Dan Dennett: Sam Harris Is WRONG [Ep. 406]

Everybody, just a little bit of a somber note as we start the podcast, something that's never happened before. But I have to report on the passing of today's guest, Daniel Dennett, who I spoke to just a few weeks ago. I think this was his final interview on a podcast. He passed away today, I'm informed. This is Friday, April 19, 2024. And as you hear, I had such a great conversation with him. I already recorded an intro, both the video and an audio intro. And, I was just so delighted and touched by him and had so much fun with him.

He influenced me greatly even though I really only got to delve into his work in the last few, months before leading up to the interview before, of course, I knew about him. He's a world famous intellectual and contributor to many fields of philosophy and many other things.

And so it's a great shock to me.

I wanted to release this, as a, you know, token of my gratitude to him for this wonderful interview. You'll see he, you know, he just come back from a dentist experiment and, you know, we're talking and, you know, part of me feels guilty to have spent, you know, an hour and a half, 2 hours, whatever it turned out to be with, you know, the final, you know, weeks of his life. But on the other hand, it's a great gift. And hopefully, his family will see this and share in the, in the delightful conversation that we had together. I view it as a great honor and privilege to have hosted him and especially now made more experiment by his passing. It's, truly affected me, and I hope the interview will be meaningful to you as much as it was to me as well. So with that, I'll start with the intro that we had planned before theorists sad, announcement came a few days ago before the release of this episode. And, and, I hope that Bang is is happy and resting in peace wherever he is.

Thanks, Dan.

Today, on Into the Impossible, we welcome a renowned philosopher of the human mind, a man who's a legend, a cognitive scientist, professor. He's also a vocal theorists, and yet he makes common cause with people like me who call themselves practicing agnostics. Daniel Dennett is a legend, and he's known as one of the 4 horsemen of new atheism, not the apocalypse.

Free will isn't a metaphysical condition that you're blessed with or not. It's an achievement.

He's been at the forefront of discussions on consciousness, free will, and the impact of Darwinian evolution on religious belief. His incisive wit, good humor, and keen intellect made him a must get guest on the Into the Impossible podcast. He's been a major figure for decades in debates, conversations, and writings about the existence of God and the nature of belief and free will. His works are tremendously influential and they include Brains the Spell, Consciousness Explained, and many more that have provoked admiration, controversy, and challenged readers to reconsider their most deeply held beliefs about the mind and its relationship to the physical world.

It's a user illusion. It's not a bad illusion. It's a good illusion. We are not the victims. Theory the beneficiaries of this illusion.

Today, I have the opportunity to explore these topics along with your questions for this phenomenal renowned professor. So without further ado, let's jump right in and discuss this magnificent new memoir from one of the heroes of the new atheist movement. How are you doing, Dan?

I'm doing just fine. How are you?

It's a great pleasure to, connect to you. I listen to your latest book in audio format, and, it's not your voice. And so it's good to hear your actual voice. And as you know, Dan, we love to judge books by their covers because what else do you have to go on on a Bayesian reasoning sense? So I want you to take us through the book, and it is unique in terms of all the 2 or 300 books I've had the pleasure of authors appearing on theorists the Impossible podcast. This is the, probably, the first one that doesn't have a subtitle. So tell me tell me the origin of the title, the cover illustration art, and the absence of the subtitle. Take it away, Dan.

Okay. Here's the book. I take it everybody can see it. Yeah. I've been thinking. I didn't want a subtitle because I thought that's enough. I wanna talk about my thinking and how I got there, and it's not about the nonacademic, non research parts of my life. I deliberately didn't want to go on and on about adventures I've had outside of academia.

I thought this is a book to talk about what I think is how I think I think and why it's a good way to think. So it's all about the wonderful thinkers who've who've helped me. And the the first thing to say is, if you if you wanna do some good thinking, surround yourself with the smartest people you can find and talk to them. And that's that's the trick.

And I've had, the pleasure of having a lot of brilliant thinkers on the podcast, 9 18 Nobel Prize winners. And many of your colleagues and friends and people that appear in this book in one form or another, including, folks like David Chalmers. And when I had David Chalmers on, he's, he's from Australia. And I said to him, you know, David, I I if I had the rock band C, also from Australia, and I had them on and I did not ask them to play, you shook me all night long, I would be a derelict in my duties as a as a host. So I want to, you know, sort of ask you. You've had some very deep criticisms of of them, obviously always with respect and always from a, scholastic scholarly perspective. But in this book, you talk about your differences with, you know, critiques of everybody. All these guests that I've had on, Penrose, Hammeroff, Hoffman, Chalmers, Sapolsky, Harris even.

So let let's start there. Let's start with Sam Harris and and, and then we'll we'll kind of work our way through for the audience's benefit. We hear all these things, Dan. Some Clarke, consciousness is an illusion. Some say it's, it's, you know, completely nonsense. There's no such thing, and associated with free

will. That?

Who says there's no free will?

Oh, yeah. Various people say that, but I don't.

Yeah. No. I know that. So what is the the the critic of, the critique that you have of of, say, Sam in your book, freedom evolves? I I and I'm I'm I'm mentioning this because, you know, he was just on the podcast, and we had a a a very long debate in which I asked him similar questions that I asked Robert Sapolsky. And I said, you know, if if there is, you know, no such thing as free will, then how can you blame somebody for God forbid, you know, killing your pet dog? And Robert said, he said, literally, to my great shame and humiliation, I'd want them punished. So where do you come down on, say, crime and punishment in a in a world, with the free will perspective that you adopt?

I think there's a definite role for punishment and have argued that and think there's nothing antediluvian or anti scientific about it because free will isn't what Sapolsky thinks it is. I'm just astonished at how both, Sam Harris and Robert Sapolsky and some other scientists have been, I think, persuaded, pawned, really, into thinking that free will depends on indeterminism by some philosophers who who have inflated free will beyond what it actually is. Free will isn't a metaphysical condition that you're blessed with or not. It's an achievement, and it's the achievement of mature self control. You don't have it when you're a baby. You don't have it till you've reached adulthood, really. And so we don't hold you responsible for things until you're an adult and until and it it we don't hold you responsible then if you're if you're if you don't have self control. That's self control is the key notion.

