Does Free Will Exist? Sam Harris and Dan Dennett Discuss

The following is a transcript of the July 3, 2016 episode of Sam Harris’s ‘Waking Up’ podcast, entitled ‘Free Will Revisited’, in which Sam and philosopher Daniel Dennett attempt to reopen a conversation and bury a hatchet.

SH: Welcome to the Waking Up Podcast. This is Sam Harris.

Well I just got back from Banff, where I attended the TED Summit, and I spoke about the danger of superhuman artificial intelligence — I’ll let you know when that talk sees the light of day — and I brought a portable recording device to the conference on the odd chance that I might find someone worth talking to who wanted to record a podcast; needless to say, there were many people worth talking to, but not much time to sit down and do a podcast. But I did record one conversation with the philosopher Dan Dennett, who probably needs no introduction here.

As many of you know Dan and I have been brothers in arms for many years, along with Richard Dawkins and Christopher Hitchens as so-called New Atheists, or the Four Horsemen after a video by that name that we recorded in Hitch’s apartment some years back. Dan and I once debated together on the same team along with Hitch at the Ciudad de las Ideas Conference in Mexico, where we were pitted against Dinesh D’Souza and Rabbi Shmuley Boteach and Robert Wright, I believe, and Nassim Taleb got in there, somehow. I hope it doesn’t seem too self serving or contemptuous of our opponents to say that we came out none the worse for wear on that occasion, and needless to say, that video is online for all to see until the end of the world.

But as many of you know, Dan and I had a very barbed exchanged on the topic of free will some years later, and that was a little over two years ago, and we never resolved it. I came out with my short book on free will, and Dan reviewed it, and then I responded to his review, and the matter was left there in a way that no one found satisfying, least of all our readers. There really was an outpouring of dismay over the tone that we took with each other — and I must say, that was totally understandable.

I want to begin by reading the first few paragraphs of my response to Dan’s review, which includes a quotation from him, so you can hear how vexed and vexing things got. And if you’re interested you can read the whole exchange on my blog; in fact, when I post this podcast on my website I’ll provide the relevant links. So this is near the beginning of my response, written as a letter to Dan:

“I want to begin by reminding our readers and myself, that exchanges like this aren’t necessarily pointless. Perhaps you need no encouragement on that front, but I’m afraid I do. In recent years I’ve spent so much time debating scientists, philosophers and other scholars, that I’ve begun to doubt whether any smart person retains the ability to change his mind. This is one of the great scandals of intellectual life; the virtues of rational discourse are everywhere espoused, yet witnessing someone relinquish a cherished opinion in real time is about as common as seeing a supernova explode overhead. The perpetual stalemate one encounters in public debates is annoying because it is so clearly the product of motivated reasoning, self-deception, and other failures of rationality. And yet we’ve grown to expect it on every topic no matter how intelligent and well-intentioned the participants. I hope you and I don’t give our readers further cause for cynicism on this front. Unfortunately, your review of my book doesn’t offer many reasons for optimism. It is a strange document — avuncular in places, but more generally sneering. I think it fair to say that one could watch an entire season of Downton Abbey on Ritalin and not detect a finer note of condescension than you manage for twenty pages running.”

And now I have a quotation from Dan’s review here, this is Dan:

“I’m not being disingenuous when I say this museum of mistakes is valuable: I am grateful to Harris for saying so boldly and clearly what less outgoing scientists are thinking, but keeping to themselves. I’ve always suspected that many who hold this hard determinist view are making these mistakes. But we mustn’t put words in people’s mouths, and now Harris has done us a great service by articulating the points explicitly, and the chorus of approval he’s received from scientists goes a long way to confirming that they’ve been making these mistakes all along. Wolfgang Pauli’s famous dismissal of another physicist’s work as “not even wrong,” reminds us of the value of crystallizing an ambient cloud of hunches into something that can be shown to be wrong. Correcting widespread misunderstanding is usually the work of many hands, and Harris has made a significant contribution.”

So this is back to me, I say:

“I hope you will recognize that your beloved Rapoport’s Rules have failed you here —

As an aside for those of you who are not familiar with them, these rules come from Anatole Rapoport, the mathematician, game theorist, and social scientist, and Dan has been a champion of these rules of argumentation for years, and they are: one, an attempt to re-express your target’s position so clearly, vividly, and fairly that your target says, “Thanks, I wish I thought of putting it that way.” Two, list any points of agreement, especially if they are not matters of general or widespread agreement. Three, mention anything you have learned from your target. Four, only then are you permitted to say so much as a word of rebuttal or criticism. So those are the rules, and Dan has often said that he aspires to follow them when criticizing another person’s point of view. So back to my text:

“I hope you will recognize that your beloved Rapoport’s Rules have failed you here. If you have decided according to the rule to first mention something positive about the target of your criticism, it will not do to say that you admire him for the enormity of his errors and the recklessness with which he clings to them despite the sterling example you’ve set in your own work. Yes, you may assert, “I’m not being disingenuous when I say this museum of mistakes is valuable,” but you are, in fact, being disingenuous. If that isn’t clear, permit me to spell it out just this once: you are asking the word valuable to pass as a token of praise, however faint, but according to you my book is “valuable” for reasons that I should find embarrassing; if I valued it as you do, I should rue the day I wrote it, as you would had you brought such value into the world. And it would be disingenuous of me not to notice how your prickliness and preening appears; you write as one protecting his academic turf. Behind and between almost every word of your essay, like some toxic background radiation, one detects an explosion of professorial vanity.”

So, that’s how snide things got, and, I must say, this is really a problem with writing rather than having a face-to-face encounter; if any of you have ever had the brilliant idea of writing a long letter or email to a friend to sort out some relationship crisis rather than just have a conversation, you’ve probably discovered how haywire things can go through an exchange of texts, and the same can be true for intellectual debates among philosophers and scientists. And it’s especially likely to happen if either or both of the people involved are writers who get attached to their writerly maneuvers. I remember writing that quip about Downtown Abbey and it made me laugh at the time, and I knew it would make many readers laugh, so I kept it in, but lines like that just amplify the damage done. So as I told Dan at the end of our podcast, I very much regret the tone I took in this exchange, and I’m very happy we got a chance to have a face to face conversation and sort things out.

I don’t think we resolved all the philosophical issues, but we spoke for nearly two hours, but there were several important topics that never came up. As you’ll hear, we were speaking in a bar, using a single microphone, and this was at the end of a long day of conferencing. So this isn’t us at our most polished or prepared, but I thought it was a very good conversation and I think those of you who are interested in the problem of free will and its connection to ethics will find it useful.

I still think there is some sense in which Dan and I are talking past one another. The nature of our remaining disagreement never became perfectly clear to me, so perhaps you guys can figure it out.

Once again this podcast is ad-free and made possible by the generosity of listeners like yourself; if you want to support what I’m doing here you can do that on my website at samharris.org/support, and that is greatly appreciated. I also want to encourage you to subscribe to the podcast directly on your smartphone or tablet, so you automatically get each episode, and please leave reviews on Itunes or any over place you happen to listen to it because these reviews really do help. Finally if you want to hear from me in general, for instance if you want to know when the talk I gave at the TED Summit is released online, please sign up for my email list through my website.

And now I give you: Dan Dennett. In a bar. Overlooking the Canadian Rockies.

SH: So I’m here with Dan Dennett at the TED Summit in Banff, and we have stolen away from the main session, and we are in a bar and about to have a conversation about the misadventure we had in discussing free will online in a series of articles and blog posts.

You and I are part of a community and a pretty visible part of the community that prides itself on being willing to change its opinions and views, more or less in real time under pressure from better arguments and better data, and, I think I said in my article in response to your review of my book Free Will, that this is a very rare occurrence; to see someone relinquish his cherished opinion, more or less on the spot, under pressure from a interlocutor — that’s about as rare as seeing a supernova overhead, and it really shouldn’t be because there’s nothing that tokens intellectual honesty more than a willingness to step away from one’s views once they are shown to be in error.

And I’m not saying we’re necessarily going to get there in this conversation about free will, but there was something that went awry in our written exchanges, tonally, and neither of us felt good about the result.

