Sean Carroll on free will

Yes, I know some of you are thinking “Oh, no—not another post on free will!” Well, if you’re one of those, you know what to do: just don’t read any further. I write about what interests me; and the free will debate, which is a rare nexus of of science, religion, and philosophy, is interesting. Plus I’m finishing up Sean Carroll’s very nice book, The Big Picture, and have just read his short chapter on the free will issue.

Sean is a compatibilist: someone who’s a determinist but still thinks that we should retain some conception of free will. In other words, he’s with me on the view that, in a given situation, we could not have “chosen” other than we did, for the laws of physics are either deterministic or probabilistic, and the quantum-mechanical “probabilistic” part doesn’t give us any “freedom of will.” Further, Sean is adamant that libertarian free will is not on the table, for it violates the laws of physics.

To his credit, Sean doesn’t try to offer up an alternative definition of free will like some compatibilists do, but merely says that using the language of “choice” is a useful convention, even if it’s not true that we “could have decided otherwise”. We talk about making such free choices, we feel like we make them, and we (I’m included here) use language implying that we could have chosen otherwise. As Sean says, “We attribute reality to our ability to make choices because thinking that way provides the best description we know of for the human-scale world.”  This is in fact the theme of his book: if talking about “higher order” behaviors beyond the behavior of particles—things like consciousness or pain—is a useful convention, and helps us communicate or understand things better, than it’s okay—so long as we keep in mind that such behaviors result from the underlying physics.

I am not going to fight about that. I use the language of choice, and I don’t suggest eliminating it (the alternative is awkward, as I’ve discovered when trying to be scientifically accurate); but I do insist that we always remember that we could not have done otherwise, and I insist that because its ramifications for human behavior are profound, we must always keep fundamental determinism in mind. (Some readers disagree on these ramifications of determinism, but Sean agrees that they are important.)

But there’s an important way that higher level talk about free will differs from higher level talk about emotions like love. Our feeling of volition, like our feeling of love, ultimately rests on neuronal, biochemical, and physical processes that adhere to the laws of physics. And we are better able to communicate by talking about “love” and “choice” than by trying to parse that language down to the level of leptons. (In fact, you could lose your lover if you talk that way!) But there’s a difference. Knowing that love rests on chemistry (once a metaphor, now a reality) doesn’t have any clear or important implications for society.

That’s not true for free will, for the difference between pure determinism and most people’s conceptions of free will (a libertarian one) has huge implications for society. Sean is correct when he says “Where the issue [the debate between compatibilists and incompatibilists] becomes more than merely academic is when we confront the notions of blame and responsibility.”

That’s why it’s not that important to keep reminding people that their feeling of love is a reflection of their experiences and genes mediated through the laws of physics, but why it IS important to keep reminding people that their feeling of agency is just a feeling, and doesn’t mean that in a given situation we could have chosen otherwise. It’s important to let them know that their behavior feel like they had agency, but really didn’t.

So, after this, there are two points I want to make about Sean’s chapter 44: “Freedom To Choose.” The first is simply a paragraph that I don’t understand, and perhaps readers can clarify. It’s this one (pp. 380-381):

One popular definition of free will is “the ability to have acted differently.” In a world governed by impersonal laws, one can argue that there is no such ability. Given the quantum state of the elementary particles that make up me and my environment, the future is governed by the laws of physics. But in the real world, we are not given that quantum state. We have incomplete information; we know about the rough configuration of our bodies and we have some idea of our mental states. Given only that incomplete information—the information we actually have—it’s completely conceivable that we could have acted differently.

I don’t get that paragraph at all. Lack of information simply means we cannot perfectly predict how we or anybody else will do, but it doesn’t say that what we’ll do isn’t determined in advance by physical laws. Unpredictability does not undermine determinism. And therefore, if you realize that, I don’t know how you can assert that “it’s completely conceivable that we could have acted differently.” It’s surely not conceivable to either Sean or me, or anybody who’s a determinist! Maybe we FEEL we could have acted differently, but what is “conceivable” is what we can conceive of with our existing knowledge. To me this means that it’s inconceivable that we could have acted differently. Am I missing a point here?

My other question is about Sean’s example of how our conventional human-level “ability to make choices” can be undermined by physical circumstances. He uses an example of a patient who had a brain tumor that changed his personality, causing him to download child pornography. (The “disease” was Klüver-Bucy syndrome, which causes hypersexuality). The patient was arrested, and, though a neurosurgeon testified that the patient was “not in control of his actions—he lacked free will”, he was convicted anyway.

What I would have added to this is that nobody who downloads child pornography is in control of his actions, at least in the sense that they could have avoided downloading the pornography. I’m sure Sean would agree. Whether you have a brain tumor, some other cause of hypersexuality, were abused yourself as a child, were mentally ill in a way with no clear physical diagnosis, or simply have been resistant to social pressures to avoid that kind of stuff—all of this is determined by your genes and your environment. How you treat someone convicted of that crime will differ depending on those causes, but a tumor takes away no more “freedom” than does the nexus of your genes and environmental experience.

And, to give him credit again, Sean recognizes this. But at the end he still seems to think that there’s a substantive difference between a tumor that affects your neurology or other things that affect your neurology, even if both cause you to seek out child pornography. Here’s what he says on p. 384:

To the extent that neuroscience becomes better and better at predicting what we will do without reference to our personal volition, it will be less and less appropriate to treat people as freely acting agents. Predestination will become part of our real world.

It doesn’t seem likely, however. Most people do maintain a certain degree of volition and autonomy, not to mention a complexity of cognitive functioning that makes predicting their future actions infeasible in practice.

But what does predictability have to do with this? We already know that people are not freely acting agents in the sense that they are free from deterministic control by their brains. We already know that people are predestined. (I’m not sure what Sean means by “a certain degree of volition and autonomy,” unless he means something like “compelled not by a brain tumor, but by other aspects of their neurology.”) And we shouldn’t treat anyone as freely acting agents.

I’ve already given my solution to this issue. We recognize that, at bottom, nobody could have done otherwise. If they are accused of something that society deems to be a crime, you find out if they really did commit that crime. If they’re found guilty, then a group of experts—scientists, psychologists, sociologists criminologists, etc.—determine what the “punishment” should be based on the person’s history (a brain tumor would mandate an operation, for instance), malleability to persuasion, likelihood of recidivism, danger to society, and deterrent effects. None of that needs the assumption that someone is a “freely acting agent.”

I know many of you will disagree on that, or on the ramifications of determinism for our punishment and reward system. But at least I have the satisfaction of knowing that Sean agrees with me on this: if science shows the way our behaviors are determined, that knowledge should affect the way we punish and reward people.

 

143 Comments

  1. Posted August 21, 2016 at 1:26 pm | Permalink

    Your summary of your divergence with Sean on this matter perfectly matches one I’d give.

    I would suggest to Sean that the word, “free,” is not only superfluous but detrimental to the concept he’s attempting to describe. The will is the most prominent aspect of that part of reality that constrains our actions to the ones we take. Freedom is the lack of constraint. What most people are interested in is the chain of events that leads to a decision and the quality of the reasoning that goes into it; contrafactuals are really only useful as computational aids in winnowing out suboptimal choices.

    Indeed, I think it’d be fair to characterize the distinction between determinism and compatibilism as very like the one between Lamarck and Darwin. A determinist sees cognition very much in Darwinian terms, where the most effective imagined plan of action is implemented. A compatibilist thinks a top-down description is better, with the thought primary and free to influence action.

    Maybe framing the matter along those lines will help break up some logjams?

    Cheers,

    b&

    • Diana MacPherson
      Posted August 21, 2016 at 1:36 pm | Permalink

      And I don’t get that one paragraph either unless Sean is talking about understanding how quantum states influence things not regular people just having more information.

      • Tim Harris
        Posted August 21, 2016 at 7:39 pm | Permalink

        ‘Given only that incomplete information—the information we actually have—it’s completely conceivable that we could have acted differently.’ Surely, SC is making the simple, and really not so interesting, point, that had we had information other than the information we had at one particular time, we should have acted differently from the way we in fact acted at that time.

        • darrelle
          Posted August 22, 2016 at 8:23 am | Permalink

          Yes, I agree.

        • Posted August 22, 2016 at 11:02 am | Permalink

          I think Sean ultimately is….

          b&

          >

    • Posted August 21, 2016 at 1:50 pm | Permalink

      I think it’d be fair to characterize the distinction between determinism and compatibilism as very like the one between Lamarck and Darwin.

      There is no distinction between those two, compatibilists *are* determinists. The distinction is between those who reject the compatibilist account (even though they use it in everyday life all the time!), and those who consider that the compatibilist account is a useful way of thinking about the world.

      • Posted August 21, 2016 at 2:47 pm | Permalink

        I really think it is time to do away with the term “free will”, especially the “free” part. That said, I am not clear about what to replace it with.

        Still, Sean’s point (I loved his book!) is that on a certain level, the language of determinism (or quantum mechanics) is useless — at least in some cases. E.g., in the case when I must compare two different choices and study them so as to make the best one, my acceptance or not of determinism is of no use whatsoever. I still have to study the literature, talk to people, and think a lot before I make my choice — even if it was determined by my molecules that I would have to do all that. And that’s what Sean is defending, as I see it.

        Let’s put physical determinism where it should be, on the scale of QM physics.

        (Is this maybe related to our idea of consciousness? (I’ll stop there on that subject. I am not a philosopher.))

        As I have said before, after I have gone through the weighty process that a choice sometimes requires and I make a good choice, meaning it has good result, it really ticks me off not to be able to enjoy the feeling that I have done something good — rather than its having taken place in spite of me, whatever that may mean.

        So in spite of the fact that I accept the laws of physics, I intend to go on behaving as if my difficult choices are ones that I made and — especially — that I can take credit for. And I think Sean would agree with that.

        • Posted August 21, 2016 at 2:54 pm | Permalink

          … it really ticks me off not to be able to enjoy the feeling that I have done something good …

          You can appreciate and admire the way an eagle flies, or admire a classic car, or enjoy a fine wine, without having to think that the eagle, car or wine had any “choice” in how it is — so why can’t you appreciate yourself for the good things you have done, without having to ascribe libertarian “free will” to yourself?

          • Posted August 28, 2016 at 8:11 am | Permalink

            Why is it that all these discussions wind up sounding like sophism to me? (That is a comment about me, not Coel. I assume criticizing myself is not against the rules. 😉 )

        • Posted August 21, 2016 at 4:24 pm | Permalink

          “…after I have gone through the weighty process that a choice sometimes requires and I make a good choice, meaning it has good result, it really ticks me off not to be able to enjoy the feeling that I have done something good…”

          Is it the perception that a choice has positive results which triggers pleasant-feeling brain chemicals like opiods and dopamine or is it just the fact of cognition that does that? My remembering feeling good can lock me into repeating the act regardless of less than stellar results encouraging my learning to be flexible and experimenting with various solutions. I become my own stick and carrot.

          Not being sessile like plants, humans require emotions to get themselves off their arses so they don’t starve to death. A plant obviously doesn’t have free will. We don’t either, but we sure do appear that we do as we spur ourselves into mobile processes of fulfilling our needs.

          When we go into flow states, free will becomes redundant.

          • sshort
            Posted August 21, 2016 at 4:51 pm | Permalink

            Last comment is very intersting. Can you extrapolate a bit?

            I’m a little familiar with the concept of flow states and have experienced it myself. Never really thought what relation that had to “free will.”

