of the “self” is another way to look at it. (And, oddly, where Dennett in my view is not only wrong, but *inconsistent*: his own multiple drafts model, which I espouse in outline, explains why it is very difficult to hold responsible “all the way in”.)

  16. Since cheating only occurred immediately after reading a passage stating you don’t have free will, it seems unlikely to have a societal impact. The only way it could have an effect would be by reminding yourself that you do not have free immediately before making a decision which seems as unlikely as reminding yourself that you have free will before every decision.

    Assuming the default position for most people, if not all, doesn’t include thinking about free one way or another before making a decision, the argument that free will is needed for the little people seems a bit misplaced.

    For those compatibilists out there, try telling yourself that you don’t have free will all day long and see if it has any impact on your behavior. I’m guessing no. I’d say you might lose your vindictiveness, but that doesn’t seem to be a feature of WIET commentariat.

  17. What is the point of redefining free will so that it’s compatible with determinism? And given that compatibilistic definitions are diverse and often conflicting, which one is right? Or does it even matter?

    My response to this is what do we use to describe free will in the colloquial sense? I’m also not sure it follows that “free will” and “libertarian free will” are the same thing; else why the need to draw the distinction? Indeed, the Stanford Encyclopedia of Philosophy has a section dedicated on the “Ability to do Otherwise,” but that’s not the entire, nor even majority of the discussion. From the section immediately following:

    “But the failure of philosophers to work the account out in a fully satisfactory and intelligible form reveals that the very idea of free will (and so of responsibility) is incoherent (Strawson 1986) or at least inconsistent with a world very much like our own (Pereboom 2001). Smilansky (2000) takes a more complicated position, on which there are two ‘levels’ on which we may assess freedom, ‘compatibilist’ and ‘ultimate’. On the ultimate level of evaluation, free will is indeed incoherent.”

    I think everyone here would not only agree in the lack of libertarian free will, but also would disagree that we have free will at the ultimate level. But to apply this philosophy consistently, we’re going to start losing a lot of meaningful language. For example, no one is going to say that you don’t really have a 1 in 292 million chance of hitting the powerball for each ticket you buy. But, at the ultimate level, it’s untrue. A full accounting of all the inputs and the deterministic laws they follow will say that whoever wins has a 100% chance and everyone else has a 0% chance.

    Dualism isn’t strictly the same question as free will, but given that the only time I ever discuss this topic is during the lengthy discussions on this argument that is mostly semantic, I repeat my initial question: How do we address the colloquial common use of the phrase “free will” in a question like, “You have the free will to pick vanilla, chocolate or strawberry ice cream for dessert.” Replace that with “choice” and the same argument about ultimate choice applies as it does for free will. Saying something like, “There are three flavors of ice cream. Which one you have is determined by some past string of events over which you have no control. Whichever one you pick will be the one you pick” is needless to say clunky and unproductive. Give me a phrase I can use that can convey the same thing I mean by “free will” above that doesn’t violate this standard of whether it is true in an ultimate sense and I’ll be happy to use it. By the way, Sean Carroll feels basically the same way , and he sure as hell doesn’t think there’s a ghost in the machine or that thinking we don’t have free will will ultimately rip society apart.

    Just because there’s differing opinions about how the term is being defined, doesn’t mean we just get to throw it out altogether because the libertarian definition is incoherent and one we all agree doesn’t make sense. We don’t understand consciousness either (which isn’t the argument for free will that some people think it is), but we shouldn’t say it simply doesn’t exist because the same ghost in the machine some people attribute to giving us free will is often said to give us consciousness.

    1. “For example, no one is going to say that you don’t really have a 1 in 292 million chance of hitting the powerball for each ticket you buy. But, at the ultimate level, it’s untrue. A full accounting of all the inputs and the deterministic laws they follow will say that whoever wins has a 100% chance and everyone else has a 0% chance.”

      Great analogy.

  18. it’s not clear why the classical definition of free will has been replaced by compatibilism rather than determinism.

    Compatibilism IS determinism!

    What is the point of redefining free will so that it’s compatible with determinism?

    1) I beg to differ that compatibilism is a “redefinition”. The deterministic/compatibilistic meaning of “free will” is just as old.

    2) Every other meaning of “free” in the English language DOES NOT imply violating determinism or the laws of physics (e.g. free fall, free style, free speech, freed from jail, free man, free nation, etc).

    3) The *most* *common* use of “free will” in every day life is *not* *about* whether laws of physics are violated. That is the meaning in: “did you sign this contract of your own free will or were you coerced?”.

    4) That shows that the *normal* meaning of the term “free will” is about social pressure, not about violating the laws of physics.

    5) The whole point of compatibilism is to have a framework for understanding and talking about “decision making” and “choosing” in a deterministic world (something that incompatibilism does not do, despite the fact that concepts around decisions and choices are ubiquitous in our life).

    And given that compatibilistic definitions are diverse and often conflicting, which one is right?

    I don’t agree that compatibilistic definitions are diverse and conflicting (though there are people who do not understand compatibilism, regarding it erroneously as a variant of dualism, and so get it wrong).

    1. Complete agreement, nothing needs to be added.

      Also: What is the point of redefining “life” to be compatible with homoeostatic biochemical processes? Why not drop the term to avoid its unavoidable vitalist connotations? Inquiring minds want to know!

  19. I’ll just make my usual comment that (among people here) there is little or no disagreement about anything factual, only about the definitions of words.

    For example, I would say that the universe is not deterministic and that most of us do act with free will most of the time. I find it odd to say that the universe is deterministic (if you ignore the random bits), or to insist that only the trivially-false ghost-in-the-machine definition of free will is allowable.

    Superdeterminism does sound very interesting, but a bugger to determine one way or the other.

  20. I have a question about the notion that though we don’t have free will in any real sense, that’s not cause for despair, since we can work to change and improve ourselves. I think I heard Sam Harris make such a statement in a YouTube talk.

    Maybe I’m misunderstanding the argument, but I’m curious about how self-improvement is possible given that any “choice” one could make in this direction – to read some books, study, keep the company of smart people etc., are not really in our control anyway. Any thoughts on this subject would be great!

    1. “Man can do as he wills, but not will what he wills.”

      You’re right, our “choices” are determined by factors out of our control. But, when we make a choice, we can then act on it if we want to.

        1. Isn’t “acting on it if we want to” out of one’s control too?

          No it isn’t. If I wanted to get up now and go for a walk, then I am entirely able to act on it.

          That sort of thing is exactly what is under our control (provided we’re not disabled or in jail, etc).

          What is *not* under our control is whether we do want to.

          1. I am usually in fairly close agreement with you on free will, though not completely, but this surprises me. Can you explain a bit about why you think the choice about whether or not to act is qualitatively different with respect to determinism than the choice that preceded and led to the choice about whether or not to act?

            1. No, there is no difference there; our choices are determined (all of them).

              *Dualist* FW is about events leading up to the choice.

              *Compatibilist* freedom is not, it is simply about whether we can act on our choices.

              If we can do what we want, within normal constraints, then we are “free”.

              If we cannot do what we want, owing to abnormal constraints (a gun to the head; being in jail, etc) then we are not “free”.

              This sort of freedom is actually what is important to us in our daily lives.

              The dualist conception of freedom is incoherent and is not what is important to us in our daily lives.

              See Jerry’s quote from Kant and Cohen in the next post for an illustration of “freedom” that matters to us.

          2. “What is *not* under our control is whether we do want to.”

            Even that is under our control to some degree.

            A couch potato who hates exercise may nevertheless decide that for health reasons it’s in his best interest to join a gym and start working out. A few months later he finds that he no longer hates exercise; in fact he feels sluggish and out of sorts on days he doesn’t do it. So it turns out that wanting to go for a walk is something he was able to change.

            1. Right Gregory, I was going to give that example again as well before I saw you have made it.

              All sorts of ways people change their lives in terms of “self help” rely on taking some existing general desire and using reason to encourage further new specific desires.

              In fact, the vast majority of desires we form
              daily are ones that derive from the application of our reason starting from other desires. The desire to renovate the kitchen will engender a huge number of other desires (e.g. all the choices involved in the renovation) that result from our own reasoning.

              So it’s not like we are just helpless puppets always reacting to externally imposed desires for which we have no responsibility. The vast majority of the desires we attempt to fulfill are internally generated by our reasoning through our goals.

            1. Yes, we can choose what we want by training ourselves to new habits of thought and behavior that generate different wants.

              1. I’m not sure I see how this solves the problem because it seems to invite an infinite regress that doesn’t favour partial or total freedom.

                Let me explain: sooner or later, the first choice is actually the product of genes, embryological development, information obtained in cultural milieus, and countless other factors that the current body has no possible say in. Sooner or later, you can’t choose what you want because what you want is the result of events that you had no say in. But then, how can the first choice be much different? Or the second? Or the third? Pile on as many choices and wants as you want (heh), it makes no difference: every choice and every want is vulnerable to the charge of being the product of forces it had no possible control over. If one of those forces is a prior want, then that want is itself a product of forces it had no possible control over, and so on until a point when you didn’t even exist.

                And now we’ve gone that far, what does it mean to have control over my behaviour and thought? If I’m a puppeteer, able to influence my future behaviour, then causally speaking I’m simultaneously a puppet, being influenced by past behaviour. In which case isn’t the past – not past me, but the world long before I was born – the true puppeteer?

                Or have I made a mistake somewhere?

              2. “what does it mean to have control over my behaviour and thought?”

                This is the key question. Is it meaningful to say that a cruise missile’s guidance system controls the missile’s trajectory? Is it any less meaningful to say that our onboard meat computers control our behavior? If the word means anything at all, then we have at least as much self-control as the cruise missile, and probably more, since our computers can reprogram themselves.

              3. reasonshark,

                Gregory gave (as usual) a good answer.

                It’s always interesting seeing people suddenly have this special worry about infinite regresses of causation ONLY applied to humans and our choices. Wheres they have no problem stopping this regression at useful points for explaining everything else in the world. E.g. the nail in the tire caused, or was responsible for the deflation of the tire. Of course, there was a cause for the nail as well, and this regression could go all the way to the big bang…but no one ever thinks we therefore can’t reasonably identify the nail as the cause of what we want to explain. Similarly, yes my choice to grab a beer from the fridge may be part of a chain of causes back to the big bang, but why would we suddenly worry this invalidates identifying my desire for beer as the cause? It’s what best, and most pragmatically, explains my action.

              4. Is it meaningful to say that a cruise missile’s guidance system controls the missile’s trajectory?

                It’s definitely true that the operations of the guidance system cause the missile’s trajectory to be what it is, and no account of this causation would be complete without different levels of analysis, up to and including its computational functions (i.e. the sensing, calculating, and responding of the system).

                But if we’re going to tie control with causality and stop there, we end up with a minuscle part of the bigger picture.

