Robert Sapolsky on free will

June 9, 2017 • 1:45 pm

UPDATE: In a review in The American Scholar, Michael Shermer gives Sapolsky’s book a very positive review, and also has a few words of his own on free will. Shermer’s assessment:

The book is Sapolsky’s magnum opus, not just in length, scope (nearly every aspect of the human condition is considered), and depth (thousands of references document decades of research by Sapolsky and many others) but also in importance as the acclaimed scientist integrates numerous disciplines to explain both our inner demons and our better angels. It is a magnificent culmination of integrative thinking, on par with similar authoritative works, such as Jared Diamond’s Guns, Germs, and Steel and Steven Pinker’s The Better Angels of Our Nature. Its length and detail are daunting, but Sapolsky’s engaging style—honed through decades of writing editorials, review essays, and columns for The Wall Street Journal, as well as popular science books (Why Zebras Don’t Get Ulcers, A Primate’s Memoir)—carries the reader effortlessly from one subject to the next. The work is a monumental contribution to the scientific understanding of human behavior that belongs on every bookshelf and many a course syllabus.

I guess I’d better read it now!


Robert Sapolsky is a professor of neurology at Stanford, and is well known for his popular writing, especially on anthropology. He got a MacArthur “genius award,” and, like me, was raised Jewish, became an atheist, and got the Freedom From Religion’s “Emperor Has No Clothes Award“. I haven’t read much of his stuff but I know a lot of readers like it, because they recommend it to me all the time.

I’ve started liking him since a reader whose name I’ve forgotten (apologies) recommended that I read an interview Sapolsky had at Vox with Sean Illing, “A Stanford scientist on the biology of human evil.” The occasion was the publication of Sapolsky’s new book, Behave: The Biology of Humans At Our Best and Worstwhich is Amazon’s #1 best seller in Biology (I haven’t read it). The interview is wide ranging, and I’ll reproduce just one bit on a topic that interests me. I’ve put asterisks on the last three sets of questions and answers, which form the crux of the discussion and clear up some common misconceptions.

Oh, and I guess he’s not a compatibilist. 🙂

Free will is an illusion

Sean Illing

Okay, but in the book you come awfully close to concluding something very different. Specifically, in your discussion of free will, you reluctantly embrace a deterministic account of human behavior. You argue that free will is, in fact, an illusion, and if that’s true, I’m not sure how “malleable” we can be.

Robert Sapolsky

If it seemed tentative, it was just because I was trying to be polite to the reader or to a certain subset of readers. If there is free will, it’s free will about all sorts of uninteresting stuff, and it’s getting cramped into tighter and increasingly boring places. It seems impossible to view the full range of influences on our behavior and conclude that there is anything like free will.

Sean Illing

That’s a bold claim…

Robert Sapolsky

You’re right. On the one hand, it seems obvious to me and to most scientists thinking about behavior that there is no free will. And yet it’s staggeringly difficult to try to begin to even imagine what a world is supposed to look like in which everybody recognizes this and accepts this.

The most obvious place to start is to approach this differently in terms of how we judge behavior. Even an extremely trivial decision like the shirt you choose to wear today, if dissected close enough, doesn’t really involve agency in the way we assume. There are millions of antecedent causes that led you to choose that shirt, and you had no control over them. So if I was to compliment you and say, “Hey, nice shirt,” that doesn’t really make any sense in that you aren’t really responsible for wearing it, at least not in the way that question implies.

Now, this is a very trivial thing and doesn’t appear to matter much, but this logic is also true for serious and consequential behaviors, and that’s where things get complicated.

*Sean Illing

If we’re just marionettes on a string and we don’t have the kind of agency that we think we have, then what sense does it make to reward or punish behavior? Doesn’t that imply some degree of freedom of action?

*Robert Sapolsky

Organisms on the average tend to increase the frequency of behaviors for which they’ve been rewarded and to do the opposite for punishment or absence of reward. That’s fine and instrumentally is going to be helpful in all sorts of circumstances. The notion of there being something virtuous about punishing a bad behavior, that’s the idea that’s got to go out the window.

I always come back to the example of epilepsy. Five hundred years ago, an epileptic seizure was a sign that you were hanging out with Satan, and the appropriate treatment for that was obvious: burning someone at the stake. This went on for hundreds of years. Now, of course, we know that such a person has got screwy potassium channels in their neurons. It’s not them; it’s a disease. It’s not a moral failing; it’s a biological phenomenon.

Now we don’t punish epileptics for their epilepsy, but if they suffer bouts frequently, we might not let them drive a car because it’s not safe. It’s not that they don’t deserve to drive a car; it’s that it’s not safe. It’s a biological thing that has to be constrained because it represents a danger.

It’s taken us 500 years or so to get to this revelation, so I don’t know how long it will take us to reach this mindset for all other sorts of behaviors, but we absolutely must get there.

*Sean Illing

So what is true for the epileptic is true for all of us all of the time? We are our brains and we had no role in the shaping of our biology or our neurology or our chemistry, and yet these are the forces that determine our behavior.

*Robert Sapolsky

That’s true, but it’s still difficult to fully grasp this. Look, I believe there is no free will whatsoever, but I can’t function that way. I get pissed off at our dog if he pees on the floor in the kitchen, even though I can easily come up with a mechanistic explanation for that.

*Sean Illing

Our entire notion of moral and legal responsibility is thrown into doubt the minute we fully embrace this truth, so I’m not sure we can really afford to own up to the implications of free will being an illusion.

*Robert Sapolsky

I think that’s mostly right. As individuals and a society, I’m not sure we’re ready to face this fact. But we could perhaps do it bits and pieces at a time.

Sapolsky (looking like a rabbi) and his book

93 thoughts on “Robert Sapolsky on free will

  1. I am a huge a fan of Sapolsky. I am reading Behave right now. Not an easy read for a layperson, but I highly recommend it. I am learning a lot about neurobiology.