And so, again, we’ll talk about free will as well, but I think this conversation is proceeding along two levels where there’s the thing we’re talking about philosophically, which is free will, but then there’s just the way in which — I want us both to be sensitive to getting hijacked into unproductive lines that make it needlessly hard to talk about what is just a purely intellectual, philosophical matter, and one of great interest — surprisingly, great interest to our audiences. There’s no topic that I’ve touched that has surprised me more in the degree to which people find it completely captivating to think about. And I know you I both think it’s a very consequential topic, it’s unlike many topics in philosophy — this one really does meet ethics and public policy in a way that important, so.

One thing you all should know in listening to this is that we have one microphone; perhaps this is a good thing because we really can’t interrupt each other and we’re just gonna pass this microphone back and forth and I now give you Dan Dennett.

DD: Thanks Sam. This is a beautiful setting; if we can’t agree on some things here, we shouldn’t be in this business.

I’m going to go back one step further in how this kind of started: you sent me the manuscript of your book Free Will , and asked me for my advice, and I didn’t have time to read it and just told you, “No, I’m sorry, I don’t have time.” And then when the book came out, I read it, and said, “Oh, I wish you’d — “ I’d forgot that we had that exchange, and I wish you’d showed it to me because I think you made some big mistakes here, and I would loved to have tried to talk you out of them. Too late. And then, time passed, and then we had the — you said you wanted me still to say what I thought the mistakes were, and that’s when I wrote my piece for your blog and for Naturalism . And it certainly struck you wrong, and I guess I regret a few bits of tone there, but I think everything I said there is defensible.

And in particular, I did use Rapoport’s Rules, contrary to what you say. If you look at the first paragraph of my piece, I applaud the book for doing a wonderful, clear job of setting out a position which I largely agreed with, and then I said you went off the rails a little later. So, I did try to articulate your view, and I haven’t heard you complain about that articulation of your view; and I said what we agree about, and I even said what I’ve learned from that book. So I did follow Rapoport’s Rules quite well, but we can just set that aside, if you want, and get down to what remains of the issue.

One thing in particular, which I know came off awfully preachy, but I really think it was most unwise of you to declare that my position sounded like religion — sounded like theology; you have to know that you’re insulting me, and that was a pretty deliberate insult and that was in the book, and I thought, “C’mon, Sam.” So you can’t expect kid gloves: if you’re going to call me a theologian, then I’m going to call you on it and say, as I said — I tell my students: when a view by an apparently senior — an author worth reading — looks that bad, maybe you’ve misinterpreted it. And of course, the main point of my essay was: yes, you have misconstrued my brand of compatibilism. You’ve got a sort of caricatured version of it, and, in fact, as I say late in the piece, you are a compatibilist in all but name; you and I agree on so many things: you agree with me that determinism and moral responsibility are compatible; you agree that a system of law, including punishment — and justified punishment — is compatible with determinism. We’re just that close to compatibilism. I’ve actually toyed with the idea, in part provoked by you and some others — Jerry Coyne and others — to say, alright, I don’t want to fight over who gets to define the term free will. As I see it, there are two completely intention-themes out there about what free will is: one is that it is incompatible with determinism, and the other is that it is the basis of moral responsibility. I think it’s the second one that’s the important one. That’s the variety of free will worth wanting, and I think the other one’s a throwaway. And I agree with you: indeterminist free will, libertarian free will, is a philosopher’s fantasy — it is not worth it, it is just a fantasy. So we agree on so much. We have no love for libertarian indeterminism, for agent causation, for all that metaphysical gobbledygook — we’re both good naturalists — and we both agree, that the truths of neuroscience, and the truths of physics — physics doesn’t have much to do with it actually — are compatible with most of our understanding — our everyday understanding — of responsibility, taking responsibility, being moral responsible enough to be held to our word. I mean, you and I both agree that you are competent to sign a contract. Me too. Well, if you go and sign a deed or a mortgage, very often if it’s notarized, the notary public will say, “Are you signing this of your own free will?” And I recently said, “Yeah, I am.” That’s the sense of free will that I think is important. I have it. There are a lot of people who don’t have that free will, and it has nothing to do with indeterminism. It has to do with their being disabled in some way. They don’t have a well running nervous system, which you need if you’re going to be a responsible agent. I think you agree with all of that.

SH: So, I certainly agree with most of that. I think there are some interesting points of disagreement on the moral responsibility issue, which we should talk about, and I think that could be very interesting for listeners, for us to unpack those differences. I am, needless to say, very uncomfortable with the idea that I have misrepresented your view, and if I did that in my book, I certainly want to correct that here, so we should clearly state what your view is at a certain point here. But I want to step back for a second, before we dive into the details of the philosophy of free will, what I was aware of doing in my book free will — and again, I would recommend that our listeners just go back and, you don’t actually have to read my book, but you can read Dan’s review of it on my blog, and you can read my response which is entitled “The Marionette’s Lament”, I believe. Then you can see the bad blood that was generated there, and I don’t know, Dan, if you’re aware of this — you don’t squander as much of your time on social media or in your inbox — but I heard from so many of our mutual readers that they were despairing of that contretemps between us. It was like Mom and Dad fighting, and it was totally unpleasant. The thing that I really regret, which — you regret that you didn’t get a chance to read my book before I published it, which, yeah, that would have been a nice thing, for both of us — but what I regret is that when you told me that you were planning to write a review of it, I kept urging you and ultimately badgering you to not do that and have a discussion with me because I knew what was gonna happen, at least from my point of view, was that you would hit me with this ten-thousand word volley, which at a dozen points or more I would feel you misconstrued me or gone off the rails, and there would be no chance to respond to those, and to respond in a further ten-thousand word volley in a piecemeal way would just lead to this exchange that was very boring to read and yielded a much bigger sense of disagreement than what was necessary. If I have to spend ninety percent of my energy taking your words out of my mouth, then this thing begins to look purely adversarial, so one thing I’ve been struggling for in my professional life is a way of having conversations like this, even ones where there’s much less goodwill than you and I have for one another — because you and I are friends and we’re on the same side of most of these debates, and so we should be able to have this kind of conversation in a way that’s productive. But I’ve been engaging in people who, you know, think I’m a racist bigot as a starting point, and I want to find ways of having conversations in real time where you can be as nimble as possible in diffusing unnecessary conflict or misunderstanding, and writing is an especially bad way to do that; certainly writing the ten-thousand word New York Review of Books piece, then someone has to react to in an angry letter. So, in any case, I wish we had that conversation, but we’re having it now, and this is instructive in its own way. Feel free to react to that, but I guess I want you to also express what compatibilism means to you, and if you’ll recall the way in which I got that wrong, feel free to say that, but I’ll then react to your version of compatibilism.

DD : Well, my view of compatibilism is pretty much what I just said, and you were nodding. And you were not considering that a serious view about free will, although you were actually — almost all of it you were agreeing with. And you also, I think, made this serious strategic or tactical error of saying this is like theology — this smells of theology. Well, as soon as you said that I thought, “Well, you just don’t understand what compatibilism is.” It’s the opposite of theology. It’s a attempt to look at what matters, to look at the terms and their meanings, and to recognize that sometimes ancient ideology gets in the way of clear thinking, so that you can’t just trust tradition; if you trusted tradition and everyday meanings of words, we would have to say all sorts of silly things. We’ve learned — in fact, one of the abiding themes in my work is there are these tactical or diplomatic choice-points: you can say, “Oh, consciousness exists, it just isn’t what you think it is,” or you can say, “No, consciousness doesn’t exists.” Well, if you’ve got one view of consciousness, if it’s this mysterious, magical, ultimately insoluble problem, then I agree, consciousness in that sense doesn’t exist, but there’s another sense, much more presumable I think, which of course consciousness exists, it just isn’t what you think it is. That was a central theme in Elbow Room , with regard to free will, and in Consciousness Explained , with regard to consciousness. My view, my tactic — and, notice, those two views they look as if they are doctrinally opposed; they’re not. They’re two different ways of dealing with the same issue: does free will really exist? Well, if free will means what Dennett says it means, yes, and you agree. If it means what some people think, then the answer is no.

SH: Yeah, I understand that, but I would put to you the question: there is a difference between explaining something and changing the subject. So this is my gripe about compatibilism — and this is something we’ll get into: I assume you will admit that there is a difference between purifying a real phenomenon of its folk-psychological baggage, which I think is what you think compatibilism is doing, and actually failing to interact with some core features that are just ineliminable from the concept itself?