                For instance, what about the fact that the internal workings were set up by a team of engineers and mechanics? They have control over the design of the guidance system that will manage the missile. Where does the man who fired the missile come into it? They have control over the team that set out to design the guidance system. And, if we go down a few scales, could we also say that the physical traits of the guidance system control it, since it wouldn’t describe its trajectory without it? After all, they also control the efforts of the team trying to make a guidance system. The quantum nature of matter controls the physical traits that are expressed, the history of the country and the individual controls how the commissioner responds to international threats, and the atmospheric conditions control the trajectory that the missile is trying to create. If control is going to be let in, then consistency demands its full contribution is acknowledged.

                There’s no refuge in internal processes, either. Compare the autonomous missile to one operated by joystick back at HQ. We say that the HQ controls the missile, which controls the trajectory. Put like that, the autonomous missile seems to be a version with HQ inside it rather than outside of it, in which case the difference between being a puppet and being a free agent is where HQ is. And if we’re conceding that much, then it seems we control our actions because circumstances – both internal and external – control us.

                I’m not saying “control” is the wrong word, but the transitive nature of causality leads to a notable consequence of extending it this way.


                It’s always interesting seeing people suddenly have this special worry about infinite regresses of causation ONLY applied to humans and our choices. Wheres they have no problem stopping this regression at useful points for explaining everything else in the world. E.g. the nail in the tire caused, or was responsible for the deflation of the tire.

                It’s not about stopping the regression for a practical explanation, and not just because you can’t technically “stop” a causal regression. It’s about the questionable conclusions of extending control backwards. Gregory Kusnick above makes the claim that even our wants are under our control, in contrast to Coel who says we can’t choose our wants but choose what we want. However, going back far enough, there comes a point where nothing that’s about to effect our existence is within our control but we are controlled utterly by it.

                This seems to me to be equating control to being a cause of something. If we cause ourselves to improve, then we control our wants and actions. But then we cannot deny that we are, in turn, controlled by prior causes, with all that this implies. At the very least, the rhetorical clout of an empowering claim that we have control is diminished somewhat.

                The other issue is whether control, or lack of it, is transitive or intransitive. If a genetic mutation controls the development of my inner dispositions, which in turn controls my being disposed to self-help, which in turn controls my shaping up, which in turn controls my longevity, then it seems contradictory to say that the genetic mutation didn’t control my longevity, albeit indirectly. But then if we want to argue that control is intransitive, such that genes don’t “control” my disposition but my disposition “controls” my actions, then consistency demands that, in a causal universe like ours, “control” is a particular subset of causality, not a different process, and so the causal implications remain. Otherwise, it would be akin to saying birds aren’t physical because rocks drop but birds don’t: we’d be highlighting an interesting but strictly irrelevant difference.

      1. Deciding on whether or not to act on a previous choice is just another choice in an unending network of choices.

        If that is all you meant to say, I agree. If you meant to say that the choice of to act or not to act on a previous choice is somehow not also “determined by factors out of our control” in the same manner as the initial choice, I disagree.

    2. I personally find the idea of no free will rather liberating. This is because it gives me a mechanism to fogive myself when I make mistakes (which is often). The mistakes are not things I willfully did, nor could they have been avoided. Instead of fretting about them I can observe how they affect the people and environment around me and make a judgement as to how awful they were. This information then becomes part of my internal “algotrithm” which hopefully will help me make less awful mistakes in future.

      1. Rory,

        What do you mean the mistakes are not something you “willfully did?’

        Did you not want, or mean to take the actions that led to those mistakes? Presumably you did as you willed (wanted) to do, otherwise you wouldn’t have taken the actions in the first place.

        Should a drunk driver not feel badly about having killed someone free will!??

        And if we are not to feel badly about past actions because of no free will, should we not feel negative emotions about possibly choosing to do harm to someone currently or in the future? Because, after all, free will means never having to say you are sorry?

        You seem to be fantasizing a Spock-like existence. Human beings are motivated by emotions and desires, and it would seem pointing those emotions in the right direction is helpful to compel changes in behaviour.

  21. I’m still baffled by why compatibilists continue to argue that we NEED some notion of free will. If you’re going to argue that in the comments, I’d appreciate your telling me why we have to have such a notion rather than just rejecting the idea and embracing determinism.

    Adopting compatibilism *IS* EMBRACING determinism! Anyone who has not embraced determinism is not a compatibilist!

    By the way, I don’t think we “need” a concept that is anything like *dualist* free will. Society would not change one bit without it. Everything would be just fine.

    We do however need concepts about “choice” and “decision making”. That means deterministic decision making, in exactly the same way that a chess-playing computer makes “choices”. Whatever language we choose to use, we do need concepts for that sort of deterministic decision making, because that is what our brains are doing all the time.

    It’s when incompatibilists decide how they want to talk about *that* that they’ll start seeing the merits of compatibilist language.

  22. It seems to me that this thing you call “free will” is not something I ever thought I had. No doubt I am misunderstanding something, but what you seem to mean by no free will is that there are limited options at any given time. Those options having been limited by all the events that preceded them.

    Personally, I’ve never thought that free will meant unlimited options. I’m free to chose to go to the grocery store. I do not have the option of flapping my wings and flying there. I don’t see that as a limitation on my free will though it clearly is.

    I do know, though, that you are wrong if you think slaying the concept of free will represents some kind of victory over religion. Many religions are completely fatalistic and in Christianity a major wing, Calvinism, is based almost completely on the idea that we have no free will. God has decided everything. Many, many religionists believe that we are just playing out a drama that god has written. It is not a stretch for them to accept that the laws of physics would be the way god manages that trick.

    I was raised in a more or less Calvinist tradition and there was always the question of personal responsibility. How to reconcile the fact that god had predetermined everything while still holding us accountable for our actions? I’ve read your posts where you raise similar concerns in the face of your deterministic outlook. What bothers me is that you seem to resolve it with the same hand-waving dismissal the Calvinists do.

    If the lack of free will has no impact on how we live, if, in fact, we all think and act as if we have it even those, like you, who are dogmatically convinced that there is no free will, what is the point? It sounds to me like a purely semantical argument with little practical value.

    1. In fact most mainstream religions are inconsistent with free will, although their adherents usually assert that we have it anyway. Every xian denomination I’m familiar with claims god is omnipotent, omniscient, and has plans for us.

    2. I warmly agree – yes, ‘what is the point?’ And since you were brought up in a Calvinist tradition, I recommend – to you and everyone who is interested in the problem of determinism and ‘free will’ – James Hogg’s ‘The Private Memoirs and Confessions of a Justified Sinner’, one of the great novels of the 19th century, and Sorley MacLean’s poem ‘Tiodhlacadh sa Clachan’ (Scottish Gaelic: ‘Funeral at Clachan’):
      There was not a man in the audience
      but took his creed from Geneva.

      There was not a man in the band
      who did not subscribe to the whole creed:

      but before they left the graveyard
      many a man understood the real distress.

      Almost all the company understood
      a thing that one would not whisper to himself alone:

      that not a third of a third believed
      in the lasting Hell of their creed.

      Also Christopher Marlowe’s ‘Dr Faustus’, which is, among other things, an indictment of the Calvinist theology Marlowe had studied at Cambridge.

  23. I’ve never entered into a conversation about free will before, but after several years of reading various takes on it and thinking about it, I’ve come to the conclusion that I am a compatibilist.

    I agree that libertarian free will is silly, and I also agree that a better understanding of determinism amongst the general public could lead to positive changes in our justice system. (I disagree vehemently with Shariff and Vohs, and I use the Vohs and Schooler (2008) paper in my Science Communications class as a great example of flashy research with completely overblown substance – studies built on the responses of 30 (or 120) college students cannot produce meaningful results, fundamentally.) But I do think that some redefined notion of “free will” is still a useful concept; unfortunately, I don’t know exactly what that definition should be.

    My problem with pure determinism is that we don’t understand causality well enough to make sense of all its implications. Randomness and uncertainty are essential properties of our physical reality and I don’t think we have the slightest clue as to how those very small-scale properties add up to influence such a messy and gigantic deterministic process like: “given my environment and everything that has happened to me ever, I have no choice but to write these very words right now.” That statement is certainly true in the libertarian sense of “choice,” but I can’t convince myself that it’s true in the deterministic sense; i.e. the concept of “choice” is void because it couldn’t have happened any other way.

    That deterministic sense of the idea always seems to me to assume that there is a traceable chain of causality to events, to “choices”; but how can that be? We often don’t even really understand what it means to say “A causes B causes A” or “A causes C and B which also causes C”. There are some mathematical ideas about what these things mean, and some statistically useful ways of analyzing systems following these models, but our understanding is not what I would consider deep. I see some redefined notion of “free will” as a stopgap for this hole in our knowledge. And that, I think, is the point of redefining free will to make it compatible with determinism. Or, maybe just using the term “free will” as a placeholder for an idea, or collection of ideas, that we can’t yet properly define. There’s an interesting and important collection of phenomena that we don’t really understand (I’m talking about how causality works here), and we should give it a name.

    1. I’ve come to the conclusion that I am a compatibilist. […] My problem with pure determinism is that we don’t understand causality well enough …

      Just on a point of wording, if you are trying for a form of “free will” by departing from “pure determinism” then you are NOT A COMPATIBILIST!

      Compatibilism means accepting 100% pure unadulterated determinism!

      It’s this sort of thing that enables Jerry to complain that there are multiple versions of “compatbilism” and that it is a refuge for people who dislike 100% pure determinism. Neither is true!

      1. I’m not necessarily suggesting a departure from pure determinism. I’m not ruling that out altogether, but I think pure determinism is meaningful only so far, after which we can’t really make sense of anything yet, causally speaking. That doesn’t mean determinism is wrong (it certainly isn’t in many senses); my point is that it alone seems insufficient to describe ideas of causation that we tend to collect under the umbrella-term “free will.”

        1. But if we’re taking a *compatibilist* meaning of “free will” then 100% pure determinism is entirely sufficient.

          People always mistake “compatibilism” as a hankering after some sort of non-deterministic thing.

  24. What is the point of redefining free will so that it’s compatible with determinism?

    I wasn’t aware there was a preferred definition. The point is that it is doable, while some claim it isn’t.

    What is the point of using a specific kind of solitaire so that its compatible with free time?

    The problem is rather when people start to discuss non-physical notions of brain function. What is the point of that? (O.o)

    If history says that “free will” is such a religious idea, fine. Then later ideas are redefinitions.

    But if “free will” is a folk physics concept, there isn’t any “redefinition”. It is just an old “effective theory” of the brain.

    At any rate, tell me what this means, especially the part I’ve bolded.

    By finding Hossenfelder’s comments, eric has already found the answer. A review is in Wikipedia, , and note that Bell found it unlikely:

    “Although [Bell] acknowledged the loophole, he also argued that [superdeterminism] was implausible. Even if the measurements performed are chosen by deterministic random number generators, the choices can be assumed to be “effectively free for the purpose at hand,” because the machine’s choice is altered by a large number of very small effects. It is unlikely for the hidden variable to be sensitive to all of the same small influences that the random number generator was.[3]”

    This shows exactly why physical or compatibilist free will is such a good model, it is ensured to work because quantum physics does. But it also shows that quantum physics is inundated with unnecessary philosophical terminology.

  25. Odd that freewill proponents claim animals have no free will, and conclude that it began with skyguy endowing it to humans.