            • Posted August 22, 2016 at 11:02 am | Permalink

              Your consciousness tends to micro-manage a lot, which is inefficient. Much better to let it observe, analyze, and do future planning and projection — just as business management should be focussed on. If you’re trying to consciously control every little muscle contraction, you should be in the beginning stages of acquiring a new physical skill — and still be focussed more on high-level perceptions of the overall effect.

              Try it: lift a pen by figuring out which of each and every muscle needs to move in which direction by which amount. The result will be a complete and total uncoordinated mess. But instead visualize the pen at a certain orientation in front of your eyes and let your body do the rest, and you’ll do it in a nice, smooth, fluid, graceful motion.

              As such, I’d argue that consciousness itself is at least as much overrated as business management….

              Cheers,

              b&

              >

        • Posted August 22, 2016 at 11:19 am | Permalink

          In my view, at least wrt the professional debate in philosophy, the topic should move to “moral responsibility” and “the nature of the self”, because that’s really what people are debating.

      • Posted August 22, 2016 at 10:56 am | Permalink

        Set aside for the moment our disagreement that your position is deterministic.

        I’m proposing that the decision-making process is loosely Darwinian. A number of potential courses of action are deterministically created in and analyzed by the brain. The course that is a best fit for that environment survives to be implemented by the body. Those individuals that do a better job at this tend to have more great-grandchildren, whose own decision-making processes are similar to those of their great-grandparents.

        For the purposes of this discussion, I would suggest that that is reasonably sufficient and complete.

        But I would further argue that, while it may or may not be consistent with compatibilist “Free Will,” it is not a complete description of compatibilist “Free Will” — and that the extra stuff you need to add on for compatibilism is, at best, superfluous if not outright physically impossible or logically incoherent.

        Specifically, if I’m understanding compatibilism correctly, there’s a need to understand decision-making in terms of some sort of teleological goal, that there’s some ideal optimum future condition that the self is choosing to guide itself towards. You could be forgiven for making that sort of mistrake for the same reasons that Richard Dawkins identifies the appearance of design in biology — but the heart of the debate would be both whether such design is real and how useful it is to pretend that there’s a designer.

        If you’d agree with me that it’s counter-productive to pretend that there’s a designer who accounts for the appearance of design in biology, you should consider agreeing with me that it’s counter-productive to pretend that there’s a decider who accounts for the appearance of choice in cognition.

        Cheers,

        b&

        >

        • Posted August 22, 2016 at 11:44 am | Permalink

          There is nothing extra that we compatibilists are asking for. Everything about compatibilism is encapsulated in a chess-playing computer.

          Yes, we talk in terms of goals and choices. The chess-playing computer needs a program such that it has an aim (winning the game), and such that it weighs up various options in relation to that aim, and then deterministically selects (chooses) the one that best attains the aim.

          There is nothing the slightest bit mysterious about this: we can *build* and program devices (chess-playing computers; aircraft autopilots) that manifest compatibilist behaviour.

          And everything about such devices is entirely in line with deterministic physics. You don’t need an added-woo module to make an aircraft autopilot work.

          • Posted August 22, 2016 at 11:57 am | Permalink

            Everything about compatibilism is encapsulated in a chess-playing computer.

            If that’s the hill you wish to make your stand on…well, a chess computer is typically the archetypal example of something that lacks “Free Will,” however the term is being defined.

            As a simple sociological matter, I can’t conceive of how you could even hypothetically get any significant proportion of the population to agree that “Free Will,” whatever it is, is a property shared with chess computers.

            May I suggest?

            Take the term, “Free Will,” and banish it from your lexicon. Argue that human cognition is the same in principle as the processing performed by chess computers, but don’t muddy the waters by declaring that therefore chess computers have “Free Will.”

            Because you’re exactly fighting the same losing battle as over “Elan Vital.” We can (reasonably) fully explain biology without worrying about how much “Elan Vital” an amoeba does or doesn’t have; similarly, we can (reasonably) fully explain cognition and decision-making without worrying about how much “Free Will” a chess computer does or doesn’t have.

            Cheers,

            b&

            >

            • Posted August 22, 2016 at 1:03 pm | Permalink

              Hi Ben,

              Firstly, the general public’s attitudes on this are a whole mess of inconsistency, so I wouldn’t put much store on that.

              Secondly, sure, we can drop the term “free will”, no problem. But we really cannot drop the concept of “choosing”, and we need to properly interpret it in a deterministic world. That’s why we need compatibilism.

              Third, yes, you can explain a chess-playing computer without the term “will” or “choose” — but then you could explain Darwinian evolution without the concept “life”. It’s just that in both cases the concepts are useful.

              • Posted August 22, 2016 at 1:24 pm | Permalink

                We might actually be making some progress!

                “Choice” is a potentially problematic term, but far less loaded than, “free will.”

                Which is why the approach I take is to, as you’re now doing, make explicit that the human decision-making process is in the same basic class of phenomena as chess computers and autopilots: we are computational engines that process inputs in ways governed by our internal states that result in specific actions.

                We are reasonably reasonable, such that, given a certain environment (input), we will predictably react in a certain way somewhat unique to each individual — which is to say that a certain person is very likely to almost always have the chocolate ice cream but another almost always the vanilla, but there are (typically) circumstances in which some other factor can override that particular preference.

                When doing our decision-making, we imagine a multiverse of different options and we imagine freely wandering that landscape, and that whole process gives us the general perception of Jerry’s tape-rewinding analogy. It is a very powerful perception, incredibly visceral, so much so that many people aren’t even really consciously aware that all that happens only in the imagination. Most people when pointing to what they’re doing when they say they’re exercising their “Free Will” are pointing to this phenomenon…

                …but the reality of what’s going on is just a super ramped-up very complex version of what the chess computer does: taking inputs and filtering them through circuitry (that itself is changed by the environment) to control some action.

                If you can buy off on that — and it sounds like you’re this =><= close to doing so…

                …then welcome to the dark side of incompatibilism. But no worries — we have cats! And boots! And food! And speciation!

                Cheers,

                b&

                >

              • Posted August 22, 2016 at 1:31 pm | Permalink

                I certainly agree that we are essentially a souped-up version of a aircraft autopilot, following the laws of physics just as that computer is.

  2. Diana MacPherson
    Posted August 21, 2016 at 1:34 pm | Permalink

    Sub 😧🔫

    • Mark Sturtevant
      Posted August 21, 2016 at 3:52 pm | Permalink

      Very funny! But you had no choice…

      • Diana MacPherson
        Posted August 21, 2016 at 4:06 pm | Permalink

        😜

  3. burt simon
    Posted August 21, 2016 at 1:38 pm | Permalink

    That first confusing paragraph makes sense as follows. The deterministic outcome depends on a complete knowledge of the state of the universe. We only have access to a small subset of that state. So it is possible that there are two different states of the universe that agree completely on the part we have access to. If the choice we make is different for the two different states of the universe, it appears that based only on what we have access to, we could have made either choice.

    In probability theory, this is called “conditioning with respect to a sigma filed”. The sigma filed characterizes the information you have access to.

  4. Posted August 21, 2016 at 1:46 pm | Permalink

    I agree with Sean (or Dan), for similar reasons (as previously discussed on this site). We do differenciate between different types of force. A gun to the head, a brain tumor, a dire financial situation, peer pressure, just one’s own volition.

    If we consequently and consistently eleminated “free will” and properly see it as “physics all the way down” we collapse all those meaningful distinctions.

    Incompatibilists have argued that Compatibilists were engaged in wordplay and semantics, but as I see it: everyone does. You can quibble forever about the “free” and whatever is a “will”, but as you see Incompatibilists don’t get rid of “choosing”, “wishing” and “wanting” either. Everyone sticks to some useful conventions.

    Further, we have no problem to divide our understanding into different domains, like sociology, psychology, biology, chemistry or physics and understand that each vantage point provides useful models (and is useless for other purposes). Free Will is such a useful model.

    • Andreas Geisler
      Posted August 22, 2016 at 3:02 am | Permalink

      Meaningful in what sense?
      Linguistically? Sure, they’re meaningful. But they don’t change the outcomes.

      Free will exists, otherwise we would not be having this discussion. However, it seems clear that “Free will” only exists as “The feeling humans get that they consciously decide (some of) their actions”.

      That feeling doesn’t seem to correspond to a factual state of affairs, though.

      Sure, that’s enough to justify the linguistic stuff, but we should not let linguistics get in the way of adapting to what we know.

      Criminality becomes largely a matter of starting points, outcomes, detrimental cognitive patterns, etc. etc.

      Many of which can be diagnosed and treated already.

      Jailing people for crimes mainly caused by their being born into poverty is not what I would call fair.

      I also think it’s a waste of resources.

      And then we can go on to argue that allowing poverty is a waste of resources too.

      Do you see how this works?

    • peepuk
      Posted August 22, 2016 at 6:18 am | Permalink

      “Free Will is such a useful model.”

      The concept of God can be a useful model, but still that doesn’t mean that believing in a God is a justified belief.

      It may be sometimes useful, but we must not forget that the believe in Free Will is at the same time very harmful and blocks progress in our judicial systems.

      Hard determinism is a justified belief because it corresponds to what we know about the nature of reality and it gets rid of the false belief that people deserve to be punished.

  5. Posted August 21, 2016 at 1:47 pm | Permalink

    First, I’ll suggest that PCC-E is mellowing considerably towards compatibilism. Here I’ll quote:

    “As a determinist, I use “choose” here as shorthand for “what one does if one’s brain is not impacted by social or
    government pressure””.

    That is pure compatibilism: a notion of a range of possibilities given a range of environments, with “choice” being the deterministic selection of one of them, computed by an brain influenced by external pressures.

    Second:

    “I don’t get that paragraph at all. Lack of information simply means we cannot perfectly predict how we or anybody else will do, …”

    Given the lack of information there is a range of possibilities that the brain has to consider. The lack of information means that the world could be in various states, each of those would produce a different outcome, and the brain has to consider all of those in order to compute its response.

    We have a notion of “choice” precisely because our brains have evolved to make decisions in a world where we lack full information, and so have to weigh up the range of possible states that the world might be in, in order to make a decision. That’s why we think about a range of possibilities, and think of people “choosing” from that range.

    “I insist that because its ramifications for human behavior are profound, we must always keep fundamental determinism in mind.”

    Perhaps, Jerry, you could do produce a series of posts about actual changes you suggest in society. By that I mean differences of the order “this person should not be in jail whereas currently he is”, rather than mere differences in commentary: “this person is in jail for deterrence” as oppose to “this person is in jail for retribution”.

    • Tim Harris
      Posted August 21, 2016 at 8:56 pm | Permalink

      The difference between compatibilists and incompatibilists has always struck me as being mostly about words, but on a deeper level about what constitutes an acceptable description or explanation of human behaviour. Since we are influenced in our actions by the ideas and ideologies, religious or otherwise, we entertain, as well as things like scientific or scholarly practice, it has not always seemed helpful to jump at once to ‘physics’ as THE explanation. I am very glad to see that Jerry is making things clearer than they seemed to be before; but although it is certainly ultimately true that a tumour ‘takes away no more “freedom” than does the nexus of your genes and environmental experience’, it obviously doesn’t do (and I am certainly not accusing Jerry of this)to suggest that there are no distinctions to be made between someone who has been brought to child pornography by the effects of a tumour, someone who has been brought to it asa result of childhood experiences, and, let us say, Sir Jimmy Saville, with his glittering career. Agreeing that we are ultimately determined does not preclude the making of distinctions and striving to create adequate explanations.