DD: Let me surprise you by saying, I don’t think there’s a sharp line between those two, and I think that’s quite obvious. That, whether I’m changing the subject — I mean, I’m so used to that retort, about any line along this — so no, I think that’s just a debater’s point, we should just set that aside. Saying, “You’re just changing the subject,” is a way of declaring a whole manifold, a whole variety — spectrum — of clarificatory views, which you’re not accepting because you’re clinging to some core part of what free will is. You want to claim that free will — the core of free will — is its denial of determinism, and I made a career saying that’s not the core. In fact, let me try a new line on you, because I’ve been thinking, why doesn’t he see this the way I see it, and I think that the big source — the likely, big source — of confusion about this is that when people think about freedom, in the context of free will, they’re ignoring a very good and legitimate notion of freedom, which is basically the engineering notion of freedom when you talk about degrees of freedom. My wrists, my shoulder, my elbow, those joints, there’s three degrees of freedom right there, and in control theory it’s all about how you control the degrees of freedom. And if we look around the world, some things have basically no degrees of freedom, like that rock over there, and some things, like you and me, have uncountably many degrees of freedom because of the versatility of our minds; the capacity that we can be moved by reasons on any topic at all — this gives us a complexity from the point of view of control theory, which is completely absent in any other creature. And that kind of freedom is actually, I claim, at the heart of our understanding of free will because it’s that complexity, which is not just complexity, but it’s the competence to control that complexity — that’s what free will is. What you want, if you’ve got free will, is the capacity — and it’ll never be perfect — to respond to the circumstances with all the degrees of freedom you need to do what you think would be really the right thing to do. You may not always do the right thing, but — let’s take a dead simple case: imagine writing a chess program, which, stupidly, was written wrong so that the king could only move forward or back or left or right, like a rook, and it could not move diagonally; and this was somehow hidden in it so that it just never even considered moves, diagonal moves, by the king. Completely disabled chess program, it’s missing a very important degree of freedom, which it should have and be able to control and recognize when to use and so forth. What you want — I mean, let me ask you a question about what would be ideal from the point of view of responsibility: what does an ideal responsible agent have? Is it not mainly true beliefs, a well-ordered set of desires, the cognitive adroitness to change one’s attention, to change one’s mind, to be moved by reasons — the capacity to listen to reasons, the capacity for some self-control? These things all come in degrees, but our model of a responsible adult, someone you would trust, someone you would make a promise to — or that you’d accept a promise from — is someone with all those degrees of freedom and control of them. Now, what removes freedom from somebody is if either the degrees of freedom don’t exist — they’re blocked mechanically — or some other agent has usurped them and has taken over control: a marionette and a puppeteer. And so, I think that our model of a free agent says nothing at all about indeterminism; we can distinguish free agents from unfree agents in a deterministic world or in an indeterministic world; determinism and indeterminism make no difference to that categorization, and it’s that categorization which makes the moral difference.

SH: So yeah, I agree with almost all of that, I just need to put a few more pieces in play here. I think there is an importance difference, nevertheless I agree that there is no bright line between changing the subject and actually purifying a concept of illusions and actually explaining something scientifically about the world — but in this case, the durability of free will as a problem for philosophers, and now scientists, is based on people’s first person experience of something they think they have. People feel like they are the authors of their thoughts and intentions and actions, and so there’s a first person description of this problem and there’s a third person description of this problem, and I think if we bounce between the two without knowing that we are bouncing between the two, we are losing sight of important details. People feel that they have libertarian free will, and when I get emails from people who are psychologically destabilized by my argument that free will doesn’t exist, these are people who feel like something integral to their psychological life and wellbeing is being put in jeopardy; and I can say this from both sides because I know what it’s like to feel that I could’ve done otherwise — so let me just, for listeners who aren’t totally up to speed here: libertarian free will is anchored to this notion of “I could’ve done otherwise.” So if we rewound the universe to precisely as it was a few moments ago, I could complete this sentence differently than I did. Whether you throw indeterminism or determinism or some combination thereof, there is no scientific rationale for that claim. If you rewound the universe to precisely its prior state, with all relevant variables intact, whether deterministic or indeterministic, these words would come out of my mouth in exactly the same order and there would be no change; I would speak this sentence a trillion times in a row, with its errors, with its glitches. So people feel that if they rewound the movie of their lives, they could do differently in each moment, and that feeling is the thing that people find so interesting about this notion that free will doesn’t exist because it is so counterintuitive psychologically. Now, I can tell you that I no longer feel that; subjectively — my experience of myself — I’m aware of the fact that it is a subjective mystery to me how these words come out of my mouth. I’m hearing these words as you’re hearing these words — I’m thinking out loud right now. I haven’t thought this thought before I thought it, it’s just coming, and I am subjectively aware of the fact that this is all coming out of the darkness of my unconsciousness mind, in some sense. There is this fear of my mind that is illuminated by consciousness, for lack of better word, and I can be subjectively identified with it, but then there’s all the stuff that is simply arriving — appearing in consciousness, the contents of consciousness, which I can’t notice until I notice them, and I can think the thought before I think, and my direct experience is compatible with a purely deterministic world. Most people’s isn’t, or they don’t think it is, and so that’s where, when you change the subject — the analogy I used in my article that responded to your review, which I still think captures it for me, I’ll just pitch it to you once more, is the notion of Atlantis; people are infatuated with this idea of Atlantis; I say, actually Atlantis doesn’t exist; it’s a myth; there’s nothing in the world that answers to the name of Atlantis; there was no underwater kingdom with advanced technology and all the rest; and whoever was — Plato was confused on this topic, just spinning yarn. And you , compatibilism — your variant and perhaps every variant takes a different approach. It says, no, no, actually there is something that conserves much of what people are concerned with about Atlantis and, in fact, it may be the historical and geographical antecedent to the first stirrings to this idea of Atlantis, and there’s this island of Sicily, the biggest island in the Mediterranean, which answers to much of what people care about with Atlantis. And I say, actually, what people care about is the underwater kingdom with advanced technology, and that is a fiction. So you and I are going to agree about Sicily — ninety nine percent of our truth claims about Sicily are gonna converge, but I’m saying the whole reason why we’re talking about Atlantis in the first place is there’s this other piece that people are attached to, which, by you purifying the subject you’re actually just no longer interacting with that subjective piece.

DD: Yeah, that’s well put. I think the analogy is less instructive — I don’t think it’s entirely fair, but let’s leave it at that. Your position is that you can see very clearly that what people really care about is that free will should be something sort of magical. And you’re right: a lot of people — “if you don’t think free will’s magical then you don’t believe in free will.” And that’s what I confront and say, well, I got something which isn’t magical, which is perfectly consistent with naturalism, and gives us moral responsibility, justification for the way we treat each other, the distinctions that matter to us like who do we hold responsible and who do we excuse because they don’t have free will — it gives us all of the landmarks of our daily lives and explains why these are what matter. And indeed, though, if the mystery — if the magic — is that important to people — and I agree with you, that magic doesn’t exist, and if we’re gonna tie free will to that, I would say, “No, free will doesn’t exist.” Now, you said something very interesting, you said that the reason people believe in this is because they feel it or they think they do, they sort of intuit they could’ve done something different in exactly the same situation. I agree with you that that’s what they think, but I don’t think that it is a forlorn task to show them that that’s not really what they should think about this — about the very feelings they have. Their sense that they are, as Kant says, “acting under the idea of freedom — “ that’s right, they are. And that’s the only way an agent can be. This is a fairly deep point, that an agent has to consider some things fixed and some things not fixed — you can’t decide, otherwise the whole setting of decision making depends on there being that kind of freedom. And so it’s no wonder, in a way, that people who are impressed with that decide that what they experience is a sense of utter freedom. They don’t need utter freedom; what they need, and can have, is the sense that in many very similar circumstance — circumstances which differed maybe only in a few atoms — they would’ve made another decision, and as soon as you allow any tiny change when you rewind the tape, the whole business about determinism falls out of the picture. And that’s why in — actually several places, I’ve gone to considerable length, probably too long, to trot out examples where we have a decision maker, which in a demonstrably deterministic world is playing chess, and it loses the game, and it’s designer says, “Well, it could’ve castled.” What do you mean it could’ve castled? What the designer means is it was just the luck of the draw; a chess program, like any complicated program is going to consult a random number generator — or a sudo-random number generator — at various points, and this time it chose wrong. However, it chose wrong because when it got a number from the sudo-random number generator, it got a one rather than a zero; flip a single bit and it would’ve made the other choice. In other words, it’s not a design flaw. An agent could be, as it were, impeccably designed — you couldn’t improve the design of the agent — so that’s what justifies saying, “Yeah, I could’ve done otherwise.” Half the time, or more, it would’ve done otherwise, this is just bad luck on this occasion. Normally it would’ve done otherwise.

