There’s no free will

June 8, 2011 • 4:26 pm

On Sam Harris’s website, he’s just put up a short but cogent piece on free will—or rather, its absence, “You do not choose what you choose.”  One excerpt:

For [many people], freedom of will is synonymous with the idea that, with respect to any specific thought or action, one could have thought or acted differently. But to say that I could have done otherwise is merely to think the thought, “I could have done otherwise” after doing whatever I, in fact, did. Rather than indicate my freedom, this thought is just an epitaph erected to moments past. What I will do next, and why, remains, at bottom, inscrutable to me. To declare my “freedom” is tantamount to saying, “I don’t know why I did it, but it’s the sort of thing I tend to do, and I don’t mind doing it.”

And this is why the last objection is just another way of not facing up to the problem. To say that “my brain” has decided to think or act in a particular way, whether consciously or not, and my freedom consists in this, is to ignore the very reason why people believe in free will in the first place: the feeling of conscious agency. People feel that they are the authors of their thoughts and actions, and this is the only reason why there seems to be a problem of free will worth talking about.

Indeed. I’ve done quite a bit of reading about free will in the past few months, and the literature is characterized by two strains: it’s tendentious, motivated by the fact that many philosophers find the idea of determinism distasteful. And it also tries to solve the problem by redefining “free will” so that it becomes far removed from what most people think it is (see my discussion of Dan Dennett’s Freedom Evolves).

If “free will” is to mean anything, it must mean what Sam insists it does: the notion that we could have done or thought something other than we did.  There is not the slightest evidence for that proposition, and there is plenty of evidence against it, including the palpable fact that thoughts and choices arise from a materialistic body and brain, demonstrations that physical and chemical manipulations of our brain can change our thoughts and actions, and recent experiments showing that our decisions are made in our brain well before we feel that we’ve made a choice.

How can we ever show that we could have “chosen” other than we did? We can’t. And if we can’t, then assertions about free will leave the realm of the empirical and enter the realm of philosophy. There is no free will.

287 thoughts on “There’s no free will

  1. Indeed, as with gods, I take this one step further: the very notion isn’t sufficiently well defined to even be able to address the question.

    Take the classical religious perspective of dualism. Your brain doesn’t have free will, but your soul does, and it drives your brain.

    So, is your soul deterministic or random?

    Either way, all you’ve done is push the mater farther up the chain.

    Whatever “free will” is supposed to be, it has to be a third alternative to determinism or randomness. And nobody’s yet been able to explain to me what this alternative is or how it’s supposed to be neither deterministic nor random.



    1. The brain is material and is determined, but what about the soul? Hah! you didn’t take that into account. Atheist’s are so easy to outsmart!

        1. Here, let me try:

          “‘Twas brillig, and the slithy Toves / Did gyre and gimble in the wabe: / All mimsy were the borogoves, / And the mome raths outgrabe.”

          Nope, sorry. I seem to have a problem with Poe, as well.



            1. Then this ebony bird beguiling my sad fancy into smiling,/
              By the grave and stern decorum of the countenance it wore,/
              ‘Though thy crest be shorn and shaven thou,’I said, ‘art sure no craven./
              Ghastly grim and ancient raven wandering from the nightly shore–/
              Tell me what thy lordly name is on the Night’s Plutonian lee!’/
              Quoth the Raven, ‘Never free.’

  2. It’s funny, if there’s no free will, then rapists and murderers are just doing what they had no choice in. We jail them for something they had no say in and we couldn’t have not jailed them because we had no say in it……..not that funny really.

    1. Question here is: Why do we jail rapists and murderers?

      The common wisdom seems to be ‘because they deserve it, as they could have chosen otherwise, because of their free will’.

      I think that we cannot justify punishment as an ethical practice by saying ‘because wrongdoers deserve punishment’.

      Not only do I find that to be a non-sequitur, I actually hold it to be immoral. Punishment is, by its nature, inflicting harm of some kind upon another human person. To do harm with a sense of righteousness is a dangerous error. That’s how vengeance works.

      Just punishment should be something more than just organized vengeance with highbrow decorum.

      The only valid defenses of punishment I’m aware of are to do with modifying behavior. Such that the best that can be said of punishment is that, grudgingly and with regret, it is the lesser of alternate evils.

      The jump from ‘free will’ to ‘therefore, punishment is justified’ just seems like a gigantic non-sequitur to me.

      So regardless of how we define it or whether or not we have it, I just don’t understand the urgency behind many of the arguments regarding free will.

      Either way, have it or not, I just don’t see that the consequences of either answer are as important as everyone makes them out to be.

      1. As noted below, deterrence and incapacitation make certain forms of punishment valid even in the absence of free will. I still hold (despite Tim Martin’s objections) that the idea of responsibility becomes an elusive one in the absence of free will.

        But here’s another angle: what are we to make of skills and talents? For instance, why should we revere Bach, or any of the musicians highlighted here by Jerry and say something like: “gosh, they really knew what they were doing”? In a sense, they didn’t. I’m not afraid that this robs the music itself of quality, but it does mean that it doesn’t make sense to give any special respect to the conglomeration of material processes that resulted in the music. To borrow Sam’s “forces of nature” analogy, a river might carve out something beautiful like the Grand Canyon, but we don’t say: “gosh, that Colorado River really knew what it was doing.”

        What’s the difference?

        1. Hmm… are you talking about our last conversation, or something previous? What exaactly am I objecting to..?

        2. I’ve discussed this before with other people: I just don’t understand some of the implications people read into Free Will. I’ve been accused a number of times of deceitfully mugging at Socratic ignorance to score cheap points, but I’m being very, very sincere: There is something in what you’re saying that I really, genuinely do not understand.

          Why should responsibility be connected to Free Will in the first place?

          What does Free Will have to do with it?

          I don’t understand how free will and responsibility are connected concepts in the first place.

          So I also don’t understand why the *absence* of Free Will should be a problem for responsibility.

          Surely all we need for responsibility is a sufficiently sophisticated decision-making agent? That agent remains responsible for its decisions independently of whether or not it has Free Will, right?

          I’m genuinely confused.

          It’s like there’s this great big fat non-sequitur in the sky and no-one else can see it… So I start wondering if I’m missing something, and I stare and I look and I hunt and I search, but I find nothing to fill in the gaping void between premise and conclusion.

          What am I missing?

          1. A better method of explaining this occured to me immediately after hitting ‘Post Comment’.

            Surely all we need to justify responsibility is will.

            Whether that will is free or not seems to be an entirely independent question from that of justifying responsibility.

          2. I’m intrigued at what it is your hunting. So I will offer the basic connection I see with a very basic scenario (there are more complex scenarios, to be certain). Free Will means you have at least two choices to make… and you are ‘free’ to choose which one.. by result of prior learning, you have a set of data against which to compare the choices. If you take the ‘irresponsible’ choice that serves the short run and the quick reward with a CHANCE that there will be no negative consequences, and those bad consequences come to fruition by direct result of that risky decision, then many will consider you to be “responsible” for the predicament you are in. If we say that the person did NOT have free will to make that decision, and such choice was predetermined, than it makes it more difficult to utilize the concept of “responsibility’ to measure the correction.

            So, I think I may have *just now* realized what you might be talking about. Regardless of the free will or not, corrective action still must be metered out to prevent it occurring again. More over, if we viewed our ‘crime and punishment’ model in that sense, and not in a vindictive sense, we might have a significantly advanced society by result. Am I close?

            but then again, maybe that is why it is a good thing to discuss if actions of criminals are free will or determined. the former causes more of a vindictive punitive system, whereas the latter fosters more of a sober minded corrective system. I think I am liking this a lot, even if it is not what you meant.

            1. If we say that the person did NOT have free will to make that decision, and such choice was predetermined, than it makes it more difficult to utilize the concept of “responsibility’ to measure the correction.

              This is the bit that makes me hunt for justification.

              You’ve stated that as if it was obvious and self-evidently true… And I’m sure that, to yourself, that’s how it appears.

              But it doesn’t appear as such to me.

              I suspect this is because we’re up against an axiom that you take for granted that I don’t – but given that I don’t know your mind, it’s hard to be sure.


              Thanks for bearing with me on this.

            2. Oh: To be clear, I think you’re getting close to the same preference that I get to – that we should have a sober corrective/deterrent system rather than a punitive one that deliberately punishes wrongdoers because they (allegedly) deserve it.

              But we could arrive at that same preference even if we assume free will exists.

              The bit that makes me scratch my head is the notion that ‘responsibility’ is somehow easier as a concept, or helped by, or in any way related to the notion of whether or not the will in question is free – because to me, the key point in establishing responsibility is the will part, not the metaphysically free part.

              Obviously the will needs to be free in the sense of ‘free from duress’ to establish responsibility – but that’s not the kind of free will we’re talking about (I think).

              1. “But we could arrive at that same preference [a sober corrective/deterrent system] even if we assume free will exists.”

                I think I am seeing the issue, or disconnect, you are having then. It seems as though you expect humankind to be sober minded when determining corrective action, and ‘responsibility’ should not bear in to that determination. I agree, but I think that is idealistic. I think it *can* be achieved at some point for humanity to reach that goal, but it is not practical. In other words, I think it would be beneficial to somehow push for society to think in terms of determinism and not ‘fault’ to correct individuals who commit crimes. Even the most average of thinkers in a position to make a decision will make a more rational corrective decision about something that has gone awry or has broken if the believe no sentient agency is involved (i.e., a machine part wears out constantly, the umbrella stand blows over whenever a gust of wind blows through, etc.) but when a person is involved in the thing that has gone awry, that same decision maker will likely be quick to throw blame and punishment on the person. This might explain the tendency for humans to go to great lengths to find a person to be at fault for something bad that happens to them, and when it is obvious no person can be blamed, they feel it is karma, God, the devil, or some other sentient being.

                Anyway, I digress… so, your disconnect may be that you think everyone should know (or at least the 98 percenters like most of us here) that responsibility does not give a pass to be barbaric, immature, vindictive, etc. when deciding the course of action to take to correct the person who is responsible for a bad thing. That regardless of the agency involved, a sober corrective/deterrent system is clearly superior for human kind.

                I agree, but because I believe most people do not think like that (even the 98 percenters), it’d be a good move to find some way to implement a determinism world view among our society.

            3. If there’s no free will, then just as the criminal has no choice to commit his crime, we have no choice but to punish him.

          3. I think you’re tilting at windmills here. I meant *precisely* what I wrote when I said the issue wrt responsibility is elusive. I don’t pretend to have it all figured out, either.

            But I will say that Sam’s “forces of nature” analogy, to bring it up again, and insofar as it is an apt analogy, demonstrates (in my mind) a problem w imputing responsibility. We would incarcerate hurricanes and earthquakes if we could (to incapacitate them and keep them from wreaking havoc in the future), but would we “hold them responsible” in the same way we hold a person responsible for his/her actions? Understand, I’m not arguing for Free Will. If anything, I’d have to admit that our idea of responsibility wrt the person is what needs re-examining.

            Frankly, I didn’t think the first paragraph in my original comment was the interesting one. I wish someone had weighed in on the other issue I raised. I’d be very intersted to hear what others think about that.

            1. I still hold (despite Tim Martin’s objections) that the idea of responsibility becomes an elusive one in the absence of free will.

              We can phrase that in another way:

              I still hold (despite Tim Martin’s objections) that the idea of responsibility becomes a non-elusive one in the presence free will.

              It’s that secondary phrasing that gets to the heart of what I don’t understand about most discussions around free will and responsibility.

              Which is, simply: What’s the connection?

              I can make a connection from ‘will’ to ‘responsibility’. That’s easy.

              But whether that will is metaphysically free or unfree seems irrelevant to the linkage between those two concepts.

              If there is a link that requires metaphysically free will (as opposed to just ‘will’ without a modifier), I can’t imagine what it looks like. My lack of imagination isn’t evidence that there isn’t such a link of course – which is why I keep coming back to this point and scratch my head whenever the topic comes up.

              Because nearly everyone just seems to accept that there is a meaningful link between metaphysically free will and responsibility… But no-one can ever explain it to me in a way that I can wrap my head around.

              I don’t to think I’m that dense… But there’s a consensus of highly-intelligent and highly-educated opinion against me on this one too… And that’s what the Dunning-Kruger effect looks like from the inside.

              So yeah… This is something I tend to fret over quite a bit whenever it comes up.

              1. I suppose we can both hope that after enough of this commentary, a critical mass will have been reached, we’ll have an epiphany, and everything will be crystal clear.  😉

                Here’s what I stumble over:

                1) I don’t find myself convinced by attempts to allay concern about responsibility that assert “responsibility is simply an independent issue.”  This is effectively what Jeff Engel asserts below: we can reduce our actions to material cause-and-effect which is not controlled by a “ghost in the machine,” yet Bach is still responsible, in a “free agent” sort of way that we wouldn’t ascribe to a river, for choosing how to arrange the notes on the page.  How is that possible?

                2) Aren’t we questioning the existence of will at all?  What is will, as opposed to free will?  Sam argues we aren’t the authors of our choices.  To my mind, while I understand (as many have pointed out) that this doesn’t undo the fact that there are always several possible ways of acting at any moment, it does make the idea of choice an impotent one.  How are we any different from any other cause-and-effect chain reaction, except in the number of, and complex relationships between (as Sigmund points out below), the inputs?  Aren’t we essentially complicated hurricanes?  So how do you make sense of responsibility by saying “it follows from having will – just not free will”?

        3. Bach really did know what he was doing, and that’s why we say it. We respect and admire skill. That’s all there. It’s independent of some sort of non-causal yet non-random act of will.

          When it comes to admiring choices, it’s a matter of respecting and admiring a personality, habit, and/or deliberative process that settles on good answers.

          The Grand Canyon is mightily impressive too, but the Colorado River hasn’t got skill, personality, habits, or deliberation, so if we admire or respect it in any sense, it’s not for those.

        4. @JS1685

          Bumped us back off to the right, because the available space to comment in was getting too narrow.

          I suppose we can both hope that after enough of this commentary, a critical mass will have been reached, we’ll have an epiphany, and everything will be crystal clear. 😉



          2) Aren’t we questioning the existence of will at all? What is will, as opposed to free will?


          And there I think you’ve identified a false assumption of mine. I thought that it was clear that ‘free will’ was not synonymous with ‘will’ – that there is this thing, called will, and that the argument is about whether that will was free or unfree.

          It is within that framework that I interpret discussions about ‘free will’.

          And now that you’ve pointed it out – that isn’t nearly so obviously apparent as I originally thought it was.

          Hmm… I might have some unexamined assumptions there.

          But anyway, I think I can express the difference using the chess-board analogy I have used elsewhere.

          At any given moment of time, the chess-board will be in a particular state (for my purposes, state includes ‘which player’s turn is it?’).

          The state of the board will entail a set of possible legal moves.

          From that set of available moves, the player (a decision-making agent) must select one.

          If that set of available moves has more than two elements, then the player has made a choice.

          In this context, that is sufficient to describe will. The following statements are synonymous in this context:

          1) The Player selected B2->B3
          2) The Player chose B2->B3
          3) The Player willed for B2->B3 to happen
          4) The Player was responsible for B2->B3

          The player’s selection may have been metaphysically free or unfree in the sense of ‘the selection of the available options was predetermined by the earlier state of the universe’.

          But in either world it still remains true that the player selected one move from the set of legal moves at the time.

          Whether or not the player ‘could have’ selected a different move in the metaphysical sense is not relevant to statements 1-4 above.

          Perhaps there’s some problems here – we could quibble over whether or not 1-4 are really synonymous.

          But for my own personal usage? They are synonyms with one another.

          1) I don’t find myself convinced by attempts to allay concern about responsibility that assert “responsibility is simply an independent issue… yet Bach is still responsible, in a “free agent” sort of way that we wouldn’t ascribe to a river, for choosing how to arrange the notes on the page. How is that possible?”

          A river is not a decision-making agent.

          One algorithm that could be used by a decision-making agent is to evaluate the available set of ‘legal moves’, simulate future outcomes resulting from those moves, then select the move that leads to the most preferred outcome.

          Now, we can argue that the decision-making agent might be metaphysically unfree. Let’s assume for the sake of argument that it is unfree.

          Even so, the fact remains that this decision-making agent does perform simulation of possible future states.

          And it remains true that the decision that is (perhaps) pre-determinedly reached will be a direct consequence of that simulation and the heuristic applied to it.

          The quality of the outcome will therefore be determined by the quality of the simulation and the quality of the heuristic (which will themselves be determined by the manipulation of symbolic representations – either bits on a CPU or a particular pattern of neural activity in the brain).

          So in this example, the ‘agent’ (simulation + heuristic) is ‘responsible for’ (the predetermined cause of) the action in question.

          In the case of Bach, one can assume that his mental processes were somewhat more advanced than a simple zero-sum game AI. But the difference should be regarded as quantitative, not qualitative.

          So Bach ’caused’ the notes to appear on the page the same way that a river ’caused’ a floating leaf to flow downstream.

          The difference is that Bach is a decision-making agent that selected from alternate possibile simulated options. That the selection he arrived at ultimately boils down to causation does not mean that the simulation didn’t happen, and it doesn’t mean that he didn’t evaluate that simulation according to some form of heuristic.

          That particular kind of ‘causation’ can be re-labelled as ‘responsibility’ to distinguish it from other kinds of causation…

          And in so doing we don’t actually lose or omit anything of practical value… Or at the very least, it seems that way to me.

          If there is something we’re missing here, I’m ignorant as to what it is.

          Hope I’m making sense, here. Thanks for bearing with me. I’m enjoying this. ^_^

          1. Jumping into this discussion…

            yet Bach is still responsible, in a “free agent” sort of way that we wouldn’t ascribe to a river, for choosing how to arrange the notes on the page. How is that possible?

            I think this is one of those issues that the human brain is, first of all, not made to handle adroitly. For example, most of us start out life as essentialists, believing, say, that “life” is some kind of *spark* that living things have and non-living things don’t. We later learn that it’s just chemistry, but it takes a lot of understanding to get this far.

            Similarly, our brains are programmed to divide the world into causal agents and non-agents, and to hold agents “responsible” for things, and so on. It’s really hard to acknowledge that it’s all determined and really be at peace with that. I think we must acknowledge that Bach and the river are one and the same. Or, if there *is* some difference, we need to very carefully define what it is and why it matters.

            I also think, and I think someone mentioned this already if not addresed it, is that we don’t really know what “responsible” means in this discussion until we define it. Just like the concept of metaphysical free will seems to slip away and become a non-concept the more you think about it, so does the idea (for me) of “holding someone responsible.” It sort of begs the question, doesn’t it. Doesn’t “responsible” imply “morally responsible”? If so, we’re right back where we started. If not, did we mean “causally responsible”? That’s difficult as well. The buck never stops with causality – you could go all the way back to the big bang.

            So what are we asking for, exactly? We know we’re all robots. Are we just trying to decide how to run a society, given that fact? Can we just… do whatever leads to the best outcome? Punish or rehabilitate criminals if that’s what leads to a happy and healthy society? Am I being overly glib, or is this a good idea?

          2. Apologies for the delay. I hate when real life interrupts my online activities. 😉

            Well, I wasn’t convinced by the chess analogy, because such a situation (a state with a set of possible “moves”, one of which “gets selected” [I might substitute “happens”]) is true for everything. But we don’t say the river, or the Plinko chip, “selected” or “was responsible” for its course.

            But I found myself more convinced by what followed. It does seem that the distinction between Bach and the river is that Bach can (could) simulate a number of possible moves and evaluate them according to some heuristic – all before the move is made. And the move “chosen” would be based on that evaluation. A Plinko chip doesn’t do that.

            So on the one hand, I think the above gets us at least a good portion of the way toward articulating the difference between a “decision-making agent” and things that don’t make decisions. I have a very hard time not seeing a difference.

            On the other hand, I also have a hard time denying that, as Tim points out above, if you go far enough down the rabbit hole, everything is the result of material, un-minded cause-and-effect. And as Tim also points out, responsibility seems to evaporate under this consideration. Which makes me wonder if the “decision-maker/non-decision-maker” distinction is only a superficial, apparent distinction, an illusion that emerges from the incomprehensibly complicated nexus of inputs attendant to “decision-makers”. I think this is what Sam Harris argues when he claims will is an illusion.

            Basically, I don’t think I’m any closer to making sense of all this. 🙁

            1. I think you’re quite correct that “decision-making” is an emergent property that arises from the capacity of the brain to simulate parts of reality.

              However, I don’t think that emergent properties are “illusions”. To make an analogy, at the level of quarks there are no solids, liquids, or gasses, but that doesn’t make ice any less solid, even though it is composed of quarks and electrons.

            2. On the other hand, I also have a hard time denying that, as Tim points out above, if you go far enough down the rabbit hole, everything is the result of material, un-minded cause-and-effect. And as Tim also points out, responsibility seems to evaporate under this consideration. Which makes me wonder if the “decision-maker/non-decision-maker” distinction is only a superficial, apparent distinction, an illusion that emerges from the incomprehensibly complicated nexus of inputs attendant to “decision-makers”.


              I don’t think I actually have much more to add to this.

              The thing I keep coming back to in all of this is the ‘responsibility seems to evaporate’ concept…

              I just fail to see it. Completely and utterly.

              As I’ve said elsewhere: I seem to be very alone in this, where those who disagree with me seem to include the majority of intelligent and educated people. That’s what Dunning-Krueger looks like from the inside. So my inability to see a connection does give me pause for self-doubt.

              But no matter which way I cut it, I can’t see how the free in free will has anything to do with responsibility – which to me is just a particularly fancy kind of causation that is linked to the calculations of any future-state simulator/heuristic engine.

              Which is to say:

              If I assume that free will exists, I am unable to provide an account of ‘responsibility’ that loses something if we remove the starting assumption.

              Which is to say: I cannot think of a framework for ‘responsibility’ that doesn’t treat the free in free will as an unneccesary assumption. To misquote Laplace: I fail to observe a need for that hypothesis.

              Interesting how it always seems to boil down to axioms and definitions in the end, isn’t it?

      2. “To do harm with a sense of righteousness is a dangerous error.”

        Just because you think vengeance is never justified doesn’t mean it’s an “error.”

        1. Well, I *think* it is in error. But then again I’m no expert, so of course take everything with a grain of salt.

          Usually when I write a post like this I pepper every statement with ‘In my opinion’ and ‘it is my view that’ and all manner of hedging.

          But it makes the comment unread-able and weak-sounding, so I usually go back and remove them all before sending.

          Anyway: I think it is in error to consider ‘harm’ to be ‘good’ because I think that the upper bound on justifying harm should be that it is, regretfully, the least evil of all the alternative courses of action.

          With that framework in mind, harming another with a sense of righteousness is indeed an error – but then again, we have to accept the framework first before I can declare that objectively, and I have not justified that framework (and I don’t plan to try that here, as I don’t have the time).

          So I suppose I could have phrased things differently.

          1. Daniel, I don’t think you should be so apologetic sounding (you didn’t apologize, I realize, but I think you should stand more firm on this idea). I think it can be clearly established that vengeance is not in any way superior to sober corrective action. It is almost always the result of a personal affront. Nonetheless, I think one would be hard pressed to justify vengeance without appealing to irrational emotional justifications.

            I am a police officer, and know the vengeance route very very well. It is typically expected of fellow officers to hate child molesters, WANT to kill a deadly criminal, and/or thoroughly disrespect the wife beater, to name a few. You quickly understand that you should not appear “soft” in that arena lest you be labeled ‘unsafe’, as if you will be less likely to pull the trigger when the trigger MUST be pulled lest someone else dies. It’s a thoroughly bullshit belief (in fact, we ‘softies’ may be MORE safe, but that is another topic), but it’s there nonetheless. If you were to watch me take a dangerous criminal into custody, you might think I am a very vengeful cop (the angry look, the forceful yelling, perhaps some profanity) but once the dangerous criminal is in custody (the threat made inert), I have no animosities. While other officers may be making some pejorative comments to the offender, I’m making sure the air conditioning in the cruiser is not too cold or not too hot. I’m no doing this in a ‘loving’ way, but certainly in a respectful way. I have learned that it is far worse to be delivering pejorative commentary to a prisoner, regardless of their crime, after they are taken into custody than to simply be impartial. Time and time again, while riding with a prisoner in the prisoner elevator to process them into the jail, I get some form of “Thank you for being polite” from that prisoner. It most certainly is far superior to remove vengeance from our values and replace it with the opposite of vengeance…. no vengeance. That does not mean a coddling love.. it simply means no vengeance.

              1. “Excellent behavior on your part. Hope that it sets an example from which your peers can learn new behavior”

                Thank you. I credit my choices on my conversion to atheism, which caused me to search for REAL solutions to problems. My conversion to atheism allowed me to seek out ethical atheists who I wanted to emulate, who were far more ethical than I (less selfish) and thus I have since striven to behave in a more ethical, compassionate, reasonable, mature, etc manner. So, given the discussion of determinism/free will, I suppose my behavior comes from data points of past experiences that make it make so much sense to be ethical that I don’t have a choice to deliberately be unethical. 🙂

            1. @ Joe Hern:

              That was…inspiring. And thanks also for counteracting some of my cynicism when it comes to cops…

              1. “That was…inspiring. And thanks also for counteracting some of my cynicism when it comes to cops…”

                🙂 Thank you as well. I do feel humbled by the gratitude, because I know it is how I SHOULD behave and feel it should not be rewarded. Whereas, the cynicism is warranted, sadly. But hopefully, as much as you will be disappointed, every time you hear of some officer doing something bad, give her or him the benefit of the doubt until you talk to another officer like myself. I often see friends on Facebook, etc. post videos of ‘wayward cops’ when the officer did nothing wrong, or when they did, the comments are thoroughly erroneous as to what they are claiming the officer did wrong rather than what they did wrong. Nonetheless, it’s really the culture of police work that is to blame for the bad image we have as the powers that be seem to do little in the way of outreach. I think it’s a fear of retribution against police officers by the criminals. I think that’s possible, but not probable, especially if we behave like we should when arresting them. So it goes, though.

              2. Joe, would you say any of it has to do with society not being willing to reimburse police officers in proportion to how much we need them? Cops, teachers, nurses–every where I look, they’re getting laid off!

            2. “I think it can be clearly established that vengeance is not in any way superior to sober corrective action.”

              It’s not a question of whether one is “superior” to the other. I think the legitimate purposes of criminal penalties include both what you call “corrective action” (e.g., rehabilitation, restitution) and retribution (“vengeance”). And I think most people feel that way too. One of the reasons I think people support harsh penalties for heinous crimes is that they believe, as I do, that people who do very bad things deserve to suffer for their wrongdoing, irrespective of any “corrective” effect the penalty may also have.

      3. @Joe Hern

        Replying out here to the side because our room available is getting very narrow.

        You’re almost there, but still not quite. You’re putting my conclusion ahead of my premise, and going backwards.

        It’s exasperating communicating this kind of thing, ain’t it? ^_^

        Let’s try another tack.

        My perception is that ‘most people’ adopt a line of reasoning that goes something like this:

        1) People have free will (barring exceptions for mentally ill, etc).
        2) If people have free will, then punishment of people who do wrong is justified.
        3) Punishment is justified.

        And again: My perception of most people who argue in favor of Free Will would point to this argument and say something like this:

        “We must have conclusion 3), it is important! And if 1) was false, punishment would be unjustified as the argument would collapse! So it is very important that 1) is true!”

        To which I would say: “Look, forget about 1). Maybe it’s true, maybe it’s false. Hard to tell. But either way, 2) is still one hell of a problem in its own right. Worry about 2) first! Then we can get around to 1).”

        So maybe 1) is true. And maybe it is false. I don’t have a horse in that race.

        However, regardless of whether or not we have free will, 2) is still a pretty big problem for the argument in my view.

        I contest 2) as a conclusion in need of an argument… And I cannot imagine what an argument for 2) would even look like.

        Now, my own lack of imagination is not evidence that 2) is false… But 2) remains a contentious claim until a valid argument founded on uncontentious premises is put in place that has 2) as its conclusion.

        Which of course throws the whole argument 1-3 into jeopardy.

        Which happens to be fine as far as I’m concerned, because I believe there are other justifications for punishment that have nothing whatsoever to do with Free Will.

        And sure, I think that the systems we would put in place from those arguments would probably be preferable to the systems that we might put in place based on 1-3 even if 1-3 turns out to be perfectly sound.

        But all the same… I still come back to 2) and scratch my head. I just don’t see how that could be considered true… Or at the very least, I don’t see how it could be considered ‘true enough’ to confidently justify punishment.

        Hope I’m making sense… And feel free to pass me by if I’m pestering you.

        It’s weird how hard it is to communicate this kind of fundamental confusion regarding abstract concepts.

        1. Okay, okay.

          From context, what I should have said is: Why is jailing rapists and murderers ethically justified?

          1. Jailing rapists and murderers is ethically justified because –
            1) It makes it harder for them to keep raping and murdering, assuming anything but the worst penal system.
            2) It may make rape and murder less appealing to other potential rapists and murderers – note this does not assume contracausal free will, just a social learning response and/or a good reason for their decision-making processes.
            3) It puts them in a place where they may be treated so as to be much less likely to keep raping and murdering once they’re out of jail, assuming any but the worst rehabilitation regime.
            4) I suppose it puts them in a place in which the harm they’ve done to society may be balances by some involuntary good done for it – the restitution model. But I have a hard time seeing this as key in the rape and murder cases.
            5) It helps keep other people from murdering, raping, etc., them out of vengeance or fear, assuming anything but the worst prison system.

            I do not claim that this list is exhaustive.

            1. Which is all fine.

              The point I was making in the comment was that punishment ‘because they deserve it’ isn’t an ethical justification – but that seems to be the one people care about most when they get wound up about defending free will.

          2. From context, what I should have said is: Why is jailing rapists and murderers ethically justified?

            Because preventing harm to several people (potential rape & murder victims) outweighs restricting one person?

      4. One can favor the existence of prisons while objecting to punishment. If a man cannot control his tendencey to assault, rape, rob or kill people, he needs to be separated from potential victims. Prisons should not be made unnecessarily unpleasant (unless you’re convinced the unpleasantness will deter crime – I see little evidence for that), but they need to have bars on the windows and doors.

    2. There are five main purposes of punishment:

      1. Incapacitation
      2. Deterrence
      3. Restitution
      4. Retribution
      5. Rehabilitation

      Even if we accept that free will is an illusion, there would still be good reason to imprison dangerous people (incapacitation). The other justifications for punishment might not remain valid (I’m thinking specifically of retribution).

        1. So, it’s the old ‘gun-to-head’ scenario. Free will talk always allows for situations where you’re not free to choose, like a gun being held to the head or watching someone rot in jail….

            1. How are you responsible if you had no say in it? I’m not talking about compatibilist responsibility, Jerry ruled that out in his post.

              1. But you did have a say in it. Who had a say in it if not for you?

                Lack of free will doesn’t mean you do things for no reason or that you don’t make choices. It means that everything you do is for a reason, including the choices you make.

              2. It does mean you have no choice. That was the point of Jerry’s post. That there is no reality to the ‘I could have chosen otherwise’ thought. You couldn’t have.

              3. I think it’s not so much that the brain could have not have done otherwise (which as a physical object, it had to “obey” the laws of physics) — as that there is no “I” that is separate or independent from the brain. It’s dualism that gets us mixed up on this subject.

              4. I think the idea is – certainly it is for me – that you still have choices when what you choose is strictly-enough* caused. The crucial thing is that it’s caused by parts of your own makeup – Sam Harris’s “the sort of thing you do, and the sort of thing you like to do”.

                * “Strictly-enough” instead of “strictly” only as a nod to the (practically) irrelevant possibility of quantum effects.

              5. Think of it like training a dog. Without training, the dog will jump on someone every time it meets them.

                With the proper training, however, the dog will stop doing this.

            1. In the context, I was discussing ethical justifications provided for punishment and their relation to the concept of free will.

              If popularity is the sole measure of ethics, then you have a point.

              But I don’t think that this is particularly common view of ethical reasoning, hmm?

              1. @Michael Kingsford Gray

                No, not from my point of view.
                Just because something ‘is’ a certain way, does not mean that describing it is any way supporting it.

                Sorted then: Evolution does not make Greg’s comment relevant given the context of ethically justifying acts of punishment.

      1. I don’t know why you think the philsophical issue of the nature of free will is relevant to the social/political justifications for punishment at all. The justification for retribution is that it serves our sense of justice. People who do bad things deserve to suffer in some way for their wrongdoing.

        1. “I don’t know why you think the philsophical issue of the nature of free will is relevant to the social/political justifications for punishment at all. The justification for retribution is that it serves our sense of justice. People who do bad things deserve to suffer in some way for their wrongdoing”

          I am stuck on your use of the word “deserves”. It *seems* to me that ‘deserving’ something means you chose to do something to ‘deserve’ it. I.e., free will. Or do you mean ‘deserves’ it in the sense that ‘society needs to see that person obtain the reward/punishment for having been ‘responsible’ for the “thing”. Then I think we might need a different word than ‘deserved’.

          1. In this context (no free will), “deserve” does not seem the proper word. Nevertheless, I’d venture it’s how the majority of us reflexively think…Just recently, in a post here about DNA evidence, it was mentioned that OJ was “where he deserves to be.”

            AAMOF, “deserve” as in “just deserts” seems to be part of our whole formal justice system. (As does “punishment.”)

        2. The adequacy of “serving our sense of justice” as a justification is at issue here. If the decision on whether and how to punish depends on the harms it will, will not, may, or may not cause, then “because we enjoy doing that to people in a way that makes us feel right” is a pretty sad proposed justification.

          1. “The adequacy of “serving our sense of justice” as a justification is at issue here. If the decision on whether and how to punish depends on the harms it will, will not, may, or may not cause, then “because we enjoy doing that to people in a way that makes us feel right” is a pretty sad proposed justification.”

            I don’t think so. I think people who do very bad things — the people who tortured and murdered those teenage boys in Syria, for example — deserve to suffer for their wrongdoing. And I think the vast majority of ordinary people would agree with me. Retribution has long been recognized as one of the purposes of the criminal justice system.

              1. Then we have no choice in punishing them. It doesn’t make any sense to argue that we should change our criminal justice policies on the grounds that free will is an illusion and people who commit crimes therefore don’t really have any choice about their actions because, by the same token, neither do we.

            1. That the vast majority of ordinary people hold a position does not justify it. Neither does a long-standing recognition. If they did, God would preside over a flat earth, slavery would be cool, and women would be wrong to disobey whichever man held authority over her.

              I do know where that mass of people over a very long time have been coming from. We feel natural rage against those who’ve done terrible things, and it’s not a desire to correct or deter the behavior – it’s a desire to make them hurt back.

              But we’re capable of moral reasoning and concluding that harm’s basically a bad thing to be minimized. If we’re going with that principle, then the harm done to the wrongdoer has to serve the purpose of minimizing harm, not jacking it up for its own sake or to make us have the warm glow of satisfied vengeance.

              1. But again, most people don’t seem to accept the moral principle you are proposing. They believe that it is just to make people who do very bad things suffer in some way for their wrongdoing. In fact, it is your view that I think most people find unjust and morally offensive.

        3. The punishment based judicial system has been in place in the US for centuries, but there’s still little to no evidence that it’s actually useful in protecting citizens, given the number of repeat offenders.

    3. Well, we lock up those who are mentally unfit to interact with society already. Human society has an interest in protecting its members from harm. One of the ways we accomplish this is by penalizing those who harm others, or by preventing future harm via imprisonment.

      The mass murderer that hears voices and experiences other hallucinations that prompt his murders may not be responsible for his actions, but we still need to lock him up. Not because he deserves punishment, but because it’s too risky not to imprison him.

      1. “Well, we lock up those who are mentally unfit to interact with society already.”

        Actually we don’t. We turn them loose to sleep in cardboard boxes and beg for handouts on streetcorners.

        We lock up those who are morally unfit (along with a bunch of recreational drug users).

        1. We imprison the mentally unfit on a regular basis, if their illness contributes to actions deemed morally inappropriate. It’s a nice thought to believe that there are no mentally ill people in prison because “we don’t do that,” but we do.

    4. I’d see people punished for deterrence, incapacitation or petty vengeance. I have no great use for fanciful excuses like “Justice”.

  3. The fight against this kind of understanding of free will is just going to continue from the religious – after all, it’s the only argument they’ve got for why their supposedly-loving god is allowed to punish us eternally for doing exactly what we were built as a result of his crap design skills.

  4. I like the way Harris constructs this argument, especially because I’ve personally plowed no ground in this area of “free will”. It is quite interesting. I will be studying everyone’s commentary, in order to synthesize some of my own philosophy.

    Any talk of “Free Will” brings to mind the story of Friedrich Nietzsche and how he arrived home from college at Easter, only to announce that he no longer believed in God, and certainly he would NOT be attending any Easter services. His sister was anguished, his mother, whose patrilineal ancestors were church ministers, unbroken back to Martin Luther himself, was nearly apoplectic. Finally, though, his mother became serene and contented. She rationalized that everything in the world was controlled and actioned by God, so if Fritz had foresaken God, it was all OK. It was all according to God’s own plan, and so why should she question that? The whole situation was in His Hands, so why need she worry?

    1. Great story. And really, only logical given God-controls-all dogma.

      Surprised the child abusing priests haven’t tried that excuse.

  5. No free will makes having good information that much more important; we may choose much more wisely if we know the truth about the options.

      1. Here’s a serious challenge:

        Is there any distinction that can be made, even in principle, between the illusion of choice and actual choice?

        That is, if you think you’re making a choice, does it even make sense to assert that you’re not actually making a choice?



        1. Hmm… Interesting…

          “Makes sense” is a bit tricky in this context.

          Could I rephrase your second question?

          That is, if you think you’re making a choice, does it even have any consequences, one way or the other, to assert that you’re not actually making a choice?

          I can answer the consequences thing, as I have above: I don’t think that the answer to the existence of the free will is particularly consequential.

          But that doesn’t mean that asking the question itself doesn’t ‘make sense’ in a grappling-with-an-inscrutable-universe-for-understanding-and-truth kind of way.

        2. Hmm… How about this: Given a choice of chocolate ice cream, or vanilla ice cream, which would you choose? At the higher level of conscious (the voice in your head that is telling you that chocolate is a nice choice) there is the illusion of choice. At the lower level, there is the memory of what ice creams you have had in the past, and how the stimulus of different chemical compounds has affected your brain. These chemical reactions have encoded into your brain a preference for chocolate, so by choosing chocolate, you are just reacting to the deterministic world. The illusion of choice exists because a) you know there is a choice of options, but b) you don’t consciously know the chemical reactions and nerve impulses that drive your memories of what you like, which are the deterministic drivers of what you choose.

          I am not sure that answers your challenge, but I do like ice cream.

          And to get meta on it: I don’t know if I have the free will to not take a Ben Challenge 😉

        3. That is, if you think you’re making a choice, does it even make sense to assert that you’re not actually making a choice?

          Yes, when you’re caught. And actually, that’s the basis of some criminal defenses…’Your Honor, in light of my client’s impoverished, abusive youth…’

          Which is probably an accurate defense at times.

        4. “That is, if you think you’re making a choice, does it even make sense to assert that you’re not actually making a choice?”

          Sure, if there can be articulated a benefit to humanity to understand determinism is reality.

            1. Now, now, the difference between counter-causal free will and compatibilist free will is much like the difference between Intelligent Design and evolutionary theory.

              One is magic. One is not.

              1. Of the four things you list, one has a mountain of evidence to support it; three do not.

              2. Oh? Do you mean that compatibilist free will is also unsupported? I thought it was fairly well established that humans perform actions and have psychologies.

      2. Come on. You can program a computer to make a choice. If you put two slices of raw meat in front of a starving ocelot, it will choose one to eat first. Out of a set of possible outcomes, one of them happened: it doesn’t matter if it was selected by a simple deterministic algorithm, a chaotic system, a truly random process or invisible elvish creatures. You still can call it a choice.

              1. Hmm…

                I don’t know.

                If we had a planetary mass of sand grains, I think it would be a reasonable statement to say that the ‘sand acts gravitationally upon other nearby objects’.

                Depends on how we’re describing ‘act’.

                When playing chess, for every possible board-state (including which player’s turn it is), there is a list of possible legal moves that might be made.

                The player in question will select one move out of the sub-set of legal moves of which they are aware.

                So long as the sub-set of legal moves of which the player is aware contains more than one option, then this is enough of a definition of ‘choice’ to match up against common usage and still remain descriptively useful in either a metaphysically free or a metaphysically unfree universe.

  6. The redefinition of free will is not so much to rescue the concept, but to counter the absurdist fatalist notions that normally come with the absence of free will. If we aren’t free, do we have moral responsibility? Yes, not in the ultimate sense but in the sense that we have the capacity to reason morally and act on that reasoning. The acceptance of responsibility changes how people behave, so in that sense how is it wrong to redefine free will into something more in line with human behaviour?

    1. Exactly. The word “choice” does have a practical, well-understood meaning in ordinary usage, so we may as well accept that meaning at something like face value, rather than muddling it up with scare-quotes and qualifiers.

      1. And I have long thought that the only USEFUL definition of “free will” is that of a relative lack of external, practical restrictions or coercions.
        I dislike the more metaphysical concept. I don’t grok it.

  7. I have always suspected that our consciousness is considered one thing much like a flock of birds. The difference is that we are made up of flocks of cells that merely looks like one thing.

    1. Except that consciousness is a verb. An action.
      A result of an assemblage of cells.
      Much as “running” is an action: a result of an assemblage of cells.
      It is vital, when discussing “consciousness”, to understand that it is a reification of a verb.
      That lack of understanding the difference between actions and their consequent reification appears to me to be at the heart of the puzzlement of “what consciousness ‘is'”.

      It famously occurs with “numbers”, as well.
      Commonly, numbers are assumed to be “things”, instead of a reification of the activity of counting.

  8. I don’t see the relevance of Harris’s argument about the inscrutability of one’s own mental machinery. Deep Blue doesn’t need to understand semiconductor physics in order to play a good game of chess. It needn’t even understand the detailed workings of its own algorithm. Yet surely it would be absurd to deny that that algorithm is responsible for the moves it makes, physics notwithstanding. Whether the algoritm happens to be deterministic or stochastic is irrelevant to that point.

    Frankly there seems to be an implied dualism in Harris’s claim that we exert no control over our mental processes — as if “we” are somehow separate from those processes. On the other hand, if “we” are identical to those processes, then I see nothing wrong with saying that those processes control our decision-making in the same general sense that Deep Blue’s deliberations control its gameplay.

    As for the feeling of conscious agency, why shouldn’t we have such feelings? What’s wrong with the idea the mechanism of consciousness goes deep enough to encompass the decision-making processes, rather then being a mere surface-level readout of inherently inscrutable causes,a s Harris would have it?

  9. I don’t know. I’m completely an armchair philosopher, but I’m just not convinced. It seems to me that consciousness is not very well understood.

    Counterfactuals do have a real effect in quantum mechanics, so that provides a model by which things that didn’t happen could have a real measurable effect. Don’t misunderstand me, I’m not arguing that qm explains free will or something, I’m just saying that it isn’t outside the realm of possibility that things that didn’t happen could have a measurable effect.

    The inability to explain how something could be doesn’t mean that it doesn’t exist. Reminds me of the religious argument “Science can’t explain x, therefore God”. I guess I’m not seeing the rational yet for rejecting free will,in some form, a priori.

    Sam Harris is right, of course, the reason that it’s still an issue (putting aside the religious arguments) is that we feel that we have conscious agency. But without knowing what that is, it’s hard to make a judgement.

  10. I still haven’t managed to grasp the concept of no free will. If I come to a fork in the road, I can go left or right. Sure, my intendend destination along with all the history that went into deciding to go there as well as any other pre-existing data that supports or conflicts with my intentions will influence which direction I go. But I can choose to ignore any part of that and go in either direction. Using relevant information to decide which way to go is making a choice. Just because I can’t go back into time to prove I could have made a different choice doesn’t mean no choice was actually made. I’m not trying to be contentious, just trying to understand. I’m a athiest who disagrees with many others on this point.

    1. But I can choose to ignore any part of that and go in either direction.

      Yeah, and how do you do that?

      1. Yeah, and how do you do that?

        By using the same mechanisims you do when you choose to ask an unanswerable question. 🙂

    2. Ok, but the reason you would make that choice is based on the entirely deterministic biology of your brain state at that time, and that’s true even if we don’t entirely understand that state or how it resulted in you making the choice.

      It’s still a choice; no free will doesn’t mean that there’s not choices. It’s just that our choices are the result of an extremely complex but ultimately deterministic heuristic, not the capacity of “souls” to exercise “free will.”

      It honestly doesn’t change anything. Whether we have free will or not is seriously a completely irrelevant discussion. It has absolutely no consequence on anything.

      1. It is possible that I don’t hold a conventional definition of free will. To me “making a choice” is synonymous to “free will.”

        I don’t see why free will has to be defined in terms of being a function of the soul (which of course does not exist) just because the theists say so.

  11. Almost everyone is familiar with the poem entitled The Road Not Taken written by Robert Frost. It’s a testament to rugged American individualism. Most of us even have parts of it memorized…“And I- I took the road less traveled by and that has made all the difference”…. All the difference. It gives us license to be explore new possibilities. It’s inspirational and optimistic. It encourages us to embrace the unpopular, to lead instead of follow. Ultimately, the poem proposes the idea that our choices are critically important- that we must have the foresight to choose the right path.

    Except, it doesn’t do this. And, most of the other things I just said about the poem won’t stand up upon a critical re-reading. The poem doesn’t belong on a hallmark card.

    If we go right to that moment in the poem where the walker is remembering his past experience analyzing the two roads, we read the following:
    “Though as for that the passing there had really worn them about the same.”

    Prior to that, he’s looking back on that moment trying to work through the haze of memory to convince himself that one path was truly different from another, but then he realizes that he’s embellishing. The paths were the same, and he could not discern any difference looking down both for as far as the eye could see.

    In this interpretation, choice and chance are equal partners in determining our paths. One path leads to another which leads to another. As we branch down these paths, we are by definition, excluded from pursuing others. We later make sense of the direction of our lives by placing exaggerated meaning upon past choices and ignoring the possible interplay of chance. We try to control our future by navigating to the right forks in the road and pursuing the “correct” paths. And, yet, chance will inevitably throw us a curveball. We may even come to regret some choices.

    Free will? Choice? I LOVE this poem now that I understand it better… We’re not defined by our choices, we’re defined by how we EXPLAIN and RATIONALIZE our choices…

  12. Sam Harris wants to convince to you that you have no self, because that’s Buddhism.

    I don’t believe in free-will, but I do have a self, and it is the totality of my mind and body.

    I think Harris has a basic religious agenda which ultimately conflicts with the social and political goals of rational atheism.

    Perhaps people will wake up to that, perhaps not.

    1. Zen goes out of its way to be cryptic and misunderstood, so it’s entirely justified that you think this about Zen Buddhism (important to be specific there – there is no ‘One True Buddhism’).

      I think that you’ll find that Zen Buddhism doesn’t deny that there are bodies and minds.

      Zazen: Sit in a certain way (body), and breathe in a certain way (body), and count the breaths in your mind in the following way (mind).

      The existence of bodies and minds are part of the practices of Zen Buddhism.

      If by ‘self’ you only mean ‘the totality of this body and this mind’ then that’s fine.

      The sense of ‘self’ that Zen is referring to in the context of no-self is more in line with Daniel Dennett’s Cartesian Theatre.

      Harris seems (like myself) to draw significantly from the useful practices, experiences and fictional stories found within the Buddhist tradition whilst still resolutely rejecting the supernatural metaphysical claims that come with them.

      In a similar way, it is possible that Socrates as we know him did not exist outside of Plato’s imagination. But even if this is so, I find insight in the stories about Socrates – so I don’t particularly care one way or the other if he did not exist.

      I look at Buddhism the same way. Other religions too, excepting that I’ve found Buddhism to have the most useful bits left over once we remove all the supernatural scaffolding.

      Yet all the same… Given the absence of supernatural claims, I think that calling Harris’ agenda ‘religious’ is a bit of a stretch.

      Atheism makes it easier to loot religions for the good bits they happened to stumble over and leave the dross behind.

      But I don’t think that the practice of doing this is something that is or should be entailed by political atheism.

    2. “Self” is a synonym of at least three things. A certain, overviewing, more-or-less-conscious brain process that is commonly assumed located in the frontal cortex; the totality of brain processes and their tendencies; and the entirety of the organism, sometimes including extended phenomena pertaining to the organism.

  13. Consider the following scenario, one most of us have experienced at one time or another.

    You’re sitting in a comfortable chair, a good book at hand, and your loyal cat (or dog, I suppose) snoozes nearby. Kitty gets up, stretches, and walks into the kitchen. She looks at her food dish, but does not eat. She walks to her litterbox, but doesn’t use it. She walks back to you, curls up, and goes back to sleep.

    “Silly cat!,” you think. Of course as any rationalist might tell you, when an animal spends a prolonged period in one position, lactic acid builds up in the extremities, and a brief walk helps to break it down and redistribute it so the kidneys can process it more easily. And so you turn to your book.

    It’s a fascinating book, full of good ideas, and soon you are engrossed. You lose track of time, but then you come to the end of a chapter and close the book. You’re still thinking about the ideas in the book, when you find yourself standing in the kitchen. You look at the fridge. “No, I’m not hungry.” You look at the bathroom. “And I don’t have to go.”

    “Gah! I completely forgot what I got up for! I hate it when that happens!”

    My answer, of course, is that you didn’t “forget” anything. You went to the kitchen for no conscious reason at all. No consciousness, no free will, just physiology.

  14. It’s very encouraging to see Harris come out four square against contra-causal (libertarian) free will. Plus he draws out the progressive, humanistic implications of denying free will for our criminal justice system, both in his latest book and the first blog he posted on this. I hope others in the atheist/humanist/freethinker community will join Sam and Jerry in debunking free will,

    Of course the implications of this debunking extend to many other aspects of our lives – to all the beliefs, attitudes and social policies premised on the false idea that people could have done otherwise in actual situations. Taking in the fact that we don’t have free will helps to generate compassion: there but for circumstances go I. And understanding the causes of behavior helps us gain control over it. The myth of being uncaused causers blocks access to these moral and practical virtues.

    What’s not to like? Well, you can’t take credit the way you might want to, or assign blame. But taking the self off its pedestal is just what this culture needs.

    1. “And understanding the causes of behavior helps us gain control over it.”

      Harris’s argument (as I read it) is that we don’t and can’t have control over it. Control implies choice and the ability to make meaningful decisions.

      I think it’s obvious that we do have such an ability, and that we can reasonably call it free will without embracing woo or denying physical causality.

      1. Agreed – it’s a mistake to suppose we don’t have control just because we don’t have contra-causal free will. We make choices that control our behavior; it’s just that we’re not exempt from being caused in our character, motives, deliberations, and choices, maybe with some randomness thrown in. Human agents and their choices are just as real and causally effective as the factors that cause them, see and

        Whether we should call the voluntary control of action “free will” is an open question, seems to me. At the very least, we should make sure that we don’t mislead people into thinking they could have done otherwise in an actual situation. It’s that belief that deflects attention from the actual causes of behavior, disempowering us and abetting the morally corrosive assignment of ultimate credit and blame.

  15. Professor Coyne–

    I’m curious whether you agree that the conjunction “could have” represents not some well-definable concept but rather represents some sort of flawed, fuzzy feeling that comes packed with human psychology.

    Or, to put it more simply, the conjunction “could have” is fundamentally, rather than contingently, incoherent–you can’t even imagine a universe in which things “could have” been other than they ended up.

    1. No, but “could have if” is conditional and fine. “Could have if” looks similar, but does not strictly imply “could have”.

  16. …our decisions are made in our brain well before we feel that we’ve made a choice.

    That does seem to be the thorniest problem for defenders of contracausal free will. The issue causes me to think of the answer employed by Skip Wilson’s Geraldine Jones: “the devil made me do it.” I wonder if Anthony Weiner is mulling that one over. 🙂

  17. Sam is wrong. There is a big difference from free will being absent and free will not being what the majority of people think it is. Rejecting contra-causal free will does not reject free will entirely.

    I know you aren’t convinced by Dennett’s version of free will, but Dennett’s version of free will is the version accepted by most people who investigate what free will means. This means at the very least that the matter is one open to discussion, and the case is far from closed as Sam suggests.

    I think it helps to compare free will to consciousness. Few people would deny that we have consciousness even though what consciousness is is very different from what most people think it is.

  18. How can we ever show that we could have “chosen” other than we did? We can’t.

    How can we ever show that we could have got a tail in that concrete, particular flip coin instead of a head? We can’t. Therefore all coin flippings are determined beforehand.

    1. Well, yes. Coin flipping is an RNG, albeit a very peculiar one. Coin flipping is a mechanical operation.

  19. What if, every time I have a choice, I roll a dice or flip a coin? Then, I have the “illusion” of choosing that method of selection, but laws of probability and random outcomes still apply, right? Or, say, if the coin ends up tails, how can I know it was never just always going to be tails?

    Sorry, apart from being a rationalist, I have to turn to faith and just believe in free will – I have no choice. Why? Because not having a choice makes me feel bad about the poor, worse about the rich, and all my efforts to try to succeed in school pointless – it would have happened anyway? Or not? Could I have chosen to dive into a minimum wage job, alcohol and cirrhosis of the liver?

    Makes me want to kill myself. If I would actually go through with it, it was going to happen anyway, right? And if I don’t go through with it…hmmm.

    Reminds me of Camus, Le Mythe de Sisyphe. The first line is potent:

    “Il n’y a qu’un problème philosophique vraiment sérieux : c’est le suicide.”

    I’m glad I’m bilingual, I hate reading translations.

    1. You needn’t believe in free will to get out of bed in the morning and make things happen – you only need to believe that things you want to happen are more likely to happen when you act to make them happen. Cheer up! You can be a happy and active rationalist without that awful speedbump.

    2. Your consciousness is just along for the ride. However, that does not make life not worth living. Think of life as an incredibly relevant and interesting book. The story will play out as written, but that does not make the book any less enjoyable.

  20. Interesting topic, and many interesting posts. Thanks to mredrmann for pointing out Frost’s poem: I’ll have to read it. And the phrase “We’re not defined by our choices, we’re defined by how we EXPLAIN and RATIONALIZE our choices” is very thought-provoking.

    Free-will cannot, of course, exist, because its existence would require supernatural intervention: what can possibly determine an action except natural laws? Is there some sort of psychic force that tells our muscles to grab the vanilla ice-cream instead of the cocolate one? If so, where is it? Can it be measured?

    To Jason I would reply that our brain is not only its cortex. We are only aware of events that occur in the “thinking part”, and have no access to the rest. So it is very hard not to fall into the “conscious will” trap; we think WE made a choice (our conscious selves), whereas in fact the choice was made FOR us by the rest of the brain, which we tend not to consider a part of ourselves.


    “As for the feeling of conscious agency, why shouldn’t we have such feelings? What’s wrong with the idea the mechanism of consciousness goes deep enough to encompass the decision-making processes, rather then being a mere surface-level readout of inherently inscrutable causes,a s Harris would have it?”

    What’s wrong is this: if consciousness went deep enough to encompass the decision-making process it would interfere with and slow down automatic reactions, thus rendering us incapable of surviving attacks by predator, for example.

    1. The fact that some decisions are reflexive and automatic doesn’t mean they all have to be.

      1. If consciousness could determine SOME choices, it would then be able to determine ALL choices, because it would have access to the choice-making module in the brain. Unless we postulate the existence of separate choice-making modules, of course. But that seem unnecessarily wasteful from an evolutionaruy point of view. Why would evolution duplicate modules just for the benefit of a species which likes to think highly of itself?

        1. Some days when I drive to work, I’m actively engaged in the task and acutely aware of every lane change and turn signal. Other days I’m absorbed in some internal rumination and do the whole thing on autopilot (unless unusual traffic conditions grab my attention). Does that imply that evolution must have equipped me with duplicate driving modules, one conscious and one unconscious? Obviously not. All it means is that consciousness is flexible and can intervene or not in decision-making as the situation warrants.

          That said, if thinking highly of yourself turns out to confer a reproductive advantage, then natural selection will (naturally!) favor the construction of brain mechanisms for high self-thinking.

          1. Is it really consciousness sometimes intervening in decision-making, or conciousness just sometimes being aware of always-subconscious decision-making.

            There’s a spooky experiment with a slide carousel that Dennett describes in Consciouness Explained that strongly suggests the latter.


  21. There is no reason to attribute a concept such as “free will” to the human brain that has absolutely no parallel elsewhere in the world.
    Human behavior is shaped by factors such as motives, memories etc. The behavior of a weather system is shaped by factors such as humidity, presure and temperature. Neither has free will.
    That has nothing to do with responsibility. Resposibility is a concept that comes into play when you have a system of rewards and punishments. Such a system definitely affects human behavior-again having nothing to do with a magical idea like free will.
    I don’t agree with Harris on everything but here he is 100% right.

    1. Well said! I find it exceeding strange that people who have thought their way out of organized religion, belief in a god or gods and other supernatural beliefs, still cling to the magical notion of Free Will. The kind that has to be floating unattached out there in space like a Platonic Ideal in order to have responsibility or meaning in their lives. Free will, as most argue for it, is just another way of sneaking the Ghost in the Machine in the back door. Why do we need Ghosts to have meaningful lives? Why do we need ghosts to be responsible? Responsibility, like morality, is what naturally arises in long-lived social species with the attendant memory banks. Especially so in a language-enabled species.

      No need for a Free Will Fairy.

    2. I’m gravitating toward this position, as expounded by Insightful Ape. Gravitating…not of my own “free will” mind you…it’s gravity!

      When he says, “Responsibility” it unlocks the puzzle. Think of a toddler going to and fro. No sense of responsibility, because the toddler possesses no mental sophistications yet. Thunderstorms and random lightning, some hits the ground, most do not. And then the biblical/religious image of gods throwing thunderbolts…willfully.

  22. If free will is treated as a capacity or metaphysical bogey of one form or another, then obviously we don’t have it. But that’s a piss-poor way of thinking about the issue (yes, I know it’s how most people think about it—I think most people think about the issue badly). It’s a brute fact of everyday experience that we make choices. We perceive and entertain options, we consider them rationally (or act impulsively, or consider them stupidly, or whatever), and then we choose one or the other option. To say that such a choice was free is not to say that the choice wasn’t determined by the physical laws of the universe°, but simply to point out the absence of factors like mental illness, coercion (e.g. by holding a gun to the head), or any other of a number of other *morally relevant* considerations. To say that I could have done otherwise isn’t to say that my action wasn’t predetermined physically, it’s to say that I had other (reasonable) options and that I was in a position to entertain them. So, for instance, in the case of coercion, I don’t have reasonable options other than what I do, because if I had acted differently I would’ve been shot.

    I honestly don’t care whether we reserve the term ‘free will’ for something that implies neither physical determinism nor mostly determinism with partial randomness, or whether we think about free will more along the lines of free choice as I discussed it above. In fact, given the history of the idea, it might be better to simply get rid of the term altogether.

    The reason I don’t care is that I think, once you start thinking about the issue the way I’ve outlined it, the entire problem dissolves. The entire reason that determinism seems distasteful is that, among ethical philosophers, there is the (quite proper) idea that we can’t be held accountable for what we can’t have done differently. The problems arise when philosophers start analyzing this fundamentally moral claim physically. The morally relevant sense of “I could have done otherwise” just isn’t “physically, I could have done otherwise,” it’s “I had other reasonable options available.” Once you clear up that misunderstanding, there are no reasons whatsoever left why having free will would be desirable, since all the moral issues thought to rest on its existence manifestly do not.

    °I don’t claim to understand either the physics or neuroscience that comes into play here, but I will note that “saving” free choice by pointing out that determinism is strictly false because aspects of our universe are partially random (whether it’s the firing of neurons or something lower down the physical chain) is not to save anything at all, even theoretically. Making our choices partially random (from a physical perspective) ought to be more disconcerting than making them entirely determined (from the perspective in which either of these options is disconcerting at all—I’m arguing that neither should be disconcerting because the extent to which our choices are random or determined physically is irrelevant to what matters about choice).

    1. “So, for instance, in the case of coercion, I don’t have reasonable options other than what I do, because if I had acted differently I would’ve been shot.”

      But that’s precisely the point. You chose not to die because you are hard-wired to prevent death. In what sense is that a free choice? You might think that choosing vanilla over chocolate is different, but it is not: some neuron-firing pattern MADE you choose vanilla instead of chocolate.

      1. And the implicit duality in language strikes again!

        “Some neuron-firing pattern MADE you choose vanilla instead of chocolate”.

        That should be re-worded as:

        “Some neuron-fring pattern MADE you physically articulate the action (be it verbal or actually physically picking up) vanilla instead of chocolate.”

        Because in this context, the neuron-firing is the choice itself. It didn’t ‘make the choice’. It is the choice, ticking over mechanistically.

        Note: I’m pretty sure I’m agreeing with you here, so that’s not an attack – just an interesting observation at the way dualism creeps into language unintended.

        1. You are right, of course. And thank you for pointing it out. It’s funny how language betrays us…

      2. Some electron-firing pattern made Deep Blue move the pawn instead of the rook. Does that mean the controlling algorithm had no role in the process?

        1. No; the controlling algorithm caused the electron-firing pattern, but Deep Blue had no choice in the matter.

          Actually, Deep Blue is quite a good illustration of why free will cannot exist: let’s say I play Deep Blue twice, each time making the same moves. Them either Deep Blue will respond with the same moves or it won’t. If it does, it will be because it is predetermined to make them. If it doesn’t, it will be because the algorithm contains a procedure which makes use of a random-number generator. In either case, Deep Blue is not free.

          1. Actually, I would be very surprised if Deep Blue didn’t include some means of learning from experience – so there’s a third option for why Deep Blue might play the same game twice.

            Doesn’t change your point, of course. Just being complete. ^_^

            1. Urgh: “so there’s a third option for why Deep Blue might not play the same game twice.”

              Silly negations, they’re always escaping from my otherwise carefully-crafted sentences. >.<

          2. Sounds to me like you’re arguing for dualism. Deep Blue’s algorithm has a causal role in choosing the moves it will make, but Deep Blue itself (whatever that may mean) has no choice but to follow its algorithm.

            1. As noted above in my reply to pierro: Dualism is intrinsic to language, and possibly it is intrinsic to human psychology. (Why yes, I have been reading Stephen Pinker recently, how could you tell?)

              The problem is that explaining things using non-dualistic language is clumsy. It’s very, very arduous to write, and even worse to read.

              So the algorithm doesn’t ’cause’ the bits to flick. The algorithm is the bits in the act of being flicked. Or perhaps more accurately, the algorithm is the pattern and sequence formed as the bits flick from one state to the other. (In another sense, the algorithm could also be the act of uttering a symbolic representation of the pattern in which the bits will flick, but perhaps I’m getting too abstract there).

              But saying all that is very cumbersome, and doesn’t roll of the tongue (fingers?) nearly so easily as the shorthand: ‘The algorithm causes the bits to flick’.

              The second form is also much easier to read and understand.

              It is also the form of phrasing that will leap to mind when attempting to type the sentence into a blog comment box.

              1. So the algorithm doesn’t ’cause’ the bits to flick. The algorithm is the bits in the act of being flicked.

                Clearly there’s more to it than that. There’s an astronomically vast space of possible patterns or sequences of bit-flipping due to purely physical causes, and the astronomically vast majority of such patterns are simply random noise that cannot reasonably be described as algorithmic.

                Even for that tiny minority of patterns that are algorithmic — that actually compute something interesting (such as chess moves or human cognition) — a bit-level analysis tell us nothing useful about what’s really going on. The algorithm operates at a much higher level of abstraction, exerting causal influence downward to the physical level.

                This is not mere verbal shorthand, nor is it dualism; it’s an accurate description of the emergent complexity of well-organized systems of flipping bits. To deny this sort of multilevel causality, it seems to me, is to assert that all systems of bits (or molecules) are equivalent and can be adequately described at the purely physical level.

              2. @Gregory Kusnick

                … The astronomically vast majority of such patterns are simply random noise that cannot reasonably be described as algorithmic.

                White noise can of course be generated algorithmically (I hope this renders properly…)

                01. FUNCTION SimpleWhiteNoise
                02. INPUT targetLength INTEGER
                03. OUTPUT B ARRAY[BYTE]
                04. BEGIN
                05.   Set count to zero
                05.   WHILE count < targetLength DO
                06.   BEGIN
                07.     Generate random byte sequence b
                08.     Append b to the byte array B
                09.     Increment count by one
                10.   END
                11. END

                This will be a fairly good white noise generator… But of course, this is not what you meant.

                But let’s take the SimpleWhiteNoise algorithm and break it down.

                Depending on the language this pseudo-code is implemented in, there will be various means by which this algorithm may be reduced to machine code, which will in turn be reduced to bits that will be fed, in sequence, to a CPU.

                In one sense, that sequence of bit-flicking is the algorithm in action. It’s just a different method of expressing the algorithm symbolically.

                Of course, it is only once it the algorithm is expressed in terms of bits inside a CPU register that it actually does anything. So perhaps that is ‘the’ algorithm.

                Then again, the same pseudo-code could be implemented on a different framework, leading to a different sequence of bits on the CPU. And it would be ‘the same’ algorithm.

                Which is ‘the real’ algorithm? Interesting question… My taste leads me to the general case, that ‘the algorithm’ is the articulation of a sequence of operations in any symbolic form. That covers all the bases. But again, personal taste probably has a lot to do with that. Many answers could be valid there.


                In your ‘white noise applied on the CPU’ example?

                That’s still a sequence of operations on the CPU. White noise applied to a CPU in a fixed sequence of register-sized chunks is still an algorithm.

                It would very probably turn out to be a useless algorithm. But it would be a symbolic representation of discrete steps, all the same.

                Even for that tiny minority of patterns that are algorithmic — that actually compute something interesting (such as chess moves or human cognition) — a bit-level analysis tell us nothing useful about what’s really going on.

                Actually: No.

                What you are describing is the act of debugging a memory dump.

                There are people who are qualified and employed to do that.

                Thankfully, I’m not one of them. Top-tier OOP languages with a built-in IDE debugger for me, thankyouverymuch. That may not be very manly, but fuck it, I ain’t no masochist.

                Debugging memory dumps…. Eeruggdxvbble… *shudders*

                it’s an accurate description of the emergent complexity of well-organized systems of flipping bits. To deny this sort of multilevel causality, it seems to me, is to assert that all systems of bits (or molecules) are equivalent and can be adequately described at the purely physical level.

                Not all bit-strings are equivalent, in that they are different sequences.

                But any bit-string may be used as an algorithm on a CPU. It will just be the case that most of them are useless, perhaps even harmful.

                I’m going to fall back on the definition I use above.

                The algorithm is the symbolic expression of a sequence of steps.

                If that sequence happens to be in a high-level programming language, machine-code, or the byte sequence that is applied to the CPU… It doesn’t matter. It’s still ‘the algorithm’.

                So the algorithm doesn’t ’cause’ the bitwise code. It is the bitwise code. Just as it is the pseudo-code as well.

                If anything causes the bit-flicking, it’s the compiler, or maybe the operating system running the process, or perhaps even the activity on the motherboard itself.

                But it ain’t the algorithm doing the causation of the bit-flicks – expressing it that way is just a short hand.


      3. No, that’s not the point.

        The point is that there is a very important distinction between doing something because someone forced you to do it at gun point (“against your will,” as we say) and doing it because you wanted to (“of your own free will,” as we say).

        In both cases there’s a causal neuronal deterministic story to be told. But when you do it because you’re forced to, it’s not your fault. When you do it because you wanted to, then it is your fault.

        This is an important distinction, and it this distinction that most philosophers rely on when discussing “free will.”

        1. And of course I write up a long post and someone has already said the important things that I wanted to say much more elegantly and concisely. Thank you.

          1. I get paid to argue this stuff, so I might have a leg up.

            I assume you know, but just in case: The account you’re defending is what is known as the compatibilist account of free will, and it is accepted by most philosophers. You do a good job defending it.

            Bottom line, you’re right, and far too many skeptics/atheists/materialists around here are having trouble grasping the point you’re making.

            1. I hope to get paid to argue philosophy some day (albeit philosophy of science primarily), but that’s still a ways off (I’ll be a senior in college next year).

              Of course now I’m thinking about the philosophy job market, so I’m going to go to a corner and sob for a little while.

            2. Yes, the job market is scary, but people do make it through and find positions. If you can’t handle the uncertainty, then find something else. But my feeling was that I wouldn’t be any worse off spending years doing something I loved even if it didn’t land a job in the end. And so far it’s been working out for me (knock on wood . . . ).

              1. Well this is definitely what I love to do and would love to spend my life doing, so the poor job market isn’t really a factor in my choosing to go on in this. Plus, as you noted, the worst case scenario is that I spend a good 10 or so years (counting both college and grad school) doing what I love.

                For whatever reason, while I find the job market incredibly scary intellectually, it tends to have little emotional impact on me (hyperbolic declarations of sobbing to the contrary). Which I think is a good thing, in this case…

              2. Technically speaking – if we’re willing to jettison “free will” as a concept lost to the folks using it for that non-causal, non-random spookiness, but we insist on the compatibility of determinism with real choices (and insist that that’s what’s needed for moral responsibility), would the classic account of ‘compatibilism’ cover the position?

                Or take it one step further – if we abandon the notion of real choices to the spooky sorts, but insist on the compatibility of determinism with moral responsibility, would that be covered by the term?

            3. Certainly sounds good.

              Mind you, I could never take up professional philosophy on account of the fact that I couldn’t have the discipline of all the reading.

              I read exactly as much philosophical works as I feel like reading, and not a jot more… And it’s nowhere near enough to qualify me as a professional philosopher.

              I mean… Kant? Really?

              I’ll stick to the armchair, methinks. ^_^

        2. I prefer to phrase it differently.

          When we’re talking about ‘free will’ I don’t think we’re talking about thinks like freedom from duress or freedom of expression or even freedom without limit. These are different concepts in that it is trivially simple to evaluate whether we could or do possess them.

          There’s not really a debate to be had on those issues.

          So if we’re having a debate, it has to be about something other than freedom from duress.

          1. Well I don’t think there’s a debate to be had about free will either, except for the fact that people stubbornly insist on going on about it. 😛 And since I think a key reason that they stubbornly go on about it is that they suffer from the confusion I mentioned (about what constitutes the morally relevant sense in which our actions are free), whenever the debate comes up, that’s what I go to.

            It’s all well and good to argue that there’s no free will, but it’s good to show why the “loss” of free will is of no moral relevance. That’s the aim of what I wrote.

            Also, while you’re right that there’s no debate to be had about whether we sometimes make free (from coercion, etc.) choices, there is very much a debate to be had about from what exactly our choices need to be free in order for us to be held morally accountable. Once we’ve shifted to that debate, we’ve hit on the really important and substantive issue that lies behind the whole mess of the free will debate.

            1. I agree with you on the second debate: Whether our will is metaphysically free or not is irrelevant to the discussion of how we attribute moral responsibility to action.

              However, that’s the conclusion of a different discussion, and I try to assume that the people I’m arguing with are genuinely interested in the free will question for its own sake rather than as a means to the end of justifying punishment.

              It doesn’t usually work out that way of course – but I try to start with the generous assumption and take the discussion at face value.

      4. “But that’s precisely the point. You chose not to die because you are hard-wired to prevent death. In what sense is that a free choice? You might think that choosing vanilla over chocolate is different, but it is not: some neuron-firing pattern MADE you choose vanilla instead of chocolate.”

        My example of coercion is an example of a choice that isn’t free, not one that is free. But, leaving aside this small misunderstanding, I really don’t think you understood what I wrote. My point is precisely that whether a choice is free or not is not a question of physical determinism. It’s not a matter of “physically, I could have done otherwise.” If THAT is what freedom of choice hinges on, then you’re precisely right that choice isn’t free. But, as I argued, that’s a terrible way of thinking about choice because whether I could have done otherwise physically is morally irrelevant. Since the whole issue of free choice/free will* is a moral issue (the key question is, “in what circumstances may I be held accountable for my actions”), saying that we don’t have free choice because we’ve defined free choice to be something physically impossible does nothing whatsoever to clarify our thinking about the issue. It’s enough to note that our actions are (at least primarily) physically determined, leaving no room for any sort of metaphysical capacity known as free will, then to note that this doesn’t mean that we have to get rid of responsibility and accountability (because hopefully we can agree that we shouldn’t get rid of these). Once we’ve done that, we can move on to the important moral issues and stop worrying about the largely irrelevant (in this context) physics and neuroscience.**

        I don’t disagree that free will, as conceived by religious thinkers (and some atheistic philosophers who really ought to know better) does not exist. And I think that there’s enough baggage from philosophical and religious writings on the issue that we should scrap the term. Mainly, I am proposing two things:

        1. The reason that people find the concept of free will appealing is that it is a quite solid moral principle to say that we can only be held responsible for what we could have done otherwise. However, it is a mistake to interpret this principle physically. If you interpret it physically, then determinism does threaten morality. However, there is a decidedly non-physical interpretation of it that is all you need to justify holding people accountable for their actions.
        2. Freedom of choice should not be interpreted as freedom from being physically determined. It should be interpreted as freedom from impingement in your decision making process of certain MORALLY RELEVANT physical impingements (be they guns, mental illness, or something else). [Granted, this does make the “solid moral principle” I mentioned a tautology, but I think this actually clarifies our thinking, because it gets us focused on looking at specific cases and asking “is this person responsible for his actions in this case,” which is the crucial moral question.]

        I can somewhat understand the reticence to accept this talk about free choice, precisely because of the legacy of free will in all its absurdity. However, this is not some radical shift in the way we use the word free to talk about different actions. After all, physically, the actions of black slaves from the 18th century were no more or less physically determined than the actions of their white slave-owners, but they were clearly less *free.* My use of freedom in talking about choice is very similar to (though not quite identical with) that sense of freedom, the sense in which slaves lack it and slavedrivers have it.

        This is a bit more scattershot than I would’ve liked it to be, and maybe it rambles a bit, but I got the important ideas out there. Hopefully it clarifies what I was trying to say, because it seems like you interpreted me as somehow coming down on the side of free will (or against physical determinism), which I definitely was not doing.

        *For reasons I explained, I prefer free choice to free will terminologically, as there is less baggage.

        **Of course lots of neuroscience is relevant to specific cases (e.g. cases of mental illness), but the general principle “our actions are determined by the physical process of neurons firing, not by some spooky capacity called free will” is decidedly NOT relevant precisely because it applies in all cases and is thus entirely useless in making the sorts moral distinctions between different cases that are essential to moral thinking.

        1. Aaron:
          My reply was poorly worded, but I didn’t misunderstand you. I was trying to draw a parallel between the coerced choice of someone who is threatened with a gun and the apparently free choice of someone who has the vanilla ice-cream instead of the chocolate one.

          In my opinion, you are confusing free will (or free choice, if you prefer) with freedom to enact those choices. I can put it no better than Schopenhauer: “Man can do what he will but he cannot will what he wills” (quoted in Sam Harris’s article).

          1. I’m not sure what you think my confusion is. Yes, there is a parallel between the coerced choice and the free choice, namely, they are both causally determined (at least to such an extent that there is no room whatsoever for free will as a metaphysical capacity/force/something to creep into the equation).

            But that parallel holds for every action, which means that it is useless for drawing the moral distinctions between coerced and free actions, which is where I think the real meat of the issue lies. My point is that, when we talk about free actions, we should be talking precisely about the morally relevant consideration of whether we were free to enact our choices. And this is just asking whether we were coerced, or mentally ill, or otherwise driven to act the way we did for some reason that absolves us of accountability for our actions.

            If my confusion is simply that I think it’s worthwhile to call such choices ‘free’ (because, as I’ve said, there is a relevant, everyday sense of the word ‘free’ that does apply to such choices), that’s not a confusion at all. If you mean something else when you say I’ve confused free will and freedom to enact our choices, then I’m at a loss, because I’ve tried very hard to explicitly deny that we have free will in the standard theological sense and have argued that the morally relevant sense in which our actions and choices are free is the latter sense.

            In fact, unless I’m misreading you, the fact that you are making the distinction between freedom to enact our choices and freedom from physical determinism makes it seem to me that you aren’t actually disagreeing with the main thrust of what I wrote, and that to the extent to which we disagree, our disagreement is terminological. But I could be wrong about that.

            1. Where I disagree with you is here:
              “But that parallel holds for every action, which means that it is useless for drawing the moral distinctions between coerced and free actions…”

              No moral judgement can be passed on the the law of gravity preventing me from flying. It can on someone preventing me from boarding my plane. The fact that our will is not free does not mean that curtailing the realisation of that will is morally justifiable.

    2. “It’s a brute fact of everyday experience that we make choices.”
      It is also a brute fact that the sun goes around the earth. We see it with our own eyes everyday. Illusions are powerful.
      Besides, the alternative to free will is not necessarily determinism, just as the alternative to something designed doesn’t have to be something by accident. Free will is only a fancy way of calling human behavior unpredictable.

      1. You also misunderstood the point of what I wrote, which was not to argue for free will (which in fact I explicitly rejected).

        Upon reading your reply to my post, I had options. I could respond to it, or I could let it pass without responding. I thought about which action I should take, thinking of reasons for or against doing so. One reason for responding is, of course, that you misunderstood what I said, and I’d like to clarify it. One reason against responding is that I (and physicalist) have already posted something with the relevant clarification. So I started typing a response to what you wrote, and I will soon his the “post comment” button so that you can read it.

        This is making a choice. As in, in our daily lives, this is what we delineate with the term ‘choice.’ It is an interesting question to ask about the physics and neuroscience of choices. But, when I (largely ignorant of neuroscience) call this a choice, I am not saying anything about the neuroscience behind it. I am certainly not saying that our neurons are interacting with something metaphysical called free will (because that’s just silly).

        Just because we can tell a physical/neuroscientific story about the choices we make doesn’t mean that we can’t tell an equally true psychological story. There are multiple ways of describing the situation, useful for different purposes.

        Where, in what I have written, is the illusion? The only way I can see that you make the fact that we make choices into an illusion is to reject all psychological talk altogether, and doing that takes the limits of absurdity and blows them to smithereens (if that’s really what you want to do, I certainly plan to hold you accountable for it :P).

  23. dunno, just a thought here – but perhaps it’s like the difference between weather and climate, where weather is analogous to a single thought and climate is analogous to patterns of thoughts (e.g. habitual thoughts or behaviours). Whereas there may indeed be no accounting for the instantiation of any particular thought (and hence no possible way of demonstrated that you could have thought differently in that instance), once could, all the same, disrupt or interrupt the pattern of thoughts or behaviours (I believe that this is the basis for cognitive behavioural psychology, as well as the point of various meditation practices).

    Once one becomes familiar with the patterns, they develop the ability to be aware of the individual instances of that thought pattern, and may dirupt it while it is going on, or even as it is beginning.

    Still, this doesn’t necessarily account for why that thought began to emerge in the first place, but it does suggest that one can ‘willfully won’t’ (or, will the condition of free-wont).

  24. Be all that as it may, I’m glad you feel good enough to choose to start posting freely again. I had no choice but to say that.

  25. To those arguing that Harris’ version of “no free will” implies “no choice”, I would say this… It isn’t thst you don’t have a choice, it’s just that you aren’t making the choice when you think you are. I can sit here and think of dozens of reasons why murder, rape, etc. are not good things, either for myself, the victims, ot society as a whole. By thinking through and learning these lessons beforehand, I am influencing the “choice” my brain will make later when circumstances arise where murder is an option (like the line is too long at Starbucks). The choice is in what information we bring in and how our higher brain processes it. We can certainly think about the road less traveled after the fact, try to remember what things made it a better road, uncover clues to look for at the next fork in the road, etc.

    Does that make any sense or am I just moving the goalposts back?

    1. Why do we have a choice in what information we bring in? Isn’t it more consistent to say at that juncture, too, that we only think we have a choice about that?

      Can some folks not help watching reality TV?!

  26. Jerry, you’re implicitly presupposing dualism again.

    You say: “If “free will” is to mean anything, it must . . . that we could have done or thought something other than we did. There . . . is plenty of evidence against it, including the palpable fact that thoughts and choices arise from a materialistic body and brain [and] demonstrations that physical and chemical manipulations of our brain can change our thoughts and actions.”

    Notice, you just said that physical manipulations of our brain can change our thoughts and actions. Thus, I could have done or thought something else, if only someone had physically manipulated my brain.

    So it is the case that “we could have done or thought something other than we did” which is your stated criterion for free will.

    You seem to be assuming that freedom requires that we be able to change thoughts or actions without any physical changes. Why assume such an absurdity?

    The common people assume this because they believe that they are non-material souls; so if physics does something then it can’t be them doing it.

    But you’re a physicalist, so you think that physics doing something is just the same thing as your doing something. Clearly if we have a change in the physical state of our brain, we can have a change in what we do. Equally clearly, we can change the physical state of our brain.

    The only mystery lies in a deep un-purged assumption of dualism. You need to rid yourself of it.

    1. “Notice, you just said that physical manipulations of our brain can change our thoughts and actions. Thus, I could have done or thought something else, if only someone had physically manipulated my brain.”

      Yes, but that’s the same as saying “If things had been different, then they would have been different”. My objection to free will (and Jerry’s, I guess) is that it implies “If things had been THE SAME, then they could have been different”.

      “Equally clearly, we can change the physical state of our brain.”

      Clearly? That’s a usage I hadn’t come across. Who is the “we” that can change the physical state of “our” brain? How does that work?

      1. How does it work? Well, let’s start with a simple example:

        I am currently changing the physical state of my fingers — it’s called typing. I’ll now change the physical state of my brain by looking up at the ceiling for a moment. There, the state of my brain changed; new patterns in the visual cortex, new memories, all sorts of changes. Really wasn’t hard. I do it all the time, actually.

        On your earlier point, there’s a fair bit of confusion (shared with many high-profile philosophers, I might add) about how to make sense of your claim “If things had been THE SAME, then they could have been different”.

        I think it’s helpful to step away from the free will question for a moment, and ask ourselves how we generally understand claims like “X could have happened, but it didn’t.”

        Someone says, “The tornado could have hit that hospital, thank goodness it didn’t!” Is that person denying determinism/materialism? No. All she’s saying is that if things had been slightly different, the tornado would have hit the hospital.

        The claim that something could have happened is always a claim that something would have happened IF things had been slightly different. (Try any example you like. “The cat could have escaped, but he didn’t.” = “The cat would have escaped if it had seen a bird outside the open window.”)

        This being the case, even the determinist can readily accept your claim. It is the case that things *could* have been different, even though things were the same. This simply a long-winded way of saying that things could have been different.

        Of course, what we can’t accept is the claim “If things had been the same, then they WOULD have been different.” But that’s rather a silly thing to want to say, isn’t it?

          1. Not quite.

            “Could have” = “Would have if”
            “Could have” = “Could have if same”
            “Could have” =/= “Could have if different”

        1. Obvious comment here is that it is not ‘you’ that is changing ‘your’ brain.

          The brain is just changing over time in response to stimuli.

          Part of those changes includes the virtualization of an inner narrative voice that is labeled ‘me, that does the thinking’.

          So there isn’t some dualistic ‘you’ that is ‘changing your brain’.

          And I know that’s not what you meant, so: Yay for unintentionally implied dualism (again!).

          Language is a tricky bastard of a thing, ain’t it?


          1. Of course there is a “me” that’s changing my brain. It’s my brain that’s changing my brain. I change myself. My brain changes itself. I change my brain. My brain changes me.

            Once we reject dualism, these formulations all become synonymous.

  27. If you can really really imagine humanity from the outside, like the components of an evolving organism, it is much easier to see our decisions as much controlled by free will as genetic mutations which either kill off or further a species. If the majority of individual components do things which cause worse environments for the other components. that portion of the organism will ‘die off’… the hope is that, if the desire is for the organism to carry on, the portion that dies off is not the dominant portion.

    So, imprisoning recreational drug users, supporting a two party political system, promoting corporate soiling of the environment, etc. etc.. may be things which will eventually kill off a portion of the over all organism, etc. etc…

    Anyway, if we imagine our independent decisions are like that of genetic ‘decision’ making, it’s easy for me to imagine no free will. Also,I like imagining the complexity of options available to us at any given micro-second that are not perceptible to us (other than in retrospect) and we feel like we are making a decision between a few options, ignorant of the countless options available to us each moment. Thus, the illusion of free will.

  28. There is no free will

    We are ever condemned to hear new creationist wingnut idjuts some of which will run for president every generation

    recycling old arguments that go back to the days of Paley or before again and again most of which were refuted by Darwin himself

    proving that old wine can continually be put into new wineskins

    If your brain is incapable of perceiving snakes in the grass as more than mere metaphor

    the rest of us will have to endure this eternal recurrence of bad thinking again and again

    and ever be vigilant in continually disinfecting new mutant viruses in the memosphere that infect the same kinds of brains every generation

  29. What I want to know is how the concept responsibility plays out if one does not choose how to act.

    1. A cause is a difference that makes a difference. The information which an individual has could have been different and the action would then have been different. The determinists are saying that the information which an individual has absolutely determines its actions.

      That is a simplistic scenario applicably for drosophila. Humans, however, regularly find their information equivocal (allowing no clear action) and we regularly teach pupils and students how to get more information in these situations.

      Of course, this is only slightly less simplistic. A programmer could write an algorithm like:

      1. if information = a, then action = A;
      2. if infromation = b, then action = B;
      3. if informaton = equivocal, then collect more information.

      Still no free will anywhere to be see, but my hunch is that an increasingly complex algorithm gets us somewhere completely different from the simplistic determinism we started with.

      We do not need to consider an awfully complex algorithm here. For simplicity, let the third line of the above read:

      3. if information = equivolcal, then toss a coin or employ a mental randomization.

      (Maybe we have a randomizer in our brains.)

      Eventually, whether we call a decision making algorithm “free will”, “choice”, or “deterministic action” seems pretty much a matter of taste.

      The determinists will claim that even the flipping of a coin is caused naturally and therefore no extra-natural agency gets in the way. The opponents will say that the outcome of a random process cannot be known in advance and therefore the decision to toss a coin is a truly free choice.

      Their motivations seem to be at cross purpose. While the determinists want to shut all doors for extra-natural causes and all the unscientific humbug that comes along with it, their opponents want to open the door for responsibility and all the ethical issues that come along with it.

      1. 3.
        IF (information = equivocal AND cost of gathering more information >= permitted cost of inquiry)
        THEN Randomize and exit
        ELSE Gather more information

        1. Yep,
          I guess you hit on the algorithm that sends researchers with comfortable grants off into endless loops of inquiry.


        1. Thanks Diane,

          guess you mean the analogy with the doors, not the algorithms.

          In my opinion there are two doors, but someone has painted “free will” on both of them. Behind one door lies extra-natural causation and superstitious nonsense. I’d keep that door shut. Behind the other door lies responsibility for ones own actions and all the ethical issues involved. I’d keep it open.

          P.S.: Only now saw the last paragraph of the original post by Jerry Coyne:

          “How can we ever show that we could have “chosen” other than we did? We can’t. And if we can’t, then assertions about free will leave the realm of the empirical and enter the realm of philosophy. There is no free will.”

          Next time you are in the dilemma of taking a hard choice, Jerry, just toss a coin. You will thereby have shown that you could have done otherwise. Of course, that does not open the door to non-scientific nonsense. But I think it is couter-productive if the quarreling parties keep pretending that it is one and the same door, which one side wants to shut and the other side wants to open.

    2. In a nutshell – it doesn’t.

      The problem is that you haven’t specified whether the choice must be free, or whether or not a metaphysically unfree choice will suffice to establish responsibility.

      I tend to struggle to explain what I mean by a metaphysically unfree choice.

      Here’s an example.

      I am playing chess with an opponent. At any point in time while it is my move (t), the board will be in state (s). A board in state (s) will have a number of legal moves, which we will assume has more than one element at the time in question. At time (t) I select move (q) – which is an individual element in the set of (Q).

      Now: Was the act of selection pre-determined?

      Perhaps it was. Perhaps it wasn’t.

      But in either case, the following statements remain true:

      1) Board-state (s) permits set (Q) of legal moves.
      2) At time (t) with board state (s), Daniel selected legal move (q).

      To my mind, 1) and 2) are sufficient to demonstrate choice.

      If the act of selection was pre-determined? Then we can say that the choice was metaphysically ‘unfree’.

      If the act of selection is not pre-determined? Then we can say that the choice was metaphysically ‘free’.

      But in either case, free or not, the move that was selected was one of a set of legal moves for that particular board-state. That’s a good enough definition of ‘choice’ for me.

      So even if we assume free will does not exist, we still have recourse to discussing metaphysically unfree choice.

      And such a framework still allows for responsibility under the following definition:

      The utterance ‘ was responsible for ‘ is equivalent to the utterance ‘ selected ‘.


      3) Daniel was the player that selected action (q) at time (t) from the available legal actions (Q).
      4) Daniel was the player that was responsible for action (q) at time (t) from the available actions (Q).

      So we can have a framework that can connect responsibility to choice even in a metaphysically unfree universe.

      So whether or not the universe is is free isn’t relevant, as the definitions in play above apply equally well in either case.

      Furthermore: In practice, I think this definition of ‘responsibility’ may actually work…

      Although I invite any actual philosophy majors with a working understanding of the philosophy of causation of actions to kindly hand me back the tatters of my argument once they’re done with it, please. ^_^

      1. Damnit!

        That definition should read:

        The utterance ‘[agent] was responsible for [action]’ is equivalent to the utterance ‘[agent] selected [action]’.

        I used the greater-than/less-than brackets to bound ‘agent’ and ‘action’ in my initial post, so of course they got stripped out for looking like HTML entities.


            1. Yep.

              I recently discovered that we can use & n b s p ; for indentation formatting if you like.


              However: Curved, square, and parenthesis brackets are easier to type (single keystroke) than the HTML encoded angle-brackets. So I’ll just try and remember to use them instead from now on. ^_^

  30. I disagree – there *is* free will and it is as Harris says it is. Determinism should not be confused with the ridiculous notion that everything had been preordained at the moment of the Big Bang.

    Claims that we could calculate/predict everything if only we had enough information do not reflect reality. At the quantum level things are only predictable on average – for example, we cannot possibly predict exactly what will happen to a specific particle. Given such fundamental uncertainty, it is incorrect to insist that everything is predictable and that a human must have no free will because everything is somehow already determined. I find the “no free will” argument as plausible as the “there is a god” proposition – the only difference is that “god” is called “determinism”.

    1. Yep.

      One of the core problems with discussions of free will is the whole ‘defining free will in a way everyone can agree on’ thing.

      I find that’s always difficult. I tend to get only three different kinds of definition.

      1) Definition such that it is trivially true that we have free will.
      2) Definition such that it is trivially true that we do not have free will.
      3) Definition that is just a restatement of another philosophical concept (usually determinism or dualism, but perhaps something else too).

      If there’s a fourth kind of definition for free will, I haven’t heard it yet and can’t imagine what it would look like.

  31. Didn’t Yakov Smirnoff already make this point 20 years ago?

    “In Soviet Russia, decisions make you!”

  32. I really have to disagree. If free will is to mean anything then our choices have to come from US. Adding a random element – the only alternative to determinism – detracts from that. The most important distinction is between the internal constraints on our choices – which are parts of us – and external constraints. This is the very distinction that we all recognise in everyday discussions. If you do something because you want to do it (internal constraint), it is freely chosen. If you do something because someone has a gun at your head (external constraint) then it isn’t.

    Deterministic formulations of free will retain the most important parts, while losing nothing that could be worth having.

    1. I like that comment. Especially because it seems to solve issues of jurisprudence. If someone has no appropriate internal constrains on behaviour, (s)he must be constrained externally.

      There is still the possibility that internal constraints leave room for different choices.
      Therefore, “Deterministic formulation of free will” sounds strange.

      1. I’m growing fond of this conception. It could serve to illustrate different meanings of freedom.

        Someone in a dictatorship, for example, may be forced to do one thing but feel quite distinctly will otherwise. He says that his will is free and means that he can be forced to _do_ something, but not to _will_ it. That’s what common people mean when they say that their mind, thoughts, or will is free.

        Sam Harris quotes Albert Einstein and suggests other examples, where people have the external freedom to do what they will. Without extenal constraints, however, they see their internal constrains as a sign of unfreedom – how decadent.

        Coudn’t resist the last quip.

  33. Heard a climber discussing this. He said most of the time he thought about where he was putting his hands and feet, but sometimes in complex situations he just did it. His comment was that thinking is a useful executive function to check but sometimes it is slow and clumsy and gets in the way.

    I think the problem is more to do with the stimulus response model of behaviourism that is badly flawed. We are looking at incredibly complex and co-evolved systems. Margulis has written interestingly on symbiosis on these matters. I think freedom does evolve, but it isn’t a platonic thing out there, with its opposite platonic thing determinism.

  34. I think the ‘free will’ question is one that is most in need of a consistent answer from the atheist community. Attacking atheists on the idea of free will is one of the few weapons that apologists have that seems to resonate with the wavering believing population.
    Generally the attack goes something like this:
    “If you don’t believe in free will then you think that people have no choice in the bad things they do and so there is no reason to punish criminals – it’s not their fault. Therefore atheism will lead to unrestricted behavior and the breakdown of decent society.”
    I think the problem here is that the concept of “free will” here is mixing up two different things.
    One is the idea that there is an independent ‘you’ that has decision making abilities independent from your physical body. Essentially this is classical dualism where the ghost in the machine is the chooser.
    The alternative meaning of ‘free will’ is the ability to choose an action. This is not independent of the body and the physical environment. For instance I might offer you a choice of different foods to eat. The choice you make (say curry or chicken soup) might depend on physical factors such as your current hunger, physically encoded memories of your past experiences with these foods or memories of things you’ve heard about my cooking skills. Your emotional state will also play a role in your choice.
    While the addition of all these factors will make one choice inevitable (there is no free will!) the complexity of the many factors involved make it difficult to predict in advance the side in which the calculus points. This means you ‘feel’ like you are making an independent choice – that you are exercising ‘free will’.
    The best analogy to free will I can think of is a childs toy that reacts to environmental conditions. For example it has a microphone and will stop moving if it hears a sound. It also has a light detector that allows it to switch on a motor connected to steerable wheels – so it can move towards a light. On a basic level this is analogous to the human condition of free will. The difference with human (apparent) ‘free will’ is that instead of two inputs and factors that affect our actions we have many, many more – memories of experiences, mood, plans for future objectives, sources of current information etc. If we had a childs toy with similar levels of inputs and factors affecting its choices of action then I suspect it would appear to be exercising ‘free will’ in the way that a conscious human does.
    The old accusation that atheists think people are like robots is actually true – the trouble being that the robots we have so far built are not advanced enough to mirror the appearance of ‘free will’ that happens in humans. It is like the old analogy of the brain being like a computer. For children these days this is a fine and easy to understand comparison while it would have been a terrible analogy just forty years ago.

    1. “The old accusation that atheists think people are like robots is actually true – the trouble being that the robots we have so far built are not advanced enough to mirror the appearance of ‘free will’ that happens in humans. It is like the old analogy of the brain being like a computer. For children these days this is a fine and easy to understand comparison while it would have been a terrible analogy just forty years ago”.

      I agree

    2. Yes, there are many many inputs and factors and hugely many (a quadrillion) synaptic connections in the brain: It is a chaotic system, with some quantum noise as well, deterministic but unpredictable, with strange attractors.


      1. Would you agree then that concepts such as rights, ethics, independent / critical thought may all be illusions ?

          1. illusion |iˈloō zh ən|
            a false idea or belief : he had no illusions about the trouble she was in.
            • a deceptive appearance or impression : the illusion of family togetherness | the tension between illusion and reality.
            • a thing that is or is likely to be wrongly perceived or interpreted by the senses : Zollner’s illusion makes parallel lines seem to diverge by placing them on a zigzag-striped background.
            be under the illusion that believe mistakenly that : the world is under the illusion that the original painting still hangs in the Winter Palace.
            be under no illusion (or illusions) be fully aware of the true state of affairs.

            “not accurately reflecting what actually is”

            1. Phos —

              You have an inordinate fondness for quoting from a dictionary.

              If I’d wanted a dictionary definition, I could as easily have consulted my own.

              But a moment’s reflection would show that dictionary definitions are inadequate to this discussion. Do you really think the dictionary definitions of “free will”, “consciousness”, and so on would suffice?


              Define “illusion”.


              1. I believe that authorities are usually good places to start but not necessarily to end. They usually cover most of the basics.

                My words : illusions are incorrect perceptions of what actually is.

                Are victorious male lions morally wrong to kill the cubs of the vanquished former leader of a pride ?

              2. Now define “incorrect”.

                Hint: Is Newton’s theory of gravity “incorrect”?

                Your last question is obtuse. Lions are amoral.


              3. Newton’s theory is a useful approximation but incorrect theory of gravity.
                It is incorrect because it is deficient in its explanatory power.

                “Lions are amoral”.

                I know what the dictionary says but what do you mean ?

              4. So, you’d conclude that Newton’s theory of gravity is an illusion, then?

                As to lions, they simply lack the human capacity for deciding whether things are right or wrong. It’s casuistic to project “morality” onto non-human species. It seems to me that this would be so whatever the source of morality.


              5. The theory is real. The concept that it adequately defines gravity is an illusion.

                With respect to lions.

                In the past chemists defined oxidation as a chemical reaction in which oxygen combined with an substance. With further understanding chemists appreciated that when oxygen combined with a substance the substance lost electrons so combination with oxygen was understood to be an example of the more general and comprehensive concept of loss of electrons.

                It seem to me that this concept can be similarly transfer to the activities of all species whether those activities are simple or complex. No product of evolution has free will,all products of evolution function in a manner to achieve their continued survival.
                In this context how can the activities of lions or any other animal be , in essence, different from that of humans ? How could the activities of lions be amoral and that of humans moral or immoral?

            2. Nice distinction. Which theory of gravity is, then, correct, not deficient in its explanatory power, and not illusory in the same sense.

              all products of evolution function in a manner to achieve their continued survival

              That’s mistaken or sloppily recounted.

              How do art, music, philosophy, war, literature, science, sports, and so on ensure humanity’s continued survival?

              this concept can be similarly transfer[red] to the activities of all species whether those activities are simple or complex

              Where are lions’ art, music, philosophy, war, literature, science, sports, and so on?

              Evidently, there are emergent functions of humans’ complex brains and complex societies that are not conceptually transferrable to the activities of all species.


              1. “Which theory of gravity is, then, correct, not deficient in its explanatory power, and not illusory in the same sense”.


                “Evidently, there are emergent functions of humans’ complex brains and complex societies that are not conceptually transferrable to the activities of all species”.

                are emergent functions exempt from the process of natural selection ?

                If all aspects of all animals,including humans, are selected by the same process for the same reasons how can some be good or bad, moral or immoral? Aren’t the characteristics of the organism simply what is required for its survival in its environment?

              2. Yet these “illusory” theories are still incredibly accurate models of reality, are they not?

                are emergent functions exempt from the process of natural selection? … Aren’t the characteristics of the organism simply what is required for its survival in its environment?

                No… but no.

                There are no exemptions from the process, but not all characteristics are selected for. They may be, for example, simply epiphenomena.


              3. I didn’t say that the theories are illusory and I certainly do not imply that they are not useful. The illusion would be in believing that they provided comprehensive description of gravity.

                “There are no exemptions from the process, but not all characteristics are selected for. They may be, for example, simply epiphenomena”.

                are the epiphenomena unaffected by evolutionary process ?

                If you accept that evolution is a ” blind watch making process” with no end game in mind are you unwilling to accept that ethics, morals, rights etc are exactly like lions killing cubs and, in the substantial scheme of things, of no greater or lesser inherent value than predation etc in that their real purpose is to afford a specie survival and they may be replaced by other characteristics if the circumstances arise?

  35. I like the reasoning here –
    “Things happen. One after the other.”

    But do we really need any idea of free will or a lack of free will?

  36. even if we had free will, it would still not explain why we make the choices we do.

    for example, here’s a thought experiment: Imagine your soul is switched with Hitler’s soul, so now you’re Hitler, everything is the same; memories, experiences, etc. the only thing that is different is that now your soul is steering the wheel. Now if your soul would make different choices than Hitler’s soul, then the soul is arbitrary, or souls are simply made good or evil regardless of what they have experiences. If you make the same choices, well then the choices are caused by the memories, experiences of the person your soul is occupying and therfor your free will DOESN’T explain why you make the choices you make.

    1. Sorry – seems to me meaningless as A/ there is no soul… & B/ what does it mean to switch souls even if they existed? If I switch souls/minds with Hitler then I AM Hitler, so the idea is – pardon me – daft.

  37. I’m not at all sure that I understand the problem that is being discussed here, and in Sam’s essay. At one level there seems to be the assumption that every event has a cause, and if we could consider all the physical inputs to my brain, there would be an entirely causal account for each and every movement, every thought, every action that I perform. So, in one sense, we could just cancel out by all the words that express our humanness, words about thought, decision, hopes, fears, wishes, dreams, etc., and just speak about inputs and outputs.

    On the face of it this seems implausible, not because I have any particular distaste for determinism, but because one of the advantages of big brains and the ability to reason is that it provides us with the ability to think, conceive meaning, and then act according to the thoughts and meanings that we have or understand.

    Just basing the lack of “free will” on the idea of counterfactuals — “Instead of doing X I could easily have done Y” — as Sam seems to do — doesn’t really take us far. Of course, once something is done, you can’t not do it. That goes without saying. So, if I write the word ‘dog’, I do it because that’s the word I chose. It seems odd to say that I didn’t choose it, but that it was somehow chosen for me by a deterministic process. Meanings don’t work like that.

    Now, I know that consciousness is a very difficult thing to understand, and not only because philosophers are bloody minded and find the idea that consciousness is just a epiphenomenon that accompanies our thoughts and actions that are somehow determined, but because it would be hard to explain consciousness as simply a byproduct of evolution. If something this prominent didn’t give us selection advantage, what on earth do we have it for?

    Now, I don’t know this for sure, and perhaps there are ways to check this empirically, but it seems to me that all our actions are underdetermined by inputs. In other words, it is possible for the individual to take the the complex meaning systems in which our lives are interlaced and respond to inputs in completely unpredictable ways, precisely because we are meaning creating beings. So that for any inputs, there are a maze of different output choices that we can make. Is this contracausal free-will? I don’t know. But it seems far more likely that it is the ability to make these kinds of “creative” responses to inputs which gives human beings such an immense seleciton advantage, and why human beings are now at the point of decimating the world by their sheer success.

    Of course, this doesn’t mean that there is originating free will in the sense that we are absolute creators. But the ability to choose — which is what Sam wants to deny — seems to me to be built into the big brain and the cultures that big brains have created. Our repertoire of possible responses to a given set of inputs — supposing it was possible to quantify those and describe them exhaustively — is much larger than Sam’s notion of choice seems to provide for. In that sense, I think, we do have free will, and I don’t know why a scientist — whose whole raison d’être is to find the correct answers to questions by means of examining the evidence — would think otherwise.

    However, I’d be glad to hear some kind of response to this. I think the whole issue of “free will” and “consciousness” (which is related) is far more complex than Sam’s ultra simple analysis suggests.

    1. Is the question really whether we CONSCIOUSLY (think) we have free will or not? Is that not a meaningless, or circular question? some people think we have it, others do not, each behaves accordingly.

    2. I agree with your analysis. Free will is not a binary thing, it’s about the power to analyse and control. There is more of it in animals with bigger brains.

      1. Thank gawd you and Eric came along. People are acting as if this is science when it just reeks of philosophical word games.

    3. but it seems to me that all our actions are underdetermined by inputs

      But don’t forget, Eric, that the prior state of our brain (a quadrillion synaptic connections!) is also an input (really, hugely many inputs).


      1. Inputs? Do we know that? What are these “states of the brain” an input to?

        Obviously, I’m not saying that there are no causal relationships between ourselves and the world, nor between our past decisions and our present ones, and then on to action. But when we are talking determinism, what are we measuring, and is everyone sure that inputs and outputs are in the same language… physical inputs… ?outputs…

        I just not at all confident that we are sure what we are talking about. As Searle puts it, in our own experience, there seems to be a gap between the physical sensations and action, a gap not filled with causal chains. I think this gap is filled with meaning, interpretation of meaning, decision and action. In simple contexts there seems to be something taking place in our brain before we are even told to do an action. Well, yes, I suppose that’s true. But I can decide today to go into the city tomorrow, and give reasons for doing so. Now, I might have been under hypnosis and my reasons are just rationalisations, just as blindsided people make rationalisations for things they can’t see, but which obviously register somehow in the brain. So the brain “sees”, and I may be programmed to do something. But I can make that distinction because I’m not always programmed. And my actions are not always anticipated by my brain, as they are sometimes in experimental situations.

        It doesn’t make sense to me to suppose that we have all this brain capacity, consciousness, and that this doesn’t make any difference to the relationship between stimulus and response, a kind of difference that can’t be accounted for simply in terms of causality. It may be true, but it is not obviously true.

        1. I meant, inputs to the “decision”.

          I think this gap is filled with meaning, interpretation of meaning, decision and action.

          Yes, the brain is iterating through a set of states before it reaches a decision. The prior state and external inputs contribute to each successive state. This is what we consciously perceive as meaning, interpretation, reasoning, and so on.

          Why do you think that consciousness is a “brain capacity” rather than an epiphenomenon?

          I’m inclined to think that consciousness is a partial near-real-time memory of what’s going on subconsciously. Driving to work on “autopilot”, not being able to remember if you’ve locked the front door, saying one thing but thinking that you’ve said another (which strikingly happened to me when I’d ordered one dessert but was expecting another), dreaming, etc., etc., and most of all Dennett’s slide carousel experiment, are, to my mind [sic!], strongly suggestive of this.


  38. Sorry, I’d have to disagree on this, in at least two senses that both involve some of the weirder aspects of modern physics.

    First, most people are familiar with the many-worlds interpretation of QM. The idea of alternate presents a standard trope of SF, and the idea of multiple futures is obviously relevant. What most people don’t think about is that the applicable equations are time symmetric; ergo, there are alternate pasts, where we go left instead of right, and yet contribute to end up standing here.

    The second sense is in an informational one. Per the Conway-Kochen theorem, the nature of quantum entanglement implies that the information about any arbitrary point of the present is potentially tied not only to information about any arbitrary point of the past, but information about any arbitrary point of the future. From the traditional Newton-era vantage of a omniscient calculator who has all the data, there is no free will; however, due to the interrelation, there is free will in the sense of a free variable, that parts of the information frame are as underspecified as two linear equations on three variables.

    Of course, that does not mean that one cannot describe that free will probabilistically; but it’s the probabilities that are mathematically determined, and thus deterministic, and not the epiphenomenon of the appearance of particle arrangements. And of course, the modern description of physics could be inexact… but then you’re no longer arguing from the best description of evidence presently available.

  39. For those wondering about the characterization of free-will I suggest reading the essay at the Stanford Encyclopedia of Philosophy:

    There is scientific evidence that we have the ability to suppress spontaneous impulses which researchers call “free won’t.” For me that is an example of free will; i.e., choosing to or choosing not to is a freedom to choose.

    It’s all in how one characterizes free will.

  40. “…the notion that we could have done or thought something other than we did.”

    This is the incompatibilist (libertarian and hard determinist) conception of free will. Most philosophers (and scientists, it seems) fall into the compatibilist camp, though. Modern compatibilists generally equate free will with moral responsibility, and try to figure out how, given determinism, agents can be said to be morally responsible for their actions.

    If free will is defined as “the notion that we could have done or thought something other than we did,” then it is incompatible with determinism by definition. David Hume (a classical compatibilist) argued that this kind of free will was incoherent and not worth wanting.

    I agree. The compatibilist conception of free will is perfectly satisfactory to me. This is not what most religious people mean, though, when they say that God gave us free will. Part of why determinism needn’t be disturbing to a rationalist is that our actions are only *in principle* predictable. Nobody actually has sufficient knowledge to flawlessly predict our behavior. As Dan Dennett put it, this is beyond our epistemic horizon.

    However, if you are a religious believer – even a liberal one who believes God’s sole creative act was as the “first cause” of the universe, then every event must have been knowingly determined by God – including human actions. For obvious reasons, this does not settle well with many people.

    1. Except our actions don’t seem to be predictable either in principle or in practice. Populations can be predictable, yes. Countries can be predictable. The economy could be, in principle. Individuals? Haven’t seen evidence for that.

      1. It would be impossible to tell empirically if our individual actions were predictable *in principle, but not in practice*. That’s the point of the practice/principle distinction.

        However, if determinism is true – and it does seem to hold for the physics that relates to our everyday experience – then our actions *must* be predictable *in principle*.

    2. Simply preferring one “conception” over another is not a scientific way to determine (pardon the word) correctness.

      1. If this were an empirical (i.e. scientific) question, I would agree entirely. However, this is philosophy. The point is to see if any concept – or “model” – of free will is coherent. In this case, the question is, in what sense can a person be said to be “free” and “responsible” if he/she really could not have done otherwise?

        It’s not really something that can be settled with data, since determinism only claims that people’s actions will be predictable *in principle* (i.e. if you were omniscient), and not necessarily in practice (in the same way that the weather is a deterministic system that appears indeterministic in practice because of all the unknowns).

        1. And yet I don’t think an omniscient observer is even possible in principle, so it seems odd to define free will in relation to an entity that cannot exist. It’s as if one says we don’t have free will because Santa already knows if we’re naughty or nice.

        2. If this were an empirical (i.e. scientific) question, I would agree entirely. However, this is philosophy.

          Exactly. And this is why it disturbs me when scientists come down adamantly on the determinist side; many readers are going to perceive this as implying there’s a scientific level of evidence somewhere. It’s especially worrisome as it seems to be addressed largely in order to contradict religious dogma rather than due to any compelling need/desire to investigate the phenomenon itself.

  41. Inherent in the concept of free will is that we can observe ourselves and learn from our actions. For example, I realize that obsessing over a particular woman is unproductive, so I decide to cease doing so. It’s an essential component of self-awareness.

    “If “free will” is to mean anything, it must mean what Sam insists it does: the notion that we could have done or thought something other than we did. ”

    This is absurd. It says that I don’t have free will, but if I make decisions based on the flip of a coin, I do, since the flipping of the coin is random.

      1. It does make a difference if we use the definition “I could have done something else”. Since the coin is random, I clearly could have done something else (if it came up heads instead of tails).

        1. No, because tossing a coin is not truly random in the sense I think you think it is.

          If you were in exactly the same situation – we’ve rewound the tape to the same instant – you’d place the coin just so, and flip it just so, and all environmental factors would be just so, and you’d catch it just so… and it would show the same result as before.


  42. Sam Harris is correct. Free will is impossible.

    I’m a hard incompatibilist (rather than a hard determinist), meaning I think free will (logically) impossible in both a deterministic universe (entirely causal) as well as an indeterministic universe (one where acausal events happen). This is the position, I think, Sam agrees with. He takes it to the level of neuroscience, which is fine…but we do not even have to take it that far to determine the logical absurdity of free will (and holding people responsible).

    ‘Trick Slattery

  43. While I don’t subscribe to any notion of free will, brains of various sorts can engage in volition. I don’t know if we can give some logical computational algorithm or anything of the sort to define it, but it is well within the boundaries of determinism. While we can’t choose what to think, we can still make choices, ie we could have done things differently.

  44. “I could have done something different” may be translated into “somebody else in my place could have done something different”, and “I could have been a different I”. I reject the second, but I can still assign responsibility by assessing the options available to a generic human being in a given situation.

    I think that free will has no explanatory role even for people who accept it. When people say “I could / could not have done something different” they are accepting (or not) to take blame (or pride) for something they did; they are not providing a causal (empirically verifiable) explanation of their actions.

  45. Harris is certainly right that a non-conscious source of free will is not what most people mean when they argue for free will. The obvious implication is that therefore, the idea of freewill is incoherent and we should stop worrying about it. But there are at least two points in the way of such a leap: 1) While the “folk” understanding of free will involves a conscious agent, this does not obviate any idea of non-predestined, non-random action. 2) There is already at least one account of consciousness that explicitly separates the source of free will-actions and consciousness, that being epiphenomenalism (probably most often associated with David Chalmers.) This seems to be what Harris is talking about at some points here but he doesn’t address it head-on, and it would be useful to hear more from him about his position on it.

  46. I think of it in terms of an extremely complicated computer than runs a big factory. Can a computer make choices? It absolutely can choose, for example, to increase the heater power to heat up chemical vat 17D a little more. It makes these choices based on its programming, how it’s wired, the sensors connecting it to the physical factory, its experiences of what happened last time it did something, etc.

    But does this computer have Free Will? Not in the sense that religious people mean – the computer does what it does based on physical laws of cause and effect. Even if you throw in a little quantum indeterminacy in there, such as a geiger counter that it occasionally factors into its decisions, that’s still not Free Will.

    Does anyone here argue that the brain is any different? The computer has the trivial definition of Free Will, meaning that there’s nothing external preventing it from turning up the heat on vat 17D, but internally it obeys cause-and-effect of physical laws, meaning that it doesn’t have contra-causal (libertarian) Free Will. It makes choices, like the decision to turn up the heat, but those choices are the result of physics.

    I get the sense that pretty much everyone here agrees with this, but we’re struggling over the language to describe it.

    1. Does anyone here argue that the brain is any different?

      Computers aren’t self aware. We are. So, yes, I’d say there is something going on in the human brain which is different.

      1. Actually – that depends on the computer.

        It’s possible to create computer systems that evaluate their own performance and usage statistics, and make changes accordingly.

        For example, the factory computer could monitor its own CPU temperature and notice that it is warming up despite the fan being on full bore.

        It could then decide how to respond to this:

        1) Use the air conditioner in the server room to make the available air cooler.
        2) Dial back the intensity of CPU processing until temperature drops back to normal.

        A computer can also monitor its own processes (as well as the operating system processes) and its own event logs.

        A computer can also attempt to fix any problems that emerge on the fly.

        So unless there’s some concept of ‘self-awareness’ that I don’t know about, a computer is arguably capable of being more self aware than we are!

      2. H.H., I guess I didn’t mean to say that there is NO difference between the factory automation computer and a brain. For example, the computer is made of plastic, silicon, aluminum, etc., while the brain is all the stuff of biochemistry. But those don’t amount to anything that would make a difference in the analogy. Similarly, I think everyone here would agree that human self-awareness, whatever it is, is a result of the physical processes of the brain, and nothing else. Same with the computer – whatever decisions it makes are the result of physical processes in either case.

        Again, I think everyone here would agree on this much, and the question of whether there’s Free Will would have the same answer in each case. Unless you’re saying that the different degrees of self-awareness between humans and computers is somehow responsible for this extra Free Will component, but I’d be surprised if anyone said that.

        1. I agree. Both the human and the computer can be self-aware, and so both can have free will. And that holds even if you have an algorithmic view of the human brain’s operation (as opposed to Roger Penrose’s view of the brain as a quantum computer).

  47. Is there any conceivable experiment that could be carried out, the answer to which would tell us whether the universe was deterministic or random? I don’t think there is. No matter what sort of outcomes we obtain from whatever experiment, we can’t know whether those outcomes were predetermined or not.

    Ergo, from a scientific point of view, determinism and non-determinism are indistinguishable, and the question of which is true is meaningless, unanswerable, and irrelevant to anything that we do – i.e., issues of responsibility and choice can’t depend on determinism or nondeterminism.

  48. “If “free will” is to mean anything, it must mean what Sam insists it does:…” Unfortunately, Sam has not established sufficient agreement for this to be true any more than for the characterization of species to be mean anything it must mean precisely what Dr. Coyne says it is. Let the discussion continue, but insisting something must be as someone says it is is a tactic I find at Biologos, not something I expect at WEIT.

  49. How can we ever show that we could have “chosen” other than we did? We can’t. And if we can’t, then assertions about free will leave the realm of the empirical and enter the realm of philosophy. There is no free will.

    Perplexing, since this is tendentious too. What Sam says is tantamount to an effective theory, noting that we can do choices. (Say, go right or left.)

    It is one thing to claim that the idea that we could somehow control how to choose, “choose” how to choose (like having a dualist homunculus controlling the brain), is untestable.

    But it is another to claim that there isn’t a choice (of left or right, say), or that it could not be very sensitive to the environment. (I.e. not repeatable in near enough the same circumstances.) Both of these are testable, and supports the idea of an effective theory.

    There is “free will”. It’s just that it isn’t “free” in the philosophic sense.

    1. Well… at least the first supports an effective theory.

      Which we assume most of us anyway. The later depends on if we are confused or encouraged to note that we may avail ourselves of the fuller distribution of apparent choices. I.e. the sense of effective theory would be personal.

  50. First, I do not see the connection between “Why Evolution is True” and the idea that there is no free will. It seems to me that a lot of confusion in this arena revolves around the reality of a materialistic brain and that there is no “Ghost in the Machine.” So the question that gets glossed over in these debates is just who is the “I” that has no free will.

    In my view, elaborated in my new book,* the “I” is an Avatar that acts on behalf of its brain. The Avatar is real, not virtual, existing as a collection of circuit patterns of nerve impulses that are unique to each individual’s consciousness. These patterns, and thus one’s personhood, evolve continuously in real time as a function of experience. Most importantly, this Avatar can dictate much of its own programming. Prior programming does influence how one makes decisions, choices, and willed actions. The real issue is proving that the Avatar has no free will. It certain does direct certain programming options at the expense of others. That is, it chooses. Who is to say the Avatar has no power (i.e. freedom) to make certain choices over others, or even more fundamentally, choices in programming itself to weight the Bayesian probabilities certain options.

    * My new book, Atoms of Mind, The “Ghost in the Machine” has just been released by Springer. The book lays out in clear language an explanation for a material basis of thought for the three kinds of human mind: non-conscious (as in brainstem/spinal reflexes and neurohormone control), subconscious, and conscious. The book is intended for everyone interested in popular science, psychology, philosophy, and religion.
    The book is available at An e-book copy can be obtained free via participating libraries or a paperback version is available for $24.95.

Leave a Reply