      • Posted August 22, 2016 at 3:49 am | Permalink

        As I see it, attributing moral responsibility is entirely pragmatic. We threaten a person with sanctions because it can modify their behaviour. We don’t threaten a non-functioning car because it won’t!

        If the brain tumour case is more akin to the latter then we don’t attribute moral responsibility.

        • Tim Harris
          Posted August 22, 2016 at 4:00 am | Permalink

          Yes. That is well said.

        • Posted August 22, 2016 at 4:44 am | Permalink

          How do you know that you can modify someone’s behavior who doesn’t have a brain tumor? Some people you can’t, and that, too, has been determined by their genes and their environment. I alluded to this in the post. We can’t see a tumor, but they’re just a driven by their own neurology as someone with a tumor, and many of these people are immune to the threat of sanctions. That you have to admit.

          • Posted August 22, 2016 at 8:21 am | Permalink

            We don’t know that we can modify *everyone’s* behaviour by threats or praise. Indeed we know that some people will not be deterred by the threat of jail and will still commit crimes.

            But we know that *overall* many people’s behaviour will be modified, and lacking full information we don’t know exactly which whose, so society operates by applying notions of blame or praise to everyone.

            • Posted August 22, 2016 at 11:21 am | Permalink

              What we *don’t* know enough about, IMO, is if that this is *the most effective way* to do so, especially in conjunction with our other values and policies. Especially if we want to make the latter uniform. (Which might be decided because of a metavalue.)

          • Gregory Kusnick
            Posted August 22, 2016 at 12:27 pm | Permalink

            Sure, there’s more than one way for brains to malfunction. That’s just entropy; there are more ways to be broken than unbroken.

            But determinism doesn’t oblige the justice system to treat all brains as broken, any more than it obliges aircraft inspectors to treat all autopilots as broken.

            • Posted August 22, 2016 at 12:37 pm | Permalink

              Except, by definition, if you’re guilty of a crime, you are broken.

              The inspectors might not assume that every autopilot is broken, but they will assume that, if an autopilot was operating the aircraft at the time of an incident, then the autopilot was broken. (Or, of course, should have been disengaged…but, obviously, an even better autopilot could have been left engaged and acted optimally to avoid the incident.)

              b&

              >

              • Gregory Kusnick
                Posted August 22, 2016 at 12:53 pm | Permalink

                Sorry, Ben, but I completely reject the notion that anyone appearing before a criminal court is by definition mentally defective. That’s Stalinism, and is the worst possible argument for a deterministic model of justice.

              • Posted August 22, 2016 at 12:59 pm | Permalink

                Er…I probably could have phrased that better. Of course, guilt would have to be sanely established, and we should always understand that legal proceedings are far from perfect and even those legally guilty may well in reality have nothing to do with that of which they’ve been convicted.

                But, within that context, if we’ve decided that somebody is guilty, we’ve also decided that that person is “defective” in the sense you used earlier.

                It may well be the case that, once having determined guilt, different approaches are called for those who suffer from brain tumors as opposed to those who do not. But we should not privilege those with brain tumors for deserving of compassion over those not so “lucky” as to have a brain tumor. If you would offer compassionate medical treatment to cure somebody of a brain tumor that caused criminal behavior, you should offer similar compassionate treatment of whatever form to anybody else who behaved criminally, regardless of the proximate cause of the behavior.

                Cheers,

                b&

                >

              • Gregory Kusnick
                Posted August 22, 2016 at 3:16 pm | Permalink

                I’d still be very reluctant to sign on to any view that classifies conscientious dissidents such as MLK or Ghandi as “defective” or “broken”.

                Also if we take hard determinism seriously, “deserving of compassion” is no more meaningful than “deserving of blame”.

              • Posted August 22, 2016 at 3:25 pm | Permalink

                I’d still be very reluctant to sign on to any view that classifies conscientious dissidents such as MLK or Ghandi as “defective” or “broken”.

                As a society, we understand full well that an inflexible legal system lacking in judgement is unjust. Which is why we have all the checks and balances, ranging from prosecutorial discretion to jury nullification to the appellate courts to executive clemency to legislative and even Constitutional overhaul.

                No determinist is even vaguely hinting that any of that is inappropriate in light of the reality of determinism. Indeed, conscientious objection would remain a very effective means of remedying unjust laws.

                Also if we take hard determinism seriously, “deserving of compassion” is no more meaningful than “deserving of blame”.

                That depends entirely on what you see as the function of the criminal justice system.

                If you see it as a means of controlling the dispensation of due punishment and vengeance, then, sure, your objection applies.

                But if you instead see it as a way of society protecting itself from harm from antisocial individuals, then compassion is the only narrative that can even begin to make sense. Are not antisocial individuals themselves members of society? Should we not extend societal protections to all members of society?

                Cheers,

                b&

                >

              • Gregory Kusnick
                Posted August 23, 2016 at 12:09 am | Permalink

                I have no problem with compassion; my problem is with the argument that says we must view all lawbreakers as mentally defective in order to have compassion for them. I’m perfectly capable of feeling compassion for mentally competent people who simply made bad decisions.

        • Richard
          Posted August 22, 2016 at 4:54 am | Permalink

          You’re obviously not familiar with Basil Fawlty….

      • Posted August 22, 2016 at 11:13 am | Permalink

        [A]lthough it is certainly ultimately true that a tumour ‘takes away no more “freedom” than does the nexus of your genes and environmental experience’, it obviously doesn’t do (and I am certainly not accusing Jerry of this)to suggest that there are no distinctions to be made between someone who has been brought to child pornography by the effects of a tumour, someone who has been brought to it asa result of childhood experiences, and, let us say, Sir Jimmy Saville, with his glittering career.

        Sam Harris has some wonderful discussions on that topic that should be more than ample to convince you that there really isn’t such a difference.

        b&

        >

        • Tim Harris
          Posted August 22, 2016 at 5:29 pm | Permalink

          Yes, I know well that there is not in the end such a difference. There are nevertheless useful distinctions to be made.

  6. Posted August 21, 2016 at 1:48 pm | Permalink

    It’s impossible to abandon pragmatic compatibilism when navigating the vagaries of daily experience, but the operational schematics of the GPS that make it possible could care less. My pedantry-infused pet peeve resides with those who allow the word “freedom” to become conveniently elastic to the point of being situationally unfalsifiable.

  7. JohnH
    Posted August 21, 2016 at 1:55 pm | Permalink

    “Given only that incomplete information—the information we actually have—it’s completely conceivable that we could have acted differently.” Could he mean that prior to the action a number of conceivable actions are possible based on the information that is accessible to us (the observer)? The more information we have, the lower the number of possible or conceivable actions become. When the information approaches the limit of total information the number of conceivable actions becomes one. It depends upon the level of information that the observer has at the time of the action that determines the number of conceivable actions that might have taken place.

    • JohnH
      Posted August 21, 2016 at 4:48 pm | Permalink

      The same argument can be made substituting agent for observer, and it is the level of information that determines the illusion of free will in terms of the number of conceivable actions that the agent can or might perform.

  8. Gasper Sciacca
    Posted August 21, 2016 at 2:07 pm | Permalink

    It seems that Carroll has modified his original thoughts on “emergent” qualities that create phenomena like “will”. I think he doesn’t want to get too far away from the current vogue of thinking everything is deterministic.
    I for one do not believe that our thoughts are authored by trajectories of atomic particles. Rather our thoughts, our will, so to speak, are the software created by other more primitive software. Physics is just the hardware that supports this process.

    • Posted August 22, 2016 at 1:56 am | Permalink

      In fact, we are all determinists on some levels (contexts) and compatibilists on others.

    • Posted August 22, 2016 at 11:12 am | Permalink

      The distinction between software and hardware is arbitrary and convenient. In reality, it’s all hardware. Every flipped bit in software is a description of a corresponding change in physical state — a RAM cell with a different voltage, a pit in a CD, and so on. By design, the bits of hardware we label as soft are very flexible and can typically be changed on the order of millihertz or faster; however, logically, it’s no different from breaking out the (really, really, really tiny) soldering iron.

      Cheers,

      b&

      >

      • Posted August 22, 2016 at 11:24 am | Permalink

        I’ve thought for a while that software is a sequence of easily accessible state transitions and that the state, pretended to be constant, is the hardware. But I am not sure that works.

        • Posted August 22, 2016 at 11:30 am | Permalink

          Depends on what you mean by, “works.”

          If you mean an accurate description of the low-level functioning of the system, absolutely.

          If you mean something that a programmer is typically going to think of when slinging code, absolutely not.

          (There are, of course, edge cases where reality comes in and slaps you in the face with a wet and stinky mackerel, and reminds you that it’s ultimately all hardware. But most of the time most of us can be perfectly blissful pretending otherwise.)

          Cheers,

          b&

          >

  9. BobTerrace
    Posted August 21, 2016 at 2:24 pm | Permalink

    Subscribe.

  10. JonLynnHarvey
    Posted August 21, 2016 at 2:44 pm | Permalink

    From Bob Dylan
    Ah, my friends from the prison, they ask unto me
    “How good, how good does it feel to be free?”
    And I answer them most mysteriously
    “Are birds free from the chains of the skyway?”

  11. Posted August 21, 2016 at 2:49 pm | Permalink

    Maybe a rough idea about what Carroll means here: “We have incomplete information; we know about the rough configuration of our bodies and we have some idea of our mental states. Given only that incomplete information—the information we actually have—it’s completely conceivable that we could have acted differently.”

    I’m thinking about things probabilistically. With complete information, i.e. “the quantum state of the elementary particles that make up me and my environment,” we have a well-defined probability measure governing our instantaneous actions: Pr(X).

    But given only incomplete information, we only have access to some conditional probability measure, Pr(X | Y = y). This is a random quantity in its own right since we can’t have complete information in practice at any previous time either. So in this sense then, the “deterministic” action (i.e. the quantum + deterministic action governed by the joint probability measure Pr(X)) does not have to agree with the conditional probability measure.

    Is that free will? I don’t know, and this is what has always bothered me about the discussion. The language of probability alone is insufficient to describe causal action, and any conversation about free will or lack thereof is ultimately a conversation that liberally tosses around notions of causality as given. I am not aware of a robust enough theory of causality to adequately treat the problem though.

    • Posted August 22, 2016 at 10:31 am | Permalink

      Good instincts! A robust theory of causality.

      • Posted August 22, 2016 at 11:00 am | Permalink

        Actually, I’m quite a fan of Pearl’s! It’s the closest thing we have to a coherent theory of causality, but it still has a long ways to go before it is robust enough to treat causal situations without a well-defined intervention, in my opinion.

        • Posted August 22, 2016 at 11:25 am | Permalink

          It is also a bit metaphysically confusing (on purpose, as Pearl’s colleagues at CMU that I knew pointed out) – I think we have to know what the relata of the causal relation are before one can clean up some of the practical uses too.

        • Posted August 22, 2016 at 8:27 pm | Permalink

          I’m not sure what you mean, but I suspect Pearl might reasonably deny that situations without any well-defined (or definable) interventions *are* causal.

  12. Jonathan Livengood
    Posted August 21, 2016 at 2:56 pm | Permalink

    On the tricky paragraph, perhaps the thought is something like this.