SH: So, I agree with all that. I think you’re not acknowledging, however, how seditious those facts are — the degree to which they undermine people’s felt sense of their own personhood. So, if you tell me that but for a single charge at a synapse I would have decided I didn’t wanna have this conversation with you or I wouldn’t have proposed to my wife, right, and my entire life would be different, acknowledging the underlying neurophysiology of all of those choice points and how tiny a difference can be that make the crucial difference, that suddenly brings back the marionette strings; now, no one is holding the strings — the universe is holding the strings — but that is not what people feel themselves to be. This feeling that if you had had just one mouthful more of lunch, something very different — you would make a radically different decision six hours from now than you are going to make, that is a life that no one, virtually no one feels they’re living.

DD: Well this is going in good directions. I think you’re largely right and exactly wrong in what you just said. I think you’re right that this is a subversive idea to many people. They’re so used to the idea that unless they are completely, absolutely undetermined, then they don’t have free will. Now the trouble with that is if you look closely at that idea — you see, if they were absolutely undetermined, that wouldn’t give them free will either. That’s a red herring. So let’s look at what does matter. It’s interesting that you say that if I thought that some tiny atomic change would’ve altered the course of, you know, some big important life decision — let’s look closely at that; because what I think we should say is it is indeed true that there are times when a decision is a real tossup. When you’ve thought about it, thought about it, thought about it — you’re going to have to act pretty soon, and you just can’t make up your mind. In cases like that, and it may be something that’s morally very important, the idea that when you do make the decision, had a few atoms been slightly different you would’ve made the other decision, I don’t find that upsetting at all because that’s one of those situations. And it doesn’t mean that when the evidence and reasons are preponderantly on one side — no, then you’d have to large change in the world for a different decision to come out. Sometimes the indeterminists, the libertarians, in fact, as a sort of signature — a lot of their views say that there has to be an absolutely undetermined choice of some importance, somewhere in the causal chain of your life for your action to be responsible. Now, thus — I had long thrust into their faces the example of Luther who says I can do no other; he’s not ducking responsibility, he’s saying, “Believe me,” — it wasn’t had the light been different or wind not been blowing I would’ve — no, he’s saying, “I was determined to do this,” and yet he’s not saying it’s not a free decision. They — some of them — amazingly to me, fall for the bait and say, “Oh, well that’s only because it must have been the case that somewhere in Luther’s life there was a moment — it might have been in his childhood — when there were two paths, A and B, and he chose A, which led to him nailing the theses on the door, and at that moment it was absolutely undetermined that he’d choose A. I think that’s the craziest fantasy imaginable. It doesn’t depend on that. So I agree with you that when we think about how chance, luck, enters into our live that can be very unsettling, and we should not hide from the fact that there are times when it’s a tossup, and we may rejoice in the decision we make or may bitterly regret it, and the fact that it was not in our control — maybe it’s a tragic fact, but it’s not a fact which disables us for responsibility.

You’re playing chess, to take a deliberately trivial case; you’re considering two possible moves, for the life of you, you can’t see what the better one is. You sort of mentally flip a coin, you don’t know — it works out great. You are very likely to retrospectively decorate that with the claim that that’s what you’ve determined. Nah, you’re kidding yourself; you’re just taking responsibility for a little bit of lucky, random coin flipped in your decision process. That does not — in fact, not only does that not disable you for free will, I think an important human point about free will is that free responsible agents recognize that when they act, they are going to be held responsible whether or not they are in complete control — and they can’t be in complete control — of the decision making that goes to making up their minds.

SH: Well, now I think we’re getting into some very interesting territory where we might actually disagree because I think, perhaps, your notion of moral responsibility is something that I don’t agree with. I think I can do a kind of compatibilist maneuver on moral responsibility and get most of what we want out of it, but I think something changes with my view of free will. But I want to unpack one point you made earlier that might’ve blown by people too quickly: the reason why indeterminism doesn’t give you freedom is that if you made a choice that was truly not determined by your past conditioning, your past attitude, it would be precisely the occasion where you would say, “I don’t know what came over me.” It wouldn’t be representative of who you’ve been up until that moment, so it has to be in part of the causal stream that you recognize to be lawfully you in each moment. But I want to ask you one question before I talk about why I think it might all be a matter of luck all the way down, and that moral responsibility is something we have to redefine in the way that you are eager to redefine free will — I want to ask you: this notion that the tiny, micro-adjustment in the universe only really spells the difference in cases where it’s a decision that could go either way, that it was fifty-fifty anyway and you just flipped a bit — it wasn’t a diversion of that great consequence because it was not something that you were fully committed to, but I think there probably are many actions, maybe decisions are the wrong category, but certainly actions, where the difference between a life-changing action and not is just a matter of a tiny piece of real estate in the brain being otherwise; so the difference between thinking something and saying it out loud to the person you’re thinking it about, or the difference between sending that angry email you wrote to your boss or to your best friend and just deciding to scrap it — that can be one of these tiny moments, where but for a little more sleep the night before, your life would be very different.

DD: On that very point: imagine how many Brits there are today who didn’t vote, and didn’t vote for the most tiny and trivial reasons; they just, “Oh, I think I’ll have another piece of toast instead; I don’t really need to vote.” And they’re kicking themselves, of course, now, and life has those moments and we factor that in when we consider what it is to be responsible. And some people do better on that sort of self-control than others. I think that regret plays a big role in our thinking about free will; when we have done something that we bitterly regret and that we do view as, “That wasn’t really what I (or at least the ideal I) would do,” and we’re ashamed of something we’ve done, we often hide behind second-rate ideas. And I think that’s part of the sort of emotional distortion, the waves of distortion, that mean that we really can’t take the folk notions at face value — they’ve got a lot of baggage on them.

SH: It’s interesting, it’s seeming to me at this moment that we have been talking past each other to some degree. That you decide to redefine free will — “Here’s the free will worth wanting” — and then you talk about degrees of freedom and agents that can operate lawfully and meet their aims, whereas I talk about free will being an incoherent concept; the libertarian notion of free will doesn’t map onto determinism, it doesnt map onto indeterminism, and it doesn’t map onto any combination thereof, and we both agree about that, but we have a difference response to it. I think —

DD: What is your response to that?

SH: Well, my response is that the free will you think you have doesn’t exist, and then we can go on to talk about all these other things that we care about, but here’s where I’m going to push into this area of moral responsibility where we may find a disagreement. So you take the classic case of Charles Whitman, the shooter in the clocktower killing, I think, fourteen people at the University of Texas, and one of the early and famous mass shootings in American history; and it turns out that he wrote this, essentially, suicide note saying, “I don’t know what’s wrong with me, but I’ve been flying into a rage,” and he killed his wife first before he went and killed all those other people, and he said, “I don’t know why I did this, I love my wife. You might want to do an autopsy on my brain after you kill me to find out what’s wrong with me.” And, in fact, that’s what was done, and they found a glioblastoma that was pressing on his amygdala, and it’s just the sort tumor in the sort of place where you think, okay, there’s something exculpatory about that, right — he was a victim of his biology, and that wasn’t Charles Whitman shooting, that was Charles Whitman plus brain tumor shooting. So, when that kind of case emerges in court, it affects our ethical notion of if he had survived and it was time to punish him, we would have given him brain surgery, had the surgery been available, and not put him in prison for the rest of his life because he was yet another victim of this bad luck incident. Now, my argument in my book Free Will , which I think you don’t agree with, is that a complete understanding of neurophysiology — should we ever attain — is exculpatory in that same sense. That basically it is brain tumors all the way down. So if you can tell em that you fully understand the charge on that one synapse that led me to hit send on email as opposed to restraining myself — that charge is something which I didn’t author. That charge is the tiniest brain tumor ever found, and that is the reason why I hit send.