  2. Interesting fellow. I read everything he writes. I once spent a very memorable afternoon in Kenya many years ago, helping him tranquilize baboons with a blow gun.

  3. “Our entire notion of moral and legal responsibility is thrown into doubt the minute we fully embrace this truth, so I’m not sure we can really afford to own up to the implications of free will being an illusion.”

    NO, NO, NO! This is totally WRONG.
    Carrots and sticks work precisely because our brains work largely deterministically.

    1. Not totally wrong: moral responsibility rests on the notion that one could choose to be either “moral” or “immoral”, and there should be no RETRIBUTIVE justice under determinism.

      1. Yes, in the case of retributive justice, I agree.

        “Look, I believe there is no free will whatsoever, but I can’t function that way. I get pissed off at our dog if he pees on the floor in the kitchen, even though I can easily come up with a mechanistic explanation for that.”

        But getting pissed off at the dog is in no way incompatible with a denial of free will- Sapolsky’s brain was determined to react that way. And getting pissed off is potentially a deterministic deterrent against future bad doggy behavior.

        1. On the one hand, getting pissed off is a waste of energy. Spock, for example would simply raise and eyebrow and attend to the floor. On the other hand we are not very much in control of our emotions. Getting pissed off (hurling a curse for example) might be a good way to relieve tension. When someone is training a dog it’s important to avoid showing emotion. You’d be better of carefully suppressing angry outbursts and simply perform the needed behavioral modification procedures.

          1. But we’re not Spocks, we’re humans. We are deterministically emotional. We can be caused to be less emotional and more rational … by being reprogrammed, by what we call learning, by self-programming, because we are caused to be self-programming systems. So, being pissed is something most humans are caused to do, but Buddhist monks exhibit it less.

            And to open up the heads of a human and Spock, we’d see causal inputs – they’d just be a bit different, and the caused outputs would too.

            1. It depends on which particular aspect we’re interested in. Are we only obtaining as objective an account as possible, as close to an omniscient view as possible? Then by all means point out the causes that result in humans getting angry. Are we trying to impose our desires on the system? Then once we know that humans get angry, we either try to change things so that, say, we reduce the incidence of anger if we don’t want it, or even increase its incidence if we do want it. And if we’re talking about the rationality of getting angry, then let us use reason to measure and study the phenomenon.

              Perhaps we should be more like Spock, in any of those cases.

      2. That does not follow, logically. Retributive justice is punishment for the kind of person one is, rather than just the act committed. That can be subject to carrot and stick as well. The fear of retributive justice can logically be a deterrent that can affect for example how one socializes loved children.

        I am not endorsing the retribution, I am pointing out the argument has a hole.

        1. That, by definition, becomes behavior modification.

          Some types of justice have no behavior modification potential, such as execution or life in prison.

          But even more moderate forms of punishment can be more than required for optimal behavior modification.

        2. Does retributive justice actually work?

          In the US we see capital punishment and absurdly high incarceration rates, but also high crime-rates.

          I think there is sufficient reason for doubt.

          1. So do I. And the fact that it doesn’t work should be sufficient reason to reject it on consequentialist grounds, without requiring a commitment to incompatibilism.

            1. Unfortunately, most people will reply that they deserve to be punished because they did these things out of their freewill.

              It doesn’t matter to them that it doesn’t work.

              1. Perhaps, but if such people are immune to rational persuasion, why should we expect that arguments from determinism will be any more successful than arguments from consequences?

        3. I disagree with your definition of “retribution” and also with Martin X’s response. Instead I propose: retribution is inherently backward-looking, not forward looking. A science fiction book I am currently listening to provides a classic example.

          In the story, to simplify a bit, Earth is threatened with invasion by far more advanced, ruthless aliens. But humans invent a doomsday weapon that can destroy both civilizations. The inventor holds “the button.” The aliens retreat. For a long time, deterrence holds. But suddenly, the aliens attack. The earth is doomed, no matter what the button holder does. In fact, if she pushes the button, Earth dies sooner.

          In that situation, do you push the button? If so: that’s retribution.

          Is retribution inherently stupid? I don’t think so.

          1. The scenario collapses as soon as the threat is ignored, because not only is deterrence in this instance ineffective, but there’s no possibility for future interactions in which a reputation for carrying out threats could reinforce the deterrance effect. You’d need reasons outside of retribution to justify the suicide-murder, such as preventing a species of serial murderers from launching further attacks on others in the future. Otherwise, it amounts to little more than pointless spite.

          2. The thing is, if your potential adversary can read your disposition pretty well – a not uncommon scenario – then deterrence depends on that “pointless spiteful” commitment. In the sci-fi story, that was how the earth got to live for 60-odd years (the aliens attacked when the old button-holder retired and a new kinder, gentler soul took over).

            Related: timeless decision theory. A decision maker who can commit to a policy, and view the overall effects of the policy, may be better off than one who always responds to each new situation on its own terms.

            1. The very concept of “society” depends entirely on there being some semblance of equality and interdependence between the members.

              No need to think of aliens. What of farm animals here on Earth? Or even human slaves? How ’bout a slave revolt that threatened to permanently poison the local water supply?

              When the balance of power gets as extreme as your alien scenario, the basic ground rules of morality that humans have evolved to deal with simply don’t apply. It might make for an interesting space opera, but it has no bearing whatsoever on modern society.




              1. I agree that in alien sci-fi scenarios like that, our established human morality doesn’t apply. But that doesn’t mean we can’t learn lessons about decision theory from considering extreme scenarios. And then we can infer that a simple act-consequentialism is not the best idea. If you always and only look forward, you don’t see the whole picture.

              2. It’s actually almost always a bad idea to extrapolate from the outliers. Take them into consideration, yes, but to sacrifice the median for the sake of the fluke?