    The ordinary, everyday properties that an agent has (or that we describe the agent as having) supervene on the properties that the fundamental *parts* of the agent have. For any difference in macro-level properties, there has to be some difference in micro-level properties.

    But the converse is not true. The properties of the fundamental parts of two agents might be different without the agents differing with respect to their macro-level properties.

    Now, suppose we ask, “Could an agent under such and so macro-level description have done X in such and such a situation?” It could be that there are *many* things that could be filled in for X that would be true at the macro-level in virtue of the fact that we could spell out the micro-level in many subtly different ways — all of which are consistent with the macro-level description.

    Suppose, then, that we watch Suzy as she skips a rock out across a lake. On the foregoing picture, to say, “Suzy could have chosen not to throw the rock,” is to say that there is a micro-level foundation consistent with our macro-level account of Suzy right up to the moment before she chooses to throw the rock (maybe in terms of her height, weight, physical strength, and so on) on which she does not choose to throw the rock.

    The above account makes what we regard as possible-for-an-agent-to-do relative to a description. The follow-up idea, I think, is that which description one selects as the *best* is going to be relative to what we know (or believe).

    Given your starting point, I agree that the above isn’t very satisfying. From your perspective, it’s enough to know that there is a level of description for *this* agent on which what she does is determined exactly by her history and the laws of nature. (Here, I want to press on how that sort of determinism should lead us to say that she could not have done otherwise. But that is a longer discussion.)

    • Posted August 22, 2016 at 11:06 am | Permalink

      Good points. burt simon makes some similar points. You’ve made a good jumping off point to explain where I think Sean went wrong.

      The problem is focusing an external observer’s knowledge and options, when what matters are Suzy’s. What happens if Suzy tries to observe the quantum state and “calculate” what she’ll do? Well if she has a playful or contrary nature, she’ll immediately contradict the prediction. Her Inner Agent trumps her Inner Scientist. But on the other hand, if she just decides, then there’s no need to do a calculation. Or if she wants, starting from knowing her decision, she can back-calculate what the conditions were previously.

      There’s more. The earlier quantum state leads to the action only by way of the brain processes, including conscious ones, that constitute Suzy-making-her-decision. And according to a theory of personal identity very popular among Reductionists, what makes those processes conscious, and what makes them Suzy, is a certain characteristic pattern of information flow. So, what happens if an external observer duplicates all the information about an earlier quantum state which includes a complete description of Suzy, and then faithfully duplicates the process by which that information transforms into Suzy’s decision? Simple: he makes another Suzy, or perhaps better, he extends her into additional territory.

      So the external observer can predict her decision – but only by making her make the decision twice over, and using the first as the model for the second.

  13. peltonrandy
    Posted August 21, 2016 at 3:12 pm | Permalink

    sub

  14. Jstackpo
    Posted August 21, 2016 at 3:26 pm | Permalink

    Sub

  15. Posted August 21, 2016 at 4:36 pm | Permalink

    One of my so-called poems to be shared with proponents of “Free Will”, and others, if they “choose” to read it:

    DEAR SPECIAL CREATIONISTS,

    Although humans, Homo Sapiens,
    (now called Homonins,)
    and our cousins, the apes,
    (we, and they,) are Hominoids,
    the ancestors we both derive from
    are Hominoidea.

    Preceding relatives, small
    mammals called Lemurs,
    stayed awake at night in trees.
    From them, we acquired
    binocular vision, eyes
    in front of our faces,
    instead of on the sides.

    Before that, our ancestors
    were fish and reptiles.
    Some fish had lungbladders
    and could breathe out of water.
    Some fish had stronger fins,
    and some developed longer,
    bones to navigate in sand.
    Our lungs, legs and arms
    we owe to them.

    Many millions of years
    before that, our ancestors were
    one-celled organisms.
    Who knows what they did,
    except divide and multiply?
    (Which led to more complex
    multi-cellular organisms.)

    Even before that, our ancestors
    were algae (scum),
    swimming, or not,
    in primordial soup,
    producing the oxygen
    that permitted development
    of all later life on Earth.

    Bacteria and viruses
    preceded most other life forms.
    (Mitochondria in human cells
    evolved from bacteria.)
    Human bodies can’t survive
    without bacteria, and they
    can’t survive without us.
    Some of our most essential
    functions are controlled by them.

    In this symbiotic entity
    we choose to call human,
    which of us controls?
    How do we coordinate
    the actions of us all?
    When we think, speak,
    write or touch, which
    one of us is in charge?
    Are we a committee?
    Are my bacteria or viruses
    writing this?

  16. keith cook + / -
    Posted August 21, 2016 at 6:02 pm | Permalink

    …” Given only that incomplete information—the information we actually have—it’s completely conceivable that we could have acted differently.”
    I take this to mean, simply, how we would act if we had all information available (quantum/brain states) but for me, no more or no less determined.
    I think the above is the door being left open for a possibility of a different decision being acted on because of all information being used, not partial information.

  17. JonLynnHarvey
    Posted August 21, 2016 at 6:15 pm | Permalink

    It’s not just a question of how much info we have.

    Some people at least feel that had more agency in some actions than in others. I suppose this could be a difference in what level the laws of physics are working- through a sense of agency or bypassing it, but it still makes a difference in moral accountability.

    In fiction, some villains do bad things with their entire “will”, others are half-hearted: experiencing some remorse, but along with a weakness of “will”.

    Shakespeare’s Richard III is knowingly “willingly” malicious from day one. Hamlet is not. Macbeth is slowly seduced into a vortex of evil-doing by the machinations of both the witches and his wife. Once he gets started, he cannot break the pattern.

    In modern fiction about serial murderers, Hannibal the Cannibal is more like Richard III. Norman Bates is a bit more Hamlet-like (except for the remake of “Psycho” with Vince Vaughn where his more knowingly malicious.)

    etc. etc.

  18. Posted August 21, 2016 at 6:21 pm | Permalink

    I interpret Sean’s account that it is conceivable that we could have chosen otherwise in the same way we say that any system with X degrees of freedom could have different outcomes when run with the same known inputs.

    Of course, it is the variables we can’t capture that determines the outcome, but given out vantage point, we don’t know anything about those variables.

    Take this scenario: you approach a traffic signal at 45 mph and your are 100 feet away when the signal turns yellow. There are no vehicles between you and the intersection and you know that if you stop at this light, you may miss your train. So you stomp on the accelerator and fly through the intersection just as the light turns red. The following day, you approach the same intersection at the same time driving the same speed, but you stop.

    Now, it’s conceivable given the known variables about the situation that you might stop or you might gun it. In any individual situation, it is determined by the laws of physics whether you’ll stop or go. But there’s no way to know this, thus either outcome remains conceivable.

    This isn’t just limited to human choice either. Take weather models. There’s many conceivable outcomes but only one determined one. When we say there’s an 80% chance of rain, it’s still conceivable that it will either rain or it won’t. Yet, perfect knowledge of all the factors at play would make the chance either 100% or 0%.

    • ChrisH
      Posted August 22, 2016 at 8:45 am | Permalink

      There are a whole lot of “unknown” inputs that fuel our decisions.

      So gunning it on one day but not on the following. What other motivations were there? Worried about work that particular day? A bit tired so couldn’t to be bothered to jump the lights on day 2? There’s a universe of other inputs!

      I don’t believe in libertarian free will for that very reason. Further then that? All a bit over my head.

      WRT the weather analogy. Well, sort of. It’s a binary possibility that it will rain or not, yes or no. Maybe it’s a wording thing.

  19. infiniteimprobabilit
    Posted August 21, 2016 at 6:40 pm | Permalink

    “if you’re one of those, you know what to do: just don’t read any further.”

    That implies, of course, that we have a choice. 🙂

    Think I’ll check out the next post – ooooh, kittehs!

    cr

    (A useful convention, as Sean says).

  20. Posted August 21, 2016 at 8:19 pm | Permalink

    There was a man who murdered his neighbor and at his sentencing he addressed the Judge: “Your Honor, I do not claim that I had no choice but to commit this crime or was not free to choose differently. In fact, I was not only free to choose differently but actually did make every possible choice that was compatible with the laws of physics. The proof of this is the actual existence of many worlds where much better choices were made. An unbelievably huge number of such worlds. My expert witness, Sean Carroll, will confirm this. In the overwhelming number of these worlds my behavior has been exemplary and I am consider to be a person of high morals. In fact, in one world I studied science and developed a new theory that was honored as being in the class of Newton, Maxwell, Darwin and Einstein. In the light of this, I ask that your sentence will honor me by granting me free meals for life.”

  21. Posted August 21, 2016 at 9:06 pm | Permalink

    I basically agree that in both passages, Sean seems to be changing his mind, or the subject, or just barking up the wrong free*. But then, I’m a determinist who doesn’t believe that we “couldn’t have done otherwise”.

    By the laws of physics, it can’t both be that the quantum state of me+environment at some time t before my decision was Q, and that I do something at t+dt other than A. But, the laws of physics don’t dictate that the state of me+environment had to be Q. The laws of physics don’t dictate boundary conditions. There are many histories consistent with the laws of physics. Some of them have me doing A, some doing B; most, of course, don’t have me at all.

    The following argument would be valid, if its premises were true:
    Necessarily, Q.
    Necessarily, if Q then A.
    Therefore, Necessarily A.

    But the first premise, in this context, is false. We do not have Necessarily, Q. All we have is Q.

    The modal fallacy is this pattern of reasoning:
    Q
    Necessarily, if Q then A.
    Therefore, Necessarily A.

    It’s tempting (when given in more plain language, anyway), but still a fallacy. There is no good reason to believe we couldn’t have done otherwise – even assuming the utmost determinism.

    *I stole the “barking up the wrong free” joke fair and square, from Peter Tse.

    For some barking up the right tree, I recommend Yudkowsky’s essay on what “can” means, as applied to agents.

    • Posted August 22, 2016 at 11:27 am | Permalink

      Pick an appropriate boundary condition for your life (or your “self”, if you prefer). Then the idea is that with that boundary condition (which was outside your control) you could not do otherwise or self originate.

      • Posted August 22, 2016 at 3:54 pm | Permalink

        I pick the boundary condition as my decision to reply to your comment, plus everything else that was going on at the same time (relative to my inertial reference frame). That doesn’t seem to be outside my control…

        • Posted August 23, 2016 at 1:04 pm | Permalink

          Everything in your reference frame is under your control?

          Besides, you misread my instructions. (My fault.) I meant what I said absolutely literally. The relevant boundary is your *life* (or the origin of your self) because those affect (“determine”) the resulting “you” and are outside, by hypothesis, of your control. This is why Kane needs something called a “self-forming action”, which is independent somehow of everything that has gone on before, and yet is integrated into what you are. Impossible, as far as I can tell, but he does try.

          • Posted August 24, 2016 at 11:25 am | Permalink

            *Something* in my current time slice is under my control. That’s enough to block the variations of the Consequence Argument seen on WEIT.

            As for Kane’s self-origination, well… some people crave hands absolutely lacking in bacteria, no matter how harmless. Slightly more on topic, some people crave in-principle unpredictability, when we already have enough practical unpredictability to foil any rival that can fit in our universe. And some people crave self-origination. Not every desire should be honored, much less shared.

    • Posted August 22, 2016 at 11:28 am | Permalink

      But, the laws of physics don’t dictate that the state of me+environment had to be Q.