  26. What is the point of redefining free will so that it’s compatible with determinism?

    The point of developing notions of “freedom” and “choice” in a deterministic world is very well illustrated by PCC-E’s very next post, in which he quotes Nick Cohen:

    What is Enlightenment? he asked in 1784. It was nothing less than the freedom to argue for your own ideas without being forced to comply by authoritarians: the general who says: “Do not argue – drill!”; the taxman who says: “Do not argue – pay!”; the priest who says: “Do not argue – believe!” To overcome them, you must first understand that “the public use of one’s reason must be free at all times and this alone can bring enlightenment to mankind”.

    I challenge any incompatibilist to make a meaningful interpretation of that without adopting compatibilist notions of “freedom” and “choice”.

  27. Vohs and Shariff:

    “What will our society do if it finds itself without the concept of free will? It may well reinvent it.”

    I think that is a descriptive statement, not a prescriptive one.

  28. The irony… What good exactly does a “free will” do you, if it can be so easily manipulated merely by contemplating its existence (i.e. the “psychological tests”)?

    If I can alter your mental state (by showing you some reading materials) and this translates directly into altered behavior/decision-making, how exactly is this supporting the “free will cause?” Who is making those decisions, you or me (or both)?

    If “empathy” is the naturalistic response to a divinely-inspired moral code, then “preparedness” is the naturalistic response to a mystical notion of free will. I.e. be prepared in advance to make the decisions that reflect your values.

    Who could argue with the notion that poor preparation leads to poor decision-making?

  29. Let’s start with the given that there is nothing mystical or magical about brains – why do we feel that free will is less real than any other human mental state? Love, sadness, free will, motivation, elation, are all mental states evolved to make out species successful.

    I don’t know if there is a a brain master module that evaluates the output of other modules and makes a decision. I don’t know if this module is what I perceive as “me”. But I certainly perceive that I am evaluating possible outcomes and that I am making a decisions. Isn’t this enough to declare free will as real?

    Something unrelated. Why is it that physicists feel like they can address free will. Whatever computation is taking place is not on the quantum realm but on neuronal circuitry relying on physiology. If anything biological systems do anything they can to avoid quantum uncertainty.

  30. The point of compatibilism, as I see it, is to come up with good explanations of human behavior. “The laws of physics” or “genes and environment” aren’t sufficient for that purpose. Simply saying that our behavior is determined doesn’t give an account of why we behave this way instead of that way.

    Any explanation that seeks to give such an account must take notice of what goes on in people’s minds. Even a detailed account of neurological events is not, I claim, an adequate explanation, any more than a detailed account of memory chips and logic gates is an adequate explanation of what a web browser does.

    A good explanation of human behavior will necessarily invoke such abstractions as thoughts, intentions, choices, and so on. As it happens, our cultural and linguistic evolution have bequeathed us a workable folk psychology that includes such concepts. Compatibilism is the claim that this folk psychology need not be discarded wholesale. Rather, many of those folk concepts map fairly well onto the necessary abstractions of a naturalistic theory of behavior (even if traditional metaphysical interpretations of those folk concepts are mistaken). All of these abstractions, of course, supervene on physical processes (just as “genes and environment” do), but that doesn’t mean we can improve our theory by throwing them overboard and invoking physics in their place.

    Physicists are (by and large) comfortable with the idea that abstractions described by their theories (particles, fields, spacetime, etc) correspond to real features of the world they study. Similarly, biologists accept that genes, species, selection and so on are real phenomena, not just symbols in a theoretical formalism. We should be equally prepared to accept the reality of the entities described by any successful theory, including theories of human behavior. The alternative is the instrumentalist notion that science isn’t in the business of describing reality, but merely of manipulating symbols in order to make predictions — a notion that failed spectacularly in the behaviorist program of the 1950s and ’60s.

    A quote from Hossenfelder’s blog post (my emphasis):

    It doesn’t mean that you are not making decisions or are not making choices. Free will or not, you have to do the thinking to arrive at a conclusion, the answer to which you previously didn’t know. Absence of free will doesn’t mean either that you are somehow forced to do something you didn’t want to do. There isn’t anything external imposing on you. You are whatever makes the decisions.

    This is compatibilism, and stands in direct opposition to the brand of incompatibilism put forward by Jerry and by Sam Harris, which explicitly asserts that we are not the maker of decisions, and that decisions aren’t real.

    1. “The point of compatibilism, as I see it, is to come up with good explanations of human behavior.”

      Science is far better equipped to do that than any kind of philosophy. Good science is value-free, bias doesn’t help. Compatibilism doesn’t add anything except wishful thinking.

      1. Good science is value-free,

        So science (or a good scientist) doesn’t value truth, reason, honesty, logic, openness, skepticism, coherence, cooperation, etc…?

        Would you like to see how science is done without any underlying values?

        Are you sure you’ve thought this one through?


      2. No, there is no wishful thinking involved. Compatibilism is as rooted in empiricism as anything else. It looks at the observable differences between living, conscious agents and non-conscious or inanimate matter and asks “how shall we talk about these differences?” It seems to me that incompatibilism tries to ignore these differences.

      3. I agree that minimizing bias is an important component of good science. It follows that good scientists should be open to reasonable criticism from anyone, even compatibilist philosophers.

    2. Any explanation that seeks to give such an account must take notice of what goes on in people’s minds. Even a detailed account of neurological events is not, I claim, an adequate explanation, any more than a detailed account of memory chips and logic gates is an adequate explanation of what a web browser does.

      Only out of practical consideration. There’s no added ingredient to turn a collection of memory chips and logic gates into a web browser. Since that’s the totality of what it’s made out of, and accounting for its behaviour through time and space, so long as we didn’t make any mistakes, it would explain the internal workings 100%. What it wouldn’t be is comprehensible to our slow, limited, and inattentive minds, but at least it wouldn’t be instrumentalist.

      This is where “black box” thinking becomes helpful by being practical. Instead of going into detail about every last component, you bunch them up into discrete modules and “box” them. You ignore the minutae and just proceed, for the current purpose, as if it’s granted they’ll behave a certain way, without necessarily caring how or why. This is exactly what is going on when our folk intuition packages other people’s brains in “thoughts”, “moods”, and “perceptions” rather than draws a detailed schematic of the whole nervous system, which would be more accurate but more evolutionarily expensive (and hence unlikely to evolve).

      The problem is you’re left with black boxes that folk psychology doesn’t care very much for because it’s designed not to care. Folk psychology is instrumentalist by design because it didn’t evolve to be scientifically rigorous, but to pick out enough from a specific environment to aid gene propagation. And that’s a problem because folk psychology isn’t only designed as a (parochial or not) truth-extracting organ, but as a survival tool, and one constrained by evolutionary limits that might result in parochial or pragmatic assumptions rather than scientifically defensible ones.

      We don’t accept folk biology without rigorous application of real biology and other interdisciplinary fields – in which case, folk biology is epistemologically irrelevant and we’re just lucky if we don’t have to change too much. Folk psychology is no different.

      1. There’s no added ingredient to turn a collection of memory chips and logic gates into a web browser.

        Sure there is: it’s called software — which is not itself a physical object, but a way of organizing physical objects to perform specific functions. If we seek a complete explanation of the browser’s behavior — not just of the physical transformations it goes through, but the reasons for those particular transformations and not others — we must include this functional level of description.

        Folk psychology is no different.

        I agree. Converting folk biology into rigorous science did not automatically entail that we jettison such concepts as life, species, and organisms. We just clarified them, stripped them of superstitious woo, and put them on a naturalistic footing. And folk psychology is no different.

        1. Software is not an added ingredient. The memory chips and logic gates have an immense number of mathematically possible arrangements, only a microscopic portion of which will act as a web browser. Every single one of those arrangements is a total consequence of the nature of the components as they interact with each other. Without realizing the total determination of the system and how any and all function is a product of the physical characteristics, you start to imply ghost-in-the-machine workings. Software is basically delicate hardware.

          Converting folk biology into rigorous science did not automatically entail that we jettison such concepts as life, species, and organisms.

          But it did entail handing over authority to science, on the basis that it had the actual arguments. The proof of this being: it’s biology that clears out folk biology, not the other way around.

          Besides, a lot of what we now consider straightforward biology was (and still is) violent towards our folk biology, especially but not uniquely when it came to human exceptionalism. The same thing happened in physics: despite centuries of progress, our folk intuitions as shown by psychology experiments are stuck somewhere around the impetus theory of the middle ages.

          We didn’t find out much worth finding out until, say, scientists investigated how the reproductive system worked, and fossils began to turn up. Some people still find it hard to concede that we are in any way related to jellyfish, however distantly. Casual uses of “life” are able to delve into vitalistic implications in a way scientific uses simply can’t. And while there is a concept called abiogenesis in both folk biology and actual biology, they are so radically different from each other that they have no meaningful connection at all.

          1. Saying that “software is delicate hardware” overlooks one of the key properties of software, which is substrate independence. When I download a program from the web, it’s read from a magnetic disk, converted temporarily to stored charges in RAM, converted again to optical pulses in a glass fiber, then to electrical pulses in a wire, then finally back to stored charges in my tablet’s internal flash memory. Similar conversions happen repeatedly every time the program executes. At any given moment, different portions of the executing program may reside in different types of storage media, and shuttle back and forth on demand. Yet it remains the same program throughout.

            Our best explanations of software recognize this substrate independence and treat the software as distinct from the physical state of the hardware, allowing us to talk sensibly about how software can be downloaded, backed up, loaded into memory, translated into machine code, and so on. There’s nothing spooky about this; it’s simply an effective theory of what software is and how it works. The software is the information being represented, not the physical objects doing the representing. And it’s that information, that particular way of organizing the hardware, that empowers the system to do something useful.

            Nobody denies that it’s possible (in principle) to give a gate-level account of the machine’s activity that would be causally complete in its own terms. But the utility of all that activity would be completely opaque at that level of description. The only way to discover what useful function the machine performs is to have access to the software-level description, or to reconstruct it by reverse engineering.

            1. Exactly Gregory! You have highlighted the central property surrounding free-will – which is that it is “software related” rather than “hardware related”. This is not dualism, for software is describable as a real world entity – albeit solely in mathematical terms.
              As total materialists, incompatibilists constantly hark to the fact that decisions reside in the human brain – a physical entity totally subject to the laws of physics…. as if this settles the entire issue. But the reality is, that like the computer, the brain is just a “execution engine” for software. It is the characteristics of the software that we must explore in this debate. We certainly still remain materialists, and non-dualists if we do so.

              A simple example: a program to sort a list into alphabetic order residing in a computer. Can we say that the hardware is “sorting hardware”? – NO. Can we determine that the computer sorts alphabetically just by examining its logic gates ?– NO. We must understand “the nature” of software and what it may be capable of (which is far more sophisticated that the mere hardware would have us guess)
              It is FUNCTIONALITY that we need to understand. The functionality is in the software. And as I previously pointed out computability (Church Turing) has nothing to do with causality – functionality is all that matters in issues of causality.
              The “self” is an aspect of the software. So is consciousness. So is “ultimate responsibility” in forming our decisions –by software that itself is self forming – hence we get free will.