DD: Oh, that is a very useful — Tom Wolfe has this passage where he says what we’ve learned from neuroscience is that we’re wired wrong, “Don’t blame us, we’re wired wrong.” No, what neuroscience show is we’re wired; it doesn’t show we’re wired wrong . Some people, like poor Whitman: wired wrong. So what you’re basically challenging me to say is, “Well doesn’t that mean that everybody is wired wrong.” There’s no such thing as being wired right for free will. That is what I think you’re now claiming; you’re saying, “It’s brain tumors all the way down.” Well, I find that extrapolation completely — I’m not moved by it at all. I don’t think it is a logical argument. I think it is a mistaken extrapolation. It’s like a mathematical induction gone wrong. The fact that Whitman — and I find it, in fact, fascinating that this is a very standard argument from the libertarians. They’ll take out a case of somebody with horrible brain damage, and say, “Well, surely this is a case of a person who is a victim, as you say, and not an agent.” Right, I agree; well then, we’re all that way. But no, that’s precisely what we understand, that we’re not all disabled. Now, nobody’s an angel — nobody’s perfect — so if anything short of perfection counts as being disabled to the point of exculpatorily disabled, then you’re right, but that’s a very strange view. The idea that you couldn’t be able enough to be held responsible is the crux of the issue right now between us; I say that the boundaries are all as porous, that as we learn more about neuroscience, as neuroscience teaches us more, we may very well, probably will, move some people that are now exculpated into the guilty, not excusable category, and others will move, but we’ll still keep the distinction between those who are basically wired right and those who are wired wrong.

SH: I’m not disputing the fact that people have different have different capacities, right; so people have different degrees of freedom, and if you have a brain tumor in the wrong place, your capacities can be undermined. There’s nothing that I said thus far that ignores the very important difference between voluntarily and involuntarily action or the ability to restrain your impulses or, I suppose, to just acting out everything that arises in your mind at the level intention. So, there are different capacities, but here’s the ethical problem, and the reason why I think more information begins to make every case look more like Charles Whitman: because, everything is as it is in a way that no one can take responsibility for. You didn’t pick your parents, you didn’t pick your genes, you didn’t pick the environment in which your nervous system was sculpted in response to it’s inputs. The only variables there are in the system are your genes and the way in which they’re played upon by the environment, and this includes ideas, this includes conversations had and not had. So to bring us back to this conversation: you are not in control of how persuaded your are or not by what I say. So I say something, it either strikes you as stupid or incredibly incisive, or somewhere on that continuum, and you don’t pick that; it is entirely dependent on the state of your brain, which is entirely dependent on every moment preceding, so we are being played by the universe. We are little corners of the universe that are just like the rest of the universe except for all of these other functions that we can talk about like voluntary behavior and involuntarily behavior, impulse control, et cetera, and there is something exculpatory about that. So, again, just to give you a little more information here, you can take the evilest person, the most easily incriminated person you can think of — let’s see, I think I use Saddam Hussein or one of his sons in my book. I mean this is the prototypical, mustache twirling, evil person. If anyone is responsible for his actions, he is, but if you just roll back the timeline of his life, at one point he was the four-year-old who was destined to become Saddam Hussein. So you look at that four-year-old and he might have the genes for psychopathy, say, he might have bad parents, he’s got a bad society, or certainly one that’s destined to influence him in ways that predispose him to psychopathic violence. So, you have an unlucky four-year-old — that is fully exculpatory. If we could help that four-year-old we would; we would intervene, we would put him in a new family, we would give him the right drugs if we had them to combat his psychopathy, and you just roll forward in his life at a certain point, as if by magic, we hold him responsible for being the true author of his actions, and yet at no point does he actually become the author of his genes and his environment and all of the causal connections. And so I’m saying to you is that we might want to still hold people responsible — I think we do. I think you and I should sign contracts and we should keep promises and we should be help responsible for breaking those promises. But —

DD: Why, if nobody’s ever responsible?

SH: Because it’s pragmatically useful to do that. Punishment makes sense if it actually influences behavior in a way that on balance leads to human flourishing. So I don’t think we throw out everything you are worried we’d throw out in the criminal justice system, but I think there is something — I think the role of luck goes all the way down.

DD: Your examples, I think, have a flaw, and I’m trying to think the cleanest way of saying what I think it is. Let’s talk about control. One of the things you said is, “Yeah, you can’t control your genes, you can’t control your environment,” — that’s right. And as a sailor, I can’t control the properties of the water, I can’t control the wind, but I can control the boat. I can’t control how hard the wind blows, but given how the wind blows, I can control the boat. Now, maybe you couldn’t control the boat because you don’t know about how to control a boat, but I do, and I can control the boat. And your argument is trying to remove the very idea of control from the world, to say nobody ever controls anything — not really . And I think that’s a reductio ad absurdum . Of course we control things, and, for that matter, an airliner can be controlled by a computer — the computer is really controlling it — and if the pilot turns off the computer then the pilot is controlling the airplane. And the idea that you can’t control all the factors — that’s irrelevant. I mean, of course you can’t. That’s why — that’s what control theory starts with, the premise that there are factors that are not in your control. The whole point of control is to respond to the factors that are out of your control by doing the right thing. And the same thing goes for moral education, you’re exactly right. Poor Saddam Hussein, terrible beginning and all that, and if we could treat him we would. Why would we treat him? Because if we treated him, if we could get in there early, we could turn him with some chance of success — and there’s good evidence of this — into an autonomous, self-controlling adult. That’s why we have moral education, and it works. And if you don’t morally educate your kids, then they’re gonna be out of control. And if they are, then shame on you as parent for not doing that moral education because you could’ve made them into a self-controlling, autonomous adult, and you let down the side, and it’s too bad for them — it’s too bad for you. I agree, these are cases where adults are not fully responsible, and I think, again, we don’t need to talk about absolute moral responsibility, nobody could be absolutely morally responsible, but you can be non-absolutely, practically responsible for who you are. After all, you make a robot — you made it, you thought about it. You send it out and it kills somebody. Who’s responsible? You are. You made it — you should’ve known better. Alright, you, twenty-year-old you, made twenty-one-year-old you, which made twenty-eight-year-old you, which made forty-year-old you. Part of being an adult is recognizing that part of our responsibility to the rest of the world is to keep ourselves as self-controlling, autonomous agents. We have a responsibility — we have a duty to maintain ourselves as self-controllers. Some people fail miserably. And it’s very interesting that we’ve had a sea change in the way the public thinks about this. When I was a kid, drunk drivers were routinely excused because, “Oh, poor Sam, he was drunk out of his mind — he wasn’t responsible.” Now we hold them doubly responsible, and we hold the bartender and his friends responsible for letting him get in that state, and I think the same thing — I think that’s a wise change, and I think that’s something we don’t want to give up. We want to keep the idea that part of getting a moral education is inculcating in an individual the goal, the motive, to become a reliable self-controller, and most people succeed pretty darn well. They’re not perfect, and what you get for that is the freedom of the state. You get to drive a car and run around and sign contracts and live a life, and you’re allowed to have a free — politically free — life. And if you can’t do that, you’re in the soup and you’re gonna suffer.

SH: Well, again, I agree there is this practical distinction, and an important one, between people we can treat as responsible agents who can behave themselves and people who are wild and unpredictable and can’t be influenced by reasons and our expectations, et cetera; so there’s an obvious difference between toddlers and older kids, and older kids and adolescents and adults, and we understand a fair amount about the physical basis of those differences. But, this is something, actually, that came up in my response to your review, which I had never thought of before and it’s just something I haven’t thought about deeply, but it seems to me ethically interesting; where you brought up Austin’s Putt, and you brought it up as an example how you really don’t want to think about free will, and in my review — or in my response to your review — I owned it as, so this actually is in line with how I do want to think about free will. And one thing that came out of Austin’s Putt for me is what strikes me as a bit of a paradox: so you take the most competent agent, so you take Tiger Woods or Tiger as he used to be, putting — Tiger woods in his prime — attempting to sink a two-foot putt. Now this is a putt that he will make nine hundred and ninety-nine times out of thousand, I would think, at least. So when he misses this putt — we’re now on the one occasion where he misses it. On it’s face, he is the person you can hold most responsible on Earth for missing this putt because if I miss the putt, you’d expect me to miss it twenty percent of the time — because I’m not a good golfer — so I am less culpable than he is; he is the most culpable of his missing the putt says the least about him because he’s someone who always makes this putt. So the role of luck here, the role of just the mere cosmic ray bombardment of his synapses, seems the most salient because for Tiger Woods to miss a two-foot putt, that is just bad luck. That’s not the Tiger Woods we know, and that’s not the Tiger Woods he knows, and his subjective experience of missing the putt will be presumably, “I don’t know what the hell happened because I should have made that putt, and I would make that putt. I’ll make that putt — watch, I’ll make it a thousand times in a row now.” So the paradox, and I wish I had a name for this paradox, but the paradox is in a case where you have a lapse in behavior, and I would argue even a moral lapse, even a crime committed, by someone who should be the absolute best candidate for responsible self-governance in that instance, it suddenly becomes the least good case for freedom of will.