                Indeed, that’s exactly what the War on Terror is an exercise in. Terrorism in the States isn’t even remotely as much as a blip. Transportation deaths and property damage barely registered an increase in 2001 despite the significant anomaly on a certain day in September of that year — and yet we’ve placed ourselves securely on a path to ruin in response to that blip.

                Or, consider gender. Yes, there are people who’re outside the standard male / female pattern in various ways, and it’s important for everybody, especially medical professionals, to be aware of those outliers. But the overwhelming majority of humans are overwhelmingly male or overwhelmingly female, and pretending otherwise is a recipe for disaster. You could do well to design a physician’s new patient information form with an appropriate way to indicate “other”, but ordering both pap smears and prostate exams for everybody regardless of gender would be insane, to say the least.




              3. That’s a great point. Or as the legal scholars like to say: hard cases make bad law. And the penal law is a prime subject on the table here. I would say that penalties should be set so as to bring about happy and fair results as much as possible, given the actual historical and sociological evidence of how people usually behave. The comprehensibility of the law is an important factor, militating toward simplicity (which also helps with fairness). Good laws will look too harsh in some cases and too lenient in others. A good law is worth following, especially by agents of the justice system, even when its verdicts would otherwise (without that larger context) be morally wrong.

              4. Punishment is typically excused as being effective for prevention, under the guise of “deterrence.” But it’s overwhelmingly been demonstrated that it serves no deterrence whatsoever; indeed, it often leads to increased rates of crime and recidivism. One particularly poignant example was when various jurisdictions made punishments for rape as harsh as for murder…and more rapists started killing their victims, because the punishment would be the same but at least there wouldn’t be a witness to testify.

                Civilized people don’t brutalize people. Punishment in any form is brutality.

                Rather than a penal system, we need to quarantine (in a dignified if simple manner) those who present an imminent danger to society and to rehabilitate those who commit crimes — and, even more importantly, to provide comprehensive education, job training, access to health care (especially mental health), and the like, to minimize the fundamental factors that drive crime in the first place.

                Structure your society in a way that you yourself would be satisfied living at any stratum, even the lowest. The cost is trivial to those in the higher strata, and the payoff to everybody, including those in the higher strata, is mind-blowing.

                All those kids in gangs running drugs? Imagine if they had top-notch schools, secure housing, good healthcare, and the rest. Still think they’d be running around shooting guns when they’re not shooting drugs? Ha! They’d be engineers, artists, doctors, craftsmen…and just think of how much wealthier you’d be with that many more people contributing to the general welfare.




    2. Erm, do you really think this has never occurred to one of the world’s leading neurobiologists who just spent a decade writing a book about the subject? Perhaps before too getting carried away with telling us in all-caps how he’s got it all wrong, you might take a moment to consider that he’s actually talking about retributive justice, even if that’s not spelled out explicitly in a snippet from an interview?

      1. Point taken. I will lay off the caps. My largely deterministic brain made me do it. It was unclear to me from the interview that he was talking specifically about retributive justice.

        I have much respect for Sapolsky- indeed, in college I was a student in one of his classes, which is probably why I reacted so strongly 😉

  4. I loved A Primate’s Memoir. I think I read another of his but APM is the one I remember.

    1. I’ve read several, and have to say that pretty much every popular science book he’s written that I’ve read has been good.

  5. If it helps…this perspective of a perfect lack of agency becomes increasingly obvious the more attention you pay to your thoughts — and mindfulness meditation is a great way to learn the skill of introspection.

    You really don’t have any control over anything — your actions, your thoughts, anything. Life really is very much a cinematic experience, and your consciousness is simply along for the ride, an observer only.

    That famous study that determines that we only become aware of decisions a few milliseconds after we’ve already made them? That can become very viscerally apparent…at which point it comes slamming home that we are rationalizing agents, not rational actors.

    …but there’s a consolation prize. The better you get at observation of your inner mental landscape…first, you tend to experience less angst at the unfolding of events; second, you tend to become more skilled at whatever it is you turn your attention to.



    1. Your comments are comforting and compelling – but how can I simply decide to do mindfulness meditation? Either it is determined that I will do it or it is determined that I will not.

      1. Compelling, maybe. Comforting, definitely not, at least for me. Hell, I _like_ to feel good about something good I’ve done on those rare occasions when that happens. Ben, when you play something especially well on your trumpet, don’t you feel good about it? And don’t you enjoy that feeling? How can you defend such satisfaction when even that was not of your choice? Don’t tell me you have no other choice…;)

        1. Ben, when you play something especially well on your trumpet, don’t you feel good about it? And don’t you enjoy that feeling?

          Fuck, yeah!

          How can you defend such satisfaction when even that was not of your choice?

          Because choice is an illusion, to begin with. And, seeing how that illusion functions leads to other insights and understandings.

          Buddhism is a mixed bag, with plenty of bullshit…but its modern Western descendants have mostly let go of the bullshit and mostly kept the good stuff. And one of the good things is the insight that there is no permanence, period, and so depending on anything for your happiness is folly. You may well eat the perfect piece of cake for dessert and experience incomparable ecstasy as a result…but will you still be happy after you’ve finished the last bite?

          The answer is not to depend on anything for happiness, but simply to be fully in the moment. Whatever is happening to you now, pleasant or horrific, is deserving of your fullest attention — including your own reaction to the experience.

          So, yes, I felt great last Tuesday when I nailed Pictures at an Exhibition. Loved every minute of it. And that joy came, stayed with me an while, and left. A different kind of joy comes to me now as I remember it — but that joy, too, will not linger long. I also felt great misery when my Dad died this past November, and that, too, came and went.

          Being at peace with the fact that nothing lasts…well, it brings a great deal of peace.

          A certain Roman “got it” a couple millennia ago: “Carpe diem.”

          Gotta run, need to practice….