              1. I would add to the above that hardware can not possibly encompass any abstractions – yet software can. In even my simple sorting example we find the abstraction of language, alphabets, names and symbols. Abstraction is an emergent property of software, not the hardware

              2. Software doesn’t actually exist, any more than numbers do. It is a very useful fiction to believe that both software and numbers exist…but it’s the exact same type of useful fiction as the schematic of a subway printed above the doors of the train.

                Use the map, but don’t fall into the trap of mistraking the map for the territory.

                For that matter, the map itself doesn’t even exist. All you actually have is a piece of paper with ink on it. Similarly, all your thoughts are nothing more than particular arrangements of the electrochemistry of your brain.




              3. Software doesn’t exist?
                How then can abstract entities that are expressed only in software lead to decisional results which themselves produce subsequent physical actions?
                Philosophers may have had a great difficulty in resolving this particular mind-body paradox, but unfortunately they knew nothing of a computers Input/Output Unit

              4. How then can abstract entities that are expressed only in software

                That’s your problem right there.

                There’s no such thing as an abstract entity expressed only in software. Go ahead and try to give an example of one and you’ll see what I mean. You should already be realizing that any example you’ll give will have an obvious physical manifestation.

                lead to decisional results which themselves produce subsequent physical actions?

                The exact same way a Rube Goldberg Contraption does….



  31. I would like to offer a different perspective on this issue. I’ve suffered from pretty sucky mental health problems for a lot of my adult life and my knowledge of the fact that we don’t have free will has a definite impact upon how I feel and how I cope.
    When I am OK my lack of free will does not bother me much at all as I am content in who I am and confident that my un-free will is going do the right thing and make sure I am OK. Also, as my state of mind is generally positive and concerned with worldly, less introspective things like watching Leeds United lose, it’s not something that I will ruminate on. However, when I am not well it causes me real issues – I feel paralysed and impotent knowing that I can do nothing to change how I feel. I constantly ruminate on the issue and during these periods it’s probably the biggest obstacle to me feeling positive and getting better.
    What’s worse is that the professionals are usually not able to help very much as they either don’t understand or don’t accept the non-existence of free will. Therapists always insist that I do have the ability to alter my thoughts and change tomorrow even after I have explained why I think the deterministic laws of the universe say otherwise. To be fair this is their job after all. They always seem to think that the laws that apply to everything else don’t apply to the mind. More surprisingly all psychiatrists that I have ever spoken to (with one exception) either don’t understand determinism or don’t accept it either.
    My sister suffers from very similar problems but coming from a different educational background (she is ‘arty’ while I have always been mathematical / scientific) does not experience the same angst regarding free will. To me a reductionist attitude to happenings of the universe and consequent non-acceptance of free will is the inevitable consequence of a scientific education. This sometimes causes me real problems and I think I am probably one of those ‘little people’ that are better off not knowing the truth! To be really honest this is the one scientific fact I wish wasn’t aware of.

  32. Hossenfelder writes well and I have tried to enjoy her blog. Like others above, it often moves far beyond me. No offense to physicists, but biology and chemistry are so much more grounded in . . . sanity. Eh, probably just simplicity.

    I liked her take on responsibility:

    “Even if you don’t have free will, you are of course responsible for your actions because “you” – that mass of neurons – are making, possibly bad, decisions. If the outcome of your thinking is socially undesirable because it puts other people at risk, those other people will try to prevent you from more wrongdoing. They will either try to fix you or lock you up. In other words, you will be held responsible. Nothing of this has anything to do with free will. It’s merely a matter of finding a solution to a problem.”

    As we have gone over at WEIT, responsible can be a bit slippery. Perhaps god holds people responsible by damnation, and we can read it as tit-for-tat. It also seems like you were responsible because you violated the objective moral fabric of the universe, which god duly punished you for, being the moral supreme being.

    We can see society as replacing god as being moral arbiter, and therefore one is held responsible for breaking given moral norms. But there is still a taint that these moral norms are a given to the world, similar to religious ones.

    As we come into a better understanding of human beings, societies will hold responsible, but the idea of “responsible for moral norms” slides away. In time, and at the end of the passage above, held responsible just means socially developed or socially controlled not to act in certain ways.

    Again, that is a meaning of social norm, but it is far away from ideas of responsible for violating a religious moral norm. For me, the idea that “we can still hold people responsible” always carries that ambiguity between those senses. Yes, in the future we will still be trying to build better selves and societies.

  33. The last paragraph of the the Scientific American article cites Voltaire’s maxim “If God did not exist it would be necessary to invent him.” The authors make a parallel to the effect that if free will doesn’t exist, we must invent it–that is, act as if it does. This is their conclusion based on the research they cite about the influence of reading excerpts that cast doubt on free will and the consequent effect on behavior, as compared with subjects who read a neutral article.

    The research seems rather lightweight to me. The claim was that subjects who read the skeptical excerpts were more likely to cheat on tests, put extra hot sauce on the nachos of those who had wronged them somehow (mentioned above), etc. For one thing, there was no mention of how much time elapsed between the reading and the testing incident, nor was there discussion of how long the effect lasted as evidenced by future tests.

    Also, there seems to be a paradox in that, while some of the research mentioned suggests both that doubting the existence of free will leads to less desire for retributive punishment by law enforcement and that it makes individuals more likely to inflict punishment (see the hot sauce experiment above). Those to things appear to conflict with one another, but that isn’t discussed in the article.

  34. Sub. Oh do I dare sub? Meh. I have no free will anyway so being who I am with all these inputs, I have no choice but to sub. Damn it!

  35. Jerry,

    I’m a compatibilist because I care about what’s true, just like you do.
    Because of this, I find it frustrating to see this falsehood continually perpetrated here about compatibilism:

    “To do that, they simply redefine the classical notion of dualistic free will so that it means something else:”


    “What is the point of redefining free will….”

    Combatilism isn’t “redefining” free will, dualist notion or otherwise. It is one of different competing concepts of free will. To say compatibilistm is “re-defining” free will is to imply falsely that there is one existing definition to re-define. And implying that some “classical dualistic notion” IS the default standard definition is false. This is been brought up many times, with many links and examples showing this to be the case.

    Even Wikipedia gets this right: “Free will is the ability to choose between different possible courses of action.” Notice it does not mention dualism, or libertarian free will, or contra-causality. And the reason Wikipedia starts with such an simple, open definition is because the article recognizes, as it goes on to state that how free will is conceived is “is a matter of some debate.”

    Any serious writing about the subject recognizes that the very definition, or the concept, of free will is the subject of debate:

    “Free Will” is a philosophical term of art for a particular sort of capacity of rational agents to choose a course of action from among various alternatives. Which sort is the free will sort is what all the fuss is about.

    Similar comments found here:

    And it’s been brought up before that even among dualist religions, like Christianity, the problem of God’s Omniscience/Omnipotence split Christian thinkers into different camps, essentially forms of compatibiism, incompatibilism, libertarianism etc. Acknowledging that an Omniscient God knows what choice you will make suggests
    the same dilemma as physical determinism. Hence many Christians have had to appeal to a compatibilist notion of free will – that yes an Omniscient Being can predict the only choice you will make…but the choice still has relevant freedom nonetheless. (vs some other camps of Christians who hold to a stricter Libertarian version, that it can not be in principle absolutely predictable or we wouldn’t have true moral responsibility – which leads THAT camp to claim even God can not predict our choices).

    Then there is the question of whether incompatibilists or compatibilists are talking about free will “as the average person sees it.” But, here again, this issue hasn’t been settled empirically. You’ve even noted this in your previous discussions comparing different studies on free will.

    In your conclusions you stated:

    “Rather, Nahmias et al. and Sarkissian et al. had identical tasks: are most people compatibilists or incompatibilists? The former says “compatibilists”; the latter “incompatibilists.” How do we reconcile these conflicting results?”

    You gave some modest “perhaps” it’s because…form of speculation. But concluded:
    “This disparity deserves further exploration.”

    Even your own discussion of studies done on people’s free will intuitions
    suggesst that, empirically, it is certainly not settled.

    So, in both the realm of religion (such as Christianity) and in philosophy a single understanding or definition of “free will” has never been settled upon. And in terms of whether icompatibilism/compatibilism/libertarianism best captures “average people’s intuitions” that hasn’t been settled yet either by studies.

    Why then do we have to keep seeing this site make the unsubstantiated charge of “re-defining” free will against compatibilism?

    1. I think it’s fairly straightforward:

      There is still a massively widespread popular belief that the spooky contra-causal version of free will DOES exist, i.e. “a precisely identical individual could have done otherwise through “will” in precisely identical circumstances”.

      If compatibilists would start their philosophical ramblings by stating clearly to all concerned up front that they completely agree that this spooky form of free will DOES NOT exist, and is logically incoherent, then I think there would be no issue.

      But my impression is that compatibilists stubbornly refuse to do this, and desperately want to take the conversation in any and all different directions; and will only address the question of this obviously non-existent form of free will when dragged to it kicking and screaming!

      That’s what’s frustrating.

      1. Ralph,

        Sorry to be blunt, but your complaint about compatibilism suggests you don’t know anything about compatibilism.

        And characterizing it at the outset as “philosophical ramblings” doesn’t suggest you’ve entered the conversation in good faith.

        [quote]If compatibilists would start their philosophical ramblings by stating clearly to all concerned up front that they completely agree that this spooky form of free will DOES NOT exist, and is logically incoherent, then I think there would be no issue.[/quote]

        Yeah, the compatibilists here have only said exactly that about a bilion times. It’s what we start with. Jerry knows this quite well and acknowledges this too.

        If you bother to read, for instance, Dan Dennett on Free Will and compatibilism, he spends much of his offert DISPELLING the concept of spooky/dualist/contra-causal free will!

        Have you bothered to actually read material by actual compatibilists?

        1. Yes, I’ve read DD on this. He’s one of my heroes, this is about the only thing where I don’t see eye to eye with him at all. Once you’ve done the easy part and kicked out spooky free will, the rest just seems to amount to musings about psychology, it muddies the waters, and I just don’t find any of it philosophically interesting.

          I admit to not having a lot of good faith left, that’s from frustrating conversations with less erudite supposed compatibilists than the likes DD and yourself that have proceeded in the way I described.

      2. Ralph, allow me to refer you to Coel’s comment at #26: “Anyone who has not embraced determinism is not a compatibilist!”

        Also mine at #35: “All of these abstractions, of course, supervene on physical processes”

        If you see instances in this thread of compatibilists waffling about the nonexistence of spooky free will, please point them out.

    2. Many philosophers use “define” sloppily; a lot of practicing scientists do better. In any case, you used “define” to mean “characterize” or “explicate”. In many contexts definitions are exclusively sign-sign correspondences (as in logic, for example) and “my definition” just means “my way of using a word”, which has no substantive content about the concepts involved at all.

  36. Jerry,

    I find that compatibilism tends to make more sense of the notion of “choice” and “could have done otherwise” in the context of determinism. Whereas I find incompatibilist arguments, such as yours and others, raise red flags of inconsistency that I think even “regular joes” can and do notice. Take the example of “fatalism” – or anything like it – that both incompatibilists like yourself and compatibilists decry as an incorrect inference from determinism. I don’t think you actually avoid giving implicit support to fatalistic-type rationalizing.