      That’s more a statement about entropy than anything else. What you’re proposing is that there are multiple microstates with indistinguishable macrostates, which is, indeed, the case. You could rejigger the positions of the air molecules in your room in all sorts of ways and you’d have no way of telling the difference; all such configurations would have the same pressure, volume, number of molecules, and temperature and be apparently homogeneous.

      And, at a macroscopic scale, where we’d use the Ideal Gas Law, those different microstates really don’t matter one whit.

      We describe that as a condition of high entropy. Indeed, it’s the very textbook example.

      What you’re describing is a case where the entropy of the system is sufficiently low that the microstate does matter. Or a condition of low entropy.

      It’s also not difficult to have systems with low entropy where it’s difficult to observe the difference between the states. Chaos theory is great for describing such systems — a tiny difference in microstates leads to radically different evolution of the macrostates.

      So, back to the top…the laws of physics did dictate the exact state of the environment, of which you are a part. There could logically conceivably have been other states that would superficially have appeared to have been identical as far as your limited powers of observation go, but, in reality, there really was just the one actual state, and that state is the direct result of the evolution of the Schrödinger Equation according to the Universe’s Hamiltonian and wave function.

      Cheers,

      b&

      >

      • Posted August 22, 2016 at 4:08 pm | Permalink

        As I suppose Sean must have pointed out in his new book (I’ve only read From Eternity to Here) we live in a time of moderate entropy, as all life forms pretty much must. The Big Bang conditions are wrong for life, and basically nothing happens during the heat death of the universe (no Boltzmann brains). And moderate entropy has both the conditions you describe – multiple realizability, and chaos – thus far, I think we agree?

        What I don’t get is how you derive “the laws of physics did dictate the exact state,” unless you mean, the laws of physics plus other boundary conditions. But of course, as I explained above, that plus makes all the difference. Interestingly, one can pick boundary conditions in the middle of history – and specifically in the middle of one’s life. And for practical purposes of decision making, that’s exactly where one should pick them.

        • Posted August 22, 2016 at 4:23 pm | Permalink

          That reads to me like you’re distinguishing between the Hamiltonian and the wave function…which, of course, is important if you want to do physics, but seems completely irrelevant in the context of this discussion.

          Sure, if the wave function of the Universe were different but the Hamiltonian remained the same, then you’re going to be making different decisions (assuming you even exist in the first place, for any meaningful definition of those terms). But that’s no different from Jerry’s example of rewinding the tape of history and seeing it play out differently.

          Again, yes — all sorts of macroscopic states are indistinguishable to us, and give the superficial appearance of being “boundary conditions,” as you put it, that look identical to us. But that’s because we’re small and stupid, not because they actually are the same.

          Indeed, part of the fun, part of the excitement in living in this universe is that we are small and stupid, and so we don’t know what the microstate actually is and can’t extrapolate the eventual anticipated macrostate — at least, not with any precision.

          But we manage to do remarkably well, with our own internal imagined copies of the Universe; well enough to function. But also poorly enough for us to be surprised and to have all sorts of chances to improve our mental models of the Universe. And, just as Sisyphus isn’t ever going to get that rock to the top of the mountain, neither are we…and that’s a good thing, because all the fun is in the climbing!

          Cheers,

          b&

          >

          • Posted August 22, 2016 at 8:55 pm | Permalink

            To keep it simple for me (my QM is weak) put it this way: the Schrödinger equation could hold for a different wave function of the universe. But I’m not trying to ignore the microscopic details of the wave function. I’m trying to point out that *they depend on us*. If we weren’t the way we are, the wave function of the universe couldn’t be the way it is. Therefore, the wave function is in no position to steal our thunder, to rob us of agency or choice. We have met the enemy, and he is us. There is (note, not could be: is) only one future, but it depends on us.

  22. Ian H Spedding
    Posted August 21, 2016 at 9:11 pm | Permalink

    If we cannot be held responsible for our actions because we had no choice in the matter, then not only is it irrational to punish offenders for crimes they commit but it must also be irrational to reward people for their achievements. So maybe it’s time to wind up the Olympics for good and have all those medals handed back.

  23. Gregory Kusnick
    Posted August 21, 2016 at 9:21 pm | Permalink

    Knowing that love rests on chemistry […] doesn’t have any clear or important implications for society.

    Doesn’t it? Seems to me there would be a lot fewer failed marriages and traumatized children if young adults took a more realistic attitude toward sexual infatuation and didn’t mistake it for some sort of spiritual destiny. From a utilitarian perspective, there’s probably at least as much net improvement in global well-being to be had here as from dispelling illusions about libertarian free will.

  24. Kevin
    Posted August 21, 2016 at 11:22 pm | Permalink

    Our knowledge is incomplete. Physics is deterministic. Free will also be an epistemological problem. We simply have incomplete knowledge of the determined universe…hence. Comparibilism.

  25. mpzrd
    Posted August 22, 2016 at 12:11 am | Permalink

    As to the mystery paragraph, you think that “predictability” is moot, which seems an odd thing for a scientist to say since scientific proof is about the ability to make accurate predictions, suitably circumscribed. There is no thought we can untangle the neural events that led to the decision, eg for wicked downloads, so as a practical matter we make no avoidable mistake by stuffing all those unobservable processes into a black box labeled “psyche” (or “personality” )(which is less presumptive than “libertarian free will”).

    As to the matter of guilt, the responsibility of the citizen with a wicked compulsion is to talk about it rather than act on it. Sure to the limited practical extent we can identify (and fix!) environmental or genetic causes (there would also be accidents during development), we should, but the individual is one of those working on their treatment so can’t evade practical responsibility for whatever the social meaning of the act.

  26. Joseph Lapsley
    Posted August 22, 2016 at 6:05 am | Permalink

    The “convenience” of the language of choice, and the degree of information awareness that went into a choice aside, it seems that a thorough-going determinism (not just structuring limitations on choices, ala compatibilism) based on a script defined by physics from presumably the Big Bang forward makes the universe fundamentally, and rigidly, teleological. That sounds more compatible with theism than atheism, and more specifically the predestination theories of Calvinism. Most theists are teleological thinkers. Darwin was not. Where is the random in random mutation, in such a scripted universe?

  27. Arno Matthias
    Posted August 22, 2016 at 6:37 am | Permalink

    In his book presentation at Google (https://www.youtube.com/watch?v=x26a-ztpQs8) Carroll says that when you want to model/explain/predict human behaviour, and you neither have complete knowledge of the relevant physical state of the universe, nor a good theory about the emergent phenomena that would arise from this particular state, then the language of “choice”, “decision” and the like, is useful.

    • Kevin
      Posted August 22, 2016 at 11:26 am | Permalink

      I would go further and say not just useful, but unavoidable. Without the ability to know future states, the future is unknown to us, which make any decision we make indisginquishable from a decision made by someone who thinks they have free will.

      We simply do not know the future well enough to avoid this circumstance.

  28. GBJames
    Posted August 22, 2016 at 7:06 am | Permalink

    I’m coming in late, having spent yesterday at Milwauke Irish Fest listening to great music. I had no choice in the matter, having married an Irish descendant.

    In any case, I’m pleased to see this post because I really enjoyed reading Carroll’s book a few weeks back.

    With regards to the issue at hand, we should remember that his central point is that the language of free will is a useful way to talk in our everyday lives. In everyday affairs we “choose to act” all the time, even if it all boils down, ultimately, to the laws of physics operating on stardust.

    When Carroll says “…it’s completely conceivable that we could have acted differently…” I think he is simply saying that while we know that ultimately all is determined, we think and talk otherwise, because it is a useful way to think and talk.

  29. Dimitris Klaras
    Posted August 22, 2016 at 7:27 am | Permalink

    To me all this “free will/choice” discussion makes no sense. People propose constantly with a “choice language” that there is no real choice. To me what Jerry, Harris etc suggest is a very simplistic and easy “determinism” which simply doesn’t work. Classic bottom-up approach with very inadequate tools of knowledge. What I suggest is to accept what comes natural to us. That there is a reality like the one we experience: Maybe our mind adds a secondary layer/interpretation that can be considered beneficial to survival, like the mapping of colors to ranges of frequency. Doing so we are able to proceed in a top-down approach to explore this reality. With this “deterministic” view we forbid that to ourselves. We are in an intellectual standstill. So the dilemma to me is this: We accept what comes natural to us and explore it or… embrace chaos willingly and spend our lives apologizing for our nevertheless “choice language” putting ourselves in this unfortunate situation where in words declare “determinism” and in practice forget it completely. Or… in everything else except the penalizing of criminals!

    • Dimitris Klaras
      Posted August 22, 2016 at 7:42 am | Permalink

      To express it differently, is a dilemma about declaring “determinism” and so eliminating a problem or declaring the problem and challenging it’s solving/exploration.

  30. Posted August 22, 2016 at 9:08 am | Permalink

    “But there’s a difference. Knowing that love rests on chemistry (once a metaphor, now a reality) doesn’t have any clear or important implications for society. That’s not true, for free will…(which) has huge implications for society”

    Three things wrong here:
    1. If one is an incompatibilist there should be no different consideration between love and free will. If free will is an illusion by incompatibilist standards, then love is certainly also an illusion. We must not “cherry pick” illusions between ones we don’t mind accepting and those we do mind. I can certainly not see how anyone can remove love out as not having a significant effect on human social and moral systems. Love significantly effects our actions and commitments to others and “action from love” needs to be reconsidered from the perspective that one is actually being manipulated by an illusion. Love based actions must therefore be considered in a way that is more game-theoretically advantageous.

    3. When we compatibilists bring up the negative impact of eliminating any concept of free will (which is significant) we are immediately accused of “arguing from consequences”. Yet the foundation argument incompatibilists themselves use for eliminating the free will concept is itself based on consequences e.g penalty and retribution.

    I just wish incompatibilists would at least be consistent in the way they present their arguments.

    • Posted August 22, 2016 at 11:36 am | Permalink

      If free will is an illusion by incompatibilist standards, then love is certainly also an illusion.

      The difference, at least for me, is that I don’t see how “Free Will” can even make sense in a supernaturalistic dualistic context, let alone the real world. I’m really not kidding when I equate it to married bachelors

      I have no trouble defining love as, for example, the condition in which another person’s happiness and wellbeing is essential to your own. But I really still don’t have a clue what “Free Will” is even supposed to be in reality, other than a religiously-tinged reification of the decision-making process.

      Of course, if you wish to point out that “love” is emergent in the same sense that the Ideal Gas Laws are, that the atoms of the brain have no more “love” in them than the atoms in a balloon have “pressure,” I’d agree with you in an heartbeat. And I’m certainly willing to grant that you think there’s some similar emergent phenomenon which is really real and best described by the label, “Free Will.” The disagreement is whether or not such a phenomenon actually exists (or even hypothetically could exist) in the first place and whether or not “Free Will” is a suitable label for it if it does.

      Cheers,

      b&

      >

      • Posted August 22, 2016 at 12:44 pm | Permalink

        “And I’m certainly willing to grant that you think there’s some similar emergent phenomenon which is really real and best described by the label, “Free Will.”

        Yes Ben, but let me warn you that there is also a similar emergent phenomenon which is called “love” and that I understand that you are about to commit significant resources and energies because of that illusion, particularly in terms of a forthcoming marriage. Be warned- our primary goal in our physical existence is expressed as FITNESS- in our survival and MAXIMIZED reproductive success. The feeling you call love is a biological reenforcement of pleasures at an EMOTIONAL level innate within you to achieve your GENES best fitness, but not necessarily your own. A better Evolutionary Game Theory based strategy is to deceive your future spouse (it’s called false signalling) and energetically try to mate with as many other partners as possible. Also be careful that expenditure of your resources is not clouded by this love illusion.