DD: I’m glad you brought that up because I wanted to raise that too. The reason — let’s take a moral case, a moral Tiger Woods case — there’s a reason why we expect the best from some and not from others, and we’re more condemnatory if the supreme court justice shoplifts, or something like that, than if somebody — some poor kid in the neighborhood does. I don’t think there’s any paradox; I think that there’s no paradox because the supreme court justice, or any serious moral agent, has taken on, in effect, the obligation to be that good. People are counting on him. People are making plans that could be life-wrecking if he doesn’t come through. They’re not doing that with everybody. They’re doing that with the ones that have the particular competence and advertise that particular competence. So it is particularly bad when they don’t live up to their own self-advertisement, and you say it’s just luck, well it may be just luck or they may look in their hearts and think, well, I don’t know whether just luck, randomness, or sudo-randomness — chaos — infects every moment of our lives, but very often there are other factors that we can point to if we go looking. Maybe Tiger Woods shouldn’t have stayed out so late last night, and he may realize that, and he may say, “I can’t imagine how I missed that,” but oh, he can indeed. And he will think it through very carefully, and he will try to improve his mental game the next day. You see, I don’t think the case is as simple as you make it out to be — I don’t think there’s any paradox there at all.

SH: Well I think you’re not connecting with the first person side of what it’s like to deviate from your own internalized norms and competencies. So, you’re the supreme court justice, right, who’s just fully committed to ethics and the rule of law, and then all of the sudden you find that you gone on a shoplifting spree, right — that is so unreflective of who you’ve been. That is a failure of some part of your system, which seems the most adventitious, it says the least about who you’ve been up until that moment. It is a departure from the norm, and as it is with Tiger Woods missing a putt that even a terrible golfer would make some significant percentage of the time, going back to prior causes doesn’t really resolve the issue. So if you ask, you know, why did he stay out late the night before? Well, then we’re back to him not authoring himself. Because if the explanation for why he decided to stay out late was because he happened to see that pretty woman across the bar, if he hadn’t seen her, he would’ve been in bed by ten, but because he is a serial philanderer, he had to stay out till four.

DD: This is all relevant.

SH: But I’m just saying that each one of those moments becomes a missed putt when you look closely at it. So one thing I hear here, and one thing I’ve definitely detected in your writing about this is you’re very concerned about the social — the societal — implications of most people getting the wrong message about free will. So if Jerry Coyne and I won the argument here, and we just announced that free will is an illusion and, in some sense, everyone is not guilty by reason of insanity, there is no control that’s good enough for free will, no one has that kind of control, everyone is just part of the universe, everyone is a force of nature, everyone is a wild animal — something very important would be lost —

DD: Well, the truth.

SH: But what I’m saying is that you can converse — I have a compatibilist maneuver for many of things you’re afraid would be lost; so, holding people accountable, holding people to their contracts, their promises, putting people in prison who are too scary to let out of prison — I think we can do all of that in a way that doesn’t preserve an illusion that anyone actually, truly authors themselves any more than Charles Whitman did with his brain tumor.

DD: Nobody ever actually, truly, absolutely authors himself or herself, but that doesn’t mean that there aren’t people who author themselves. They use found objects that they find — you know, “Do It Yourself”; there’s a limit to “Do It Yourself”. And, indeed, there’s a lot of luck involved, and I think that people understand this. People understand that the best game in town is to be a moral agent who lives by the community’s ideas of moral agency, and takes punishment gracefully when it’s deserved and may secretly harbor thoughts like, “I’m just really unlucky today. Just really really unlucky.” Tough, I was really unlucky. But the last thing I’m going to do is plead, “I’m just cosmically unlucky, your Honor.” Yeah, okay, so what. You did it, you declare yourself to be a responsible agent, on this occasion you let down the side, and it’s quite irrelevant whether you were unlucky today.

SH: Again, I think the subject there changes to the pragmatics of holding people responsible, and how our courts function, and how are relationships need to be. But again, I just think this is vulnerable to more information. So Tiger Woods’ missed putt: let’s say we had full knowledge of all of the variables, so we find that it was actually the chirp of a bird that got into his head, that, you know, his auditory cortex did its little dance, and that was enough to get him to miss this putt. The game of golf wouldn’t necessarily change because there is no way to incorporate the influence of birds into the rules of the game, so what golf would have to say at that point is, “Listen, sometimes you get lucky and sometimes you don’t get lucky with the birds, right? But we’re not going to kill all the birds so just deal with it.” That is a pragmatic response to a situation that’s just too complex, and yet it is still a fact that the bird caused him to miss that putt, and he’s not actually responsible. We’re just using a heuristic here to hold him responsible. He’s responsible in the sense that next time we will expect him to make that putt, rightfully based on his, and he probably will make it.

DD: Well you are covertly sliding back into absolute responsibility. Right, he’s not absolutely responsible because the bird played an unexpected causal role in this case, but as long as you’re not holding out for absolute responsibility, he’s still responsible. And you say, “This is just pragmatic.” Well, the whole idea of free will is ultimately, I think, pragmatic. We’re going to have a consequentialist account of why we hold people responsibility. And it’s actually — it’s not a simple deterrence and rehabilitation idea — it’s the idea that if we want to have a secure society where there’s respect for law, then the law has to be reasonable because we’re pretty reasonable people. And unreasonable laws will not be respected. So the law has, built into it, lots of excusing conditions, but it says not everything is an excuse: “We draw the line here.” Now, where do we draw that line? We draw it somewhere — and it’s not a metaphysical line, it’s a pragmatic line. And what determines where that line is, is something about what human nature is like in general, and setting it higher would excuse too many people and would lead to disrespect for the law. Setting it lower would be too punitive on people that we intuitively feel were really not that responsible for what they did. So we have this artifact — this wonderful human artifact — which is law and order. And it’s a human — well it’s not so much intelligently by humans, so much as it evolved by cultural evolution to be a system, which people can appreciate it’s value even if they can’t entirely explain how the working parts work and why they’re as good as they are.

SH: So take the prototypical psychopath, an evil person who is as responsible for his evil as possible because he’s a sadist, he gets pleasure from it, he’s now in prison, but he says if we let him out he would do it again because what he really likes to do is kill little girls and boys, and he has no regrets. This is as culpable a person as we’re ever going to find.

DD: No.

SH: Well he is in the sense that he’s not Charles Whitman who’s saying, “I don’t know what came over me,” he’s saying, “This is who I want to be. You can talk about neurophysiology all you want, I’m happy to be who I am — I love this, and this is how I want to live.” So, you might have some reason to think he’s not as responsible as you and I are, but I think this would push most people’s buttons in terms of judging him to be the author of his evil as much as anyone’s the author of anything. This guy is satisfied to be who he is, and he is just a violent, sadistic person who we should lock up. Now, if we had a cure for this condition, let’s call it psychopathy, but maybe this is something beyond that, and we could just give him the pill that made him have the epiphany, “Oh my God, I can’t believe I was this evil bastard. I’m so regretful. I’m so sorry for everything I’ve done. I want to spend the rest of my life making it up to you people,” we would — rather than punish him in a retributive scheme and just let him rot away in prison for the rest of his life — we would have a name for this condition that we now call evil, let’s say it’s called psychopathy plus, and we would just give this pill to all the people who were at risk for being this sort of evil person, in same way we give insulin to diabetics. We would just view these people who, for whatever reason, genes or environment, were destined to become these evil bastards — we would give them the anti-evil pill. And wouldn’t that be, in some sense, exculpatory?