    To appeal to an example I’ve brought up before, take your INR5 talk about free will:

    At 40:30 you talk about the virtues of acknowledging determinism and the lack of free will, with the example of an ex girlfriend. You mentioned
    beating yourself up about “the one that got away” but then on incompatibilism, you realized self recrimination and thinking “Should have done this instead of that” make no sense, “Because you could not have done this instead of that.” And: “there was no way it could have worked out” because the laws of physics determined that. This is “good” because it is comforting, taking away feelings of guilt etc.

    My reaction to that was similar to hearing the Christian take comfort in claiming God saved them from some disaster or misfortune. It may be comforting but to anyone paying attention this suggests the Christian has not noticed the implications of their own claims (e.g. if God’s will applies to their good fortune, it has implications for what befalls everyone else, including the bad).

    Similarly, your example of taking “comfort” from the fact things “couldn’t have gone otherwise” is, by wider implication, actually settling into just the type of fatalism we both want to decry…and I think it’s not being noticed because in that particular case it “feels better.”

    To see the problem, just apply the same logic to “Fred” who beat his high school girlfriend badly during a fight. Why should Fred feel guilty, or go through any self recrimination? Why even THINK that he “should” have done otherwise than beat her? After all, “there was no way it could have worked out otherwise” because the laws of physics determined that. (And, of course, this implies that no one else can recriminate Fred, telling him he “should” have done otherwise).

    Like the Christian thanking God for saving her in an airplane crash, once you unleash this logic for yourself, you can’t just wall it off from everyone else.

    But it gets worse. The very nature of determinism entails the logic goes forward as well as backward – our future is as determined as the past. So the SAME logic above can be applied to any choice we are ABOUT to make: Fred wants to steal a car, and he isn’t stopping himself. Why? Because he accepts that whatever happens is as fixed as the past, and he can no more change it than the past. So even if he chooses to steal the car, he will have no reason to feel guilt or engage in self-recrimination, or other self-censuring feelings that may stop him stealing the car. “I’m not really responsible for my choice in a meaningful way – I couldn’t really do this or that!”

    Fred or anyone else can dismiss speculation about another route of action he “could” take, or dismiss nagging feelings of guilt etc, on the same logic Jerry is using to comfort himself by appealing to determinism. What feels comforting in one scenario, is worrying fatalism in another.

    But then, I’m sure people will say “No, Fred shouldn’t feel fine about taking the car. He should feel he’d be guilty of doing something wrong.” Or the would impress upon Fred: “You STILL have a relevant choice before steering the car. Even though your actions are determined it’s STILL the case there are reasons you SHOULD NOT CHOOSE to steal the car.”

    Of course I’d agree. But then this same logic, on determinism, goes backward into the past fixed, determined past choices. It still makes sense to say “I SHOULD have done X not Y.” You can’t have it both ways: using determinist logic to absolve feelings of guilt and responsibility for past actions, but to ignore this logic and emphasize we have choice in a relevant sense for current or future actions.

    Understanding why “I should have done X not Y” is, after all, one of the fundamental ways we learn from experience!

    So as I see it, the logic Jerry appeals to in his girlfriend example can’t be simply contained to particular, fairly innocuous events and choices. It actually promulgates a form of fatalism, when you look beyond his example, that both incompatibilists and compatibilists recognize as the problem they are trying to avoid.

    This is what happens IMO when incompatibiilsts talk too loosely and imprecisely about not really having choices, or that things could not have happened otherwise etc. This isn’t a little people argument, it’s the opposite. I find that “average people” are quite up to noticing these types of inconsistencies, undermining the project we both share.


    1. I think this confuses three different issues: a purely descriptive, nigh-omniscient account of events in space-time, hypothetical idealisations, and manipulating people to behave in certain ways.

      From a dispassionate standpoint, it is the case that whatever happens will happen. Fred will steal the car if all the prior causes – including his own fatalistic deliberations – lead to him doing so. If he doesn’t steal the car, this too is a result of prior causes. So too is whether or not he feels guilty, whether or not he engages in hypothetical ideals, and whether or not anyone else convinces him not to steal the car for whatever reason. Of course, this can lead to a bizarre self-fulfilling prophecy that he steals the car because he believes there’s nothing he can do but steal the car, but from our dispassionate standpoint, what is, is.

      Of course, we as ethical beings want Fred not to steal the car. So do we have to emphasize to him that he has choice? This might manipulate him into not stealing the car, but does that mean saying it is necessarily true? After all, we don’t just say things because they’re true: we say things to get people to do things. Something being the case is quite distinct from our commenting on it or not. It could be the case that Fred does not steal the car because we tell him there’s nothing he can do but avoid stealing it, another bizarre self-fulfilling prophecy. Or it might be he steals it anyway, thus making the statement incorrect.

      And we certainly don’t need choice to discuss hypotheticals and idealisations. Perhaps Fred lived his entire life as a criminal, and thinks to himself “I should have done differently”. But then he could equally think “I should have been born a good guy rather than a bad guy”. These statements could be correct – insofar as a hypothetical is correct – but they don’t conflict with the fact that he hasn’t lived an ideal life. They might influence him to live a different life in the future, but that’s still a matter of causality.

      There’s the question of whether something is the case at a particular locale in space-time (and concede we probably don’t know). There’s hypothetical and ideal scenarios, whether or not they were idealized. And there’s the manipulations that factor into a person’s behaviour, whether or not those manipulations involve believing or talking about compatibilism,incompatibilism, libertarian free will, fatalism, etc. Your problem doesn’t follow once you parse out those individual elements.

      1. There’s hypothetical and ideal scenarios, whether or not they were idealized.

        Sorry, I meant “whether or not they were realized.”

      2. reasonshark,

        Thanks for that reply. It is actually a beautiful example of just the problems I’m getting at.

        In denying a significant sense of “could have done otherwise” – the same sense in which Jerry absolves himself of self-recrimination – look at the path you feel you’ve been led towards:

        You are stuck saying Fred couldn’t “really” do otherwise…so you leave us with simply trying to manipulate him with lies! It’s not “true” Fred really could choose otherwise than fulfill a desire to steal a car…but hey, we can tell him he’s got that choice if it will manipulate him to not steal the car.

        Can you see how this is the Ultimate Little People Argument? Just what Jerry decries compatibilists engage in?

        You imply it’s ok for us to know the truth, but we don’t have to tell it to other people if it means withholding that truth will get them to behave the way we want. (E.g. refrain from stealing).

        And this depends on a pretty amazing assumption, that we both promulgate, as Jerry wishes to, the claim “we have no free will and could never have done otherwise” and just hope nobody notices the inconstancy of our telling them to do otherwise when it suits us – e.g. tell them “you really can choose not to steal that car.”

        How likely is THAT scenario to work out?

        These are exactly the, it seems to me, ill-conceived ways of thinking about determinism that are such a liability to our generally shared goals of getting rid of what isn’t true and retaining truth, and being able to talk coherently to others about it.

        The problem is that inconpatibilists seem to get hung up on single definitions of words like “choice” and “free” and “could have done otherwise” as if they are some sort of platonic, inviolate concepts. It’s like saying today my son was gay after winning his game. But then I say “of course he wasn’t REALLY suddenly a homosexual after winning his game, it’s ultimately an illusion.” No, it’s not an illusion. He really was “gay” – but by that I mean “happy.” “Gay” actually REALLY DOES have more than one meaning. It’s not false so long as the meaning is understood.

        The same goes for “choice.” Incompatibilists continually want to say “we can use the word “choice” but it’s actually code word for an illusion: “we don’t REALLY have a significant choice between two courses of action.” But that just imagines we are at the mercy of platonic definitions.
        No, words mean what we want them to mean.
        And “having a choice” and “understanding I could have done otherwise” can, and do, have significant meaning within determinism.
        And people use those words, without requiring indeterminism, all the time.

        Jerry only gets us into these problems of incoherence pointed out above by presuming some concept of “choice” that makes “could have done otherwise” strictly untrue, which forces one into just the problems you are encountering: you’ve removed your ability to talk truthfully about things that really matter – that is coherently recommend one action over another without speaking out of another side of your mouth that “what I’m saying isn’t REALLY true.”

        1. If you reread my comment, you’ll find I wasn’t endorsing any kind of Little People Argument. There’s a difference between a full account of the world, a hypothetical ideal, and actions that manipulate others into behaving in certain ways. My point was to disentangle those three issues, because I thought you’d needlessly jumbled them up in making a case for the relevance of choice.

          My point was that there’s a difference between a state of affairs being true, and a speech act (whether affirming this to be the case, denying it, saying something else on the assumption that such and such is the case, etc.). If we strike up a conversation with Fred, then no matter what we say, the mere act of doing so will manipulate him into behaving some way, even though we don’t know beforehand how the conversation will go or what the consequences will be. I was trying to lay out a suite of possibilities, not endorse any particular one.

          As far as the nature of causality is concerned, it seems to me that telling people they have a choice depends on being clear what that implies. I’m not saying there isn’t a defensible compatibilist definition – heck, technically I am a compatibilist – but for the same reasons of clarity that drive you to point out how we are the causes of our own actions, I want to point out that prior factors in turn are the causes of that “we”.

          And then I want to see what happens, because focusing on individuals as the causes of actions seems to me… not factually wrong, but incomplete, like emphasizing the role of a virus in disease and not e.g. the factors that enabled its arising and spread in the first place. Plus, there’s straightforward having as complete an account as humanly possible.

        2. I think there is an additional problem with the terms, other than different people using different definitions. An underlying problem that leads to different people using different definitions.

          Words like “choice” entered human language long ago to designate something that humans undoubtedly do. The phenomenon really happens. But, our understanding of how that phenomenon occurs has changed, and it doesn’t match our older understanding of the phenomenon.

          This sort of thing happens all the time. Sometimes the original word is retained, sometimes it isn’t. This conflict between Compatiblists and Incompatiblists seems to me to largely be an argument over just that. Despite each camp’s claims about the other side’s claims both sides, arguing here at least, are nearly perfectly congruent on what the phenomena involved in the free will argument are and the only real differences I see are differences regarding which aspects are more important, and of course what the implications of those more important aspects are.

          To explain my own view briefly, I’ll start out by saying that I am not in the middle. I largely agree with both the Compatiblists and Incompatibles, at least as argued here over the past several years. But, I disagree with both sides regarding what they claim to be the / a main reason for their position.

          I don’t agree with the Compatiblists that a commitment to a Compatibilist philosophy is necessary, or even especially helpful, in order to use terms like “choice,” “responsibility” and similar to talk about human behavior, ethics or rights.

          I also don’t agree with the Incompatiblists that a commitment to a Incompatibilist philosophy is necessary, or even especially helpful, in order to facilitate changes in our criminal justice systems.

          1. I also don’t agree with the Incompatiblists that a commitment to a Incompatibilist philosophy is necessary, or even especially helpful, in order to facilitate changes in our criminal justice systems.