And the theory about self control that it it amazes me that that Sapolsky doesn't realize, this is one of the best ways of looking at evolution. Evolution begins with the simplest imaginable agents, single celled agents. And then we get multicellular agents. And then we get multicellular agents that aren't, you know, plants or or or fungi or or, coral polyps, but that that move. And once you've got motion, you've got control. And in order to have control, you have to look ahead. And evolution has designed things that can look ahead. Before there was life, nothing could look ahead.

Nothing at all in the whole universe could look ahead. Once you have look ahead, then you have the possibility of making choices based on what you see. Now maybe what you see isn't what's gonna happen, and then you may make choices that are bad. But over the long run, probabilistically, evolution, let's replicate the agents that are the best at self preservation, that duck the incoming bricks and then find the food before they starve, that find mates, and so forth. I haven't said anything controversial. That's textbook ho That's right. That's how evolution works. But it makes things that do things for reasons.

And once you have things that do things for reasons, you're on the way to free will. Not I wouldn't say that an octopus or a clam or even a crocodile have reasons that they understand. They don't have to, but they still do things for reasons. Trees do things for reasons. Trees don't have to understand the reasons. They do the things they do, but they do things for reasons. That's a theme in in my book From Bacteria to Bach and Bach. This is competence without comprehension.

We're the one species that so far evolved that doesn't just do things for reasons, but represents reasons to ourselves and argues about reasons and tries to reason others into behaving better and so forth. And this creates the social contract. It creates the environment for civilization where we can judge that some of our fellow human beings have reached the age of reason they can be reasoned with, and we can trust them. They're safe. We we can let them run free. In a way, we don't let lions run free or bears run free or small children. And once you're capable of listening to reasons and being moved by reasons, as Kant put it, then you can have freedom. Have nothing notice I haven't mentioned the word determinism.

It has nothing to do with determinism. Determinism and free will are completely this joint categories. There's no there's no, implication one way or the other between them.

So this is what's called a pattern interrupt. It's a way to rejuvenize, refresh your mental synapses as I know you're getting a slight charge out of hearing each word that Daniel and I say, But I need to take a quick moment to invite all of you subscribe to this podcast or YouTube channel no matter where you're listening or watching. I promise you it's causing me to up my game. You see the phenomenal guests we're getting just in the realm of consciousness, including Bang, and Sam Harris, and Robert Sapolsky just in the last 3 months. It's been a phenomenal ride. And unfortunately it's a numbers game, and I'm trying my best to up my game and become a better interviewer. This is my side hustle after all. It's a labor of love.

I don't make very much money on it, but the one thing you can remunerate me with is by subscribing. Only about 50% of you are actually subscribed or following the podcast. So please do me a favor. Subscribe and share. It really helps out and will help us grow and continue to get great guests like Dan, Sam, and Robert. And stay tuned for a special episode coming up with Don Hoffman. When I mentioned to both of them, I I hear a lot of people, as as you know, will deny the existence of free will, and and, I mean, that's a title essentially of of books by those authors. And I say, have you ever met somebody, Dan? I say, have you met somebody who behaved as if they don't have free will? That's not a that's not a psychopath.

And they can never say, yeah. I mean, nobody behaves like they have no free will.

The way you would behave if you had no free will is you'd you'd you'd sit there like a tree and just take your lumps and and not think ahead. And it's possible to talk people out of their free will because if you've got free will, you can be moved by reasons, and you can be moved by reasons good and bad. And so that's why I think books like Sapolsky's and Sand's are actually a little bit socially destructive. They they're they're acts of high class social vandalism in that they weaken our conviction, our perfectly naturalistic conviction, that we are what we obviously Clarke, forethinking, reasonable human beings who can figure out how to do things together in concert, avoid harming others. That's that's the glory of human civilization.

Yeah. It really is, and it should be it should be celebrated. And I feel like they get into these sort of, almost solipsistic or, you know, self self referential definitions where they can't admit that there are, possibilities where the very notion that they're trying to criticize undermines theory own argument. And for me Why

are they arguing for him?

Yeah. Exactly. Right.

What on earth do you think you're doing, Robert?

It's impossible not to, you know, be swayed. You know, again, when you make it personal, etcetera, that they, you know, that they don't stand by the courage of their convictions. But to their credit, I think at least at least, Robert does. And he he admits to his shame as a as a as he literally said in the interview.

Robert Sapolsky does, and those are some of the best parts of his book where he confesses that he can't sometimes he he just has to act as if he has free will. Good for him. Yeah. Of course, he does. I trust him.

And you talk in the book about the kind of difficulty in understanding from an evolutionary point, the evolutionary or selective, purpose of sense of humor, which is, you know, almost probably uniquely a a human trait. Maybe maybe there are some higher primates that that have it. But what about the the origin of music? It it would seem that, again, if you're sitting around a campfire and you're you're clucking or you're dancing or whatever, it makes you kind of theory, very unfit to survive, you know, the lions that are that are prowling around you. I always wonder, what is the evolutionary advantage or point of music and which is not unique to the human civilization, obviously, but especially in humans. What would be an advantage, if any?

Well, let's let's look at some nonhuman species. We have we have birds, for science, some of whom have remarkably wonderful bird song and that that, don't just have a caw a caw or a chirp chirp, but very elaborate songs. And there it's pretty clear that the that the, point of that is sexual selection. It's like the peacock's tail. It's also beautiful. Sexual selection, well studied by Darwin. It's not just the survival of the fittest. It's the survival and procreation that matters.

If that's the finish line. You gotta procreate. You gotta replicate. And and the the crossing that finish line means you gotta attract a mate. You gotta get somebody to mate with you. Certainly, that has a lot to do with the ornamentation and the beauty that we see in many animal species. And it it's in sort of an arms race because it depends on the, females in almost all cases, the sexual selection. The females are the ones that do the judging and the males that do the showing off.

And, costly signaling theory, Zahavi's wonderful contribution to this, is that you can't have a cheap advertisement of your own excellence. It's it it's, that will not it's not because the females will understand that these are cheaters. It's just that they won't be attracted to them. It's gotta be something difficult. It's, it's the same thing as with starting, or pranking in in antelopes that do these incredible leaps and get the lions not to chase them. They're saying, don't bother. Don't waste your energy on me. I'm too good for you.