          1. My father (bless him) used to say people were such fools to think they could or should be happy all the time. He felt there was a centering tendency so that moments of joy were balanced by moments of boredom or unhappiness. I’ve always thought that sounded a lot like Buddhism. Anyway, it sounds about right to me.

      2. It’s as much of a perspective shift as anything else. The unfolding of history happens regardless of your perspective on it. You’d start meditating the same way you’d start any other activity, like taking up an hobby or exercising or going on a trip or whatever.

        But after you’ve sat down and had a close look at your thoughts — which is what mindfulness trains you how to do in an effective manner — you will eventually probably realize that, whatever you currently think of as your self is an illusion — somewhat like the dots that aren’t there at the intersections of the squares in that one illusion. By the time you figure that out…on the one hand, you’ll understand that the conversation about “Free Will” has as much bearing on reality as one about angels dancing on pinheads…but, on the other, you’re also rather likely to be more effective at decision-making as well as less dissatisfied with whatever life throws at you. And, however much assholiness you might have in you today — we’ve all got some, some more or less than others — you’re likely to be less of an asshole as time goes on.

        It’s a rather good tradeoff, if you ask me….


        b& >

    2. Having been a practicing pseudo-Buddhist for a few decades, I can only concur. Once one discovers that the self is an illusion, the whole free will thing becomes bit of a non-issue, or obvious that it must be so.

      I do think that there are some mental health issues here too. For some people, finding that they have no self could really pull the floor boards out from under them; just as no free will might for some hold the threat of being helpless in the face of dangerous mental states.

      It does mean shifting some mental furniture about, at least for some people. (I’m one!) Clearly dealing with reality is preferable to holding onto illusions as way of coping, but it is a skill, and takes deliberate conscious effort (of a kind).

    3. This has been discussed repeatedly, but still: what you describe here? That is body-soul dualism. One could just as well invent an essence of pocket calculatorness and claim it is only along for the ride while the body of the pocket calculator, which somehow isn’t what matters about the pocket calculator, does all the actual calculation. Exact same logic.

      1. Except that our experience isn’t one of pocket calculatorness, so that doesn’t require explanation, while our experience of dualism does.

        1. I have never had an experience of dualism and cannot imagine what such an experience would be like. Some kind of floating above your body looking down on it, as in some near-death experiences that people describe?

            1. But that’s not dualism. I have more agency than a rock because I have internal thought processes that the rock does not have. It does not mean that I am a little homunculus riding along in a body while helplessly watching it do stuff (dualism). It means that I am the body doing that stuff, having that agency.

              1. Gotta run, but…the cheat-sheet answer that makes little sense without the accompanying text is that what we typically identify as our selves is our thoughts, especially our self-reflective thoughts about ourselves. But our thoughts rarely last more than a few seconds at a time, a few minutes at most, before some other thought forms and takes shape. And thoughts are commentary that the mind mechanistically creates as a sort of voice-over of what’s going on. Decision-making and all forms of control happen elsewhere in the brain and are only indirectly accessible to the conscious mind. The conscious mind is just a cluster of symbols, really…

                …but, within the virtual reality of the mind, we’re gods with absolute power. Imagine a pink elephant floating in front of you, and it’s there — you can see it.

                Just got the call; gotta head out the door. Sorry to cut this short, mid-thought. But that might give you some sort of an hint of where it leads…and check out “10% happier” (iPhone app I use, book by Dan Harris I’ve not read) if you want a good woo-free introduction to mindfulness.




              2. Ben Goren,

                I do not think that people generally, or even you, identify as their/your thoughts. If I were to punch you, you would most likely say “ouch, why did you punch me?” I’d be very surprised if you said, “ouch, why did you punch the body that I am a helpless and passive captive of?”

                Either way it is irrelevant though, because what counts is that we are our bodies. The consciousness is merely an emergent process of such a body, it is not the human itself, just like an essence of pocket calculatorness is not the true pocket calculator.

              3. Alex, your example is perfect to illustrate my case.

                You do think to yourself, and perhaps say out loud, “Why did you punch me?”

                And that is a thought!

                Think for a moment about some major surgery, such as cardiac bypass. The body is experiencing physical trauma that makes a punch to the face inconsequential in comparison. But the mind is blissfully unaware. (To be sure, that’s because the anesthesiologist has shut down the physical workings of the brain that create the mind.)

                And also think for a moment about a particularly engrossing example of storytelling — a book or a movie, or maybe even something like hypnosis or an haunted house. Your body is perfectly safe, but your mind could be in anguish for the duration. (Again, it’s physiology at the bottom, without question.)

                Or consider a punch to the face in various contexts, from the sucker punch you’re obviously implying to an un-forceful “Hey, bro, how’s it hangin?” across the chin in the locker room to a boxing match to Hollywood stuntman acting out a scene. The constant is the fist making contact with the face; the variable is the thoughts, the mind’s interpretation of the action, the symbology associated with the physicality.

                Now, consider also how powerful this self-idenfication is of thought and self that you offered up what you thought was a knock-down counterexample…of self-identification with a thought.




          1. This, so very much. Dualism is all over Western culture – and sometimes leads me to a wrong interpretation of something before I see my mistake – but I have never found it in my experience either.

          2. Here’s a trivial example.

            Imagine a pink elephant, about six inches tall, floating at eye level midway between you and the computer monitor.

            You can do that, right?

            If you have trouble…dredge up a memory from somewhere. Maybe something especially memorable, like your first kiss; maybe something mundane, like eating breakfast this morning. You’re able to re-experience that memory, at least fuzzily, right?

            Where is the elephant? Where are you when you’re remembering these things?

            You’re having an experience of a floating miniature pink elephant / first kiss / breakfast, yet clearly there’s nothing objective in your immediate surroundings that matches that experience.

            But here’s the thing: your direct experience right now, what you’re seeing with your eyes and so on…you’re experiencing in fundamentally the same way as you’re experiencing the elephant: in your mind.

            The advanced example: you’re thinking something to yourself right now, probably something either like, “that’s, nuts,” or, “hmmm…he might have a point.” Can you hear that thought in your head as a voice? Try reading this very sentence to yourself silently and see if you can hear it the same way you see the elephant, and then compare that voice with all your thoughts.

            Who’s speaking those thoughts in your head?

            (The punchline, of course, is that thinking is a function of the brain just like breathing is a function of the lungs — but this can help you see how you think you’re your thoughts and you are, but in a completely different way from how you probably think so.)




    4. Funny, without meditation I came to the same conclusion(s).

      I’m certainly not mindful and don’t take introspection seriously.

      I think my nihilism has more or less the same effect. You get rid of a lot of unnecessary baggage; enjoying life without the illusions.

    5. This argument from meditation strikes me as a bit like saying “If you can learn to let go of the steering wheel, you’ll discover that there’s nobody in control of your car.”

      As such it is indeed analogous to fMRI finger-twitch experiments in which subjects are specifically instructed to act impulsively, without forethought.

      But it remains a mystery why anyone should regard either case as an accurate model of focused, linear reasoning or deliberative decision-making.

      1. Gregory,

        That has been precisely my suspicion of the type of inferences many seem to draw – Sam Harris included – from meditation.

      2. No, that’s not a good analogy for your case — but it is for mine.

        Try it the next time you’re actually driving your car. How much of your time driving are you actually thinking about moving the steering wheel? And, in a safe situation, such as an empty parking lot, try to micro-manage the flexing of the muscles in your body to turn the wheel.

        Performance driving instructors will tell you to focus your attention first on where the car should go, and very closely behind that on a visceral awareness of the shifting of the weight of the car from one wheel to another — driving by the seat of the pants, literally.

        Especially considering how slow conscious thinking and analysis is, not a fraction of driving can be even remotely plausibly explained by you thinking about turning the steering wheel — and yet, the body turns the wheel in micrometer-precise millisecond-timed ways.

        If you try to analyze your physical actions during driving, you’ll quickly learn just how much of a distraction it is, how potentially dangerous it can be.

        But, on the other hand, if you simply actively observe what you’re experiencing as you’re driving — pay attention to what you’re seeing, what you’re hearing, and, especially, what you’re thinking, all without trying to actively control any of it at anything other than the most general, highest level, you’ll drive better (however you personally define, “better”) than otherwise.

        This is the phenomena that’s often referred to as “flow,” by the way. And it has an especially cinematic quality to it. Try it now…type something out and just watch your fingers moving on the keyboard without worrying about what they’re doing. It should almost seem like you’re watching a movie with the actor leaning back slightly to make room for the camera where the eyes would normally be positioned, and feel just as weird.




      3. Ben,

        In Daniel Kahneman’s terms, you’ve just said that System 1 does almost all the work, while System 2 only makes a few very high level decisions. That’s not exactly news, and its bearing on free will is hard to see, at best.

        Rather than saying the “self” (meaning the mental “self”) is an illusion, I think Dennett is closer to the truth in his philosophy-of-mind work. The self is a construction. If you stop constructing it, it won’t exist.

        1. Thing is, we can very easily see how “we” are not in control of the steering wheel the way one would have to be for “Free Will” to make sense. And if your decision to drive to work has the control of the steering wheel as an essential sub-element, “Free Will” is pretty much gone from that as well. But then, when you closely examine your decision to drive to work…it turns out to be no different from your “decision” to rotate the steering wheel at a rate of so many degrees per second at such-and-such a moment in time. It, too, is made up of all sorts of overlapping and intersecting cognitive functions beyond conscious control, and the consciousness only retroactively makes up a very incomplete story to explain it all…a story, to be sure, in which the self was the Decider In Chief.

          As for the sense self…well, your sense of self is the same today as always, right? Even the same as when you were a very young child? But your actual self today is radically different from your child self, even if it feels the same.

          That means that either you’ve got a dualistic Christian-style soul that’s a constant…or it’s the sense that’s constant even as the self itself changes. And if you think of the sense of self as the sensation that arises from introspection, the mirror-in-a-mirror-in-a-mirror reflective feeling, that quickly and obviously is seen as the actual answer.

          From color science, we know that your perception of “red” is the same as everybody else’s, within certain physiological variations, and is extremely closely related to the actual physiology of the eye — with some other footnotes that also have well-known physiological influences. It’s no stretch at all to suggest that your self subjectively feels exactly the same to you as to everybody else — just as your self subjectively feels the same today as when you were a toddler, and as the color red is the same today for you as then.




          1. Kahneman is studiously noncommittal about the implementation details of Systems 1 and 2. But I tend to think – and you seem to be saying here – that System 2 is entirely built out of System 1 components.

            Yes. And? I just don’t see how that’s relevant.

            Everyone constructs their own self, but those constructs could, like you say, be very similar. Similarity judgments also depend on who’s doing the comparison, and for what purpose.

            1. I don’t know enough about Kahneman’s Systems to directly address your characterization. What I do know is that I’ve yet to observe a single instance of my own consciousness acting in a causal command role. Rather, it always provides an after-the-fact voice-over running commentary analysis of stuff that’s clearly inaccessible to direct conscious observation. And my observations are consistent with similar reports from others who’ve similarly carefully analyzed their own internal mental landscape.

              It’s a very consistent pattern.

              Joseph Goldstein likes to quote one of his teachers saying, essentially, “If you wish to understand your mind, sit down and observe it.” Those who do so come to very similar conclusions. Those who do not come to these conclusions invariably have never performed introspective contemplation at all or have only dabbled slightly and quickly abandoned the practice.

              As others have noted, this sort of observation does not have a particularly great reputation amongst the scientific community. However, all objective studies in cognitive neuroscience paint a consistent and / or identical picture of brain function — many disparate and largely independent functions that communicate and interact, with consciousness only one of the many bits and the last to react.