            I largely agree with everything you just said, but this part is where I really get hung up in the Incompatibilist position–I don’t see what the debate on free will has to do with criminal justice at all, other than in the indirect sense that we can modify behavior through the environment, since behavior is determined. We know that reinforcement and punishment drives human behavior and that behavioral patterns can be modified. Discussing that the Universe is deterministic distracts from the argument that we should not have a retributive justice system. People can (and have) argued against retributive justice even while believing in free will. There’s a perfectly justifiable argument that revenge is not a worthy ideal simply because as humans, we’d prefer people not seek out revenge against us. There’s no need to bring free will into the discussion. All this aside, bringing up determinism inevitably leads to a discussion something like this:

            Me: “There’s no free choice. The criminal could not have done otherwise, thus we shouldn’t seek retribution since he couldn’t help himself.”

            You: “Well, I have no choice either but to seek retribution and I’m the judge.”

            Obviously, this sort of conversation can quickly go down the rabbit hole and soon have nothing to do with the topic at hand–criminal justice. For evidence that these conversations become prolonged debates that are 95% semantics, see every post Jerry has ever made on this site about free will.

        3. reasonshark,

          Ok, though I have quibbles with that, I still don’t see any answer to the problems I highlighted in Jerry’s logic, where he appeals to determinism to stop thinking about what he could have done differently, and stop self-recrimination and guilt. As I’ve said, once you’ve unleashed that logic, it doesn’t stay in a nice, neat little box, mollifying only fairly innocuous incidents in one’s life.

          If Fred uses the same appeal to explain why he feels no guilt or remorse for beating up a woman, is that a good thing? If Bernie Madof upon being caught says to the press “I see no reason to feel guilty or suffer any self-recrimination for what I did. Nor do I bother questioning whether it would have been better not to scam people of billions. Physics being what it is, I couldn’t have done otherwise!”

          Do incompatibilists like Jerry start writing newspaper pieces declaring “Good for Bernie, we need more of this attitude!” If not, why does Jerry get to use that logic, but not Bernie?

          And, again, for the conversation with Fred. If Fred says “I’m going to steal the car because I can’t do otherwise, and either way I don’t feel guilty about doing so, because my actions are determined.”

          In order to make a case to Fred that he ought not steal the car we HAVE to have grounds for explaining that he could “do otherwise,”
          How can you advise him to do otherwise without establishing that possibility to begin with for him?

          If you take the view that entails “doing otherwise” is always strictly an illusion, is false, then you are stuck with a little people argument where you are lying to him to tell him he ought to choose an alternative course of action. This just crumbles into incoherence, ultimately, where we are constantly lying to one another.

          OR…you have some truly meaningful, TRUE sense in which Fred can choose to do otherwise – one whose definition takes into account determinism and doesn’t deny determinism, but in which is it TRUE to say “you can do otherwise.” In that case, you don’t have to lie to Fred, you can make a true case, and we don’t go around lying to one another.

          And it seems to me you are pretty much on board with our having a valid sense of “choice” “could do otherwise” and “responsibility” within the context of determinism.

          But then, as I’ve said, that same logic will apply to any past actions. Which means that would turn back to show Jerry is perpetuating a problematic and unhelpful line of thinking when he says he doesn’t need to think about “what I could have done instead” to lead to a better situation, or that he shouldn’t feel guilt etc. It would be TRUE and USEFUL to say Jerry could have done otherwise. This is, as I’ve said, what it means to learn from experience in the first place! And we aren’t Spock – our feelings of guilt, shame etc from doing the wrong thing are motivations to do the good thing, just as good emotions of compassion, joy and love trigger good behaviours. (Which isn’t to suggest one live with guilt forever, but certainly negative emotions if brought on by our doing something significantly wrong, can compel change).

          So, to sum up, I have brought up this problem before, how incompatibiists like Jerry often speak in ways that create problems for making a coherent case for how we are to act given determinism, and often tacitly endorse fatalistic type thinking as a by-product.

          1. “If Fred uses the same appeal to explain why he feels no guilt or remorse for beating up a woman, is that a good thing?… Do incompatibilists like Jerry start writing newspaper pieces declaring “Good for Bernie, we need more of this attitude!” If not, why does Jerry get to use that logic, but not Bernie?”

            I don’t have an answer to that. Self-fulfilling prophecies are weird, and if Fred et al actually say that about a past event, it’d be hard for a determinist to argue that they’re factually incorrect. After all, they said X was bound to happen, and lo and behold it was. It also seems equally compelling to point out that the action was bad, whether or not it was pre-ordained, though that’s less because of the nature of causality and more because it harms people. On the third hand, if the message is “And I’ll do it again because I believe I was preordained to do so,” then we’re at least informed that this is a person not committed to reducing harm and that therapy, incarceration, or similar will reduce his chances.

            I suppose the closest thing to an answer – at least intellectually – is that as far as a future event is concerned, someone confidently claiming that they are determined to do X doesn’t actually know whether they will or not, and can’t know until the moment has passed by. I suppose there’s also hypothetical reasoning, competing impulses, and that self-awareness issue I mentioned lower down this page. And since we think and reason and act on what we know, what we sense, and what we want, that’s basically choice. Fred almost certainly has more on his mind than “I’m determined” – maybe he enjoys it, or else why would he as a human being with a brain do it?

            But yeah, I’m not sure what taking comfort in a deterministic universe is actually about. It’s not like bad things disappear just because events are determined by physical law.

            1. ‘It’s not like bad things disappear just because events are determined by physical law.’

              That remark hits a nail, one that needs to be hit, very firmly on the head.

              1. Right, but do we hold a person morally responsible for the bad things or not?

                That’s why it is better to say that the compatibilists are really claiming that *moral responsibility* is compatible with determinism, and the incompatibilists *not*. See Fischer and Ravizza’s _Responsibility and Control_ where the former is the explicit aim.

    2. “Similarly, your example of taking “comfort” from the fact things “couldn’t have gone otherwise” is, by wider implication, actually settling into just the type of fatalism we both want to decry…and I think it’s not being noticed because in that particular case it “feels better.””

      I agree.

      I suppose one could say that when a situation is in the distant past, and effectively out of reach of our decision making process, that it’s ok to take solace from determinism. But it’s surely so easy for a complacent mindest to then leak into our active decision making.

      One may certainly choose to take a fatalistic view of life, but I don’t see it as a logical consequence of determinism at all. It’s perfectly logical reasonable to keep trying to do better.

      “This is what happens IMO when incompatibiilsts talk too loosely and imprecisely about not really having choices, or that things could not have happened otherwise etc.”

      I agree with this too.

      I think the words “choice” and “decision” are used in imprecise and misleading ways. Although our choices may not be “free” in any spooky dualistic sense, our brains process data to generate an output – and what are we to call that output if not a “choice” or “decision”?

      The worst fatalistic misinterpretations of determinism seemed to be under the misapprehension that thinking carefully could not possibly generate a better output that not thinking at all.

      1. I think part of the problem is the way self-awareness seems to sabotage a deterministic understanding of ourselves. Or, to put it another way, the the problem is the observer effect with a vengeance.

        The very act of trying to see ourselves as (admittedly highly elaborate) cogs in the machinery of the universe changes how that machinery operates. A rebel doing so will immediately want to contradict what he sees in order to prove a point, completely failing to realize that he’s blown the experiment rather than disproved the hypothesis.

        Certainly, the easiest way to make this obviously fit with determinism is to fulfil a self-fulfilling prophecy you yourself predicted, which is where the easy association with fatalism comes in.

  37. Superdeterminism primer.

    A couple of things to note first of all:

    (1) Superdeterminism is generally not taken seriously in the physics community – it is objectively a fringe idea, I think Gerard t’Hooft is the one major figure working on it.

    (2) As a “loophole” for Bell’s Inequality, superdeterminism really has nothing to do with philosophical free will as discussed here. It concerns statistical independence, “free” meaning random. It concerns the statistical validity of experimental results.

    Brief historical background:

    Prior to QM (see Laplace) probabilistic processes were seen as only pseudo-random, but ultimately deterministic. For example, rolling a die is effectively random because the tiny variations in forces are so complex and subtle that we can’t predict the outcome. But, in principle, all the inputs are deterministic, and a sufficiently sophisticated model of the muscle action of the roller, surface of the table, etc., could in principle predict the outcome.

    In the 1920’s, QM introduced something fundamentally different, a physical theory incorporating irreducible probability distributions for the properties of QM objects, deterministically unknowable EVEN IN PRINCIPLE. In the mainstream Copenhagen Interpretation, the probability distribution does not just represent our lack of knowledge of a “true” underlying deterministic state. QM says that reality EQUALS indeterminate probability distributions.

    This led to Einstein’s dissent (the EPR paradox), and a philosophical impasse until Bell came up with a way to actually test things with entangled particles. Bell set up the null hypothesis of a generic deterministic “Hidden Variable Theory” (the kind of thing Einstein preferred), by which two entangled particles might carry their “true” (just as-yet unknown) underlying states from their origin to the detectors. Bell then showed that QM predicts a statistical distribution of observations that cannot be consistent with ANY possible (local) Hidden Variable Theory. Alain Aspect (and later others) did experiments that show violation of the Bell Inequality, i.e. they show that no (local) Hidden Variable Theory can explain the results, and that QM’s irreducible indeterminate probability distributions for properties are indeed correct. (This is also where “spooky action at a distance” comes in, not so much in QM itself, but in trying to form a classical intuition for the weird behavior of entangled particles that follow the predictions of QM perfectly.)

    There were some widely acknowledged potential “loopholes” in Alain Aspect’s original experiment, all of which have since been closed quite convincingly with more and more sophisticated experiments, e.g. see Jerry’s recent post

    “Superdeterminism” is not usually mentioned among the recognized loopholes, honestly because it’s not generally taken seriously. Of course, one of the assumptions of Bell’s Theorem is that the statistics are valid. In practice, that means that the detector settings are independent, i.e. random with respect to the state of the hypothesized Hidden Variables when the entangled particles are emitted. It is sometimes misleadingly said that the “free will” of the experimenter is at stake, but really it’s more mundane – it’s just a question of whether the detector settings are independent. Superdeterminism challenges the independence (randomness) of the detector settings. Effectively, it is saying that the universe somehow “fixes” the detector settings so that they are not independent from the Hidden Variables, i.e. we are getting a biased and statistically valid random sample.

    But there seems to be no plausible mechanism for such a thing (t’Hooft seems to be working on one, but I don’t understand it). It seems to amount to a conspiracy on the part of the universe to trick us. It is literally equivalent to the following: Suppose we roll a die repeatedly for a large number of trials, and we see “6” come up about half the time. We would conclude that the die is loaded. But superdeterminism says that this may in fact be a fair die; and the many “missing” outcomes of “1” through “5” that we would expect to see in a random sampling from a fair die are being hidden from us by a mysterious mechanism.

    It seems that superdeterminism must ultimately imply that ALL experimental statistics are untrustworthy, and thus all experiments and effectively all of science is meaningless. Although “spooky action at a distance” is difficult to accept, superdeteriminism seems to be a far more preposterous conspiracy on the part of the universe. It seems to have no more explanatory power than “God did it”. It’s equivalent to saying that the universe is really 6,000 years old, but that God set things up to perfectly to look much older to test our faith.

    1. Here’s an interview with Gerard ‘t Hooft talking about Superdeterminism.

      Again, to emphasize – he’s a heavyweight, major player in physics, highly imaginative; but in this narrow respect (Superdeterminism) he’s seen as way out in left field, this interpretation of QM is not at all mainstream.

  38. “What is the point of redefining free will so that it’s compatIble with determinism?”

    We need some version of “acting freely” to distinguish those offenders who should be rehabilitated from those who should not. For example, we need to rehabilitate the reckless driver but not the (first seizure) epileptic driver. Both might have caused deaths, but the former chose to act recklessly.

    1. All we need to know is whether someone breaks the law who is capable of recognising reasonable potential consequences of their actions.

      It’s about acting responsibly. That all depends on the mental capacities of the actor and the relevant circumstances. Obeying the law is generally the responsible thing to do.

      An epileptic seizure is a relevant circumstance. Any history? No? Then no responsibility. Reckless driving? Oh, the driver was a chimpanzee. Then no responsibility. No reckless driving. Just a freak accident.

      Free will go hang.

  39. My following comment surely will not be popular with this group and if there are responses at all they will likely be personal attack, but it may give someone a thread to unravel. We all behave as if we have free will and we prefer to think that we are free to make choices whether that is in fact true or not. If denied by external entities to do as we choose (or think that we choose) we resist. So in a practical sense it does not matter (to almost all) of us whether something called free will exists or not, so why raise the issue. Well it does matter to a very small subgroup which can profit from the discussion in some way. Guess who that group comprises. Qui bono? We have little or no understanding of free will, consciousness, qualia or our relentless quest to survive and no idea where to begin. The simplest explanation is that we do not know and may be incapable of knowing in principle, e.g. dogs and mathematics. Bringing in science to explain these questions does not help in the least. In fact similar mysteries attach to science. We are able in specific circumstances to make useful predictions (e.g. build machines) but we have no proof that in fact we have the deeper answers how the world actually works, nor do we (except a small subgroup) actually need that deep knowledge to benefit from the predictions. If you think I am a Luddite, I made a good living making technological predictions (it’s called engineering) and applied the same foundational knowledge as those who accumulate knowledge for “science”.

  40. I’m just happy professor Ceiling Cat got my e-mail and it turned into an interesting discussion. 🙂

    I’m working on a Physics PhD (just started!) but I work on Relativity, so the question on Bell’s theorem is out of my reach for now, until I have time to branch out a little and perhaps look into Foundations of Quantum Mechanics, which is where the meat of the question is (not FW, since that’s dead, but Superdeterminism).

    Oddly enough, there is a debate among physicists about whether or not the study of Quantum Mechanics Foundations is even worth it (I think it certainly is). Here is a paper making the case for it (fun, no math, but not light reading either, ~30 pages):


    1. Thanks for posting that arxiv link, Hector.

      There is also the so-called retrocausal loophole:

      I recommend Huw Price’s book, Time’s Arrow and Archimedes’ Point.

      Seems to me quantum behavior could be a consequence of time-symmetric magnetic interactions between relativistically-circulating point charges (i.e. the Hestenes zitterbewegung interpretation of quantum theory).

  41. Bee has long been my favorite physics blogger. Read back through her archives for much more fascinating, clearly explained goodness!

  42. Most of the points raised in the incompatibilist position presented in these discussions are entirely straw man arguments. As a convinced compatibilist myself let me deal with them quickly:
    1) As a compatibilist I could not care a jot about “the little people”. Consequentialist implications of the free will debate are amusing but irrelevant. In return I would hope that incompatibilists themselves stop arguing from consequences themselves – e.g. arguing for changes to penal policy because “free will does not exist”
    2) Calling the compatibilist argument “dualist” is a total misrepresentation of our position (and a cheap shot) – our explanations of mental activity are COMPATIBLE with scientific determinism, that is why we call ourselves “compatibilists”
    3) Making allusions that our arguments have “religious overtones” is totally incorrect and unworthy – we are on the whole- fellow atheists
    4) Claiming that seeking better definitions of terms in the free-will debate is somehow proof that our position is weak is also a very poor argument. All science involves redefinitions of terms that describe phenomenon, as we get greater understanding…. eg. gravity was once a mutually attractive force between mass, now is a distortion of space-time by mass.
    The worse thing I see in the incompatibilist argument is that it is grossly reductionist. It argues that because neural activity is a wholly deterministic at the neural level itself, that nothing above the neural level need be considered. This is equivalent to arguing that as an AND gate does nothing but a logical conjunction, there can be no such thing above it such as a computer or computation.
    Free-will is an emergent capability – emergent from very sophisticated cognitive processing capabilities in the human mind – abstraction, associative processing, learning, simulation, pattern recognition. Arguing that this phenomenon can not exist demands that the argument be carried out at the level that these higher processes exist – not at the “AND Gate” level. It must encompass explanation that includes the nature and processes of consciousness and the self… where free will would emerge. Argue at this level and I will listen.

    1. AFAIK noncompatibilist determinists don’t dispute the value of emergent property terms. Its often fine and useful to treat minds, temperature, and wave motion as phenomena separate from their underlying substrates.

      I think the disagreement is in whether ‘free will’ is the term we should be using to describe the emergent but fully deterministic mental activities that occur in the human brain. Compatibilists (AIUI) say yes, noncompatibilists say no.

      I tend to side with the noncompatiblists because I think laypeople associate the term ‘free will’ with at least some amount of nondeterminism. Thus using ‘free will’ to describe this deterministic system is just confusing laypeople and not helping them understand the system properly.

      Why do I think this? Because most laypeople in America are dualists. They believe their mental activities occur after brain death and thus are not fully determined by it. Thus using the term ‘free will’ to refer to a fully deterministic phenomenon, the way the compatibilists want to, is just confusing things more, not bringing clarity to the discussion at all.

      1. “I tend to side with the noncompatiblists because I think laypeople associate the term ‘free will’ with at least some amount of nondeterminism.”
        But this is my point Eric… it is not appropriate to discredit a phenomenon because “the man in the street” has a misconception of it or an improper definition.
        If this were our criteria we would also have to reject the existance of gravity or evolution etc etc

  43. Consider re-framing the issue:

    The key question being addressed in “free will” discussions is _predictability_. Can my “decisions”/behavior be predicted? How much control do I have over my own future behavior? Can I even predict _my own_ future behavior, let alone that of others?

    The answer, which is 100% compatible (sorry for “overloading” that term) with a pure-physics-based model of a decision-making system, is that the future behavior of _any_ chaotic system can only be predicted to the extent that the system can be accurately modeled.

    Knowing the starting conditions of a chaotic system, you might be able to model it for a certain amount of time before deviations from the real system and the model accumulate to an unpredictable degree.

    We can accurately model weather systems a few days out… Beyond that, even for a highly-instrumented, few-degrees-of-freedom, relatively-simple-to-model weather systems, it becomes IMPOSSIBLE to model accurately beyond a few days.

    Human behavior results from a physically-based, yet EXTREMELY COMPLEX system, one which includes the brain: A system which incorporates vast amounts of information from the environment (which includes a vast number of other brains) and processes this information in a vast tangled collection of neurons and synapses. There are effectively an INFINITE number of degrees of freedom in the “human system”: Instincts, emotions, cognition, nature, nurture, etc.

    EVEN IF you could know exactly what the current state of a human decision-making system was, including the atomic state of the brain, the signals in flight heading towards it from sensory systems, the state of the environment surrounding it etc., you MIGHT be able to model that system for a few nanoseconds before ANY model of ANY realizable fidelity became hopelessly out of sync with the physical system being modeled.

    I.e. you may not have “free will” (physics is physics!), but your decisions and behavior CANNOT be pre-determined or predicted because of the inherent, unpredictable, chaotic nature of the natural world.

    This is ironic, considering that the entire purpose of the human brain is to optimize behavior by finding enough order in its environment to be able to predict ITS future state, even accounting for its own effects on its environment… (So really, when you try to predict someone’s decisions (even your own), you’re building a model of something that itself is a model.)

    1. The above comment is referring to _absolute_ predictability with mathematical certainty.

      Of course, we can build relatively rough models of individuals and take good (and often accurate) guesses as to what they’ll do or how they’ll behave, and even influence that behavior.

      1. Everything you say is totally correct Jason, human agency is essentially a chaotic system and has very little longer-term predictability. But even as a staunch compatibilist myself I don’t feel that lack of predictability is all that there is to free will. Our outcomes resemble quantum physic’s “wave function”. Each future outcome has a large range of attendant probabilities associated with it – but certain of the possible outcomes are more probable than others. At the “decision time” the wave function collapses and one possible outcome becomes the agent’s chosen “intended outcome”. Why have we evolved to have this capability? – it is, as Dennet says, to give us evitability – the ability to avoid certain possible outcomes (which lends animals a higher survival value).

        But this behaviour cannot be said to confirm that we have either incompatibilism or compatibilism. The question to be settled is WHO is collapsing the wave function and how are they doing it (ultimate responsibility). Here, as you say, we come to consciousness- human consciousness being the most sophisticated and wide ranging of capabilities. It is ultimately an exceedingly complex form of self-awareness. But what is that self-awareness then? It is, as you say, the entities ability to model its own self as an agent in the universe.”. Not only that, but its ability to simulate its own influence on the possible outcomes… and then to CHOOSE. Choice- this is what breaks down the “wave function”. The “self” is the chooser. The ability to model ourselves puts us in, what Hofstadter calls, this “strange loop.
        So in essence our conscious mind is a vastly sophisticated information processing entity, which we can say, is analogous to an algorithmic processing complex that can simulate future possibilities for itself. Within this complex MUST be recognition of the SELF – the particular agent affecting and affected by the outcomes. And crucially, the decisional data and the criteria are to a large degree SELF FORMED in a recursive way as we grow into adults. This essentially gives to human agency the element of free will.

        1. Another way to put it is:

          A) Up until the “decision point” (wherever you place the threshold where some notion of “free will” is exercised) is that prior to that moment, no reasonable system that can exist within our universe* can predict with certainty the outcome.

          B) After the “decision point”, the outcome _could not have been otherwise_, given the states of the system up until that point.

          Shorter: The future can’t be predicted, but you can contemplate the past.


          To the “no free-will leads to the breakdown of society” types, who would “steal that car just because without free will they have no choice anyways”, I’d posit a version of Pascal’s Wager:

          Without “free will”, your choices are:

          A) unimportant, because choice itself is an illusion and you have no control of your destiny, or

          B) important, because you CAN have an impact on your own future.

          If I pick A and I’m wrong, I’m giving up control of my own destiny. I.e. OOPS!

          If I pick B and I’m wrong, I’ll never _know_ I’m wrong, and indeed couldn’t have chosen otherwise. I.e. whatever… 🙂

          *I say “within our universe” to account for the possibility that our universe exists within some meta-verse where even what looks to us like absolute quantum uncertainty, isn’t; as if our universe is simply executing within a virtual universe machine that can be rerun from scratch, with the same random seed, to reproduce our universe’s events perfectly.