DD: Yes it would, but I want to look at a case, which I think is better. Not a psychopathic torturer, but a Bernie Madoff — a calm, calculating, greedy, heartless, fraudulent, person, with a fine education, good upbringing, and no clear signs of any pathology — psychopathology.

SH: I think that’s the same case, it’s just he’s a more normal person, but I’m going to invent something now, it’s called Madoff Syndrome. It’s the kind of person who, even with the benefit of an education and a great upbringing and good relationships and all of the right environmental influences, he still is malignantly selfish enough to show up like Bernie Madoff; he can lie to people with alacrity and he can pursue his own aims; he has a kind of time horizon for his gratification, which allows him to forget about the fact that he’s running a Ponzi scheme that is guaranteed to blow up and he’s really not motivated by that — he’s discounted that future pain to the point where he’s no longer motivated by it. So, if we could completely understand the neurophysiology of what it was to be Bernie Madoff, and just by dint of luck there was an intervention that could cancel it — there’s now a pill, the anti-Madoff pill, that maybe it’s just a designer pill just for Bernie Madoff, maybe it would work for no one else, but it’ll work for him, and you give it to him and he becomes as regretful as you’d ever want him to be and no longer capable of that behavior. He says, “I cannot believe who I was. I just watched that documentary about me, and I do not recognize that guy.” It seems to be the same case to me.

DD: But you’re leaving out a lot of the recursive cycles of the whole situation. If we had a Madoff pill, then people would know there was a Madoff pill. And so a lot of people who knew that they could do this until they had to take the Madoff pill, would factor that in to their behavior. Let me —

SH: That’s a different case then.

DD: No, I don’t think it is. I think it’s important not to shield off the calculations of all the effects on respect for the law and the sense people have about the security that respect for the law gives them. You have to factor that all in, and the fact that there’s a Madoff Syndrome doesn’t change that. And it’s small comfort to the rest of the citizenry that when we diagnose a Madoff Syndrome case, we give him the pill, and then he’s better — not good enough. We want to have the threat of punishment there so that the people that have Madoff Syndrome behave themselves. Now, consider it in this light: most psychopaths never commit a crime. Why? They’re deterred by the penalties. And, here’s where my homely comparison with penalties in sports comes in: you wanna play football — soccer? You want to have a yellow card and a red card. And you want to use them sparingly, but when you use ’em you want to mean ’em, and it’s gonna hurt, and they’re real penalties. And would the game be better without them? No. Do we even stop to ask whether the players could control themselves better? No. If you wanna play soccer, you’re gonna live by these rules, and don’t even think about coming up and saying, “Oh, not me, I’m a hot-headed latin who can’t control his emotions,” — tough. You play by the rules or you get your red card. And this is an enabler for a pastime that many people enjoy and think is a pretty good thing, football. You can’t play football without rules, and you can’t play football without rules that have penalties. And those penalties are not just — it’s not for rehabilitation, and it’s not just to deter that individual in the future. It is to preserve the common mutual knowledge of what to expect on the pitch. And that’s the way the law works in general. The implacability of the law is a very interesting feature because it means, say, judges are often put in a position of wanting to temper justice with mercy, and they often do. But there must not be a law that says, “…and by the way judges, under various circumstances you ought to temper justice with mercy.” That subverts the law. it subverts the respect for the law, which is one of its main features. And I think if you look at it in terms of the rules of the game, or law and order, criminal behavior, then you can see that that simply heightens or clarifies the situation in all moral behavior. Holding people responsible — there is this curious — you might think it’s paradoxical; I don’t think it is — that we don’t want people to count on the mercy of their judges. We want them to be grateful when they get mercy.

SH: Again, I think we’re having two conversations here — there’s two topics in play — and we’re not necessarily noticing when we’re bouncing between them. So there are just the practical constraints of jurisprudence or the criminal justice system or the world as we currently understand it, and we have to figure out how to function in it — we have to figure out how to play the game in a way that makes sense. So yeah, I fully agree that punishment makes sense if it deters any significant number of people from crimes, and even in a state of total information we may still want to reserve certain punishments because those are the best levers to pull in terms of influencing human behavior, but then there’s just the ethical case and what is true of the world and how those two interact. And I think, just to go back to Madoff for a second, if we understood Madoff Syndrome, and this was a real condition which could be easily cured, and families who were aware of Madoff Syndrome were giving their teenage boys the Madoff pill before they went to business school —

DD: Why would they bother?

SH: Well, because it’s gross negligence otherwise. It’s like strapping your child into a seatbelt when as a parent it’s your responsibility to do that. You’re unleashing your Madoff prone children on the rest of society if you don’t give them this pill.

DD: It’s not optional.

SH: Well, everything’s optional. You could starve your kids to death, but then society will hold you responsible for that. What I’m saying is that because we don’t understand Madoff Syndrome and we have no cure for it, Madoff looks like a fully culpable moral actor who just belongs behind bars. But if the state of our knowledge of his condition changed radically, and there was a way to intervene, it would begin to look like neurosurgery for Charles Whitman, just a different case of it. And I’m saying that on some level everything becomes like that, from Tiger Woods’ putt to your reaction to my making point. I’m making this point and you are underwhelmed, right — that is a state of your brain which we dimly understand, but it’s not one that you are responsible for — you as the center of your subjective life. And now to connect this conversation back to where I think the illusion of free will or the sense of free will is really the motivating force on this topic for everyone: if we connected you to the right brain scanning technique of the future, and we could know in advance everything you were gonna think and do before you subjectively could, right, so you’re about to utter something, which we have in the lab already transcribed — to be confronted with that precognitive record of your behavior moment after moment would be fundamentally undermining of people’s felt sense of what they are as agents. If the guy with the white coat knows what you are going to do before you are going to do it, from my point of view, we have every reason to believe that is a neurological fact about us. You are not the first to know what you are about to think or do if we can get ahold of your neurophysiology in any real detail; well then that will completely shatter this sense for people that they are the author’s of their actions. And it brings us back to the psychological case for free will — not the pragmatics of it, not the this is how you have to design a game so that it’s fun or so that it’s orderly or so that we know what to expect. It’s this is who people feel they are in each moment.

DD: Sam, I think you’re just wrong about the effect —

SH: The effect of that?

DD: Yeah, I think so. We have to look more closely, and here’s where, actually, what looks like a empirical but boring detail matters. Some people think that the difference between prediction in real time and prediction about the discovery of things that happened in the past — that the difference between them is important to science. It really isn’t. People sometimes say that there’s no predictive power to evolutionary biology. No, there’s plenty of it. For instance, I predicted if we go to any island in the world and start examining the birds there, we can say a lot about what their DNA is going the show us. It’s not predicting the future in any interesting sense. We’re predicting that we’ll find certain fossils in the ground — it’s about the past. So what happens in neuroscience right now, for instance, in the Soon and et al. experiments, and Patrick Haggard’s experiments, it’s not real time prediction. They have to massage that data with a number crunching program for quite awhile in order to generate the so-called prediction, which they then check against what the person actually did. They get it right, you know, sixty-eighty-percent of the time, but they can’t predict in real time. If they could, then of course they could make real money by playing rock, paper, and scissors with the person with their head in the scanner — they can’t do that. But suppose they could —

SH: But that’s just a technological wrinkle; there’s no reason to think they couldn’t. But let me just ask you one question to sharpen this up, which I think will make it clearer: if I showed my phone now, and I had a transcript of everything we said in this conversation, right, so I knew in advance everything you were going to say, I knew what I was gonna say, I knew what you were gonna say, down to every syllable, and, on some level we know that’s true of us. The perfect mind reading instrument would get it five hundred milliseconds early, right, before you have a sense of even how you’re gonna form the word. That’s the undermining case, and I think there’s just — as a matter of determinism and indeterminism, however you want to knit those together at the level of cells — we know that’s true of us. Consciousness just does not get involved early enough for us to feel like we’re pushing the river.

DD: I think there’s two confusions in that: one of them is that your example has — it’s sort of irrelevant, but since you trotted out the example. Many years ago Donald MacKay — Dan MacKay — showed that, as it were, the Laplacian Demon can’t predict mere behavior if he’s gonna interact with you, because then he’s gotta predict his own behavior too, and to do that he’s going to have to have a complete description of himself, which you can’t have, as Turing showed. So, there is a theoretical limit to how good that — if your phone had the whole transcript on it, I guess what it would show is that —

SH: So make it a conversation with somebody else.