              That the subjective experience is a perfect match for the objective observation really should give reason for substantial confidence in this matter.




              1. “I’ve yet to observe a single instance of my own consciousness acting in a causal command role.”

                Is it your contention then that this sentence describing your conscious experience of introspection was generated entirely by unconscious processes, and that your conscious experience played no causal role in shaping its content? If so that would be a remarkable coincidence. How are these unconscious processes supposed to know what consciousness feels like in order to talk about it coherently?

                And how is this notion of consciousness as purely passive observer meaningfully different from the homunculus in the Cartesian Theater? If that’s what introspection is telling you, you shouldn’t trust it.

                Finally, how did consciousness evolve if it has no causal consequences on which natural selection could act?

              2. Consciousness is the output of an one-way function. It is created by the brain, but nothing in consciousness feeds back into the brain — any more than the outputs of a computer (typically) feed back into the inputs.

                Dennett’s derision is a bit too quick in this case. Cartesian Dualism can be ruled out as an explanatory theory just as surely as Aristotelian Metaphysics, no question. But that does not mean that either is a poor model of human experiences. Stop pushing your cup and it’ll stop moving across the table, just as Aristotle says it will. And the perception of consciousness really is very cinematic, just like the illustration on the Wikipedia page for Cartesian theater.

                That doesn’t mean that there really is an homunculus in the brain, and it especially doesn’t mean that the homunculus is in command of anything — especially since observation reveals that it’s purely reactionary. But the perception of consciousness really is rather well represented conceptually as an homunculus.

                This again goes back to the fundamental point that the self is an illusion. The homunculus could be said to be the heart of the illusion.

                Go back to that illustration on Wikipedia. Now, try to see your own head without resorting to a mirror. Look for your head, and you will not find it — though you might manage to see a bit of nose and eyebrow and maybe lips and tongue. Is not your visual perception a perfect match for watching the projection of a pair of video cameras positioned where your eyes are? Which is exactly what’s going on, in fact….




              3. Why would the brain bother to calculate an output function that never serves as input to any downstream process?

                And how, on your theory, do you account for our ability to talk coherently about the experience of consciousness, if that experience doesn’t somehow feed back into our speech centers?

              4. I can’t answer “why” questions. Why is the water boiling? Is it because it’s at that place on the phase transition graph, or because of the heat input from the large resistor, or because you’re thirsty for a cup of tea?

                And brain circuitry doesn’t have to be wired through consciousness for the types of feedback you’re wondering about. Much of cognition is linguistic and auditory — as if there’s a lecturer droning on inside your head telling you about everything that’s going on. You certainly don’t need to be conscious of a loud noise for the rest of your body to react to it, so why should you need to be conscious of the auditory hallucinations of thought for your brain to process those signals?

                …if I may be so bold, it seems you’re the one reifying consciousness into something dualistic, with this line of challenge. If consciousness is a product of the brain — as it emphatically must be — then why should you expect it to be able to alter that which creates it?

                There’re recursive feedback loops in the brain, yes — and the sensation of recognition of self is a good example. But, even then, the consciousness merely observes….

                Pushing back a bit further…for consciousness to have this special causal power that “Free Will” in its various forms requires, there must be something in the brain that initiates this causality. But there isn’t anything like that; it simply processes inputs from the senses and translates that into various reflexive actions.

                Hell, this goes on all the fucking time in everybody’s life. I just caught myself drumming an hemiola for no reason whatsoever. How on Earth does any theory of causality account for that sort of thing? Where did that action come from? Clearly not my consciousness. But how was that action any different from when a drummer does the same thing consciously in a practice room? The answer when you sit down and have a look at your mind becomes obvious: consciousness is purely observational, and really is nothing more than the voice-over script the mind creates to give meaning to our lives.

                …now, care to explain why I was just unconsciously scratching my thumb? Or why I didn’t realize I was holding down the back arrow to type, “unconsciously,” until after I had typed the question mark and the cursor was halfway through the word, “scratching”?




              5. “If consciousness is a product of the brain — as it emphatically must be — then why should you expect it to be able to alter that which creates it?”

                I expect that precisely because I don’t view consciousness as something distinct from brain activity. I view it as integral to brain activity, fully embedded in the causal web. Just as a computer program alters the state of the hardware it runs on, mental events — including consciousness — alter the state of the brain.

                I’m under no burden to explain every quirk of your behavior. But you do have a burden to defend your claim that consciousness plays no role whatever in explaining any behavior.

                And that’s really all I have time for today.

              6. Luckily, I don’t need to consult introspection to check on the causal effectiveness of consciousness. Plain old cognitive psychological research, such as a recently replicated finding on Motor Priming, show this. “The finding shows how information that’s not consciously perceived can affect our behaviour, and that it can have an opposite effect when subliminal than when consciously perceived.” Conscious perception reverses the effect of this type of priming. Of course, you can always come up with a gratuitously complicated hypothesis that conjures additional unconscious processes to do all the work – but that’s not what the researchers are saying.

              7. Complicated? That different inputs (presentations of the data) should produce different outputs? And that different inputs are perceived differently? That should be the least surprising result.

                But I don’t see anything in the study that established that there was a causal feedback loop explicable solely by conscious willful decision-making — only that there are correlations between ways information is presented and the resulting decisions as well as the way the subjects experienced the experiment.

                It’s also telling that you would dismiss direct observation of your own cognition. If you lived in a cave and somebody told you that the Sun actually rises in the West, not the North, would you refrain from stepping outside the cave to observe the Sun’s motion for yourself, content with limiting yourself to the shadows that move across the cave’s floor?




              8. Ben, earlier you said “nothing in consciousness feeds back into the brain.” But psychologists varied whether a priming was conscious, and got different results. That’s feedback for ya.

                If you want to dial down your claim to “nothing in conscious *decision*,” there’s plenty of psychological research there too. You could start with Libet’s “free won’t”. He asked subjects to veto an urge to act. He measured the readiness potentials that indicated the urge, but then the measured potentials flattened out (compared to previous experiments) and the subjects didn’t act. If you could establish that conscious intention never makes a difference, you could probably join Kahneman on the Nobel prize winners list.

              9. Paul, none of those sorts of psychological investigations are even tangentially relevant to making the distinction between our two positions.

                Consider a computer and its monitor as an analogy. You’re pointing to analyses of the interactions of the elements on the display as evidence that the display is an integral part of the computer’s computation. The arrow moves around the screen and menus pop down and icons follow the arrow and so on and so forth.

                I’m pointing out that the display itself is an output device, merely reflecting the internal state of the computer. The computer knows damned little about what’s shown on the display and cares even less. Even in highly contrived situations where it’s important, it has access to the video framebuffer — no need for a camera pointed at the display. At the same time, of course, the display is the primary, if not only, way that the user has any clue what’s going on in the computer.

                It should be obvious why analyzing what you can see on the display isn’t going to tell you which model is the better fit — that the display is where the action is at, or that the display is an output peripheral. To answer the question, you need a different kind of analysis.

                As with all analogies, this is far from perfect; it’ll easily break down if you push it just so.




    6. Ben,

      I agree with Gregory that the inferences some, including yourself, drawn from meditation about our identity, consciousness and reasoning can be dubious.

      “That can become very viscerally apparent…at which point it comes slamming home that we are rationalizing agents, not rational actors.”

      As profound as that may sound, it doesn’t seem correct.

      If we ONLY rationalize ad hoc, then it would not make sense that we can often predict the behavior of people – before they act – by appeal to a model where they are rational actors.

      I know my wife doesn’t want to eat raw chicken, and a model where she reasons from this desire toward choosing actions that result in cooking chicken will predictably follow. If I did not treat her as a rational agent, I could not predict, or make sense of this behavior. And if I ask here why she took the steps she did in cooking the chicken, her rational will essentially match up with the one that predicted her as a rational actor. There are countless ways in which we use the rational actor model to predict the behavior of people. We do it all day long.

      Yes, there are *some* experiments that show people will rationalise ad hoc about their behavior, and that their ad hoc reasons will not seem the most accurate account of their behavior. But, as we’ve gone through before, it is extremely incautious to leap from such experiments to “Therefore ALL human reasoning is of this ad hoc character.”

      Because you can’t just do that and leave hanging all that is left needing explaining: the countless examples where the rational actor model predicts our behavior, and the post reasoning matches the predictions. If, for instance, my wife’s actions in cooking chicken did not derive from rational considerations arising from her desire to not eat raw chicken (or some similar desire), I would ask for the alternative explanation – and demand it explains and predicts her behavior as well or better. I’ve yet to see such a replacement.

      1. I assume the reply would be that while your wife’s brain is a rational actor, your actual wife (i.e. the soul-essence riding along and helplessly watching the brain decide to cook a chicken) is not.

        It would not be phrased like that, of course, because then it would be obviously ridiculous, but that appears to be the logic of “my brain has decided for me” and “we are the slaves of our chemistry”.

    7. I don’t buy it, if only because introspection is unreliable as a tool for the mind sciences. The argument is much more straightforward when you discuss the cause-and-effect nature of any and all psychological acts.

      The instant there’s at least one causal precedent for someone’s thoughts and actions, causally speaking there’s nothing to take credit for because the thoughts and actions are effects dependent on those impersonal causes that precede one’s own existence. There’s no way to cite oneself as a personal cause in the past and claim credit for that, because that just delays the problem rather than solves it. And if there weren’t any cause, the thoughts and actions would be the products of randomness, which again renders it impossible to claim credit sensibly. Exactly the same argument holds for any object in the universe – supercomputer, Ford Anglia, sunflower, sandstone, helium atom – no matter how complex its internal mechanisms are.

      That’s why I think the “but it feels like we have free will” canard is nonsense, too. You can’t feel like a state that cannot possibly exist. The closest one could get would be to an open-minded ignorance reified onto the world at large, which is like saying you see impossible shapes in a mist.

      1. If you take “it feels like I have free will” to mean “it feels like I can defy the laws of causality”, then yes, it’s obviously nonsense. But if we take it to mean
        “it feels like my intentions have causal consequences in the world”, then it seems like a straightforwardly correct statement to me.

        Yes, our intentions themselves have causal antecedents, but that doesn’t invalidate the claim that our intentions matter precisely because they’re part of the causal chain, and not an epiphenomenal sideshow.

  6. This topic is generally a bit above my mental pay grade but I wonder if the fact that we are concious of our thoughts (mostly) isn’t what complicates things.

    It just seems like Neurological Calvinism to me.

  7. It was cool to see him speak her in Colo Spgs. He was very reserved however, on talking about his atheism, given the reputation of our little town. He needn’t have worried, given the types of folks that showed up.

    I noticed a Jerry quote on the handbill he had printed up for the talk (which centered on schizotypal disorders in religious behaviors across cultures): “In religion, faith is a virtue. In science, faith is a vice.”

    I think he’s a fan, Jerry. 🙂

  8. So what is true for the epileptic is true for all of us all of the time? We are our brains and we had no role in the shaping of our biology or our neurology or our chemistry, and yet these are the forces that determine our behavior.

    Yes, it’s seizures all the way down.

  9. It might be worth noting that 2,500 years ago, in ancient Greece epilepsy was a sign you were closer to God. Hercules was supposed to have it, and that was a good thing. (This has not been shown in any of the modern Hercules movies from Steve Reeves, to Dwayne “The Rock” Johnson.)
    Babylonians and others, however, thought it demonic.

    When Hippocrates was the first to offer a natural explanation for epilepsy, his treatise on same was entitled “On the Sacred Disease” precisely because it was widely considered to come from God.
    Saith Hippocrates “”I am about to discuss the disease called ‘sacred.’ It is not, in my opinion, any more divine or more sacred than other diseases, but has a natural cause, and its supposed divine origin is due to men’s inexperience and to their wonder at its peculiar character.”


    As for free will, it not only seems intuitive due to issues of moral agency, but also due to instances where there seems to be a lack of moral agency in some cases more than others.

    The fellow who has made a full recovery from alcoholism and consequently no longer beats his wife is generally accounted to have a higher level of free will now then he had a few years ago.

    As I posted here many moons ago, that is why there is a perceived difference between the “commons” and the “nobles” version of “Tristan and Isolde”, with only the latter having T&I helplessly in the grip of an aphrodisiac.

    I suppose it partly boils down to a sense of how you define the self (and its boundaries). There is a sense in which the molecules in my brain ARE me in a way that a gun to my head is not. It is true that the self is profoundly interdependent with the world around it, and is not a self-subsistent entity, but it is still the case that some of our actions have a sense of agency behind them whereas others do not.
    Thus, if you reject that notion that quantum physics allows a “free will of the gaps”, free will remains a necessity of some forms of moral rhetoric.

  10. Started reading the book, mostly enjoy it so far, but I’m bothered by the time he spends and the credence he seems to give to thinly supported or questionable assertions in psychology such as subconscious or implicit racial bias, or really a lot of the stuff that has recently been found to be highly questionable.

  11. Saw his Stanford lectures on YouTube and probably saw every talk he gave, and was amazed, also loved his “blog” at Mind & Matter at WSJ. He’s one of the people I wish had a more active role.

    Check out his talk on religion vs shizophrenia and other mental illness.

  12. Thanks for making me (us) aware of Sapolsky’s book!

    Sounds like a good read.

    As an aside: Being a good neurologist doesn’t equate to being a good philosopher. One poorly examined (or unexamined) assumption and you can be off in the wrong direction.

  13. Sapolsky is amazing.

    Still not at all concerned why people would be upset why free will is an illusion. People need to grow up. And learn that what we don’t know about tomorrow is just as incredible as anything free will could give us.

  14. We are our brains and we had no role in the shaping of our biology or our neurology or our chemistry

    Hang on a second. A professor of neurology thinks that brain activity plays no role in regulating brain chemistry or shaping future brain development?

    I suspect something is getting lost in translation here.

    1. I should think it was obvious from the use of the pluperfect. Our biology, neurology, and chemistry were shaped by forces that had nothing to do with us. Most obviously, go back a year before you were even conceived and the crisscrossing dominoes that will lead to your birth and development were not in any way caused by you.

      After all, our stomachs had no role in shaping our digestive system. Both are the products of non-digestive forces. ;P

      1. I grant that the overall architecture of our brains (and digestive systems) is the product of evolutionary forces that predate our individual existence. But to imagine that this explains the detailed structure of any individual brain (or stomach) is a drastic oversimplification. Such structure is the product of an interplay between development and lived experience; what happens in an individual brain does indeed affect the future development of that brain (and I imagine the same is true for stomachs). Brain activity is obviously not the ultimate cause of brain structure, but to say that it plays no role whatever in shaping such structure is simply false.

        Now I’m sure Sapolsky knows this, but he’s allowed Illing to put nonsensical claims in his mouth, and then endorsed those claims as true. That’s careless of him.

  15. I am ashamed to say that I had never heard of Robert M. Sapolsky before he was mentioned here. With all the glowing reviews, I’ll rush right over to Powell’s in Portland, OR tomorrow to buy one or more of his books.

    As always, I find what Ben Goren has to say fascinating. I wish I could do as he suggests. I’ve tried to develop
    “mindfulness” at times in the past without success. Can’t seem to shut down perceptions of the outside world (I’m one of those persons who can’t screen out any conversations to focus on just one. Makes for very interesting verbal goulash.) I can’t seem to get my mental machinery to stop or slowdown either. If I were to wake up from sleep numerous times on any given night, I would awake from a dream and, I almost always remember them. Often, they resolve issues I had when awake.

    However, since I realize “I” am a symbiotic conglomerate of organisms that when we die, all will be reduced back to a molecular state that can possibly reform into something else, I don’t worry a lot about the notion of “free will” as it applies to us.

      1. No.

        Mindfulness is the state of being aware, including being aware of one’s thoughts. The opposite is the state of being lost in thought, which is the condition most people (including most meditators) spend most of their time in.

        If you think back on your most recent emotional outburst that you later regretted, you might recognize that you were lost in thought the whole time and mostly just verbalizing the outrage you felt. Since then, hopefully, you’ve replayed the situation and thought of things you wish you didn’t say, or said differently.

        What if you had developed the skill to recognize when you’re lost in thought and to think about and analyze those thoughts as they’re happening? Might you not have managed to take at least some of the edge off your outburst? You probably wouldn’t have handled the situation perfectly, but you likely would have handled it better.

        The sitting-on-the-cushion time that meditators practice is a development of that skill, just like time musicians spend in the practice room with the metronome and pitch pipe is preparation for the concert.

        What, exactly, in any of that is even remotely wooish?




    1. I’ve tried to develop “mindfulness” at times in the past without success. Can’t seem to shut down perceptions of the outside world (I’m one of those persons who can’t screen out any conversations to focus on just one. Makes for very interesting verbal goulash.) I can’t seem to get my mental machinery to stop or slowdown either.

      This will obviously come as a surprise to you…but your observations are exactly the point of mindfulness meditation.

      It’s not to stop thinking, silence the mind, shut out the world, or anything like that.

      It’s simply to observe.

      In other words, you were doing it right!




  16. I must say, that Jerry aside, almost all the scientists and philosophers whose views and accomplishments I most respect (on ALL subject matters – not just the free will topic)seem to turn out after all to be compatibilists. Many also seem to sport beards.

Leave a Reply