          Ultimately though, the “free will” issue is inevitably subsumed first into a question about the nature of consciousness, and then into a larger metaphysical “nature of the universe” question.

  44. I tend to think of ‘free will’ as something like ‘centre of gravity’. Both are abstract ideas and don’t exist as identifiable material objects or processes, but both are useful fictions and enable us to handle our daily lives and interactions.

    Perhaps ‘consciousness’ is the centre of awareness (over time) too. Not ‘real’ but abstract.

  45. I feel that resistance to the idea of the nonexistence of “free” will is inextricably linked to the notion that, somewhere in us, we have an autonomous,”agent”, a “soul”, what have you- the average person is loath to accept the idea that they have no real control over their decisions, some being too “taken in” by the illusion that they do; some because they are ignorant of the nature of cause-and-effect at the level of their thinking, and some because the idea of no “free” will is incompatible with their religious programming. “Free” will is an essential part of every religion’s tenets- there would be no point in punishing anyone in a Hell, were they not ultimately the “agent” behind their sinful actions, and why reward people eternally in Heaven who would have gone on to do the good that they did anyway? Even Buddhism, with its refusal to take a stance as to whether there is a “God”, relies on the individual making a “self-willed” decision to escape suffering by adopting Buddhism’s tenets (ironically, one of its goals is the cessation of desire, but is not the initial decision motivated by a desire itself, the desire to escape suffering?).

    Most people, in my experience, don’t think deeply about much of anything; the marvelous illusion our minds create that there is a “me” who, like a man in a control booth, pulls levers and pushes buttons, his decisions unaffected by outside circumstances (the definition of “free” is, “Not affected by any outside condition or circumstance”- have you ever made a decision that was not, in some way,influenced by some condition or circumstance?)is a “given” for them, and, with our societies having been shaped over millennia by the assumption that there IS “free” will, the instances are few where they are moved to challenge, or even examine, this notion (I like to say, “The ego, like everything else in the universe, works perfectly: it thinks it is itself.”).

    Take away the “free” will, and you take away the “controller”- what does that mean? No one’s responsible for their behavior, anymore?
    I don’t personally believe in even compatibalist free will, but you don’t see me running amuck; to those who claim the notion of “free” will, even if it doesn’t exist, is necessary to preserve morality in this world, I could point out to them that the moral state of the world doesn’t seem to be very good right now, even WITH a majority of people believing in “free” will! It’s a situation kind of like the Middle East: there have been attempts to create “peace” in the M.E. for 75 years now, but no real peace is apparent and things seem to be getting continually worse. Why is this? The answer is that there are people in the M.E. who don’t WANT peace, because they still cling to outmoded religious and ethnic hatreds. In a similar vein, our criminal justice system has a greater percentage of our population locked up than any other country on Earth, yet crime rates don’t appear to be falling to any meaningful extent. Why is that? The answer is that our justice system is still clinging to the concept of “free” will as a basis for assigning blame; the system is so overloaded that no real attempt is being made to “rehabilitate” offenders and, if it is, it’s being made using the old idea that a person can just “will” themselves to change their behavior, given the right opportunity.
    I’ve gone on too long, but I want to end with my belief that EDUCATION is the key: I do not run amuck today (even though I WAS a law-breaker, and a hell-raiser, for a while) because I was exposed to the proper information as to what is needed to get along in this world with a modicum of peace. It’s said that the big difference between mankind and animals is our ability to form images in our minds, and act upon those images as if they were real- we need to USE this capacity, as it’s the only thing that will keep us from acting like the animals we are, whose brains and social instincts were not designed for the world we live in today.

  46. If you look at a chess game, you have transformation rules that dictate the movement of pieces. You can describe a chess game as a series of moves, and the moves of each piece can be explained in accordance with historic transformation laws (King Pawn to 4). So you can “explain” a game of chess based on the initial conditions, and laws governing the movement of the chess pieces, and the actual sequence of moves deployed, consistent with the movement rules, through the game.

    What can’t be explained by the initial conditions plus the rules of transformation must be random, e.g. the piece moved, and the particular location the piece moves to within the scope permitted by the movement rules, right? How do we know its not random?

    If you are looking for historic transformation laws (X -> Y), you will never be able to distinguish between random events and intentional behavior. [Note: what makes it intentional act is that it is driven toward an end, which as Baconian science dictates, we don’t look for in natural science, especially outside of social science.]

    Indeterminism does not entail “free will”, but it is equally possible that there are teleological generalizations that may be developed with further study that we do not today distinguish from random events. We certainly are not looking for teleological principles, or funding research to look for these principles. But what I am saying is no more speculative than the determinist argument. Quite the opposite, because what I am saying is consistent with a set of conceptual assumptions that our understanding of the world is founded upon.

    We know that free will exists, because people enter into legally binding agreements of their own free will every day, and if you go to court to challenge one of these agreements, you do not call a physicist or a neurologist to prove that you were under duress. This is not to postulate some metaphysical entity, but to say we have a concept of free will versus coercion, and it has a well defined use in our legal system. How physics could prove our definition is wrong is unthinkable. Definitions are not right or wrong (except relative to a socially defined convention), they are useful or not useful. It is like claiming physics disproves “1 + 1 = 2”.

    Nietzsche was pretty clear about this when he said that now we have eliminated free will, it is pressing on us to eliminate the idea of no-free-will. There is only strong will and weak will. But the point–if we accept a reductionist metaphysics based on universal scientific laws of mechanics–is not that the will is free or not free, but that there is no will at all from that perspective. Which suggests that we are leaving something out.

    Science tells us how something came to be, where it is or where it likely will be (where the artillery shell will land if fired at such and such an angle–if the cannon doesn’t misfire–or what the like result of a particular chemical combination will be). It also tells us what the material composition of something is, and the measurable qualities of a substance (specific gravity, molecular mass, chemical properties). But science does tell us what something is (which relates to form or essence) nor what the purpose of something is (which is related to the question why). Science cannot prove that a particular object is a screw driver for example, that would be something a lawyer might do. It certainly doesn’t answer the question of who, either.

    The questions of “who”, “what” and “why” cannot be reduced or eliminated to questions of how and questions about the material composition and properties of what something is. Certainly, the question of free will relates to who someone is, and what they understand their purpose here to be, which is outside of what you would predict or measure in a traditional physics experiment.

    If you could scan someone’s brain and tell that they will be an accountant at IBM in 10 years, great, it would change everything. But that will never happen in our lifetimes, if ever happens. Even the Rock/Paper/Scissors experiment that was popularized, no one has tested (to my knowledge) if it can be gamed through biofeedback (is a neurological poker face possible?).

    1. Everything must be basically an epiphenomenon flowing from the initial conditions of the Big Bang, in which case, how do you distinguish anything from other things? In determinism, there are no discontinuities, yes? If you are going to assign it a will, it must be a cosmic will (along the lines of Spinoza and Einstein). But you don’t really have anything like a will at all, you have passive matter being acted upon by physical forces.

      On the other hand, if you indeterminacy, then you have real discontinuity. Indeterminacy means that there is room, nested within the framework of physics, just as within the rules of chess, for the emergence of higher ordered phenomenon describable in terms of teleological principles (“John left because he had to use the bathroom”, “Here Kasparov sacrifices his queen in order to set up the series of moves at blah-blah”).

      I am not going to haul out all the arguments for holism in biology, other than what makes a whole is the maintenance of particular boundary conditions, and the coordination of the parts within the whole toward some end. So I think we will get there as we move from the molecular level to trying to understand the level of viruses, cells, mitochondria, and related systems.

  47. Let’s take it another direction. You have physical forces that act on matter, in accordance with physical transformation laws.

    Are ideas one of the forces described by physics? No.

    Yet we know that human behavior is different based on whether someone believes in free will or believes in determinism.

    So it would appear we have discovered another force in the universe, the power of ideas.

    Likewise, can physics explain why for every number, there is a greater number? Have we conducted an experiment that proves this to be the case?

    Yet, with a system of normative practices, like mathematics, we can transform the world through technology.

    Say one child counts “1, 2, 3, 4, . . .”

    and another counts “1, 2, 4, 5 . . .”

    Does physics prove that the first one is counting correctly, and the second is counting incorrectly. Their behavior is explained–presumably in both cases–by a series of mechanical processes. Where does the concept of a “right way” emerge? How can it emerge?

    Hubert Dreyfus (a Heideggerian) wrote a book called What Machines Can’t Do decades ago about why AI would fail, back in the day when people thought it was 5 years to Commander Data. I still don’t think the conceptual problems of modeling the human brain on a abacus have been addressed by naturalists, and while there are smart physicists like Bohr and Penrose, many today are simply dogmatists who refuse to consider other possibilities. I think they are enamored with a particular picture, and are somehow afraid that if the picture is wrong or incomplete, then somehow things will revert to medieval Catholicism.

    1. Where does the concept of a “right way” emerge? How can it emerge?

      Well, it’s simple: just rule your life based on 1,2,4,5 and you’ll figure yourself. Or you can imagine that. It’s like natural selection, but for ideas in the virtual universe in our mind; the idea that 1,2,4,5 is “the right way” will die (in reality via your body or in your imagination) sooner than the idea that 1,2,3,4,5 is “the right way”.

      1. I both agree and disagree. I can perfectly imagine a hunter gather society that got along well without counting or a concept of number. (And I believe that such a society has been discovered in fact in the Amazon or something).

        However, I do agree that the normative cultural systems that prevail generally do so because they confer survival benefits to the groups that adopt them. However, this is an explanation of why there is a right answer in the practice of counting but not one for creative writing.

        But children learn to count, and when they do, they don’t decide to count a particular way because they see that it has greater survival advantage. They internalize an objective or at least intersubjective norm, a “right way”–and what does physics have to say about objective norms? [War and demographics generally drives what system of norms prevail, but does not explain the existence of objective norms, such as valid logical or mathematical forms of proofs.]

  48. When a claim gets so tenuous that you need to write a furious exhortation to maintain it, it’s time to give up. Determinism has been falsified long ago. And without it, classical randomness is sufficient to explain all acts of freedom. As hundreds of people have probably already explained in ways far clearer than I can.

    Religious leaders can convince their subjects that they have been taken over by God and are unable to resist their commands. Anti-religious leaders, however, (who have been around for centuries — you’re not new on the scene!!!) have gone so far in denying everything godlike that they have nothing to substitute in such a threat. Even something like “the laws of physics” would be recognized as godlike by their well-trained subjects.

    And if that’s as far as it went, it would be a good thing that they would have deprived themselves of a tool of tyranny. But they found a way out. Instead of giving a name to a putative controlling power, they turned the semantics upside-down and spoke of a power of NOT being controlled — “free will”. Thus they were now able to be tyrants also, telling their subjects they have no free will to resist obedience, while maintaining the veneer of nihilism.

    Science is not, and has never been, an anti-religious enterprise. Those who use religion to deny science are getting away with hypocrisy because skeptics like you refuse to participate in the struggle to build a new modernity.

Leave a Reply