        Just joking Ben, as a compatibilist I don’t believe we can base our life strategies on physical levels of interpretation. There are such emergent phenomenon as love and free will and they are an essential construct of our functioning as human beings human beings. These constructs, taken together, form a coherent whole. Separating out some as illusions, some not makes the whole system incoherent. We are physical entities indeed, obeying the laws of physics and subject to determinism. But our human lives are much more than this.
        Wishing you all all the happiness that love, (as we feel it at a human level) can afford
        Cheers
        Howie

        • Posted August 22, 2016 at 1:02 pm | Permalink

          Thanks for the warm wishes…and, again, if I knew what phenomenon you were pointing to with the term, “Free Will,” and could agree that that term was an appropriate label…well, of course I’d be in agreement with you.

          The difference is that we do agree (at least roughly) about the term, “love,” but do not agree about, “Free Will.”

          When you understand how I differentiate between those two phenomenon, you’ll understand how I can embrace the one but reject the other.

          Cheers,

          b&

          >

  31. T Hupy
    Posted August 22, 2016 at 9:23 am | Permalink

    I agree with you and had same questions about this. Sounded like Sean was saying that we can talk as though we have free will because we feel we do. But seemed he was also saying that it was an emergent phenomenon.

    • Gregory Kusnick
      Posted August 22, 2016 at 11:12 am | Permalink

      He is saying that it’s an emergent phenomenon, that as such it’s as real as baseball, and that’s why it’s legitimate to talk about it as a real thing.

      • peepuk
        Posted August 23, 2016 at 9:17 am | Permalink

        Does he belief that the belief in Free Will is a physical object? That would be very strange.

        This piece makes the usual compatibilist error, thinking that the perception of X is the same as X.

        Look at the wikipedia entry of sound; you will discover 2 definitions:

        a) In physics, sound is a vibration that propagates as a typically audible mechanical wave of pressure and displacement, through a medium such as air or water.

        b) In physiology and psychology, sound is the reception of such waves and their perception by the brain.

        Only the first explanation is about the real sound; the second one is about our perception of a real sound; this part is not physical; it is emergent or imagined.

        Free Will may be an emergent property but that doesn’t make it real.

        • Gregory Kusnick
          Posted August 23, 2016 at 11:11 am | Permalink

          I think most people here would dispute your claim that perception isn’t physical.

          And if only physical objects are real, where does that leave phenomena like entropy, natural selection, or radioactive half-lives? Or baseball, for that matter?

          • Posted August 23, 2016 at 1:07 pm | Permalink

            peepuk might be meaning that perception is not physical sensu Bunge (and others) which is as opposed to chemical, biological, social, etc. and *not* as a synonym for “material”. This is why the latter is useful, to boot.

          • peepuk
            Posted August 23, 2016 at 2:08 pm | Permalink

            We know that our belief in Free Will is produced by physical processes in our brain.

            Our brain produces a lot of things that don’t exist (God, Santa Claus, soul); there is absolutely no reason to belief that Free Will is an exception.

            Do you really belief that Free Will is (like a baseball) a physical object?

            • Joseph Lapsley
              Posted August 23, 2016 at 4:12 pm | Permalink

              How does reducing everything to physics, a form of physics-Calvinism, explain anything? Historians can’t point to biology to explain history, why should biologists and chemists point to physics?

              • peepuk
                Posted August 24, 2016 at 10:34 am | Permalink

                It doesn’t mean biologist, economist or psychologists should start with fundamental physics; that wouldn’t be very practical. But all knowledge they gather is ultimately constrained by what’s physically possible.

                And if you are not a dualist all the things that objectively exist in the real world have to be made of physical stuff.

            • Gregory Kusnick
              Posted August 23, 2016 at 4:51 pm | Permalink

              The game of baseball isn’t a physical object; it’s a phenomenon of human behavior.

              So is the fact that people label some instances of behavior as freely willed, and that there is broad cross-cultural agreement about which behaviors should be so labeled. (So that’s something free will has that God and Santa Claus don’t have.)

              Putting aside the semantic quibbles about whether it’s really “free”, the fact remains that such classification exists and says something useful. So in that sense, yes, “free will” is as real as baseball (the game, not the actual ball).

              • peepuk
                Posted August 24, 2016 at 10:49 am | Permalink

                Ok, I understand.

  32. Posted August 22, 2016 at 11:01 am | Permalink

    The incompatibilist dismissal of ANY form of free will always reminds me of that endearing anecdote that Dennett tells about the questioning of Lee Siegel’s publication of her book on magic:

    Siegel: “I’’m writing a book on magic, I explain, and I’’m asked, Real magic? By real magic people mean miracles, thaumaturgical acts, and supernatural powers. No, I answer: Conjuring tricks, not real magic. Real magic, in other words, refers to the magic that is not real, while the magic that is real, that can actually be done, is not real magic.”

    Free will is something that “can be done”.
    It is incompatibilists that want free will to be “real magic”, only because they WANT it to be impossible. The fact that a free-will “magic” can actually exist (be “real”) but only in the terms that compatibilists use, makes incompatibilists the people who don’t want to deal with “REAL reality”.

    • peepuk
      Posted August 22, 2016 at 1:11 pm | Permalink

      I want the real thing and I want it to be really possible, but science tells us that’s not possible.

      It could be worse; me thinking that fake Free Will is real Free Will.

      Or even worse; we could mislead people; selling fake Free Will as real Free Will.

      • Gregory Kusnick
        Posted August 22, 2016 at 11:30 pm | Permalink

        I don’t see the appeal of “real” free will. If what you want is for your conscious deliberation to play a role in your decision-making, you have that now, and there’s nothing fake about it.

        If you don’t want to be a puppet, all you have to do is notice that there’s no puppeteer, no external agency whose purposes override yours.

        If for some reason you want decisions that come out of nowhere, unmotivated by your thoughts, desires, or anything else that goes on in your brain, you can have that by flipping a coin.

        • Posted August 23, 2016 at 1:09 pm | Permalink

          Some people do not want to be a “puppet” to states outside their control period, not just agents. In which case, their formative environment, the current external state, etc. are just as bad. (I wouldn’t mind being an unmoved mover too, to have self-origination, but I think it is by now dangerously silly to think it is possible.)

          • Gregory Kusnick
            Posted August 23, 2016 at 5:39 pm | Permalink

            This still doesn’t answer the question of why self-origination is something worth wanting. What does it buy you, other than bragging rights?

            If what you want is the ability to redesign yourself according to some specified set of values, which values should those be? If it’s the values you hold now, how is that meaningfully different from wholesale self-modification (which might actually be possible with advanced technology)? If it’s the values you would hold if you were self-originating, that’s not just impossible; it’s incoherent.

  33. Shwell Thanksh
    Posted August 22, 2016 at 8:03 pm | Permalink

    It’s only tangentially related, but I just finished an sf/fantasy novel I really enjoyed, and I always love to pass that along when I can.

    The central conflict (without any spoilers) revolves around attempts to thwart an evil villain genius whose nefarious plan will ultimately result in the loss of free will for all humanity, as a side-effect of his acquiring god-like powers.

    It’s an enjoyably well-written, rollicking, steampunk/fantasy clockwork gangster sort of British yarn, and I recommend it despite the need to suspend disbelief on the free will fuzziness of the premise.

    Angelmaker, by Nick Harkaway

  34. Posted August 22, 2016 at 9:33 pm | Permalink

    I am not sure what SC means in that paragraph, but I am willing to bet it has something to do with coarse graining.

  35. Posted August 23, 2016 at 1:02 am | Permalink

    To claim one has free will is to be in denial that physics is “driving” your decisions. Also, to believe or speak of having “ability” has the same flaw. If you think you have capacities or abilities, you are in denial that physics is driving whatever happens in your life.

  36. peepuk
    Posted August 23, 2016 at 4:20 am | Permalink

    What evidence would convince a compatibilist that Free Will doesn’t exist?

    And what evidence would convince a incompatibilist that Free Will does exist?

    • Posted August 23, 2016 at 9:54 pm | Permalink

      Yes, that is exactly the right question to ask. I am a free will agnostic (I simply don’t know, and neither pro nor con has made a convincing case to me), but if someone could write a program that always predicts my choices and I cannot outwit it, I would agree that I have no free will.

    • strongforce
      Posted August 23, 2016 at 11:09 pm | Permalink

      +1

    • Posted August 24, 2016 at 10:17 am | Permalink

      At the risk of breaking the Roolz, I humbly submit that these questions would make for a great “above the fold” post. I would be very interested in reading Prof CC’s answer and an entire comments section worth of answers as well.

    • Gregory Kusnick
      Posted August 24, 2016 at 6:19 pm | Permalink

      I’ll take a stab at answering peepuk’s question.

      To start with, as a compatibilist I’m already convinced of the nonexistence of libertarian free will. I don’t need any further evidence of that.

      So I take it that the question is in some sense about language: what does “free will” mean in ordinary usage (as distinct from philosophical or theological debates), and when so used, does it refer to some real aspect of human behavior?

      To find out, let’s imagine we have a large database of specific instances of behavior, tagged with various factors relating to the circumstances of each instance. Citizen volunteers (not philosophers or theologians) have gone through the database and labeled each instance as either freely willed or not.

      We then run an analysis on the data to see if the “free will” labels correlate with any objectively defined factor (or combination of factors) in the circumstances of each instance. Suppose they do; let’s call that factor F.

      Now we’re in a position to say, empirically, that when people talk about free will, what they’re referring to is factor F. This constitutes evidence that something called “free will” exists: factor F exists, and we have evidence that people call it free will. I would hope that evidence of this sort would be convincing to incompatibilists.

      Conversely, if despite our best efforts we fail to identify any factor F that correlates significantly with the “free will” label, that would be evidence against the existence of anything called “free will”, and compatibilists would be obliged to accept that the label is semantically empty.

      Two additional points to note about this thought experiment:

      First, what people believe about the mechanism of free will — whether it’s natural or supernatural — is irrelevant to the question of whether the label picks out a real phenomenon.

      Second, in the event that factor F does exist, saying that people ought not to call it “free will” is like saying they ought not to split infinitives or end sentences with prepositions. It’s just a peeve about usage, not an argument against the existence of factor F.

      • Posted August 24, 2016 at 6:34 pm | Permalink

        First, what people believe about the mechanism of free will — whether it’s natural or supernatural — is irrelevant to the question of whether the label picks out a real phenomenon.

        Second, in the event that factor F does exist, saying that people ought not to call it “free will” is like saying they ought not to split infinitives or end sentences with prepositions. It’s just a peeve about usage, not an argument against the existence of factor F.

        This is exactly why I like the analogy with Elan Vital.

        If we re-jiggered your proposed experiment to instead probe the nature and / or existence of Elan Vital, you may very well come to a conclusion of its existence as an actual real phenomenon that takes the form of a continuum where bacteria have barely any and humans have more than any other species. You still might not know what it is, but you’d have a clear indication that there’s something consistent and real it’s describing.

        But would you be okay proposing that we should therefore keep Elan Vital as part of our modern biology-informed lexicon? What would it really mean to insist that, yes, Elan Vital is a meaningful concept because it roughly correlates with biological complexity? Would it be just a pet peeve about usage to insist that we should drop the concept of Elan Vital entirely in favor of an accurate description of science as we actually understand it?