DD: Well, we’d have to be very careful how we wrote that. But you say that you think that would really undermine the idea that we had free will. I think it really depends on the details of the context. Let me build up with a simple case: you give me two simple arithmetic problems to do. I do them, I hand you the result and you say, “Ta-dah!” and you show me a piece of paper, you had written down the very results that I had written down. That doesn’t cut any ice because it’s sort of obvious. You know, I’m pretty good at arithmetic, you predicted I’d get the answers right, and I did. Now, if you asked me to write a poem: I sit down to write a poem, and when I hand it to you, you take out a piece of paper and show that you’ve got the same poem that has already been written down on the piece of paper. Well, what I’ll be sure is that you played a magic trick of some sort on me — but yeah, that would be very, very unnerving because I believe, and believe for good reason that the sort of complexity of the cycles going on in my brain are such that it is beyond feasibility to be able make that kind of prediction.

SH: — but even just five hundred milliseconds in advance. So if we had the poem reading technology hooked up to your brain, and we got the poem — we got each word of the poem half a second before you got it, and we could prove that to you, isn’t that the same case?

DD: No, I don’t think it really is. First of all, yeah let me just — I don’t want to get into a lot of science fictional imaginings about very, very extreme and physically dubious possibilities in principle — so, let’s sort of set that aside, I think we should, but if you want to go back to it we can. The idea that a conscious author has to be conscious of the creative process that generates each word — that’s an extraordinarily extreme and unlikely view of authorship. And if we look at history we see that Mozart and other great artists, they say, “No, it’s not like that at all.” Mozart says that, “These tunes come to me and I write them down,” but he claims authorship for them, and so he should. Why should he? Well, because nobody else wrote them, and they were processes in his brain, and he controls them to some degree. He controls them not at the micro-level — he controls them at a temporally macro-level. Thus, when Venus Williams returns service, she’s gotta put that stroke in motion before it’s fully reached full consciousness — I don’t like that way of speaking, but — she’s got a couple hundred — less than a couple hundred — milliseconds to shape her response. As she’s waiting for the serve, she’s making conditional plans, and those are deliberate. She’s decided that if she can, she’s setting up to do a backhand lob down the lane, and she’s deluded in thinking when she gets the serve she was expecting and does that, that she planned that. But the fact that her response happened so fast doesn’t show that this was not a conscious act of hers. I picked this up in your book, and I thought, this is one of the things I wanted to talk to you about too because you had an unrealistic demand on what a conscious decision or a conscious bit of authorship would be. Musicians — let’s take jazz improvisation on the piano — so I’m playing around midnight, and I decided the next chorus, I’m not sure why, but I’m going to up the tempo — unusual in that piece, but I wanna try it fast. Now, do I know exactly which notes are gonna come out when? No. In fact, I know what I’ll be doing is setting in motion some control circuits that I can’t control directly, but I’ve honed them — I’ve practiced things like this. They may be bad musical habits, but they’re my habits, and I know how to get a characteristic Dennett cliché to come out of my fingers at that moment. I’m the author of that, but I’m not — you seem to be holding out for a kind of authorship that would deny that that was authorship.

SH: I think there is this shift between first and third person views of free will, and I think the first person is primary in terms of describing what people think they’re gonna lose psychologically when they give up this notion of free, and I think they lose it to some significant degree under your compatibilism because we both repudiate libertarian free will. The “I could’ve done otherwise,” you think is just not important and it’s untrue, I think most people think it’s important and it’s untrue, and so we’re playing a slightly different game there, but the punch line is the same from the both of us: no, you couldn’t have done otherwise.

DD: I have to demur slightly; I think there’s a perfectly good sense of “could’ve done otherwise,” where yes, you could’ve done otherwise, in that you’re competent to do a range of things, so you have that degree of freedom. on this particular occasion you did one thing, but you could’ve done the other. It might be just the flip of a bit that makes the difference. That is a perfectly legitimate sense of “could’ve done otherwise.” The only reason you’re ruling it out is because you’re going for absolute, atom for atom, physical replication, but that just is irrelevant to the real world of causation.

SH: Well I would say that what you’re actually promising there is not that they could’ve done otherwise, but that they can do otherwise next time. So to bring is back to Tiger Woods’ putt: Tiger Woods couldn’t do otherwise because the bird caused him to miss the putt or a glitch in his nervous system caused him to miss the putt, and if you return the universe to that state exactly, he’s going to miss that putt a trillion times in a row, but he can do otherwise in the sense that he can be expected to make his next putt because it is within his range of competency to do that.

DD: There’s a paper that [Christopher] Taylor and I wrote together called “Who’s Afraid of Determinism”, and in it we argue that this is a common confusion about determinism — that this idea of rolling the tape back and playing it exactly the same way. Let’s take our canonical example of a random event: a coin flip with a fair coin. Well, it’s not actually random, it’s determined, but it’s determined by the position of every particle in the visible universe, but so what? It could’ve been otherwise. That’s what we mean when we say — we use coin flips to implement “could’ve done otherwise” when we need it, and that’s a perfectly legitimate sense of “could’ve done otherwise,” even though determinism reigns. The coin’s coming up heads is, in an important sense, uncaused . It’s determined by the whole state of the universe at that moment, but it has no more salient cause. Now if you make a coin flipper that is balanced in a mercury bath and has very carefully calibrated arms, and you put a coin and flip it, and you can probably make a device which will flip a thousand times and it will always come up heads. Then those aren’t random, and those are caused. Now, we can contrast cases like that with cases where they are random, all without touching determinism. So: deterministic world; some coin tosses are random and some aren’t; the ones that are random could’ve been otherwise.

SH: Well in that case, a coin flip is a surrogate for true randomness. We’re talking about a chaotic system, where you can’t predict the outcome because you don’t know the initial conditions, and you couldn’t do the math even if you did, but it’s a prosaic version of randomness. If I want to randomize a decision, I’m going to flip a coin and take the outcome. The issue is psychologically — and again, you and I agree that people don’t have this freedom, I just think that people put a lot of stock in the illusion that they do, and you and I might disagree about this — people can’t own the micro-causes, and they can’t go all the way upstream and be the author of their actions. I am making an unrealistic demand on their subjectivity: no one has ever had this kind of control or this kind of ownership of their actions. But the felt sense of libertarian free will does presume it, and I would say to you that if something like this neural imaging experiment we talked about were available, people would find it a total challenge to their sense of their authorship of their poem, or their authorship of their music or their authorship of their volitional action if everything could be predicted prior to their conscious awareness of any of the relevant elements. Before you heard the sentence in your mind to speak — before you thought for an hour about which word you were gonna choose here — and I can show you that at every point you thought you were deciding, several seconds earlier we knew what your brain was gonna do, that gives your brain the marionette feeling, even if those strings are attached only to the universe of causality, not to the hands of other person.

DD: Well, it would take quite awhile for me to unpack everything in there, but I do want to disagree. I want to suggest that when you say, “Well the coin flip, that’s just a surrogate for real randomness,” you’re making that move again: “That’s not real free will,” or “that’s not real consciousness, that’s just a cheap substitute.” Well, no, in fact, it’s not a cheap substitute. The difference between the way the world runs if we’ve merely got deterministic chaos and the way it runs if we’ve got quantum indeterminacy, makes no moral difference at all.

SH: I wasn’t saying that it makes a moral difference, I was just distinguishing between those two have physical facts.

DD: Well, but then, if it doesn’t make a moral difference, people are simply deluded if they think it does. So in other words, people are deluded to think they need what you call free will — they don’t. They can have — the kinds of free will worth wanting are perfectly compatible with determinism.

SH: We completely agree about that, so this might be a good point to end on because dinner is calling and our brains are inclining toward it through perhaps no free will of our own, but I just want to say, I’m very happy that we had this conversation in the spirit of collegiality that I hope all of our conversations would happen in. Because you and I, being the control systems that we are, we’re up to the task in most contexts. So I just want to say that if there was any point in that exchange — the written exchange — on this topic, that offended you, that made me seem less of a reliable ally for you or a friend, I regret that, and it’s been a great pleasure to collaborate with you in all the ways that we have, thus far, and I’m very happy we had this conversation.

DD: And I’ll echo the same sentiment back to you. This has been instructive for both us, I think.

[ Music plays]