Look. I can throw these leaps. Only a inexpensive costly signal can send that message, and only an expensive costly signal can send a message, hey. You wanna mate with me? Because I'm really I got I got energy and time to burn. That idea, that that motive sells a 1000000 guitars a year. That's right. But some of those guitarists decide they'd rather make music than love.

Yeah. It is, sort of, evolution has has, you know, not done us physicists and professors any favors in the mating. I'm okay with mating. I've I've I've done okay. I've done my fair share. But, happily married for 16 years with a bunch of kids. But the, you know, kind of question that always comes up to me is I I had on the physicist, Michio Kaku. And, again, I'm always astonished on how, you know, kind of self referential again that that, you know, it it's believed and he claimed in in the interview I did with him that, you know, evolution takes over, and then that's how you get life from inorganic, you know, hydrogen and helium.

And it no. There's nothing of the sort. And so I wanted to ask you, what is the minimum viable product? What what is the minimum thing that evolution needs to operate on? I had Craig Venter on, and and he said anything theory has DNA, but, I mean, we can imagine things that don't have DNA. Right? So what do you think? What's the minimum viable product that mother nature could produce or even be contemplated to produce in the entire universe to originate life? And then how does evolution take over from there?

We know something about this. We can sketch out some of the requirements. There there have to be there has to be, some fairly stable, nonvolatile macromolecules because there has to be structure. The chemoton was a sort of simplest life form that was described in some detail by the Hungarian, science whose name is escaping me at the moment. But, the Kenaton. Look it up, c h e m a t o n.

Tibor Ganti?

I learned about him from, Urs Zatleri, the Hungarian Arthur with with John Maynard Smith of of the wonderful book on major transitions and evolution. And what you have to have is a you've gotta have a protective envelope. You've gotta have you've got to have a a a a tissue that that that surrounds you, a cell wall, in effect. And it has to be permeable. It has to be controlled entrance and exit for raw materials and waste products, and it has to have reproduction, and it has to have a source of energy. So metabolism, a, a protective, skin of some sort, and a reproductive system. One of the points that I think is worth making is that to get evolution started, this is still a a a deep puzzle. There are lots of but but the beautiful thing is there's more than enough theories out there.

There's an embarrassment of riches, lots of ideas. There's the DNA, the RNA first world. There's various other ideas out there. Nick Lane in in England has some excellent ideas about the, original sources of energy. We don't know yet for sure. Well, but but we're closing in. I hope I live long enough to get somebody really, really hitting the target and everybody agreeing, which could which could happen which could happen and happen soon. But the first thing that reproduces doesn't have to reproduce fast.

If it takes a 1000000 years for it to make a copy of itself, who's counting? And and so there's competition, you you can you can reproduce slow and still get the benefits of the evolutionary ratchet. The evolutionary ratchet is the key. You've gotta have replication and selection.

And, you know, speaking about, you know, replication selection, you make the case in the book and based on your, earlier works, you know, that language is almost also certainly, an evolved process, and there are higher order and lower order languages. Although we don't see, you know, Denisovans walking around, we still do see primitive languages with, you know, even non written languages that exist on the planet. I interviewed your cross town rival, Noam Chomsky, and he sort of made a persuasive case that, to me, had implications for artificial intelligence. And that was that, you know, the communication, there's so much nonverbal and and even the most durable form of of communication might be generated nonverbally. And that made me think what extent can these LLMs that you mentioned in the book as well, can they ever achieve, you know, a Turing test level? We'll talk about the deficiencies and problems you have with the Turing test later. But, could it could a LLM that doesn't have a body or, you know, doesn't have the ability to pop a circuit, you know, to cause it to feel pain when it does something wrong. Could it ever hope to evolve, you know, or or to to be present itself as a as almost, you know, human level artificial intelligence if it doesn't have embodiment?

Well, of course, it has to have experiment in one sense. You have to have some some hardware to run the software on. One of the points that I like to stress these days is that brains are not at all like Von Neumann machines. They're not they're not one of the amazing things about computers, and Turing was very clear about this, is that they have to be very bureaucratic. They have to be very rich, and they have an operating system. And they have to do you have to know exactly what they're going to do in order to program them. Their design depends on the uniformity, on the fact that there's 1,000,000,000,000,000,000 of exactly identical elements almost to the, yes, to the atomic level. Your flip flops, your your registers, and you have timing pulses, not brains aren't like that.

Brains are made of 1,000,000,000 of individualistic neurons. No 2 are exactly alike, and they don't act in quite the same time scale. So it's an entirely different underlying structure. Now could you nevertheless build such a structure in a silicon, digital computer? Yeah. Sure. You could. Because you can you can simulate it in the same I mean, you can simulate an analog machine in a digital machine. You can simulate a parallel machine in a serial machine.

You pay big prices for that in time and and and energy. LLMs are incredible energy hogs compared with human brains. Energy is really key. And some people, for instance, Terry Deacon argues that for all the wonders of computers, they ended up getting us to explore the wrong part of design space because all the computers that were designed following, the Turing and Von Neumann and the like were, in a sense, he calls them parasitic. They didn't have to worry about energy. They were provided for by their plug into the power. And this means that all the designs, all of the space that we've explored has been space that depends on there being a sort of steward, shepherd, nursemaid to take care of the machine to make sure it's energy. No no neuron, no circuit has to worry about whether it's gonna be alive or whether it gets enough energy.

Whereas the neurons in your brain are working for a living. They they will die if they don't connect. And so your brain is made of neurons and glial cells even more that that the the neurons are looking for work, and there's no HR director. There's no human resources director or Anna or neuron resource director. And when when a part of the brain dies, neurons hungry for work will take over those tasks. We don't have anything much like that in in digital computers yet. So brains are they are computers, of course. They're not radiators.

They're not for cooling the blood. They're they're control centers. They are the the control headquarters for movable arms and legs, for for mobile things. That's what brains Clarke. And so they're computers, but they're not much like digital computers. Still still, you could simulate all that in principle in prince

I wanna get your thoughts on Sir Roger Penrose, Nobel Prize winner, good friend of the show. His take is that human consciousness is nonalgorithmic, and so is not even capable of being modeled by Turing machines. And he actually believes in sort of a quantum mechanical understanding of human consciousness. He implies that not only quantum mechanics is responsible for consciousness, but gravitational forces are at work via what's called the vile curvature, which is a derivative of Einstein's stress energy tensor and gravitational, curvature tensor, g mu nu. So what do you make of these physical interpretations where the where the microtubules are caused to theory wave functions called collapse, caused by the local variance of a classical field. So quantum mechanics propitiates a, is propitiated by a classical mechanical, structure like Einstein's relativity. The g mu nu is a classical tensor. It is not quantum at all.

What do you make of these physical interpretations?

I think it's malarkey. Yeah. And I thought so. You know, I think I wrote perhaps the first review of Roger Penrose's, Emperor's New Mind, and I pointed out the problem right there. He has the wrong notion of algorithm that he's using there. He's thinking of algorithms for things. And, look, there's there's no feasible algorithm for chess. There isn't.

It's it's not an infinite game, but it's there's no feasible algorithm for it, almost certainly. Well, that means computers can't play chess. Right? No. It doesn't mean that at all. It means that they can play very good chess. It's just that the algorithms that they use are algorithms for playing legal chess. And some and how many of those are there? There's gazillions. And some of them are better than others.

There's no algorithm for being a perfect mathematician, but there's algorithms for learning a hell of a lot and doing pretty well. And don't expect that you're going to have an algorithm that guarantees truth ever. He's just setting up a preposterous standard for what a mind is. And right. So does that mean the mind is not algorithmic? No. It means there isn't the master algorithm. Even some people in AI sometimes talk about the master algorithm, but it is not a master algorithm in the sense that the the, Penrose thinks it's an algorithm for doing pretty darn well. And how many of those are there? Gazillions.

That's right. More than stars in the sky.

The big the big mistake this big mistake goes back to Descartes who wondered if he could theorists his clear and distinct ideas. And he decided he could if if God would guarantee them. And so he tried to prove the existence of God so he could trust his clear and distinct ideas. That's a hopeless quest. The best we can do is gather the smartest people around we could find, let them compete to find the truth, and see where you find intelligence, see where you find agreement. And that's the best you can do. When it's good enough, it get it gets us to the moon. It gets, robots to Mars.

It builds bridges and and cures diseases and allows us to predict eclipses years in advance. All of that knowledge is defeasible. It's not like geometry.

And even in the context, you know, staying with Einstein for a bit, my my favorite, you know, kind of counterpoint to the claims of AI, you know, apocalypse is the so called story of Einstein's happiest thought, which you may know, but I'll repeat it. So Einstein said, quote, my happiest thought was that an observer in free fall would experience no gravitational forces. And it led to the conception of the so called Einstein equivalence principle. And the reason I bring that up is because I'm curious how a computer might be expected to, a, visualize what free fall might feel that sensation in the pit of one's stomach as you, you know, crest a hill or on a roller coaster or launch on a SpaceX rocket, a, and b, whether or not said computer could identify with this happiest thought. In other words, there seems to be something, you know, sui generis, something, I don't know, that Einstein could have felt. And I don't know. I I propose that as the Keating test. You know? Can algorithms come up with completely new laws of physics, laws of nature, things that are verifiable, empirical, connected to data such as the type that my colleagues and I collect through our telescopes.

What's your take on that? Are there possible worlds where, you know, possible in areas where AI can actually create new laws of physics? Not discover, oh, well, the Navier Stokes equation behaves like this, so we should render smoke like that. No. No. No. Truly new. A Newton's, 6th law. You know, something of 5th law of thermodynamics. Can you envision that, Dan?

Yeah. I'll I'll tell you why. Yeah. All learning, all invention, all discovery is a matter of generate and test. It's all that is what evolution does. It's what we do. Right now, you've got lots of possible thoughts running through your head. Some of them are get getting thought, and some of them are dying.

They're not they're not rising to the level of you're not gonna say them and you're not even really gonna think them. But it's that's what's going on in your head. It's what's going on in my head right now. We're all cherry pickers. Now cherry pickers, first, you do it rough, then you do the quality control. You have the the fountain that generates lots of stuff, and then you have the critic, the, judger who decides what's worth further work. I think that LLMs, for instance, can be very valuable in the fountain roll, in the generation role. They can be very good at generating off the wall things that you or I would never think of.

Why? Because they're not like you and I. They're they're they're different. They're enough different that they bang come up with gonzo ideas that might for someone say someone might say, oh, I wish I'd thought of that, but I never would have thought of that. Now we all have styles. Chopin had his style. Mahler had his style. Beethoven had his style. Wonderful.

But that means Chopin doesn't have Grimsky Korsakov style or Rachmaninoff style. Don't expect Chopin to write a Gershmanin. He bang hear it just fine, but it would never occur to him. And I think that LLMs Keating on the scrapings of the Internet for years years and tremendous data mining and digesting, but not just the way we do it. They might be a great source of thinking outside the box, of off the wall ideas that we, who are humans, would just not it wouldn't occur to us, and they'd be right. Look. When we look at the history of great science, we see the the really wonderful breakthroughs are often where people come up with an idea that first C sort of daft and even outrageous, even impossible, and says, wait a minute. Wait a minute.

Wait a minute. Maybe something here. And I think that, we've now got a new generator to go with our testers. And we're still gonna rely on human testers. Well, they can do some testing too. The the AIs can do some testing. But I think we wanna keep them as smart machines, not artificial colleagues. We don't wanna give them the autonomy they could have because then they'll be dangerous.

How do we enforce that?

By keeping them parasitical, making them machines that don't have to fend for themselves, that we can unplug. In principle, we don't have to do that. We could try to make them as self sustaining, as agential, as as autonomous as we are, but we shouldn't. And one of the things that I use to point this out to you, I say, just imagine that you learned that there was some person or some institution that had your on off switch. What would you one of the highest priority goals be for you?

My my welfare.

You you're getting resting control of that switch. And if they're that smart, they're gonna be pretty good at it. And we already see inklings of that in the red team testing where where I think it was GPT 4 that conned a human being into, identifying a CAPTCHA because it didn't have eyes. So look out. We we don't want that. We don't we have enough psychopaths and sociopaths running around as it is, and AI agents will be sort of natural psychopaths if theory may because they are sort of immortal.

They're immortal up to planetary constraints. Right? The paper I pointed this out to Nick Bostrom and others. You know, the there aren't infinite amounts of, you know, iron and, nickel and so forth in the earth's crust that are easily exploitable by fellow, you know, AIs and and agents. But you're right. And it is interesting. I asked Sam Harris, and I'll ask you, I'll tell you. But I said, Sam, you don't believe that humans have free will, but you believe AI has free will. What do you think he said?

I don't know. What did

he say? He said yes. He thinks it does. So, and I love to

say it. He thinks AI has free will?

He thinks AI Harris. He said he they he believes they can develop free will. Not maybe not now, but they can develop a, free will. I mean, he's a very significant, opponent, and and he really believes that we should be extremely cautious with AI.

Well, so am I. I am I am sounding the alarm. I have a piece about counterfeit people in the Atlantic, and which is and other pieces in progress. And I've been talking about this basically with every audience that I get. We're really in danger of being lulled into fascination with large language models, things like GPT 4, to the point where we're gonna be turned into puppets because we will be reasoned with and cajoled and fascinated and seduced and lied to, and we won't know who to trust. And once we lose trust, civilization is in deep trouble. We rely on trust. Hey.

It's me again. So sorry to interrupt this fa the end of the deep dive, but I need to assign you a little bit of homework, but there's something in it for you besides the intellectual knowledge that you'll gain from joining my Monday magic messages that I send out each and every Monday. I share everything from around the universe of ideas that I explore, our exclusive content highlights, the latest science news, and, of course, the occasional fun facts you can impress your friends with magnificent dad jokes and memes. To get on the mailing list, just head over to Keating bang join. That's brianketing.com/list. And you will be entered also to win one of these chunks of space schmutz, a real 4000000000 year old piece of the early solar system. And I send one out every month to a lucky listener and subscriber to the Monday magic mailing list. But I always send one out guaranteed if you have a dotedu email address.

And to get that, go to briancating.com/edu. Thanks. Now, let's get back to boggling the brain with Dan Dennett. I agree. And, you know, the the thing that I point out, you know, how many phone numbers can you remember right now, Dan? I mean, you might be able to remember more, but, because we grew up in an age, you and I, without cell phones. But I I know my wife's phone number and my own phone number, and that's about it. And how many how many people know directions, you know, to any new place or can derive, you know, direct across town. We've outsourced these things.

So it's obvious we're gonna outsource thinking to these machines. Right? And and they're not vetted.

In the making regards. Yes.

Yeah. And they're not vetted and and and they're wonderful and they're and they're alluring because they do the work of a 100 graduate students and they get it more or less right. Except when you ask it point of fact, you know, render a picture of George Washington and you get a beautiful, you know, Oprah Winfrey like character with a with a white wig. So theory there's still obviously guardrails put in accidentally by, you know, probably some, you know, 21 year old, you know, interns at Google, and that and that'll be taken care of. But but eventually, we won't know who to trust because I trust them to, you know, get with questions I won't ask my rabbi, but, but then sometimes I'll get advice that, you know, it tells me I ask it, what books has Brian Keating written? And it'll say losing the Nobel Prize. Okay. Fine. It'll say into the impossible.

Yes. Thank you. Then I'll say a brief history of time. I'll say no. No. No. That that was, you know, professor Hawking. Oh, I'm sorry.

My apologies. So I think we're gonna have to have some way of vetting these these entities and and understanding who

we're talking about. I think the that's why I like the term counterfeit. I don't know about you, but do you look carefully at every $20 bill you get? No. Never. No. Neither do I. Why? Because we have laws in place. It's a very serious crime.

You can go to jail for 10 years if you get caught passing counterfeit money or making it. The technology is very good. Not perfect, but it's just not worth people's while to make counterfeit money. We can make it so it's not worth people's while to make counterfeit people. We can put in place the watermarking, the systems. For starters, every hardware manufacturer who makes anything that connects to the Internet could have and be required by law to have the detection. It's like color copiers in Europe where they have a lot of colored currency. If you put euros in in a high end color copier, it won't it won't copy them.

And we can have things like that, which will and and it will be particularly against the law to introduce a counterfeit person that doesn't have the watermark on it. That would be, something which is clearly not inadvertent. And when they catch people doing it, they will know they're doing it for nefarious reasons. And there will be penalties for that.

That's right. And especially when the stakes are I mean, I feel like we've already reached the tipping point. Right? This year is an election

We have.

This year is an election year. And can you imagine the robo calls that you get from Joe Biden Keating, it's not important to come out to vote. You know, I I need your help to get me out of jail. You know, it could be so nefarious, and there's a plethora of of brilliant we we tend to be very America centric. Right? And but we don't think about the brilliant engineers working in Iran or in Russia or or any of these that are malevolent and and would like to see nothing in this chaos agents sowing the seeds of discord among our democracy. And so we have already passed the Rubicon in some sense. This era, I'm I'm very worried about the election this year, you know, considering, whether or not it will be, manipulated by artificial agents, counterfeit humans as you say.

I wanna

talk about, your amazing book From Bacteria to Bang and Bang. It inspired me to wanna write a book called From Rocks to Rachmaninoff and Bang. We'll see if that if that ever comes out. So it talks about the, effects of evolution on human cognition. And in particular, I'm curious about the field of bioengineering and bioethics, and and how might our understanding of evolution and maybe a responsibility, not to viruses, but if we have ethical responsibilities perhaps to artificial humans, What, ethical quandaries could come up in the age of CRISPR and, genetic modifications? What do you see? Are you optimistic, about, you know, human's ability to to keep, the the, you know, ethical controls in place, or what would be your equivalent warning, if you have one, for bioengineering?

First of all, the biologists Clarke ahead of the AI people, and that they are they've been sensitized to this for for quite a few years, and they have the various levels of safety for not letting, artificial created or genetically modified organisms loose because they realize that they can they can replicate and and and create horrible catastrophes. And that's true also of AIs. That's one of the points I make in my piece about counterfeit people, is they can replicate. Computers are great at replicating things. And this is this Jeff Hinton and I agree entirely about this. This is this is a real danger, is that they they can replicate. And it doesn't even need evil paid actors. It doesn't need bad actors.

It just needs a few slips by second rate engineers, and, and and we'll have replication getting on. This has been a concern for years, ever since the field of artificial life got started. CRISPR and and other technologies in the pipeline are gonna make a huge difference. 1 of my, favorite biologists, Frances Arnold, is breeding proteins that don't exist in nature using artificial evolution. And she's trying to make proteins that will, you know, turn trash into butane or or fuel, and and it it's entirely possible. And her thesis supervisor said to her, but, Frances, there's there's no proteins in nature that do anything like that. Jesus asked because there hasn't been selection for them. And now there is.

And is she making progress? Well, she got the Nobel Prize, so in in chemistry. So, yes, so we got some great scientific advances in the future theory. But I think people are being suitably suitably cautious about it, and maybe the current theory about AI will get everybody thinking a little bit more carefully about some of these prospects. And there are I I was involved in a a National Academy of Sciences group that was asked about the whether we should do allow human fetal stem cells to be xenotransplanted into chimpanzees. And we decided no. No. This this is we do not want to risk creating a hybrid primate that is problematic in a way such a hybrid the the the gaps in the in the in the species that, Darwin noted, they're very important. It's very important that there be gaps, that there be boundaries to reproduction and boundaries and that we have some pretty clear cases that that the we have islands, not all archipelagos attached to each other.

And, we don't we really don't want to make, chimpanzees that have, in the womb, brains that have human human DNA in them.

Exactly. I wanna take the opportunity to talk, with you about the philosophy of science, in particular, the philosophy of physics. And that's to raise, you know, sort of a a question to you that I've asked to several other, you know, philosophers like David Albert who was visiting me last month. I say often that, you know, a lot of, social scientists, they they're accused of having physics envy. You've probably heard that, Kennard. But I actually believe that physicists have mathematician envy, and that at least a mathematician has girdled to lie back fall back on to say what is possible to be considered as part of the program of mathematics. But physics if you ask a physicist, they'll probably mumble something about popper and falsification. But, but I think that's really out of vogue these days.

So what would you suggest to me and my graduate students as a as a physicist, experimentalist in my case? What would you suggest as a good, you know, alternative definition for them to decide what is, you know, crack pottery and what is what is legitimate science for us to take, to take a deeper interest and invest our most valuable resource time into.

First of all, I'd like to draw attention to, distinction that one of my philosophical theory heroes, Wilfred Sellers, made between what he called the manifest image and the scientific image. The manifest image is that's the world we live in. It's got, tables and chairs and rainbows and colors and and and baseball and money and consciousness and free will and all those things. And then there's the world of physics or of science in general, and that's where you have the atoms and the molecules and the quarks and the, law of gravity and all of that. What's real? And as you know, there have been times when some physicists wanted to, say, well, it's the tables and chairs that are real, and physics is all sort of, just, well, as, my friend David Moser one put once said wonderful line. He said, quarks are the dreams stuff is made of. And then, of course, you have others who say, no. That's just atoms in the void, and and there aren't really minds or bodies or colors or anything.

Well, those are, I think, myopic positions that we all live in the manifest image, and we couldn't get through the day. And thank goodness for evolution. It's provided us with the manifest image. Evolution, including cultural evolution, has provided us with a a well behaved world of, middle sized dry goods, hardware, cars, boats, people, dogs and cats, colors, rainbows, baseball, etcetera. Let's call that real. Even though we can recognize that in some sense, it's all a user illusion. Nature has provided us with these smearing the boundaries, fuzzing it up. It's a user illusion.

It's like the user illusion of your cell phone. It's it's it's not a bad illusion. It's a good illusion. We are not the victims. We're the beneficiaries of this illusion, as same way you're the beneficiary of the user illusion on your cell phone. Let's get rid of the idea that the claim that this is illusory is, as it were, derogatory. Now, yippee, we've got this wonderful user illusion that nature has provided for us. And now we have software engineers who are copying nature and making other user illusions for us so that we don't have to understand what's going on inside our cell phones.

Now is the user illusion real? Well, yeah, it's real. And, you know, it's it's it's in terms of of LED patterns on the screen and little sound effects and things like that, which are all quite adjustable. Now how about the user illusion in our heads? Is that real? It's real, but it isn't what you think it is. There aren't any colors. There's no screen in your head. It doesn't have to be. It's because you got eyes. You look at the screen and or you look at the world, and the user illusion is made for you to use.

And you use it, and what it's of what it's made of is tables and chairs and dollars and music and poems and people and all the rest of the things in the manifest image. So that's that's reality, and it's also in the sense that physicists and biologists and others can understand. It's it's all sort of illusory.

One thing that's, you know, kind of struck me over the years is, how academia has changed. And your book is really a, a wonderful sort of series of time capsules. I don't know if such a thing has, has even been invented, but it's a memoir Brian it's describing in your own words, how, side quest, how, how academia has changed. That's the way I read it. And, you know, theory really delighted to see that. And I wanted to get your, you know, take on the, the future of academia. And especially in light of, you know, things like we've already discussed artificial intelligence. I often make the case, you know, why should students learn, you know, special relativity from Brian Keating when they can learn it from Albert Einstein through the 10,000,000 words that he has recorded in brains.

And we can make a LLM and a holographic rendering using, NVIDIA graphics chips to render everything down to the last wrinkle. So, our profession of the professoriate hasn't changed since the year 10 80 in Bologna, Italy. Do you feel like we're at risk, that we might be the last generation, the to to profess in the way that we do? Brains, you know, or or is it more resilient? I I felt like COVID would be the end of professorship if it were vulnerable. But but, but it's it's it's resilient. So what do you make of academia? How much has it changed, and what do you see we for the future? Are you advising a bright young graduate student? Should she go into this field?

Well, the immediate future, I think, is we're gonna have to douse the flames of polarization between the woke and the anti woke. That's a great distraction and a and a great mistake. Well Keating, as most things Arthur forward by well meaning people, but, I think deeply problematic. I blame it a lot on the postmodernists, who said we don't have to worry about truth. No. We need to worry about truth. Truth matters. And we have to be the defendants of truth, and we have to be the defendants of academic freedom.

And we have to recognize that there are hard truths. Well, let me say something a bit surprising, maybe. Is truth, as it were, all you need to worry about? No. There are truths that we don't need to assert and that we don't need to discover. And if you wonder about that, just ask yourself if there's truths about yourself that you don't think the world would be better knowing. Certainly, you wouldn't be better knowing them, and, nor would anybody else be better. Secrets secrets have their place, and they have and and we just don't have to explore them. And there are areas of scientific research that we could just don't do it, man.

Don't do it, lady. It's these are these will just make more trouble than they're worth. And find another topic for your curiosity. But the truth does matter, and it's truth. And people aren't entitled to their own truth. They're not even really entitled to their own beliefs. If their beliefs are stupid enough and ill informed enough and if they're radically, victimized by disinformation, then at some point, we should hold them responsible for that. I think this is one of I think this is one of the hardest things to figure out.

How do we make people responsible for their own beliefs? And we all rely on our informants. I have my informants that I trust. In every field, there are the scientists, philosophers whose opinion I trust. And I may have made some mistakes. I may have had some curious informants whose views I shouldn't trust. And I'm on guard for that. And, of course, in my in my books, I've gone after a few that I think shouldn't be trusted to the extent that they Clarke, academic bullies, for instance. You know Goodhart's Law? Goodhart's Law is that when a symptom becomes a target, it ceases to be a good symptom.

Publisher perish is a good example. Yes. We come up with something which is a pretty good symptom of excellence, and then people game the system. And nature gains the system. Evolution discovered Goodhart's law 1,000,000,000 of years ago, and there are plenty of examples where nature brains the system. It's don't whenever you make laws, for instance, you have to expect that there are gonna be loopholes that if there are any loopholes that'd be found and you can't make a law without loopholes, expect some peep you made the law because people wanted to do something, and they shouldn't. They're still gonna wanna do it, and so they're gonna look for loopholes. So Goodhart's law is a is a very important principle, and it's it's a basic principle of nature.

And it governs academia as it governs all other things. So don't expect a perfect fix. We're just gonna have to roll with the punches and keep fixing things as we go and recognize that people, some well intentioned, some not so well intentioned, are going to gain the system when they can.

Yeah. So sticking with academia, but only tangentially, do you remember a former student of yours named Jonathan Blackley? He was a physics major, and he is a video game designer. He's actually credited with designing the original Xbox. He was a physics major at Tufts in in the 19 eight eighties late 19 eighties. And he, used to call used to call him, you took a he took a class with you, and he said he was called by you an example of the scientific mind. But, but the thing that he wants to thank you for are well, 2 things. 1, he wants to thank you for recommending the Chef's Choice knife sharpener that he still has from 30 plus years ago. So he thanks you for that, but he also wants to thank you for leaving Twitter.

Now I, you know, I didn't do that research. Why did you leave Twitter? What you you criticize Elon's. You know, you don't wanna be a part of it. Can you explain what did you mean by that?

Oh, I think it's obvious. And I don't have any deep reasons for leaving Twitter. I just I just thought that Elon Musk was the worst sort of loose cannon, and that he was not taking seriously the problems that Twitter is causing. And, I was involved in Twitter because Deb Roy was originally an MIT professor, wonderful thinker, roboticist, bang AI person, and computer scientist. And and and Deb was actually the, I theory, the vice president for research at Twitter for a while. And Deb encouraged me to get on Twitter. And, he's he's left. I've left.

I think I think Deb has left. He's, he's certainly very worried about the harm that can be done by social media. And, he started out being very optimistic about it, and I sort of convinced him that it was more problematic than that. There's a paper we did together in Scientific American called our transparent future, where we where we talk about how transparency is good, but we don't want perfect transparency. Perfect transparency is more than corrosive. It's absolutely destructive of responsible agency.

David Brin here at UC San Diego and, elsewhere has written a lot about that. It's a transparent society. So, Dan, I've got a few questions from the audience besides, that one that you answered. The first one is, how do we raise children from, in this time where they have such a small attention span to focus on things like you did in your grand career in an environment where they're basically flooded with dopamine releasing stimuli. What advice do you have for parents? This comes from, one of my viewers, nunan 3347.

That's a very good question, and I don't know if I have any I don't think I have any original wisdom on that. I'm I'm worried about it. I have grandchildren, and I'm happy to say I see the grandchildren really getting interested in books, not all of them, but some of them, and, really interested in making things and not just spending their time, doing video games, although and social media, although they at least some of them do quite a bit of that. I think this isn't this is a problem, and I encourage people to create periods of potential boredom for your children where where, nevertheless, you put them in a room with some things. They wouldn't look at them if they had a phone or a television in theory. But, if they just have to stay there, they probably will look at them.

Although theory was a study at at Stanford, you know, one of the, not not too dissimilar from the prisoners survey, a study I think where they they gave children, not children, but they gave freshmen or, you know, sophomores at Stanford the opportunity to to be in a room with out, any cell phone, for 30 minutes, or they could use the cell phone. They still had to stay in there, but they had to endure, a significant electric shock. And something like half of them took the electric shock so that they could use their cell phones, and that was 30 minutes. So I'm not that saying

Oh, that's a recent experiment, C. I and that Stanford has done some other wacky experiments in the past.

They have. That's right.

But, the famous prism experiment.

That's right. That's right. So another listener, viewer, zaggy 21. Reminder, you can always ask my, esteemed guest questions at, my YouTube channel, comment pay community page is doctor Brian Keating or Twitter or LinkedIn or, Instagram, anywhere you like. What does Dan think that Hitchens would have thought about or become given, if he had survived to be in this era of culture wars? Would he have remained a champion of free speech and liberty, or would the woke mind virus have changed him in some way?

Oh, I oh, I think he'd be, entirely on the side of free speech. I learned a lot from Hitch. I didn't get to know him well. It was only during that era of the Four Horsemen that I spent some time with him. But, he, one of the things he taught me is that you could be outrageous with impunity.

If you're British, Bang, if you're British, you bang.

Yeah. Yeah. Yeah.

Something with the accent.

But, when I when I wrote Breaking the Spell, my book on religion, which is the most ecumenical and mild of the 4 books

Yeah.

A lot of people who thought they were very smart and and are very smart told me that, you know, I was gonna have to have bodyguards, and I was gonna have to, change my my my phone number and everything and and and really start protecting myself that the religious right was really dangerous. To satisfy my wife Clarke, to placate her, I I took some significant precautions for a while. And then hits just went all over the bible belt without tempering his speech at all. And he he he did just fine. So that was an important lesson to learn, that, I I think that the religious right is a sort of, it's what's sometimes called bang apple growing, a distress crop. When you feel that you're you're losing, you get desperate, and you start and you start striking out. And we're in that desperate period now, because religion is losing a lot of ground and fast. And so the the ones that see that happening all around them are getting desperate.

And we have to calm them down and ease them into their reduced influence in the world.

Talk about and we'll wrap up with, just the 2 more questions for me if I can beg your forbearance for just a few more minutes, Dan. And that's, related to, Richard Feynman. And you mentioned Feynman and meeting him, you know, towards the end of his life and and having some interactions with him. And the relationship that I really want to, ask you about maybe is is his famous claim called the cataclysm question where he said, what statement can conveys the most information in the fewest words about the universe? And he he claimed it was the atomic hypothesis that everything's made of little atoms that are whirling around and moving at tremendous speeds and pair up and make interesting combinations through various permutations, etcetera, etcetera. I wanna ask you, if you had to, you know, sort of speculate on the most powerful statement in science, philosophy, could be from your career, something that humans have a right to have a little bit of chutzpah, a little bit of swagger Keating invented, discovered, or come upon. What would that be? What is sort of the paradigmatic example of the majesty of the human mind?

No. I I have no difficulty with that question at all. By the way, it was his books that really influenced me. I didn't have that much interaction with him, but I thought, surely, you're joking, mister Feynman. And, the second one whose title Suits from Man. Right now, I think What

do you care what other people think? Yeah. Mhmm.

Yeah. Everybody should read those books, in part because he's so good and so willing to share his tricks.

Yeah.

And when I wrote intuition pumps, I wanted to do the same sort of theory. Look. A lot of this is just tricks. You can all do them. Here here are ways you you can be smarter if you do these tricks. I think, as I've said, that if I were to give a prize for the best idea anybody ever had, it would be Darwin's. Because it's Darwin's idea that ties the world of science to the world of art and culture and humanity, that it is Darwin's idea, which is a strange inversion of reasoning. It's it's the idea that you that intelligence isn't the source.

It's the effect of mindless, purposeless churning, And that turns everything upside down, and it's still theory, and it's even more wonderful.

That's right. And it's always fun for me to point out that, both Charles Darwin and, Albert Einstein were deeply suffering from what we call the computers syndrome. And, Darwin famously said, I am very poorly today and very stupid, and I hate everyone and everything. One lives only to make blunders. I'm going to write a little book for Murray on orchids today, and I hate them worse than everything. So farewell, and in a sweet frame of mine, I am ever yours, Charles Darwin. And, and I I Einstein, I came upon this. I'm giving a TED talk, which will undoubtedly get at least a logarithm of the number of views that your wonderful famous TED talk got, and you discussed that in your new book.

But I'm giving a talk about the computers syndrome entitled, am I good enough to have the imposter syndrome? And I came up

I love it. That's a hard time.

Thank you. Thank you.

I hope I hope, as I say, it does, gets a fraction of of your TED Talks views. We'll put a link to that talk in the show notes. But Einstein said, I consider myself an involuntary swindler, and I am not deserving of all the attention people give me. But on that note, I wanna finish with the last question kind of tied into, another quote by Richard Feynman. Feynman said, you know, science is the belief in the ignorance of experts. And my podcast is is called Into the Impossible, and it's named after sir Arthur c C, who said the only way to know the limits of the possible is to go beyond them into the impossible. So that's where I got the name of this podcast. But he also said a few other things, including, for every expert, there's an equal and opposite expert.

I like to hit my department chair with that every now and then. But he also said the following, and that's how you close your book with a chapter called what if I'm wrong. And that's to ask you this comment on this question or this this statement by sir Arthur c Clark. He said when an elderly but distinguished scientist says something is impasse says something is possible, he is very certainly right. But when he says something is impossible, he is very likely to be wrong. I wanna ask you, what have you been wrong about? What have you changed your mind about Bang, if anything?

Oh, gosh. I've changed my mind about quite a few things. And sometimes, in order to I would be seduced by a wisecrack. In my first book, I said about bang idea about the language of the brain. I said, it seemed to have all the virtues of replacing the little man in the brain with a committee, which was, I thought, a pretty good gag. And then later, I realized, no. No. That's right.

Replacing the little man in the brains with a committee is exactly the way to go. Homuncular functionalism, as it's often called, thanks to Bill Likin, who gave it a name. The idea that we big human agents are made of, actually, about a trillion smaller agents, human cells, and a lot of cells that Arthur, even more cells that aren't human. And that this is one this is the road to understanding what what we are. We're colonies of agents. We do replace the little man in the brain with the community. That's how you make make progress.

Well, Dan Dennett, this has been a true treat for me. It's first time meeting. I hope we get to meet in person someday, because the simulacrum, should should be replaced wherever possible. I can't thank you enough for this wonderful interview, and I can't wait to share it with the world.

Thank you so much, Dan, for being here. Good. I I look forward to seeing what you edit out of it. You know, what I that is I'm interested in the finished product, not the I don't worry about what you edit out

of it. Absolutely. Well, thank you, Dan. Have a good day. Feel free.

Feel well. Bye bye.

Hey theory, fellow magicians. If you made it all the way to the end, I know you're gonna love my interview with Sam Barris. And click here for a playlist of the best episodes from the past few weeks. And see you next Sunday on Into the Impossible.