        Obviously, I submit that “Free Will” has no more place in modern cognitive neuroscience than “Elan Vital” does in modern biology — whether or not you can get repeatable consensus on consistency with some objective measurement.

        Cheers,

        b&

        >

        • Posted August 24, 2016 at 7:31 pm | Permalink

          Wiki says:

          Élan vitale was coined by French philosopher Henri Bergson in his 1907 book Creative Evolution

          So, not part of everyday speech originally, but a pseudo-scientific hypothesis by a philosopher. Terms like that, which have gotten an authoritative definition from a single author, allow a meaning distinct from the general public’s everyday usage. For run of the mill terms, Gregory’s approach makes more sense.

          For example, for a long time many or most people thought whales were fish. If asked for definitions they might well propose “large fish with a horizontal tail” or some such. But, discovering that there is nothing answering to the definition is not the same as discovering there are no whales.

          • Posted August 24, 2016 at 8:30 pm | Permalink

            So, not part of everyday speech originally, but a pseudo-scientific hypothesis by a philosopher. Terms like that, which have gotten an authoritative definition from a single author, allow a meaning distinct from the general public’s everyday usage.

            Well, if that’s the direction you want to take the discussion…you’ve also got to account for the fact that the general public’s everyday usage of the term is in no small part the solution to theodicy. And, indeed, in which the actual phenomenon us rationalists agree exists is the exact example typically used in everyday discourse of what “Free Will” is not.

            So, either we take my original interpretation of Gregory’s proposal, in which case we’re left with an actual phenomenon for which the label, “Free Will,” is as useful as, “Elan Vital“; or we take your interpretation of going with the common definition, in which case “Free Will” is as much a theological construct as anything.

            That’s the circle you’re trying to square.

            And even if you win this battle…then what? Do you really want to try to convince religious people (80%+ of Americans!) that they really do have “Free Will,” only “Free Will” is the exact opposite of what they think it is, and the fact that they do have “Free Will” that isn’t “Free Will” says nothing about why bad things happen to good people?

            Really, just how, exactly, is that conversation supposed to play out without sounding like a bad parody of Monty Python?

            Cheers,

            b&

            >

            • strongforce
              Posted August 24, 2016 at 9:15 pm | Permalink

              Ben, over the last two recent WEIT blogpost devoted to Free will, I keep thinking that your analysis and comments are so devastatingly and undeniably correct, surely this will settle the argument. But alas this appears not to be the case.

              This last comment to Paultorek is resplendent. It really should be the last word. My appreciation for all of your efforts and the insights I have gleaned.

              rick

              • Gregory Kusnick
                Posted August 24, 2016 at 9:25 pm | Permalink

                “It really should be the last word.”

                We can hope!

              • Posted August 25, 2016 at 1:41 pm | Permalink

                I think one can confidently predict that the current microstate of the Universe will evolve into a macrostate in which people will continue to discuss whether or not our current microstate would have evolved differently from how it actually did evolve, Laplacian determinism be damned.

                Hope all you want…but only if you want your hopes dashed like Babylonian babies against the stones.

                Cheers,

                b&

                >

  37. Andy
    Posted August 24, 2016 at 8:24 pm | Permalink

    On the basis of this, I bought Sean’s book, which is really well written and I’m enjoying a lot! So, thank you!

    To me, the key point is this bit, “Given the quantum state of the elementary particles that make up me and my environment, the future is governed by the laws of physics. But in the real world, we are not given that quantum state. We have incomplete information;”

    And, in a nutshell, that’s it.

    Ultimately, if you want to do science, you need something testable. If your pet theory relies on an infinite array of many other worlds or things that are not even -in principle- measurable, then it might well be right, but in an important sense it’s not really scientific.

  38. Posted August 25, 2016 at 7:24 pm | Permalink

    Determinism goes back to Democritus, and Aristotle may have been the first compatibilist, although that’s controversial. The phrase “libera voluntas” goes back to Lucretius, who wasn’t bothered about theodicy. He was, though, some kind of libertarian, so, point for you, I guess.

    The authority of the person who coins a phrase only lasts while the language community defers to them. We still talk about erotic love even though few believe in the god Eros. So, getting back to “what would convince you”: if you can show me that the general public defers to theologians on the reference of “free will”, then I’ll abandon the term. This includes the uses of “free will” in the penal code. After all, Jerry argues that we should change our jailing practices not just because some of them are cruel (which I agree some are), but also because nobody is morally responsible. And Sam Harris argues that every human action is comparable to those that are driven by brain tumors, which conflicts with the current practices of the courts.

    • Posted August 25, 2016 at 7:31 pm | Permalink

      Er, that’s supposed to be a reply to Ben’s comment of August 24, 2016 at 8:30 pm. And confirmation of his later prediction 🙂

    • Posted August 29, 2016 at 12:12 pm | Permalink

      This includes the uses of “free will” in the penal code.

      Do you hold similar standards for other examples of specific jargon? The law distinguishes between assault and battery, between libel and slander. Do you? Are you disturbed by the fact that judges in courtrooms don’t hand out blue ribbons to their favorite pies? Do you call 9-1-1 when somebody says something about murdering a performance of Tchaikovsky? Do you ever take a constitutional that’s just a walk and has nothing whatsoever to do with the highest law of the land?

      Cheers,

      b&

      >

      • Posted August 29, 2016 at 8:34 pm | Permalink

        It’s a plausible hypothesis that “free will” is jargon, when used by lawyers, and not closely related to everyday usage. It’s also a plausible hypothesis that it’s not. Jerry argues that because free will (everyday sense) doesn’t exist, some related legal practices need to change: which implies that there is some pretty close relationship there. You might disagree. I agree that there’s a relation, but I also think most lawyers and judges view “legal free will” in compatibilist fashion. I think everyday people define free will in a mostly-ostensive fashion, pointing at the experience of choice, with little in the way of verbal definition or deference to ancient etymology.

        The experience of choice might be deeply deceptive, and I can easily imagine some followups to the Libet experiment might turn out certain ways, which would show all “decisions” to be epiphenomenal. Or they could show otherwise. May the best science win.

        • Posted August 30, 2016 at 7:57 pm | Permalink

          The legal term is very clearly and explicitly defined to distinguish from actions performed under duress. Indeed, courts will recognize that you can, indeed, choose to disobey a gunman’s orders, to use the classic example. The definition in this case gives you the option of obeying the gunman’s orders without suffering additional legal consequences, in a somewhat ironic twist that gives more legal freedoms in a situation described as one lacking such.

          All that is pretty clearly unrelated, except most distantly, from Jerry’s legal musings, as well as from philosophical, theological, and everyday usage of the phrase.

          Cheers,

          b&

          >

          • Posted August 31, 2016 at 8:50 pm | Permalink

            So you’re saying, that if the judge asks a defendant, “did you take the money from the vault of your own free will,” he technically legally shouldn’t say “No Your Honor, I did it because the gunman ordered me to.” He should say, “Yes but, I was under extreme duress”? If that’s right, then you’re right, the legal definition of free will is very different.

            Wikipedia writes:

            The basis of the defense is that the duress actually overwhelmed the defendant’s will and would also have overwhelmed the will of a person of ordinary courage (a hybrid test requiring both subjective evidence of the accused’s state of mind, and an objective confirmation that the failure to resist the threats was reasonable), thus rendering the entire behavior involuntary.

            Wait, what?

            • Posted August 31, 2016 at 10:48 pm | Permalink

              In a court of law, you’d use the term as defined by the law. So, no; he didn’t take the money of his own free will because he was under overwhelming duress such that a person of ordinary courage would not reasonably have acted otherwise.

              But, don’t you see it, right there? The counterfactual?

              If the person were of extraordinary courage, he might have chosen the path with a strong chance of death instead. But the law recognizes the unreasonableness of demanding that the least of us must act as the greatest might hope to, and so we don’t punish people for making reasonable choices in unreasonable situations.

              All this is in stark contrast to the gunman who, for the sake of the argument, we shall assume does not himself have anybody threatening to kill him. A determinist will recognize the inevitability of the chain of events that leads him to point the gun at his victim’s head, but that chain of events does not involve some other person directly and immediately threatening his life (or the lives of his loved ones, etc. — again, just for the sake of the argument). As such, there is no duress. And, so, legally, he acted of his own free will — even despite the fact that the religious and philosophical notions of “free will” are entirely incoherent and utterly inapplicable to the law.

              Again: does Jupiter have any sort of “free will” in following its orbit? Because the gunman and victim have exactly the same sort of “free will,” whatever it is, in their own actions. The difference is not in the fundamental nature of action, but whether or not one person is threatening another.

              Cheers,

              b&

              >

              • Posted September 1, 2016 at 8:20 pm | Permalink

                On your view of what “free will” means philosophically, which both hard determinists and libertarians would agree with, sure, it’s way different from the legal meaning. On a compatibilist view, it’s not so different after all. The “will” part means the agent has goal-states, and the “free” part means they can reason through alternative plans, evaluate likely consequences, and select their favorite, thereby setting that plan in motion. We think these features (or something in this neighborhood) suffice to explain, without explaining away, the experience of deciding. That experience being the point of origin of the whole “free will” concept.

                Some libertarians, and some hard determinists, have argued that no, the everyday experience of deciding requires something more: a dualistic soul, or a causa sui, or both. They appeal to particular details of experience, or they do cognitive psychology, trying to show this. And when I read those arguments I think “Hooray, at least they’re engaging the actual argument!”

              • Posted September 2, 2016 at 11:13 am | Permalink

                On your view of what “free will” means philosophically, which both hard determinists and libertarians would agree with, sure, it’s way different from the legal meaning. On a compatibilist view, it’s not so different after all.

                See, that’s the whole problem with the compatibilist position.

                The overwhelming majority, especially historically, have been libertarians, with basically all the rest as determinists. We’ve been having this discussion for millennia.

                And, just very recently, a tiny and obscure set of people who call themselves, “compatibilists” have decided to invent their own self-admittedly-different definition of the term. And then they get very upset when nobody has any clue what on Earth they’re talking about.

                We think these features (or something in this neighborhood) suffice to explain, without explaining away, the experience of deciding.

                Nobody’s disputing the fact that we have decision-making processes that we experience intimately. Everybody from antiquity to modernity on every side of the discussion agrees that that’s something we all experience.

                The “free will” debate isn’t at all about whether or not we have that perception, or even explaining the mechanism.

                The “free will” debate is all about whether or not the “free” part of the term is illusory.

                Here — let me distill it down…with an experiment! We’re all good empiricists here, right? Should we not be willing to give credence to observation, even if the observation counters intuition?

                So: pick a number between 1 and 10. The number doesn’t matter; just pick it and write it down or whatever else you need to do to commit yourself to that number.

                Got your number?

                Mine is, “4.”

                As a determinist, it is my observation and conclusion that there is no other choice I could possibly have made at the moment I typed the “4” — because we ran the experiment, and there’s the results. We could re-run the experiment, of course, but we can’t exactly duplicate it. And, we know from the rest of physics that, if we could magically reset every parameter back to exactly as it was, I’d type the exact same “4.”

                A libertarian would disagree, of course, and cite the fact that he can imagine having picked other numbers as evidence that such a choice was possible. By the same reasoning, I can imagine having grown up a farm boy repairing vaporaters and running into a crazy old guy in the mountains with a burning sword; therefore, I am Luke Skywalker.

                The rounding-error compatibilist says that I imagined various number choices before I settled on, “4,” and that this precognitive imagination is somehow more real than Tatooine, even though, after I’ve made my choice, Tatooine vanishes in a puff of hyperspace exhaust. And then gets upset when both the determinists and libertarians can’t make sense of what point the compatibilist thinks he’s making. What is it, exactly…we really imagine things but what we imagine isn’t necessarily real? And that’s relevant and / or profound…?

                Does that help?

                Now, here’s the kicker. The experimental results still stand; there’s my “4” up there. I haven’t magically rewound the clock (or whatever) to pick a different number. Neither, I presume, has anybody else who has performed this experiment.

                And I would submit that that, right there, is irrefutable empirical evidence against any and all claims of “free will,” whatever form you think it has. We can separately investigate the decision-making process, but the question of “free will” is unambiguously settled in the negative by this very experiment — exactly the same way that Michelson and Morley demonstrated the non-existence of the Luminiferous Aether. (Even if somebody from a century or two ago might see a vague similarity between the Aether and modern Quantum Field Theory.)

                Cheers,

                b&

                >

              • Posted September 3, 2016 at 3:33 pm | Permalink

                Yes, the free will debate is all about whether the “free” part is illusory. But an illusion requires two things: it requires the facts to be one way (here, determinism), and it requires the appearance to be the other way. Not: a deductive inference that people often make mistakenly, but the experience itself, has to “say something” different from the facts. Compatibilists are those who noticed that experience doesn’t portray choice as indeterministic. The reason compatibilism in its current form is recent, is probably because science changed our understanding of causality.

                Ever seen the “moon illusion”? According to this NASA website, the “illusion” happens when the moon is on the horizon. I’ve had the privilege of seeing this “illusion” in full force, when the moon was framed by a tree lined city street: the moon looked – exactly how large I couldn’t have said – but larger than the whole city! But hey, the moon is larger than my whole city! The facts, and the appearance, say the same thing. It’s when the moon is high up in the sky, that our perception is illusory.

                So maybe the NASA writer made a mistaken deductive inference, that when the moon is on the horizon, we think it has grown from its normal. And maybe some people do make that mistake. But that’s a mistake not conveyed by the experience, but made in later thoughts about the experience. If you get “what the experience says” wrong, you can either desperately try to prove that what it says is true, or claim that the experience is illusory – but it’s better to get it right. (The “moon illusion” stuff features two conflicting experiences. The “free will” stuff does too – our experiences of time and causality contain illusory elements.)

                The choice experience conveys that acts of imagination are influential on our choice, not that the contents imagined are real, or exist in some parallel universe. As I expect you’ll agree, acts of imagination often allow us to avoid bad choices. In short, determinism isn’t fatalism. Likewise, it isn’t puppetry. It isn’t mind control. It’s none of the bogeymen that libertarians claim it to be. It would be, if libertarians were right about what the experience of choice portrays. But they’re not.

              • Posted September 5, 2016 at 12:48 pm | Permalink

                it requires the appearance to be the other way. Not: a deductive inference that people often make mistakenly, but the experience itself, has to “say something” different from the facts

                You’re contradicting yourself. If you deductively infer something different from the facts, your experience is saying something different from the facts. In the famous Müller-Lyer illusion, the two lines are projected as the same length on the retina, but the experience says they’re different.

                the moon looked – exactly how large I couldn’t have said – but larger than the whole city

                Indeed, the surface area of the visible portion of the Moon is about the same as that of the entire continent of Africa. For you to perceive it as merely larger than the whole city is every bit as illusory to perceive a mountain as merely larger than a molehill.

                Compatibilists are those who noticed that experience doesn’t portray choice as indeterministic.

                Yet libertarians will positively insist, up and down, soup to nuts, that they are free to make any choice they wish, and they will cite their own experiences to support their position. And here you are telling them that their decisions are deterministic, the exact opposite of their experience, and you’re also telling them that what they’re experiencing is subjectively deterministic, when they’re telling you that’s not what they’re experiencing.

                Who are you to know better than libertarians what they’re experiencing?

                Now, you and I will both agree that the decision-making process really is deterministic, and that appearances of indeterminism are illusory. The question is why you insist on telling libertarians that they’re really experiencing things deterministically even though they insist otherwise, and why you insist on using the language of indeterminism when describing the phenomena to determinists.

                Is it not much more straightforward to directly address the misconceptions of libertarians and, separately, to describe cognition in terms unencumbered by ancient superstition?

                Cheers,

                b&

                >

              • Posted September 6, 2016 at 6:35 pm | Permalink

                Perceiving something as X (and one did not perceive either Y or not-Y), when it is both X & Y, is not an illusion.

                Descriptions of experience can disagree because different people experience different things, or because someone describes their experience with the wrong words. Overheard at a restaurant: “What is that spice? I think it’s cinnamon.” “No, it tastes like nutmeg.” “You’re right, it tastes like nutmeg.”

                Libertarians and other normal adults *can* make any choice they wish. Which means: had they endorsed that other choice, they would have embarked on that other plan. What they can’t do is . But unknown facts, by definition, aren’t part of experience, so *that* can’t be what their experience is telling them. It’s an inference they make from combining mistaken beliefs (to wit: that the past, micro- and macroscopic alike, is “fixed”) with the experience.

                To ask about abilities is to ask counterfactual questions. If she tried to ride a bike, would she succeed? Then she is able to ride bikes. If she tried to make another choice, would she succeed? Maybe the thought of the other choice would so upset her she’d have a fatal heart attack. But barring some unusual possibilities, yes, she would.

              • Posted September 7, 2016 at 10:47 am | Permalink

                To ask about abilities is to ask counterfactual questions. If she tried to ride a bike, would she succeed?

                Counterfactuals get you nowhere with respect to the question. If you dropped a ball from a different height, would it take a different amount of time to fall?

                And, indeed…that pretty much points to the basic fault in the whole concept of “Free Will,” as defined by everybody but the fractional minority of compatibilists.

                Everybody understands that, if the gunman puts the gun to your head, you’re picking vanilla even if you would normally otherwise pick chocolate — just as, if you drop the ball from the top of the ceiling it’s going to take a lot longer to reach the floor than if you drop it from ankle height.

                Libertarians assert that, lacking a gunman, some etherial “you” controls the choice between chocolate and vanilla — akin to a suggestion that, if you drop the ball from your chin, it’s up to the ball to decide whether to hit your left foot or your right foot. A determinist insists that, no, it’s still, fundamentally, no different. You might not be able to eyeball the outcome ahead of time, especially if it’s windy, but which foot the ball hits is entirely explained by Newton…as is your choice of ice cream.

                As best I understand it, the compatibilist position says that the ball has little jet engines that nudge it one direction or another, and that’s close enough to the libertarian magic that we should think of it in the same terms. Determinists note that that’s more complex, yes, and maybe harder to predict, sure…but everybody else has long since agreed that those properties are entirely orthogonal and unrelated, so why are compabilists obsessively insisting they’re relevant?

                After all, if complexly unpredictable self-organization is what compatibilists attribute human free will to, why don’t you similarly insist tornadoes have free will? Clearly, it’s because of more than mere counterfactuals and complexity…which, again, puts compatibilists in the position of wanting to both have and eat their cake.

                Cheers,

                b&

                >

              • Posted September 8, 2016 at 11:55 am | Permalink

                You get so much right, and keep coming so close, but then somehow at the last minute, the standards for what counts as choice or freedom go to infinity.

                Libertarians assert that, lacking a gunman, some etherial “you” controls the choice

                And compatibilists assert that a material you controls the choice. And incompatibilist determinists assert – what? That a material being can’t be “you”? That it can’t control anything? Libertarians often deny that “mere” matter could truly be conscious, and so have genuine preferences, or truly reason about plans or actions (or anything for that matter). The premise is wrong, but at least it explains their conclusion. If their premise were right, it would show material brains to be just as mindless and choiceless as planets or tornadoes. But if goals and desires and scenario-modelling and choice simply are brain processes and none the less real for that, it makes humans quite different from tornadoes.

              • Posted September 8, 2016 at 12:15 pm | Permalink

                But if goals and desires and scenario-modelling and choice simply are brain processes and none the less real for that, it makes humans quite different from tornadoes.

                No; you’re falling prey to the same pre-Copernican anthropocentric hubris as religious libertarians.

                “Free Will” in all its forms very emphatically is associated with some sort of specialness attributed to humans — as your statement here makes explicit.

                But we know overwhelmingly that we’re not special. Ultimately, the same forces that shape and move the tornado are the exact same ones that shape and move us: electromagnetism, overwhelmingly, with a shout-out to gravity and some footnotes for the nuclear forces and the rest of the pantheon of the Standard Model.

                Do we happen to have mental models that roughly replicate our surroundings, thereby giving us a more intimate perception of the unfolding of the universal Schrödinger Equation? Sure — but, again, so do thermostats, just with a much less detailed model.

                When you understand that no, really, your perceptive illusions be damned, you actually aren’t fundamentally any different from a tornado — which is the overwhelming conclusion of millennia of scientific investigation — then you will finally understand how incoherent and anti-scientific “Free Will,” in all its forms, compatibilist included, truly is.

                Ultimately, determinism is nothing more nor less than an acknowledgement that humans really are part of the natural world and no different from any other part of it. Libertarianism explicitly rejects that brute fact, and compatibilism is trying to stake out the proverbial halfway-to-crazy-town position between the two.

                (And, of course, to be clear: there are obvious differences between tornadoes and humans, just as there’re differences between giraffes, banana slugs, and oil rigs. And those differences are important and explain their different behaviors. It’s just that “Free Will” isn’t part of the explanation.)

                Cheers,

                b&

                >

              • Posted September 9, 2016 at 11:39 am | Permalink

                I think we’ve hit on a root disagreement, because I have no problem taking an anthropocentric approach to all value questions, including all ideas of what’s special and what’s not. We’re anthropos – where else should we be centered? From a pure physics point of view, there’s almost no difference between an original Picasso and a skillful forgery, but I have no problem assigning them wildly different values. There is no reason to expect that what humans value – including free will – must line up with fundamental categories of physics or metaphysics. That’s Sean’s point in his post Free Will is as Real as Baseball (paragraphs 7+) – and so this thread comes full circle.

              • Posted September 12, 2016 at 2:14 pm | Permalink

                From a pure physics point of view, there’s almost no difference between an original Picasso and a skillful forgery, but I have no problem assigning them wildly different values.

                From a pure physics point of view, the original and forgery have radically different pasts — and it’s the object’s past that you presumably value.

                Now, there may be practical impediments to tracing an object’s past…but, first, that says something about your ignorance and nothing about the objects; and, second, your ignorance should prompt you to consider the wisdom of the value you’re placing on that particular property.

                Sure, you do value originality. But perhaps you have some other value that means more to you than originality?

                And how do you hope to prioritize your values like this without facing reality head-on, as opposed to what you wish reality would be instead?

                Cheers,

                b&

                >

              • Posted September 12, 2016 at 5:40 pm | Permalink

                That can be the last word, until the next time Jerry brings up the issue.

  39. Posted September 6, 2016 at 6:53 pm | Permalink

    Rats, that was supposed to say

    What they can’t do is [ the other choice + hold constant all facts, known and unknown, prior to the decision ].


One Trackback/Pingback

  1. […] via Sean Carroll on free will — Why Evolution Is True […]

%d bloggers like this: