Roger Penrose: Physics of Consciousness and the Infinite Universe | Lex Fridman Podcast #85

Lex Fridman 0:00 The following is a conversation with Roger Penrose, physicist, mathematician, and philosopher at University of Oxford. He has made fundamental contributions in many disciplines from the mathematical physics of general relativity and cosmology to the limitations of a computational view of consciousness. In his book, The Emperor's New Mind, Roger writes that, quote, "Children are not afraid to pose basic questions that may embarrass us as adults to ask." In many ways, my goal with this podcast is to embrace the inner child that is not constrained by how one should behave, speak, and think in the adult world. Roger is one of the most important minds of our time, so it's truly a pleasure and an honor to talk with him. This conversation was recorded before the outbreak of the pandemic. For everyone feeling the medical, psychological, and financial burden of the crisis, I'm sending love your way. Stay strong. We're in this together. We'll beat this thing. This is the Artificial Intelligence Podcast. If you enjoy it, subscribe on YouTube, review it with five stars on Apple Podcast, support it on Patreon, or simply connect with me on Twitter, @lexfridman, spelled F-R-I-D-M-A-N. As usual, I'll do a few minutes of ads now and never any ads in the middle that can break the flow of the conversation. I hope that works for you and doesn't hurt the listening experience. Quick summary of the ads. Two sponsors: ExpressVPN and Cash App. Please consider supporting the podcast by getting ExpressVPN at expressvpn.com/lexpod and downloading Cash App and using code LEXPODCAST. This show is presented by Cash App, the number one finance app in the App Store. When you get it, use code LEXPODCAST. Cash App lets you send money to friends, buy Bitcoin, and invest in the stock market with as little as $1. Since Cash App does fractional share trading, let me mention that the order execution algorithm that works behind the scenes to create the abstraction of the fractional orders is an algorithmic marvel, so big props to the Cash App engineers for solving a hard problem that, in the end, provides an easy interface that takes a step up to the next layer of abstraction over the stock market, making trading more accessible for new investors and diversification much easier. So again, if you get Cash App from the App Store or Google Play and use the code LEXPODCAST, you get $10 and Cash App will also donate $10 to FIRST, an organization that is helping to advance robotics and STEM education for young people around the world. This show is sponsored by ExpressVPN. Get it at expressvpn.com/lexpod to get a discount and to support this podcast. I've been using ExpressVPN for many years. I love it. It's easy to use. Press the big power on button and your privacy is protected. And, if you like, you can make it look like your location's anywhere else in the world. I might be in Boston now, but I can make it look like I'm in New York, London, Paris, or anywhere else. This has a large number of obvious benefits. Certainly, it allows you to access international versions of streaming websites like the Japanese Netflix or the UK Hulu. ExpressVPN works on any device you can imagine. I use it on Linux, shout out to Ubuntu, Windows, Android. But it's available everywhere else too. Once again, get it at expressvpn.com/lexpod to get a discount and to support this podcast. And now here's my conversation with Roger Penrose.

Lex Fridman 3:51 You mentioned in conversation with Eric Weinstein on the Portal podcast that 2001: Space Odyssey is your favorite movie.

Roger Penrose 4:00 (laughs)

Lex Fridman 4:00 Which aspect, if you could mention, of its representation of artificial intelligence, science, engineering connected with you?

Roger Penrose 4:07 There are all sorts of scenes there which are so am- amazing. And how the science was so well done, I mean people say, "Oh, no, Interstellar it's the- this amazing mo- which is the most scientific movie," but I thought it's another patch on 2001. I mean, 2001, they- they really went into all sorts of details regarding, you know, getting the free fall well done and everything. I thought it was extremely well done.

Lex Fridman 4:36 So just the details were-

Roger Penrose 4:37 Yeah.

Lex Fridman 4:38 ... me- mesmerizing in terms of just-

Roger Penrose 4:39 And also things like the scene where at the beginning they have these ance- sort of human ancestors which are sort of-

Lex Fridman 4:46 Right.

Roger Penrose 4:47 Um, or sort of apes.

Lex Fridman 4:48 The m- the monolith.

Roger Penrose 4:49 Apes becoming humans. Yes. And, uh, well it's the one where he throws the bone up into the air and then it becomes this... I mean there's, that's an amazing sequence there.

Lex Fridman 5:00 What do you make of the monolith? Does it ha- have any scientific or philosophical meaning to you? This kind of thing that sparks innovation?

Roger Penrose 5:08 The monolith? Not really. (laughs)

Lex Fridman 5:09 (laughs)

Roger Penrose 5:10 That comes from Arthur C. Clarke.

Lex Fridman 5:12 This-

Roger Penrose 5:12 I was always a great fan of Arthur C. Clarke.

Lex Fridman 5:14 So it's just a nice plot device?

Roger Penrose 5:16 Yeah. Oh, the plot is excellent. Yes, yeah.

Lex Fridman 5:18 So HAL 9000 decides to get rid of the astronauts because he, it, she believes that they will interfere with the mission.

Roger Penrose 5:27 That's right.

Lex Fridman 5:28 Do you s-

Roger Penrose 5:28 Yeah, well, there you are. It's this view... I don't know whether I disagree with it, 'cause in a certain sense it was telling you it's wrong. See, the, the machine seemed to think it was, uh, superior to the human, and so it was entitled to get rid of the human beings and run the show itself.

Lex Fridman 5:47 Well, do you think HAL did the right thing? Do you think HAL's flawed, evil? Or if we think about systems like HAL, would we want HAL to do the same thing in the future? What was the flaw there?

Roger Penrose 5:59 Well, you're basically ques- touching on questions, you see. Is one supposed to believe that HAL was actually conscious?I mean, it was played rather that way, as though HAL was a, a conscious being.

Lex Fridman 6:13 Because HAL showed some pain, uh, uh, some cogni- HAL appeared to be cognizant of its, of what it means to die.

Roger Penrose 6:24 Yes. Yes, I think that's-

Lex Fridman 6:24 And therefore, had that-

Roger Penrose 6:26 ... true, yes.

Lex Fridman 6:27 ... an inkling of consciousness.

Roger Penrose 6:28 Yeah. I mean, I'm not sure that aspect of it was made completely clear whether HAL was really just a very sophisticated computer, which really didn't actually have these feelings and somehow... But you're right. It, it, it didn't like the idea of being turned off. (laughs)

Lex Fridman 6:45 How does it change things if HAL was or wasn't conscious?

Roger Penrose 6:49 Well, it might say that it would be wrong to turn it off if it was actually conscious. I mean, these questions arise if you think... I mean, AI, one of the i- ideas, it's sort of a mixture in a sense. You say if it's trying to do everything a human can do, and if you take the view that consciousness is something which would come along when the computer is sufficiently complicated, sufficiently whatever criterion you use to, uh, characterize its consciousness in terms of some com- computational criterion.

Lex Fridman 7:28 So, how does consciousness change our evaluation of the decision that HAL made-

Roger Penrose 7:35 Yes. Well, I guess I'm-

Lex Fridman 7:36 ... which is to presume-

Roger Penrose 7:36 ... trying to say that people are a bit confused about this because if they say these machines will become conscious, but just simply because it's a degree of computation, and when you get beyond that certain degree of computation, it will become conscious, then, of course, you have all these problems. I mean, you might say, "Well, one of the reasons you're doing AI is because you want to send a device out to some distant planet, and you don't want to send a human out there 'cause then you'd have to bring it back again, and that costs you far more (laughs) than just sending it there and leaving it there." But if this device is actually a conscious entity, then you have to face up to the fact that that's immoral. And so, the mere fact that you're making some AI device, uh, a- and getting a r- r- thinking that removes your responsibility to it would be incorrect.

Lex Fridman 8:25 And so-

Roger Penrose 8:26 This is a sound p- flaw in that kind of viewpoint. I'm not sure how, you know, people who take it very seriously... I mean, I had this curious conversation with, with, uh... I'm going to forget names, and I'm afraid because this is what happens to me at the wro- wrong moment. Hofstadter, D- Douglas Hofstadter.

Lex Fridman 8:45 Douglas Hofstadter, yeah.

Roger Penrose 8:46 And he'd written this book, uh-

Lex Fridman 8:48 Godel Escher Bach?

Roger Penrose 8:49 ... Godel Escher Bach, which I liked. I thought it was a fantastic book. But I didn't agree with his, uh, conclusion from Godel's theorem. I think he got it wrong, you see? Well, uh, I'll just tell you my story, you see-

Lex Fridman 9:01 Okay.

Roger Penrose 9:01 ... 'cause I'd never met him, and then I knew I was going to meet him. The occasion... (laughs) I, I realized he was coming, and he wanted to talk to me. And I said, "That's fine." And I just thought in my mind, "Well, I'm going to paint him into a corner," you see?

Lex Fridman 9:12 (laughs)

Roger Penrose 9:12 "'Cause I'll use his arguments to convince him that c- certain numbers are conscious." You know, some integers, large enough integers, are actually conscious. And this was going to be my reductio ad absurdum. And so I started having this argument with him. He simply lept into the corner. (laughs) He didn't even need to be painted into it.

Lex Fridman 9:30 Mm-hmm.

Roger Penrose 9:31 He took the view that certain numbers were conscious.

Lex Fridman 9:33 Oh, wow.

Roger Penrose 9:33 I thought that was a reductio ad absurdum, but he seemed to think it was perfectly (laughs) reasonable point of view.

Lex Fridman 9:38 Without the absurdum there.

Roger Penrose 9:40 (laughs) Yes.

Lex Fridman 9:41 Interesting. But, uh,

Lex Fridman 9:43 the thing you mentioned about HAL is the intuition that a lot of the people, at least in the artificial intelligence world, had and have, I think... they don't make it explicit, but that if you increase the power of computation, naturally, consciousness will emerge.

Roger Penrose 9:59 Yes, I think that's what they think. But basically, that's 'cause they can't think of anything else.

Lex Fridman 10:02 Well, that's right.

Roger Penrose 10:03 And so it's a reasonable thing. I mean, you think what does the brain do? Well, it does do a lot of computation. I think most of what you actually call computation is, is done by the cerebellum. I mean, this is one of the things that people don't much mention. I mean, I come to this subject from the outside, and certain things strike me which you hardly ever hear mentioned. I mean, you hear it mentioned about the left/right business, the-

Lex Fridman 10:29 (laughs)

Roger Penrose 10:29 ... how you move your right arm, that's your-

Lex Fridman 10:30 Yes.

Roger Penrose 10:30 ... the left side of the brain and-

Lex Fridman 10:32 Yeah.

Roger Penrose 10:32 ... and so on and all that sort of stuff. And it's more than that. If you... you have these plots of different parts of the brain. Th- there are two of these. These things are called the homunculi, which you see these pictures of a distorted human figure-

Lex Fridman 10:47 (laughs)

Roger Penrose 10:47 ... and, uh, showing different parts of the brain controlling different parts of the body. And it's not simply things like, okay, the right hand is controlled and sense- both sensory and motor on the left side, left hand on the right side. It's more than that. Vision is at the back, basically. Your feet at the top. And it's as though it's about the worst organization you can imagine.

Lex Fridman 11:14 (laughs) Right. Yeah.

Roger Penrose 11:14 So it can't just be a mistake in nature. There's something going on there. And this is made more pr- pronounced when you think of the cerebellum. The cerebellum has... When I first, first thinking about these things, I w- was told that it had half as many neurons or something like that, comparable. But now they tell me it's got far more neurons than the ce- cerebrum. The cerebrum is this sort of convoluted thing at the top people always talk about. Cerebellum is this thing just looks a bit like a ball of wool right at the back underneath.

Lex Fridman 11:47 (laughs) Yep.

Roger Penrose 11:48 It's got more neurons. It's got more connections. Computationally, it's got much more going on than this from the cerebrum. But as far as we know, although it's slightly controversial, the cerebellum is entirely unconscious. So the actions... You have a pianist who plays incredible piece of music, and you think of... and he moves his little finger into this little key to get it, hit it just the right moment.... does he or she consciously will that movement? No. Okay, the consciousness that's coming in has probably to do with the feeling of, uh, the piece of music that's being performed, and that sort of thing which is going on. But the details of what's going on are controlled, I would think, almost entirely by the cerebellum. That's where you have this precision and the, the really detailed ... Once you get... I mean, you think of a tennis player or something. Does that tennis player think exactly how to, which, which muscle should be moved in what direction and so on? No, of course not. But he or she will maybe think, "Well, if the ball is angled in such a way in that corner, that will be tricky for the opponent." And ex- the details of that are all done largely with the cerebellum because that's where all the precise motions ... But it's unconscious.

Lex Fridman 13:13 So why is it interesting to you that so much computation is done in the cerebellum and yet is unconscious?

Roger Penrose 13:19 Because it doesn't, it's- it's the view that somehow it's computation-

Lex Fridman 13:24 Ah.

Roger Penrose 13:24 ... which is producing the consciousness. And it's here you have an incredible amount of computation going on, and as far as we know, it's completely unconscious. So why, what's the difference? And I think it's an important thing. What's the difference? Why has the cerebrum got all this very peculiar stuff that ... very hard to see on a computational perspective, like having the, everything have to cross over onto the other side and do something which looks completely inefficient. And you've got funny things like the frontal lobe and the preta- whatever you call the lobes. And the place where they come together-

Lex Fridman 14:06 Mm-hmm.

Roger Penrose 14:07 ... you have the different parts. The control, you see, one to do with motor and the other to do with sensory. And they're sort of opposite each other rather than being connected by ne- by ... It's not as though you've got electrical circuits. There's something else going on there. So it's, i- i- it just the idea that it's like a complicated computer just seems to me to be completely missing the point. There must be a lot of computation going on, but the cerebellum seems to be much better at doing that than the cerebrum is.

Lex Fridman 14:42 So, for sure, I, I think what explains it, it's, is a h- is like half hope and half we don't know what's going on-

Roger Penrose 14:49 (laughs) Yeah.

Lex Fridman 14:49 ... and therefore, from the computer science perspective, you hope that a Turing machine can be perfectly-

Roger Penrose 14:56 Yeah.

Lex Fridman 14:56 ... can achieve general intelligence.

Roger Penrose 14:58 Well, you have this wonderful thing about Turing and, uh, Godel and Church and Curry and various people, particularly Turing, and I guess Post was the other one. These people who developed the idea of what a computation is. And there were different i- ideas of what a computa- it developed differently. I mean, Church's way of doing it was very different from Turing's. But then they were shown to be equivalent. And so the view emerged that what we mean by a computation is a very clear concept. And one of the wonderful things that Turing did was to show that you could have what we call a universal Turing machine. It's, you just have to have a certain finite device. Okay, it has to have an unlimited storage space which is accessible to it, but the actual computation, if you like, is performed by this one universal device. And so the view comes away, well, you have this universal Turing machine, and maybe the, the brain is something like that, a universal Turing machine. And it's got maybe not an unlimited storage, but a huge storage accessible to it. And this model is one which is what's used in ordinary computation, and it's a very powerful model. And the universalness of computation is very useful. You could have some problem and you may not see immediately how to put it onto a computer, but if it is something of that nature, then, uh, there are all sorts of sub-programs and sub-routines when all the... I mean, I learned a little bit of computing when (laughs) I was, when I was a student, but not very much. But, uh, it was enough to get the general ideas.

Lex Fridman 16:45 And there's something really pleasant about a formal system like that-

Roger Penrose 16:49 Yeah.

Lex Fridman 16:49 ... where you can start discussing about what's provable, what's not, these kinds of things.

Roger Penrose 16:52 And you've got i- a notion which is an absolute notion, this n- notion of computability and you-

Lex Fridman 16:57 Computability.

Roger Penrose 16:57 ... can address when things are, what mathematical problems are computably solvable and which aren't.

Lex Fridman 17:03 So-

Roger Penrose 17:04 And it's a very beautiful area of mathematics and it's a, a very powerful area of mathematics. And it underlies the whole sort of, uh, well, I would say (laughs) the principles of computing machines that we have today.

Lex Fridman 17:19 Could you say what is Godel's incompleteness theorem and how does it ... maybe also say-

Roger Penrose 17:24 (laughs)

Lex Fridman 17:24 ... is it heartbreaking to you and, uh, how does it interfere with this notion of computation and-

Roger Penrose 17:31 Well, you see, yeah.

Lex Fridman 17:31 ... consciousness?

Roger Penrose 17:32 Sure. Well, the idea is basically, the idea is which I formulated in my first year as a graduate student in Cambridge. I did my undergraduate work in mathematics in London. And I had a colleague, Ian Perceval, we used to discuss things like computational and logical systems quite a lot. I'd heard about Godel's theorem. I was a bit worried by the idea that it seemed to say there were things in mathematics that you could never prove. And so when I went to Cambridge as a graduate student, I went to various courses. You see, I was doing pure mathematics. I was doing algebraic geometry of a sort, little bit different from w- what my supervisor and people with... (laughs)

Lex Fridman 18:15 (laughs)

Roger Penrose 18:15 But it was algebraic geometry.

Lex Fridman 18:16 Algebra, yeah.

Roger Penrose 18:17 And, uh, I was interested, I got particularly interested in-... three lecture courses that were nothing to do with what I was supposed to be doing. One was a course by Hermann Bondi on E- Einstein's general theory of relativity, which was a beautiful course. He was a, an amazing lecturer, brought these things alive, absolutely. And now there was a course on quantum mechanics given by the great physicist Paul Dirac, very beautiful course in a completely different way. (laughs) It was... he was very kind of organized and never got excited about anything, seemingly. (laughs) And, uh, but it was h- extremely well put together, and I, I found that amazing too. Third course that was nothing to do with what I should be doing was a course on mathematical logic. And I got ex- as I say, my, my discussions with Ian Perceval.

Lex Fridman 19:12 Was the incompleteness theorem already deeply within mathematical logic space? Was... ex- were you introduced to-

Roger Penrose 19:20 I was introduced to it in detail by the course b- by Steane. And he... it was two things he described which were very fundamental to my understanding. One was Turing machines and the whole idea of computability and all that, so that was all very much part of the course. The other one was the Godel theorem, and it wasn't what I was afraid it was, to tell you there were things in mathematics you couldn't prove. It was basically... and he phrased it in a way which often people didn't. And if you read Douglas Hofstadter's book, he doesn't, you see? But Steane made it very clear, and also in a, in a sort of public lecture that he gave to a mathematical... I think it may be the Adams Society, one of the mathematical undergraduate societies. And he made this point again very clearly, that if you've got a formal system of proof... so suppose what you mean by proof is something which you could check with a computer. So, to say whether you've got it right or not, you've got a lot of steps. Have you carried this computational procedure, well, following the proof, steps of a proof correctly, that can be checked by an algorithm, by, by a computer. So, that's the key thing. Now, what... you have to... n- now, you see, is e- is this any good? If you've got a s- uh, an algorithmic system which claims to say, "Yes, this is right. N- this, you've proved it correctly. This is true." If you've proved it... if you made a mistake, it doesn't say it's true or false, but if you ha- if you've done it right, then the conclusion you've come to is correct. Now, you say, "Why do you believe it's correct?" 'Cause you've looked at the rules, and you said, "Well, okay, that one's all right. Yeah, that one's all right. What about that? Oh, I'm not s- oh, yeah, I see, I see why it's all right." Okay. You go through all the rules. You say, "Yes, following those rules, if it says, 'Yes, it's true,' it is true." So, you've got to make sure that these rules are ones that you trust. Is... if you follow the rules and it says it's a proof, is the result actually true?

Lex Fridman 21:27 Right.

Roger Penrose 21:27 And that your belief that it's true depends upon looking at the rules and understanding them. Now, what Godel shows, that if you have such a system, then you can construct a statement of the very kind that it's supposed to look at, a mathematical statement, and you can see by the way it's constructed and what it means, that it's true but not provable by the rules that you've been given. And it depends on your trust in the rules. Do you believe that the rules only give you truths? If you believe the rules only give you truths, then you believe this other statement is also true. I found this absolutely mind-blowing. When I saw this, it blew my (laughs) you know, blew my mind.

Lex Fridman 22:13 Mm-hmm.

Roger Penrose 22:13 I thought, "My God, you can see that this statement is true." It's as good as any proof because it only depends on your belief in the reliability of the proof procedure, that's all it is, and understanding that the coding is done correctly, and it enables you to transcend that system. So, whatever system you have, as long as you can understand what it's doing and why you believe it only gives you truths, then you can see beyond that system. Now, how do you see beyond it? What is it that enables you to transcend that system? Well, it's your understanding of what the system is actually saying and what the statement that you've constructed is actually saying. So, it's this quality of understanding, whatever it is, which is not governed by rules. It's not a computational procedure.

Lex Fridman 23:07 So, this idea of understanding is now going to be within the rules of the s- the f- the f- within the formal system that comes from us human-

Roger Penrose 23:13 Yes, you're only using those rules anyway-

Lex Fridman 23:15 Yeah.

Roger Penrose 23:15 ... because you have understood them to be rules which only give you truths. There'd be no point in it otherwise. I mean, people say, "Well, okay, this is w- uh... is one set of rules as good as any other?" Well, it's not true, you see? You have to understand what the rules mean. And why does that understanding of the mean give you something beyond the rules themselves? And that's, that's what it was. That's what blew my mind. It's somehow understanding why the rules give you truths enables you to transcend

Roger Penrose 23:45 the rules.

Lex Fridman 23:46 So, that's where... I mean, even at that time, that's already where the thought entered your mind that the idea of understanding, or we can start calling it things like intelligence or even consciousness, is outside the rules?

Roger Penrose 23:59 Yes. See, as I've always concentrated on understanding. You know, people say... people tell me ￡ point P others, "Well, we know, what about creativity? That's something a machine can't do as creative." Well, I don't know. What is creativity? And I don't know. You know, somebody can put some funny things on a piece of paper and say, "That's creative," and you could make a machine do that. Is it really creative? I don't know. You see, I, I worry about that one. I sort of agree with it in a sense, but it's so hard to do anything with that statement. But understanding, yes, you can. You can make... go...... see that understanding, whatever it is, um, it's very hard to put your finger on it. That's absolutely true.

Lex Fridman 24:35 Can you try to define or maybe dance around a definition of-

Roger Penrose 24:40 (laughs)

Lex Fridman 24:40 ... understanding?

Roger Penrose 24:42 To some degree, but I don't... I've often wondered about this. But there is something there which is very slippery. It's something to... like standing back. And it's got to be something, you see... it's also got to be something which was of value to our remote ancestors.

Lex Fridman 24:58 Right. Yeah.

Roger Penrose 24:59 Because I... sometimes I... there's a cartoon which I drew sometimes, showing you how all these... There's a... in the foreground you see this mathematician just doing some mathematical theorem. There's a little bit of a joke in that theorem, but let's not go into that. (laughs) He's trying to prove some theorem and, uh, he's about to be eaten by a saber-toothed tiger-

Lex Fridman 25:17 (laughs)

Roger Penrose 25:17 ... who's hiding in the, in the, in the undergrowth, you see. And in the distance, you see his he- his cousins building, uh, growing crops, building shelters, um, domesticating animals. And in the slight foreground, you see they've built a mammoth trap and this poor old mammoth is falling into a pit, you see, and all these people around him are about to grab him, you see. And, and... well, you see, those are the ones who... the quality of understanding which goes with all the... it's not just the mathematician doing his mathematics. This understanding quality is something else which has been of tremendous advantage to us. Not just to us. See, I don't think consciousness is limited to humans.

Lex Fridman 26:03 Yeah, that's the interesting question. At which point, if it is indeed connected to the evolutionary process-

Roger Penrose 26:08 Yeah.

Lex Fridman 26:09 ... at which point did we pick up this, uh-

Roger Penrose 26:11 A very hard question. It's certainly... I, I don't think it's primates, you know. You, you see these pictures of African hunting dogs and how they, they can plan amongst themselves how to catch the antelopes, th-... some of these, um, David Attenborough films, I think this probably was one of them, and you could see the, the hunting dogs and they divide themselves into two groups and they go in two routes, two different routes. One of them goes and they sort of hide next to the river and the other group goes around and they start yelping at these... they don't bark, I guess, whatever noise hunting dogs do, the antelopes, and they sort of round them up and they chase them in the direction of the river. And there are the other ones just waiting for them, just to get... because this... when they get to the river it slows them down and so they pounce on them. So they've obviously planned this all out somehow. I've no idea how. And there is some element of conscious planning as far as I can see. I don't think it's just some kind of a... so much of AI these days is done on, what are they called? Bottom-up systems. Is it... what... yeah-

Lex Fridman 27:22 Right.

Roger Penrose 27:22 ... where you have neural networks and they, and they... you give them a zillion different things to look at and, and then they sort of ch- ch- can choose one thing over another just because it's seen so many examples and, uh, picks up on little signals which human may not even be conscious of.

Lex Fridman 27:41 And that doesn't feel like understanding?

Roger Penrose 27:43 There's no understanding in that whatsoever.

Lex Fridman 27:45 (laughs)

Roger Penrose 27:45 So-

Lex Fridman 27:46 Well, you're being a little bit human-centric, so...

Roger Penrose 27:49 Well, I, I'm talking about-

Lex Fridman 27:50 I think I would expe-

Roger Penrose 27:51 See, I'm not with the dogs, am I? 'Cause the dogs-

Lex Fridman 27:52 No, you're not. Sorry, sorry. Not human-centric, but, uh, I misspoke.

Roger Penrose 27:56 Yeah.

Lex Fridman 27:57 Biol- biology-centric. Is it possible that consciousness would just look slightly different and-

Roger Penrose 28:03 Well, I'm not saying it's biological 'cause we don't know.

Lex Fridman 28:06 Right.

Roger Penrose 28:06 I think other examples, uh, elephants is a wonderful example too where they... this was... I think this was the one, Attenborough one, where the, the elephants have to go from... they're long... the, the troop of them have to go long distances and the leader of a troop is a female, they all g- are apparently. And this female, th- she had to go all the way from one part of the country to another and at a certain point she made a detour and they went off on this big detour. All the troop came with her and this was where her sister had died and there were her bones lying around and they go and pick up the bones and they hand it round and they caress the bones and then they put them back and they all go back again. What in the hell are they doing?

Lex Fridman 28:50 (laughs) That's so interesting.

Roger Penrose 28:52 I mean, there's something going on. There's no clear connection with natural selection. There's just some deep feeling going on there which has to do with their conscious experience. And, uh, I think it's something that overall is advantageous, um, on natural selection, but not directly to do with natural selection. Yes.

Lex Fridman 29:18 I like that. There's something going on, go on, going on there.

Roger Penrose 29:22 Yeah.

Lex Fridman 29:22 Like I told you, I'm Russian so I tend to romanticize all things of this nature. (laughs)

Roger Penrose 29:28 (laughs)

Lex Fridman 29:28 That, uh, that it's not merely a cold, hard computation.

Roger Penrose 29:33 Perhaps I could just slightly answer your question. You were asking me what is it. There's something about sort of standing back and thinking about your own thought processes. I mean, there is something like that in the Gödel thing 'cause it's just... you're not following the rules. You're standing back and thinking about the rules. And so there is something that you might say... you think about you're doing something and then you think, "What the hell am I doing?" And you sort of stand back and think about what it is that's making you think in such a way.

Lex Fridman 30:05 Just take a step back outside the s- the game you've been playing.

Roger Penrose 30:08 Yeah. You back up and you think about... you, you're just not playing the game anymore. You're thinking about what the hell you're doing in playing this game.

Lex Fridman 30:16 And that's, that's somehow... it's... uh, it- it's not a very precise description but it somehow feels very true that that's somehow understanding.

Roger Penrose 30:24 Yeah.

Lex Fridman 30:24 So this kind of reflection.

Roger Penrose 30:26 A reflection, yes. Yeah.

Lex Fridman 30:28 Yeah.

Roger Penrose 30:28 There is so- it's a bit hard to put your finger on but there is something there which I think maybe could be unearthed at some point and see this is really what's going on, why conscious beings have this advantage, what it is that gives them an advantage and I think it goes way back. I don't think... we're talking about the hunting dogs and the elephants s-... that's pretty clear that octopuses (laughs) have the same sort of quality. Um, we call it consciousness? Yeah, I think so. I've seen enough examples of the way that they behave, and the evolution route is completely different. Does it go way back to some common ancestor or did it come separately?

Lex Fridman 31:07 My hope is it's something simple. But the hard question, if there's a hardware prerequisite-

Roger Penrose 31:13 Yeah.

Lex Fridman 31:13 ...that we have to develop some kind of, uh, hardware mechanisms in our computers. Like basically, as you suggest and we'll get to in a second, we kinda have to throw away the computer as we know it today.

Roger Penrose 31:26 Yeah.

Lex Fridman 31:26 Uh, the deterministic machines we know today, to try to create a... I mean, my, my hope of course is not, but, um...

Roger Penrose 31:35 Well, actually I should go really back to the

Roger Penrose 31:37 story, which in a sense I haven't finished.

Lex Fridman 31:39 Yeah.

Roger Penrose 31:39 Because I went to these three courses, you see, when I was a graduate student.

Lex Fridman 31:43 Sure.

Roger Penrose 31:44 And so I started to think, "Well, I'm really... I'm a pretty... what you might call a materialist in the sense of thinking that there's no kind of mystical or something or other which comes in from who knows where."

Lex Fridman 31:56 Are you still that? Are you still throughout your life been a mat-

Roger Penrose 31:58 I don't like the word materialist because it suggests we know what material is.

Lex Fridman 32:02 (laughs)

Roger Penrose 32:02 And that's, that is-

Lex Fridman 32:03 Well put.

Roger Penrose 32:04 ...is a bad word because-

Lex Fridman 32:06 But there's no mystical...

Roger Penrose 32:07 It's not some mystical something which is not sci- not treatable by science.

Lex Fridman 32:11 That's so beautifully put, just to pause on that for a second. You're a materialist but you acknowledge that we don't really know what the material is. (laughs)

Roger Penrose 32:19 That's right.

Lex Fridman 32:19 (laughs) That's well put.

Roger Penrose 32:21 I mean, I, I like to call myself a scientist I suppose, but it means that, uh... Yes, well you see, the question goes on here. So I began thinking, "Okay, if consciousness or understanding is something which is not a computational process, what can it be?" And I knew enough from my undergraduate work, I knew about ma- Newtonian mechanics, and I knew, um, how basically you could put it on a computer. There is a fundamental issue which, is it important or not that computation depends upon discrete things, so using discrete elements, whereas the physical laws depend on the continuum. Now, is this something to do with it? Is it the fact that we use the continuum in our physics and if we model our physical system, we use discrete systems like ordinary computers? I came to the view that that's probably not it. I might have to retract on that someday, but the view was no, you, you can get close enough. Uh, it's not altogether clear, I have to say. But you can get close enough. And, you know, I went to this course on by Bondi on general relativity and I thought, "Well, you can put that on a computer." Of course, that was a long time before people... and I've sort of grown up with this, how people have done better and better calculations and they could work out black, about black holes and they can then work out how black holes can interact with each other, spiral around, and what kind of gravitational waves come out. And it's still a very impressive piece of computational work, how you can actually work out the shapes of those signals. And now we have LIGO seeing these signals and they say, "Yeah, there's those black holes spiraling into each other." This is just a vindication of the power of computation in describing Einstein's general relativity.

Lex Fridman 34:16 So in that case, we can get close. But we... uh, so with computation, we can get close to our understanding of the physics.

Roger Penrose 34:23 You can get very close. Now, is that close enough, you see? And then I went to this course by Dirac. Now you see, I think it was the very first lecture that he gave and he was talking about the superposition principle and he said if you have a particle, you usually think of particle can be over here or over there, but in quantum mechanics, it can be over here and over there at the same time. And you have these states which involve a superposition in some sense of, at different locations for that particle. And then he got out his piece of chalk, and some people say he broke it in two as a kind of illustration of how the piece of chalk might be over here and over there at the same time. And he was talking about this, and I... my mind wandered. I don't remember who or what he said. All I can remember, he's just moved on to the next topic and something about energy he'd mentioned, which I had no idea what it had to do with anything. And so I'd been struck with this and worried about it ever since. It's probably just as well I didn't hear his explanation because it was probably one of these things to calm me down and not worry about it anymore.

Lex Fridman 35:31 (laughs)

Roger Penrose 35:31 Whereas in my case, I've worried about it ever since. So I thought maybe that's the catch. There is something in quantum mechanics where the superpositions become one or the other, and that's not part of quantum mechanics. There's something missing in the theory. The theory is incomplete. It's not just incomplete; it's in a certain sense not quite right because if you follow the equation, the basic equation of quantum mechanics, that's the Schrodinger equation, you could put that on a computer too. There are lots of difficulties about how many parameters you have to put in, so on. That can be very tricky, but nevertheless, it is a computational process. Modulo this question about the continuum as before, but it's not clear that makes any difference.

Lex Fridman 36:16 So our theories of quantum mechanics may be missing the same element that, uh, the Universal Turing Machine is missing about consciousness.

Roger Penrose 36:25 Yes, yes. Yeah, this is the view I held is that you need a theory, and that that, what, what people call the reduction of the state or the collapse of the wave function, which you have to have otherwise quantum mechanics doesn't relate to the world we see. To make it relate to the world we see, you've got to break the quan- you've got to break the Schrodinger equation. Schrodinger himself was absolutely (laughs) appalled by this idea. His own equ- his own equation... I mean, that's why he introduced this famous Schrodinger's cat as a thought experiment. He's really saying, "Look, this is where my equation leads you into it."

Lex Fridman 36:59 Mm-hmm.

Roger Penrose 36:59 "There's something wrong," (laughs) something we haven't understood which is basically fundamental.... and so I was trying to put all these things together and said, "Well, it's got to be the non-compute- computability comes in there." And I also can't quite remember right when I thought this, but I said, "It's when gravity is involved in quantum mechanics." It's the combination of those two, and it's that point when the... you have good th- good reasons to believe... This, this came much later. But I have good reason to believe that the principles of general relativity and those of quantum mechanics, most particularly, it's the basic principle of equivalence, which goes back to Galileo. If you fall freely, you eliminate the gravitational field. So you imagine Galileo lay- lay- drawing, thr- uh, dropping his big rock and his little rock from the Leaning Tower. Whether he actually ever did that or not is p- pretty irrelevant. (laughs) And as the rocks fall to the ground, you li- have a little insect sitting on one of them looking at the other one, and it seems to think, "Oh, there's no gravity here." Of course, it hits the ground and then it realize something's different's going on. But when it's in free fall, the gravity has been eliminated. Galileo understood that very beautifully. You can see his won- wonderful examples of fireworks, and you see the fireworks, and explode, and you see this sphere of sparkling fireworks just remains a sphere as it falls down as though there were no gravity. So he understood that principle, but he couldn't make a theory out of it. Einstein came along, used exactly the same principle, and that's the basis of Einstein's general theory of relativity. Now, there is a conflict. This is something I did much, much later, so this wasn't, uh, those days. Much, much later. You can see there is a basic conflict between the principle of superposition, the thing that Dirac was talking about, and the principle of general covari- no, well, principle of equivalence. Gravitational field's equivalent to an acceleration.

Lex Fridman 39:03 Can you pause for a second? What is the principle of equivalence?

Roger Penrose 39:06 It's this Galileo principle that you can eliminate... at least locally, you have to be in a, a, a small neighborhood 'cause, you see, if you had people dropping rocks all around the world somewhere, you can't get rid of it all at once. But in the local neighborhood, you can eliminate the gravitational field by falling freely with it. And we now see it with astronauts and they don't f-... you know, the earth is right there. You can see the great globe of the earth right beneath them. (laughs) But they don't care about it. They... as far as they're concerned, there's no gravity.

Lex Fridman 39:37 Mm-hmm.

Roger Penrose 39:39 They fr- fall freely with- in the gravitational field and that gets rid of the gravitational field. And that's the principle of equivalence.

Lex Fridman 39:46 So what's the... what's the contradiction? What's the tension with superposition and equivalence?

Roger Penrose 39:51 Ah, well, that's technical. (laughs)

Lex Fridman 39:52 Well, but so we... uh, just to backtrack for a second, just to see if we can weave a thread through it all.

Roger Penrose 39:57 Yes.

Lex Fridman 39:58 So you... we were s- started to, to think about consciousness as potentially needing, uh, some-

Roger Penrose 40:05 Yes.

Lex Fridman 40:05 ... of the same, uh, not mystical, but some of the same magic.

Roger Penrose 40:08 Yeah. You see,

Roger Penrose 40:09 it is a complicated story, so, so, you know, people think, "Oh, I'm drifting away from the point," or something.

Lex Fridman 40:14 No.

Roger Penrose 40:14 But I think it is a complicated story. So what I'm trying to say... I mean, I tried to put it in a nutshell, and it's not so easy. I'm trying to say that whatever consciousness is, it's not a computation.

Lex Fridman 40:26 Yes.

Roger Penrose 40:27 Or it's not a physical process which can be described by computation.

Lex Fridman 40:33 But it nevertheless could be... So one, one of the interesting models that you've proposed is the orchestrated objective reduction.

Roger Penrose 40:42 Yes.

Lex Fridman 40:42 Which is st-

Roger Penrose 40:42 Well, you see, that's going from there, you see? So I say I have no idea. So I wrote this book. Through my s- scientific career, I thought, "You know, when I'm retired, I'll have enough time to write a, a sort of popular-ish book which I will explain my ideas and puzzles, what I like, beautiful things about physics and mathematics, and this puzzle about computability and consciousness," and so on. And in the process of writing this book... Well, I thought I'd do it when I was retired. I didn't actually. I wa- didn't wait that long because there was a radio discussion between Edward Fredkin and Marvin Minsky.

Lex Fridman 41:23 Hmm.

Roger Penrose 41:24 And they were talking about what computers could do, and they were entering, entering a big room. They imagined entering this big room. At the other end of the room, two computers were talking to each other. And as you walk up to the computers, they will have communicated to each other more ideas, concepts, things, than the entire human race had ever commuted... done. (laughs)

Lex Fridman 41:48 Yes.

Roger Penrose 41:49 So I thought, "Well, I know where you're coming from," (laughs) "but I just don't believe you. There's something missing." That's... it's not com- that... so I thought, "Well, I should write my book." And so I did. It was, uh, roughly the same time Stephen Hawking was writing his, uh, Brief History of Time, and, uh-

Lex Fridman 42:08 In the '80s at some point. The book you're talking about is The Emperor's New Mind.

Roger Penrose 42:12 Emperor's New Mind, that's right.

Lex Fridman 42:13 Which... and both are, are incredible books. Uh, the Brief History of Time and Emperor's New Mind.

Roger Penrose 42:18 Yes. It was quite interesting 'cause he got... he told me he'd got, um, Carl Sagan, I think, to write a forward-

Lex Fridman 42:24 That's a good get. (laughs)

Roger Penrose 42:25 ... to the book, you see. So I thought, "Gosh, I... what am I gonna do? I'm gonna have to... (laughs) I'm not gonna get anywhere unless I get somebody." So I said, "Oh, I know Martin Gardner, so I wonder if he'd do it." So he did, and he did a very nice forward.

Lex Fridman 42:36 So that... so that's an incredible book, and some of the, uh, this is the same people you mentioned, Ed Frankin, which, uh, I guess f- of, uh, expert systems fame, and Minsky, of course, people know in the AI world. But they represent the artificial intelligence world-

Roger Penrose 42:49 Absolutely, that's right.

Lex Fridman 42:49 ... that do hope and dream that AIs, intelligences-

Roger Penrose 42:53 That's right.

Lex Fridman 42:53 ... is primarily

Roger Penrose 42:53 Well, you see, it was my thinking, "Well, you know, th- I see where they're coming from," and I... from that perspective-

Lex Fridman 42:58 I disagree. Oh.

Roger Penrose 42:58 Yeah, you're right. But I... that's not my perspective, so I thought I had to say it. And as I was writing my book, you see, I thought, "Well, I don't really know anything about neurophysiology. What am I doing writing this book?" So I started reading up about neurophysiology, and I read up and I think, and I was trying to find out, well, how it is that nerve signals could possibly preserve quantum coherence. And all I read is that the sig- um, electrical signals which go along the nerves create, um...... affects through the brain and there's no chance you can isolate it. (laughs) So I thought, "This is hopeless." So I come to the end of the book and I more or less give up. I just think of something which I didn't believe in, that maybe this is a way around it, but no. And then you say, I thought, "Well, maybe this book will at least stimulate young people to do science or something." And I got all these letters from old retired people instead. (laughs) It's the only people who could have had time to read my book. (laughs)

Lex Fridman 43:52 So, th- I mean, but-

Roger Penrose 43:53 Except for Stuart Hameroff.

Lex Fridman 43:54 Ex- except for Stuart Hameroff. Yeah.

Roger Penrose 43:56 Stuart Hameroff wrote to me and he said, "I think you're missing something. You don't know about microtubules." Stu- he didn't put it quite like that, but that was more or less it. And he said, "This is what you really need to consider." So I thought, "My God, yes, that's a much more promising structure."

Lex Fridman 44:13 So y- I mean, fundamentally you were searching for the source of... a- a- non-computable source of consciousness within the human brain-

Roger Penrose 44:23 Yeah.

Lex Fridman 44:23 ... in the biology. And so-

Roger Penrose 44:26 Yes.

Lex Fridman 44:26 ... what are mar- (laughs) if I may ask, what are microtubules?

Roger Penrose 44:29 (laughs) Well, you see, I, I was ignorant in what I'd read. I never came across them in, in, in, in the books I looked at. Perhaps I only read rather superficially, which is true. But I didn't know about microtubules. Stuart... I think one of the things he wa- that impressed him about them was when you see pictures of mitosis, that's a cell dividing, and you see all the chromosomes. And the chromosomes get... they're get... all get lined up and then they get pulled apart. And so th- as the cell divides, the... half the chromosomes go... you know, ha- they, well, they divide into the two parts and they go two different ways. And what is it that's pulling them apart? Well, those are these little things called microtubules. And so he started to get interested in them. And he formed the view, when he was... ha- his day job or night job or whatever you call it, is to put people to sleep.

Lex Fridman 45:22 Mm-hmm.

Roger Penrose 45:23 Except he doesn't like calling it sleep because it's different. General anesthetics-

Lex Fridman 45:27 Mm-hmm.

Roger Penrose 45:27 ... in a reversible way. So you want to make s- sure that they don't experience the pain that would otherwise be something that they feel. And consciousness is turned off for a while and it can be turned back on again. So it's crucial that you can turn it off and turn it on. And what do you do when you're doing that? What do general anesthetic gases do? And see, he formed the view that it's the microtubules that they affect. And the details of, of why he formed that view is not altogether clear to me, but there... but there's an interesting story he keeps talking about. But I f- found this very exciting because I thought these structures, these little tubes which inhabit pretty well all cells, it's not just neurons, uh, apart from red blood sessil- uh, red blood cells, they inhabit pretty well all the other cells in the body. But they're not all the same kind. You get different kinds of microtubules. And the ones that excited me the most, and this is... may still not be totally clear, but the ones that excited me most were the ones that... the only ones that I knew about at the time because they were... they're very, very symmetrical structures. And I had reason to believe that these very symmetrical structures would be much better at preserving a quantum state, quantum coherence, preserving the thing without... you just need to c- preserve certain degrees of freedom without them leaking into the environment. Once they leak into the environment, you're lost. So you've got to preserve these quantum states at a level which the state reduction process comes in, and that's where I think the non-computability comes in, and it's the measurement process in quantum mechanics, what's going on.

Lex Fridman 47:21 So something about the, uh, th- the measurement process and what's going on, something about the structure of the microtubules.

Roger Penrose 47:26 Yes.

Lex Fridman 47:27 Your intuition says, "May- maybe there's something here. Maybe this kind of structure allows for the, the, the mystery of the quantum mechanics."

Roger Penrose 47:35 There was a much better chance, yes. It just struck me that partly it was the symmetry.

Lex Fridman 47:40 Right.

Roger Penrose 47:40 Because there is a feature of, of symmetry you can produ- pre- preserve quantum coherence much better with symmetrical structures. And so there is a good reason for that. And that impressed me a lot. I didn't know the difference between the A lattice and B lattice at that time, which could be important. (laughs) Now, that couldn't... that's, see, which isn't talked about much.

Lex Fridman 48:00 But that's some... in some sense details. We've got to take a step back just to say-

Roger Penrose 48:04 Yeah, yeah.

Lex Fridman 48:04 ... in case people are not familiar.

Roger Penrose 48:05 Yeah, yeah.

Lex Fridman 48:06 So this, this, this was, um, called the orchestrated objective reduction idea or Orch OR, which is a biological philosophy of mind that postulates that consciousness originates at the quantum level inside neurons. So that has to do with your search for where w- where is it coming from. So that's counter to the notion that consciousness might arise from the computation performed by the synapses.

Roger Penrose 48:33 Yes. I think-

Lex Fridman 48:33 So-

Roger Penrose 48:33 ... the, the key point... See, sometimes people say it's qu- because it's quantum mechanical. It's not just that. You see, it's, it's more outrageous than that. You see, this is one reason I think we're so far off from it-

Lex Fridman 48:47 Mm-hmm.

Roger Penrose 48:48 ... because we don't even know the physics right. You see, it's not just quantum mechanics. People say, "Oh, you know, quantum systems and biological structures." No. Well, you're s- starting to see that some basic biological s- systems does depend on quantum s- I mean, well, y- in the first place, all of chemistry is quantum mechanics. People got used to that, so they don't count that. (laughs) So they said, "Let's not count quantum qu- que- chemistry. We sort of got the hang of that," they think. But you have quantum effects which are not just chemical in photosynthesis. And this is one of the striking things in the last several years, that photosynthesis seems to be...... a basically quantum process, which is not simply que- chemical. It's using quantum mechanics in a very basic way, so you could start saying, "Oh, well, if photosynthesis is based on quantum mechanics, why not, uh, behavior of neurons and things like that?" Maybe there's something which is a bit like photosynthesis in that respect. But what I'm saying is even more outrageous than that, because those things are talking about conventional quantum mechanics. Now, my argument says that conventional quantum mechanics, if you're just following the Schrödinger equation, that's still computable. So you've got to go beyond that. So you've got to go to where quantum mechanics goes wrong, in a certain sense. Y- you have to be a little bit careful about that, because the way people do quantum mechanics is a sort of, um, mixture of two different processes. One of them is the Schrödinger equation, which is a- an equation th- Schrödinger wrote down, and it tells you how the sys- the state of a system evolves. And it evolves according to this equation, completely deterministic, but it evolves into ridiculous situations. And this was what Schrödinger was very much pointing out with his cat. He said, "You follow my equation," that's Schrödinger's equation, "and you could say that you have to... Your cat, ca- a cat which is dead and alive at the same time."

Lex Fridman 51:04 Mm-hmm.

Roger Penrose 51:04 That would be the evolution of the Schrödinger equation, would lead to a state which is the qu- cat being dead and alive at the same time. And he's more or less saying this is an absurdity. People now, they say, "Oh, well, Schrödinger said you can have a cat which is dead and alive." It's not that. You see, he was saying, (laughs) this is an absurdity. There's something missing, and that the reduction of the state, or the collapse of the wave function, or whatever it is, is something which is- has to be understood. It's not following the Schrödinger equation. It's not the way we conventionally do quantum mechanics. There's something more than that, and it's easy to quote authority here because Einstein, (laughs) at least three of the greatest physicists of 20th century, uh, who were very fundamental in developing quantum mechanics, Einstein one of them, Schrödinger another, Dirac another. You have to look carefully at Dirac's writing 'cause he didn't tend to s- say this out loud very much 'cause he was very cautious about what he said. You find the right place and you see he says quantum mechanics is a provisional theory. We need something which explains the collapse of the wave function. We need to go beyond the theory we have now. I happen to be one of the kinds of people, there are many, there is a whole group of people, they're all considered to be a bit, you know, a bit mavericks, who believe that quantum mechanics needs to be modified. There's a small minority of those people, which are already a minority, who think that the way in which it's modified has to be with gravity, and there is an even smaller minority of those people who think it's the particular way that I think it is, you see. (laughs) So it's-

Lex Fridman 52:55 So those are the quantum gravity folks, but what's, what's-

Roger Penrose 52:57 You see, quantum gravity is already not this. 'Cause when you say quantum gravity, what you really mean is quantum mechanics applied to gravitational theory. So you say, let's take this wonderful formalism of quantum mechanics and make gravity fit into it, so that is what quantum gravity is meant to be. Now, I'm saying you've got to be more even-handed, that gravity affects the structure of quantum mechanics too. It's not just you quantize gravity. You've got to gravitize quantum mechanics, and it's a, it's a two-way thing. It's-

Lex Fridman 53:31 But then when do you even get started so that you're saying that we have to figure out, uh, totally new ideas in there?

Roger Penrose 53:36 Exactly. No, it's, it's you're stuck-

Lex Fridman 53:39 (laughs)

Roger Penrose 53:39 ... you don't have a theory. That's the trouble. So tha- this is a big problem if you say, "Okay, well, what's the theory?" I say, "I don't know." (laughs)

Lex Fridman 53:47 So we're maybe in the very early days, sort of-

Roger Penrose 53:49 It is in the very early days, isn't it? But I, just making this point.

Lex Fridman 53:52 Yes.

Roger Penrose 53:53 Well, you see, Stuart Hameroff tends to be, "Oh, Penrose says that it's, that it's got to be a reduction of the state and so and so, let's use it." The trouble is Penrose doesn't say that. Penrose says, "Well, I think that." (laughs)

Lex Fridman 54:04 Yeah, right. (laughs) No.

Roger Penrose 54:04 And then we have no, we have no experiments as yet which shows that.

Lex Fridman 54:09 Yes.

Roger Penrose 54:10 There are experiments which are being thought through and which I'm hoping will be performed. There is a- uh, an experiment which is being developed by Dirk Bouwmeester, who has, I've known for a long time, who shares his time between Leiden in the Netherlands and Santa Barbara in the US, and he's been working on an experiment which could perhaps demonstrate that quantum mechanics, as we now understand it, if you don't bring in the gravitational effects, um, has to be modified.

Lex Fridman 54:42 And, and then there's also experiments that are underway that kind of, um, look at the microtubule side of things to see if there's-

Roger Penrose 54:51 Yes.

Lex Fridman 54:51 ... in the biology, you could see something like that. Could you briefly mention it? 'Cause that's a really sort of one of the only experimental attempts in the very early days of even thinking about-

Roger Penrose 55:02 Yes.

Lex Fridman 55:02 ... consciousness.

Roger Penrose 55:02 We just, I think there's- there's a very serious area here, which is what Stuart Hameroff is doing, and I think it's very important. One of the few places that you can really get a bit of a handle on what consciousness is, is what turns it off. And when you're thinking about general anesthetics, it's very specific. These things turn consciousness off. What the hell do they do?

Lex Fridman 55:26 Right.

Roger Penrose 55:26 Well, Stuart, uh, and hi- he- a number of people who work with him and others, uh, happen to believe that the general anesthetics directly affect microtubules, and there is some evidence for this. I don't know how strong it is and how-... watertight the case is, but I think there is some evidence pointing in that kind of direction. It's not just an ordinary chemical process. There's something quite different about it. And one of the main candidates is that these anesthetic gases do affect directly microtubules.

Lex Fridman 56:02 Mm-hmm.

Roger Penrose 56:02 And how strong that evidence is, I wouldn't be in a position to say, but I think there is fairly impressive evidence-

Lex Fridman 56:10 And the point is-

Roger Penrose 56:11 ... in that direction.

Lex Fridman 56:11 ... the, the experiments are being undertaken which is-

Roger Penrose 56:12 Yeah.

Lex Fridman 56:13 ... is interesting.

Roger Penrose 56:13 I mean, that is experimental. You see, so it's, it's a very clear direction where you can think of experiments which could indicate whe- whether or not it's really microtubules which the anesthetic gases directly affect.

Lex Fridman 56:26 That's really exciting. Uh, one of the sad things is, uh, as far as I'm, from my outside perspective is not many people are working on this. So, there's a very, like with Stuart even, it feels like there's very few people who are carrying the flag forward on this.

Roger Penrose 56:41 I think it's, it's not many in the sense it's a minority, but it's not zero anymore.

Lex Fridman 56:46 (laughs)

Roger Penrose 56:46 You see, when Stuart and I were originally-

Lex Fridman 56:48 It was zero.

Roger Penrose 56:49 ... taught by this, you know, we were just, just us and a few, few of our friends. There weren't many people taking s- but it's grown into, into a, one of the main viewpoints. You know, there might be about four or five or six different views that, which people hold, and it's one of them.

Lex Fridman 57:07 So-

Roger Penrose 57:07 So, it's, it's considered as one of the possible lines of thinking, yes.

Lex Fridman 57:13 You describe physics theories as falling into one of three categories, the superb, the useful-

Roger Penrose 57:18 (laughs)

Lex Fridman 57:18 ... or the tentative. I like those, uh, words. It's a beautiful categorization. Do you think we will ever have a superb theory of intelligence and of consciousness?

Roger Penrose 57:29 We might. We're a long way from it. I don't think we're even w- whether in the tentative scale. (laughs) I mean, it's, uh ...

Lex Fridman 57:39 Uh, you don't think we've even entered the realm of tentative?

Roger Penrose 57:42 Probably not. I think it's-

Lex Fridman 57:43 Yeah, that's right.

Roger Penrose 57:44 ... ? ... is. Uh, you know, well, you see, they're so co- they're so controversial. We don't have a, a clear view which, which is, uh, accepted by a majority. I mean, you say, you have people, most views are computational in one form or another. They think it's some, but it's not very clear 'cause even the, the IIT people who think, you think of them as computational, but I've heard them say, say, "No, consciousness is supposed to be not computational." I say, "Well, if it's not compu- what in the hell is it?" (laughs) What's go- what's going on? What physical processes are going on which are that?

Lex Fridman 58:19 What, what does it mean for something to be computational then? So (laughs) , is, um-

Roger Penrose 58:25 Well, there has to be a, a process which is ... You see, it's very curious the way the history has developed in quantum mechanics 'cause very early on people thought there was something to do with consciousness. But it was almost the other way around. You see, you have to say the, the, the Schrodinger equations says all these different alternatives happen all at once, and then when is it that only one of them happens? Well, one of the views, which was quite commonly held by a few distinguished quantum physicists is when a conscious being looks at the system or becomes aware of it, and at that point it becomes one or the other. That's a row where consciousness is somehow actively reducing the state. My view is almost the exact opposite of that. It's just the state reduces itself in some way which, some non-computational way which we don't understand, we don't have a proper theory of, and that is the building block of what consciousness is. So, consciousness is the other way around. It depends on that choice which nature makes all the time when the state becomes one or the other rather than the superposition of one and the other. And when that happens, there is what we're saying now an element of proto-consciousness takes place. Proto-consciousness is roughly speaking the building block out of which actual consciousness is constructed. So, you have these proto-conscious elements which are when the state decides to one, do one thing or the other. That's the thing which when organized together, that's the OR part in Orch OR. But the orch part, that's the (laughs) ... The OR part at least one can see where one's arriving at as a theory. You can say it's the quantum choice of going this way or that way. But the orch part, which is the orchestration of this, is much more mysterious. And how does the brain somehow orchestrate all these individual OR processes into a genuous- genuine, genuine conscious experience or something?

Lex Fridman 1:00:32 And it, and it might be something that's beautifully simple, but we're in, completely in the dark about.

Roger Penrose 1:00:37 Yeah. I think at the moment we're, it's, that's the thing, you know. We happily put the word orch down there to say orchestrated. (laughs)

Lex Fridman 1:00:44 (laughs)

Roger Penrose 1:00:44 But that's even more unclear what that really means.

Lex Fridman 1:00:49 Just like the word material-

Roger Penrose 1:00:50 (laughs)

Lex Fridman 1:00:50 ... orchestrated.

Roger Penrose 1:00:52 Yes.

Lex Fridman 1:00:52 Uh, who knows?

Roger Penrose 1:00:54 Materialism, yes.

Lex Fridman 1:00:54 And we've been dancing a little bit between the word intelligence or understanding and consciousness.

Roger Penrose 1:01:00 Yeah.

Lex Fridman 1:01:01 Do you kind of see those as sitting-

Roger Penrose 1:01:03 Yeah.

Lex Fridman 1:01:03 ... in the same space of mystery as what we discussed?

Roger Penrose 1:01:05 Yes. Well, you see, I tend to say you have understanding and intelligence and awareness. And somehow understanding is in the middle of it, you see. It's, I like to say, uh, could you say of an entity that it's actually intelligent if it doesn't have the quality of understanding? Now you see, I'm using terms I don't even know how to define.

Lex Fridman 1:01:32 Right.

Roger Penrose 1:01:33 But who cares? I'm just-

Lex Fridman 1:01:34 They're po- they're po-

Roger Penrose 1:01:34 ... relaying them.

Lex Fridman 1:01:35 They're somewhat poetic. So, if I somehow understand them-

Roger Penrose 1:01:38 Yes.

Lex Fridman 1:01:39 (laughs)

Roger Penrose 1:01:39 That's right, we don't (laughs) ... Exactly.

Lex Fridman 1:01:40 But they're not mathematical in nature you see.

Roger Penrose 1:01:42 Yes. You see, uh, as a mathematician, I don't know how to define any of them, but at least I can point to the connections. So, the idea is intelligence is something which I believe needs...... understanding, otherwise you wouldn't say it's really intelligence. And understanding needs awareness, otherwise you wouldn't really say it's understanding. Do you say if an entity that understands something unless- unless it's really aware of it in our normal usage. So, there's three sort of awareness, understanding, and intelligence. And I just tend to concentrate on, uh, understanding because that's where I can say something. (laughs)

Lex Fridman 1:02:19 Okay.

Roger Penrose 1:02:19 And that's the Gödel theorem, things like that. But they, it, uh, well, what does it mean to be, perceive the color blue or something? I mean, I'm the foggiest. (laughs) That's a much more difficult question. I mean, is it the same if I see a color blue and you see it? If you're somebody with, what, this- this condition? What is it called when...

Lex Fridman 1:02:38 Oh, where you assign, uh-

Roger Penrose 1:02:39 Where you get mixed up.

Lex Fridman 1:02:40 ... like a sound to a- to a color. That kind of-

Roger Penrose 1:02:41 Yeah, yeah. That's right. You get colors and sounds mixed up.

Lex Fridman 1:02:44 (laughs)

Roger Penrose 1:02:44 And- and that sort of thing. I mean, uh, an interesting subject. I mean...

Lex Fridman 1:02:47 Yeah. But from the physics perspective, from the fundamentals perspective, we don't...

Roger Penrose 1:02:53 I think we're way off having much understanding what's going on there.

Lex Fridman 1:02:57 In your 2010 book, Cycles of Time, you suggest that another universe may have existed before the Big Bang. Can you describe this idea? First of all, what is the Big Bang? Sounds like a funny word.

Roger Penrose 1:03:13 Yes.

Lex Fridman 1:03:13 And what, uh, may have been there before it?

Roger Penrose 1:03:17 Yes. Just as a matter of terminology, I don't like to call it another universe, 'cause when you have another universe, you think of it kind of quite separate from us. But these things, they're not separate. Now, the Big Bang conventional theory... See, I was actually brought up (laughs) in the sense of when I started getting interested in cosmology, there was a thing called the Steady State model, which was sort of philosophically very interesting. And there wasn't a Big Bang in that theory, that somehow new material was created all the time in the form of hydrogen, and the universe kept on expanding and expanding and expanding, and there was room for more hydrogen. It was a rather philosophically nice picture. It was disproved when the Big Bang... Well, when I say the Big Bang, this was theoretically discovered by people trying to solve Einstein's equations and apply it to cosmology. Einstein didn't like the idea. He liked an i- a universe which was there all the time.

Lex Fridman 1:04:14 Okay.

Roger Penrose 1:04:14 And he had a model which they- which there all the time. But then there was this discovery, accidental discovery, very important discovery of this microwave background. And if you, you know, there's the crackle on your television screen, which is- is already just sensing this (laughs) microwave background, which is coming at us from all directions. And you can trace it back and back and back and back. And it came from a very early stage of the universe. Well, it's part of the Big Bang theory. The Big Bang theory was when people tried to solve Einstein's equations, they really found you had to have this initial state where the universe... It was used to be called the primordial atom and things like this. Th- there's Friedmann and Lemaître. Friedmann was a Russian, Lemaître was a Belgian. And they independently... Well, basically Friedmann first. And Lemaître talked about the initial state, which is a very, very concentrated initial state which seemed to be the origin of the universe.

Lex Fridman 1:05:13 Primordial atom. That's a nice-

Roger Penrose 1:05:14 Primordial atom is what he called it. Yes.

Lex Fridman 1:05:16 Beautiful term.

Roger Penrose 1:05:17 And then it became... Well, Fred Hoyle used the term Big Bang in a kind of derogatory sense.

Lex Fridman 1:05:21 Yeah.

Roger Penrose 1:05:22 He said, "Well, didn't "

Lex Fridman 1:05:22 Just like with the Schrodinger and the cats.

Roger Penrose 1:05:24 (laughs) Yeah.

Lex Fridman 1:05:24 Right?

Roger Penrose 1:05:25 Yes. It's like sort of, it c- pick, got picked up on, whereas it wasn't his intention originally. (laughs) But it then, the evidence piled up and piled up. And my, one of my friends that I learned a lot from when I was in Cambridge was Dennis Sharma. He was a great proponent of Steady State, and then he got converted. He said, "No, I'm sorry." I had a great respect for him. He went around lecturing, said, "I was wrong. The Steady State model doesn't work. There was this Big Bang." And this microwave background that you see... Okay, it's not actually quite the Big Bang when I say not quite. It's about 380,000 years after the Big Bang, but that's what you see. But then you have to have had this Big Bang before it in order to make the equations work. And it works beautifully, except for one little thing, which is this thing called inflation, which people had to put into it to make it work. When I first heard of it, I didn't like it at all.

Lex Fridman 1:06:16 What's inflation?

Roger Penrose 1:06:18 Inflation is that in the first... I'm gonna give you a very tiny number. Think of a second. That's not very long. Now, I'm gonna give you a fraction of a second, one over a number. This number has 32 digits between... Well, let's say between 36 and 32 digits. Tiny, tiny time between those two tiny ridiculous seconds, fraction of a second, the universe was supposed to have expanded in this exponential way, an enormous way for no pater- patent reason. You had to invent a particular thing called the inflaton field to make it do it. And I thought, "This is completely crazy." There are reasons why people stuck with this idea. You see, the thing is that I formed my model for reasons which are very fundamental, if you like. It has to do with this very fundamental principle, which is known as the second law of thermodynamics. The second law of thermodynamics says, more or less, things get more and more random as time goes on. Now, another way of saying exactly the same thing is things get less and less random as things go back. As you go back in time, they get less and less random. So, you go back and back and back and back, and the earliest thing you can directly see is this microwave background. What's one of the most striking features of it is that it's random. It has this sp- sp- what you call this spectrum of... Which is, uh, what's called the Planck spectrum of frequencies, uh, different intensities for different frequencies. And there's this wonderful curve due to Max Planck. And what's it telling you? It's telling you that the entropy is at a maximum.... started off at a maximum, and it's going up ever since.

Lex Fridman 1:08:02 (laughs)

Roger Penrose 1:08:02 I call that the mammoth in the room. (laughs) I mean, this is a paradox.

Lex Fridman 1:08:05 Mammoth in... Yeah, it is. It is.

Roger Penrose 1:08:07 And so people... Why don't cosmologists worry about this? So I worried about it, and then I thought, "Well, it's not really a paradox because you're looking at matter and radiation at a maximum entropy state." What you're not seeing directly in that is the gravitation. It's gravitation which is not thermalized. The gravitation was very, very low entropy, and it's low entropy by the uniformity, and you see that in the microwave too. It's very uniform over the whole sky. I'm compressing a long story into a very short few sentences. (laughs)

Lex Fridman 1:08:40 And doing a great job. Yeah.

Roger Penrose 1:08:42 So what I'm saying is that there's a huge puzzle. Why was gravity in this very low entropy state, very highly organized state? Everything else was all random. And that, to me, was the biggest problem in cosmology. The biggest problem, nobody seems to even worry about it. People are saying they solved all the problems, and they don't even worry about it. They think inflation solves it. It doesn't. It can't, because it's, it's just, uh...

Lex Fridman 1:09:12 Just to clarify, the- that was your problem with the inflation describing some aspect of-

Roger Penrose 1:09:18 Yes. Well, if la- if-

Lex Fridman 1:09:18 ... the moments right after the Big Bang?

Roger Penrose 1:09:20 Inflation is supposed to stretch it out and make it all uniform, you see. It doesn't do it because it can only do it if it's uniform already at the beginning. It's- it's... You just have to look at... I can't go into the details. But it doesn't solve it, and it was completely clear to me it doesn't solve it.

Lex Fridman 1:09:33 But where does the conformal cyclic cosmology of s-

Roger Penrose 1:09:36 Yeah. Well, this is... (laughs)

Lex Fridman 1:09:37 ... starting to talk about something before-

Roger Penrose 1:09:40 Yes.

Lex Fridman 1:09:40 ... that singularity or Big Bang?

Roger Penrose 1:09:41 Well, I began... I was just thinking to myself how boring this universe is going to be. You've got this exponential expansion. This was discovered early in the- in this, uh, uh, century, 20, uh, 21st century. People discovered that these supernova-exploding stars showed that the universe is actually undergoing this exponential expansion, so it's a self-similar expansion. And it seems to be a feature of this term that Einstein introduced into his cosmology for the wrong reason. He wanted a universe that was static. He put this new term into his cosmology to make it make sense. It's called the cosmological constant. And then when he got convinced that the universe had a Big Bang, he retracted it, complaining that this was his greatest blunder. The trouble is it wasn't a blunder. It was actually right. (laughs)

Lex Fridman 1:10:36 (laughs)

Roger Penrose 1:10:36 Very ironic. And so the universe seems to be behaving with this cosmological constant. Okay. So this universe is expanding and expanding. What's gonna happen in the future? Well, it gets more and more boring for a while. What's the most interesting thing in the universe? Well, there's black holes. The black holes more or less gulp down entire clusters of galaxies. They clas-... It'll swallow up most of our galaxy. Then we will run into Andromeda Galaxy's black hole. That black hole will swallow our one. They'll get bigger and bigger, and they'll s- basically swallow up the whole cluster of galaxies, gulp it all down, pretty well all. Most of it, maybe not all. Most of it. Okay, and then it- that'll happen to... There'll be just these black holes around. Pretty boring, but still not as boring as it's gonna get. It's gonna get more boring because these black holes, you wait, you wait, and you wait, and you wait, and you wait an unbelievable length of time, and Hawking's, uh, black hole evaporation starts to come in. And the black holes, you just... It's incredibly tedious. Finally evaporate away. Each one goes away, disappears with a pop at the end. What could be more boring? That- it was boring then. Now this is really boring. There's nothing. Not even black holes. Universe gets colder, and colder, and colder, and colder. And I thought, "This is very, very boring."

Lex Fridman 1:11:52 (laughs)

Roger Penrose 1:11:52 Now, that's not science, is it? But it's- it's emotional. So I thought, "Who's gonna be bored by this universe?" Not us. We won't be around. It'll be mostly photons running around. And what do photons do? They don't get bored because... It's a part of relativity, you see. It's not really that they don't experience anything. That's not the point. (laughs) That photons get right out to infinity without experience any time. It's the way ph- way relativity works. And this was part of what I used to do in my old days when I was looking at gravitational radiation and how things behaved at infinity. Infinity is just like another place. You can squash it down. As long as you don't have any mass in the world, infinity is just another place. The photons get there. The gravitons get there. Where do they get... They run into infinity. And they say, "Well, now I'm here. What do I... There's something on the other side. Is there?" The usual view, it's just a mathematical notion. There's nothing on the other side. That's just the boundary of it. A nice example is this beautiful series of pictures by the Dutch artist MC Escher. You may know them, the ones called Circle Limits. They're a very famous one with the angels and the devils. And you can see them crowding, and crowding, and crowding up to the edge. Now, the kind of geometry that these angels and devils inhabit, that's their infinity, but from our perspective, infinity is just a ki- it's a place. Okay. There is- is...

Lex Fridman 1:13:18 I- Oh, sorry. Can you just, uh, take a brief pause?

Roger Penrose 1:13:20 (laughs) Yes.

Lex Fridman 1:13:20 And the- in- in, uh, the- just the word you're saying, "infinity is just a place." So, uh, for- for the most part, infinity sort of even just going back, uh-

Roger Penrose 1:13:28 Yes. I thi-

Lex Fridman 1:13:28 ... infinity is a mathematical concept.

Roger Penrose 1:13:31 I think this is one of the things-

Lex Fridman 1:13:31 Do you think there's an actual physical manifest... Like, in which way does infinity ever manifest itself in our physical universe?

Roger Penrose 1:13:40 Well, it does in various places. You see, it's a thing that w- if you're not a mathematician, you think, "Oh, infinity. I can't think about that." Mathematicians think about infinity all the time.

Lex Fridman 1:13:48 Yeah.

Roger Penrose 1:13:48 They get used to the idea, and they just play around with different kinds of infinities, and it becomes no problem. But you know, you just have to take my word for it. (laughs)

Lex Fridman 1:13:57 Got it.

Roger Penrose 1:13:57 Now, one of the things is, you see, you take Euclidean geometry. Well, it just keeps on, keeps on, keeps on going, and it goes out to infinity. Now, there's other kinds of geometry, and this is a-... what's called hyperbolic geometry, it's a bit like Euclidean geometry. It's a little bit different. It's like what Escher was trying to describe in his angels and devils. And he learned about this from Coxeter and he thought that's a very nice thing, I'll try and represent this infinity to this kind of geometry. So it's not quite Euclidean geometry, it's a bit like it, that the angels and the devils inhabit. And their infinity, by this nice transformation you squash the, their infinity down so you can draw it as this nice circle boundary to their universe. Now, from our outside perspective we can see their infinity as this boundary. Now, what I'm saying is that it's very like that. The infinity that we might experience like those angels and devils in their world can be thought of as a boundary. Now, I found this a very useful way of talking about radiation, gravitational radiation and things like that. It was a trick, mathematical trick. So know what I'm saying is that that mathematical trick becomes real, that somehow the photons, they need to go somewhere because for their, from their perspective infinity is just another place. Now, this is a difficult idea to get your mind around, so-

Lex Fridman 1:15:31 Yes.

Roger Penrose 1:15:32 ... that's why cosm... one of the reasons cosmologists are finding a lot of trouble (laughs) taking me seriously. But to me it's not such a wild idea. What's on the other side of that infinity? You have to think, "Why am I allowed to think of this?" Because photons don't have any mass, and we in physics have beautiful ways of measuring time. There are incredibly precise clocks, atomic and nuclear clocks, unbelievably precise. Why are they be, so precise? Because of the two most famous equations of 20th century physics, one of them is Einstein's E=MC². What's that tell us? Energy and mass are equivalent. The other one is even older than that, still 20th century, only just. Max Planck, E=Hν. ν is a frequency, H is a constant again like C, E is energy. Energy and frequency are equivalent. Put the two together, energy and mass are equivalent, Einstein, energy and frequency are equivalent, Max Planck. Put the two together, mass and frequency are equivalent. Absolutely basic physical principle. If you have a massive entity, a massive particle, it is a clock with a very, very precise frequency. It's not... you can't directly use it, you have to scale it down, so your atomic and nuclear clocks, but that's the basic principle. You scale it down to something you can actually perceive, but it's the same principle. If you have mass, you have beautiful clocks. But the other side of that coin is, if you don't have mass, you don't have clocks.

Lex Fridman 1:17:14 Mm-hmm.

Roger Penrose 1:17:14 If you don't have clocks, you don't have rulers, you don't have scale. (laughs)

Lex Fridman 1:17:20 S- so you don't have space and time?

Roger Penrose 1:17:21 You don't have a measure of the scale of space and time.

Lex Fridman 1:17:24 Oh, oh, scale of space and time.

Roger Penrose 1:17:25 So you have to s... If the... You do have the structure, the, what's called the conformal structure. You see, it's what the angels and devils have. If you look at the eye of the devil, no matter how close to the boundary it is, it has the same shape, but it has a different size. So you can scale up and you can scale down, but you mustn't change the shape. So it's basically the same idea, but applied to spacetime now. In the very remote future, you have things which don't measure the scale, but the shape, if you like, is still there. Now, that's in the remote future. Now I'm gonna do the exact opposite. Now I'm gonna go way back into the Big Bang. Now, as you get there things go hotter and hotter, denser and denser. What's the universe dominated by? Particles moving around almost with the speed of light. When they get almost with the speed of light, okay, they begin to lose the mass, too. They're for a completely opposite reason. They lose the sense of scale as well. So my crazy idea is the Big Bang and the remote future, they seem completely different. One is extremely dense, extremely hot, the other's very, very rarefied and very, very cold. But if you squash one down by this conformal scaling, you get the other. So although they look and feel very different, they're really almost the same. The remote future on the other side, I'm claiming, is that where do the photons go? They go into the next Big Bang. You've gotta get your mind around that crazy idea.

Lex Fridman 1:19:01 Taking a step on the other side-

Roger Penrose 1:19:02 (laughs)

Lex Fridman 1:19:03 ... of the place that is infinity. Okay.

Roger Penrose 1:19:05 Yes.

Lex Fridman 1:19:06 But, uh-

Roger Penrose 1:19:07 So I'm saying the other side of our Big Bang, now I'm going back into the Big Bang.

Lex Fridman 1:19:10 Back, backwards.

Roger Penrose 1:19:11 There was the remote future of a previous eon.

Lex Fridman 1:19:13 Previous eon.

Roger Penrose 1:19:15 And what I'm saying is that previous eon, there are signals coming through to us which we can see and which we do see. And these are both signals... The two main signals are to do with black holes. One of them is the collisions between black holes, and as they spiral into each other they release a lot of energy in the form of gravitational waves. Those gravitational waves get through in a certain form into the next eon.

Lex Fridman 1:19:44 That's fascinating that there's some... I mean, uh, maybe I'm... Maybe you can correct me if I'm wrong, but that means that some information can travel-

Roger Penrose 1:19:51 Yes.

Lex Fridman 1:19:52 ... from another eon.

Roger Penrose 1:19:53 Exactly.

Lex Fridman 1:19:54 That, that is, um, uh, (laughs) fascinating. S- I mean, um, I- I've seen somewhere described sort of the discussion of, um, the Fermi paradox, you know, that there, if there's intelligent life-

Roger Penrose 1:20:07 (laughs) Yes.

Lex Fridman 1:20:08 ... um, being, you know, communication imme- immediately takes you there. So-

Roger Penrose 1:20:13 We have a paper-

Lex Fridman 1:20:14 Do you think-

Roger Penrose 1:20:14 I have my, my cr- my colleague, Vahe Gurzadyan, who I worked with this, on these ideas for a while, we have a crazy paper on that, yes.

Lex Fridman 1:20:21 So do you-

Roger Penrose 1:20:22 Looking at the Fermi paradox, yes. (laughs)

Lex Fridman 1:20:23 Right. So, so if th- if the universe is just cycling over and over and over, punctuated by...

Roger Penrose 1:20:29 Yep.

Lex Fridman 1:20:29 ... the, uh, uh, punctuated the singularity of the Big Bang-

Roger Penrose 1:20:32 Yes.

Lex Fridman 1:20:32 ... and then intelligent or a-a-any kind of intelligent systems can communicate through from eon to eon-

Roger Penrose 1:20:39 Exactly.

Lex Fridman 1:20:39 ... why haven't we heard anything from our alien friends?

Roger Penrose 1:20:44 'Cause we don't know how to look.

Lex Fridman 1:20:47 That's fundamentally the reason is we-

Roger Penrose 1:20:48 I don't know. You see, it's, it's, it's speculation.

Lex Fridman 1:20:51 Oh.

Roger Penrose 1:20:51 I mean, th- the SETI program was a reasonable thing to do, but still speculation. It's trying to say, okay, maybe not too far away was a civilization which got there first, before us, early enough that they could send us signals. But how far away would you need to go before... I mean, I don't know. We, we've so little knowledge about that. We haven't seen any signals yet, but it's worth looking. It's worth looking. What I'm trying to say, here's another possible place where you might look.

Lex Fridman 1:21:22 Oh.

Roger Penrose 1:21:22 Now, you're not looking at civilizations which got there first. You're looking at those civilizations which were so successful, probably a lot more successful than there- than we are likely to be by the looks of things, um, which knew how to handle their own global warming or whatever it is, and to get through it all and to live to a ripe old age, in the sense of a civilization, to the extent that they could harness signals that they could propagate through, for some reason of their own desires, whatever we wouldn't know, to, to other civilizations which might be able to pick up the signals. But what kind of signals would they be? I haven't the foggiest. (laughs)

Lex Fridman 1:22:04 (laughs)

Lex Fridman 1:22:05 Let me ask the question-

Roger Penrose 1:22:06 Yes.

Lex Fridman 1:22:06 ... what to you is the most beautiful idea in physics or mathematics or the art at the intersection of the two?

Roger Penrose 1:22:15 I'm gonna have to say complex analysis. I might have said infinities and th- one of the most single best beautiful idea I think was the fact that you can have in- infinities of different sizes and so on.

Lex Fridman 1:22:27 Mm-hmm.

Roger Penrose 1:22:27 But that's, in a way, I think complex analysis. This goes... It's got so much magic in it. It's a very simple idea. You take these, okay, so if you... you take numbers, you take the integers and then you fill 'em up into the fractions and the real numbers. You imagine you're trying to measure a continuous line, and then you think of how you can solve equations. Then what about X squared equals minus one? Well, there's no real number which has, satisfies that. So you have to think of, well, there's a number called I (laughs) . You think you invent it. Well, in a certain sense, it's there already. But this number, when you add that square root of minus one to it, you have what's called the complex numbers. And they're an incredible system. If you like, you put one little thing in, you put square root of minus one in, and you get how much benefit out of it, all sorts of things that you'd never imagined before. And it's that amazing, all hiding there in putting that square root of minus one in. I think that's-

Lex Fridman 1:23:31 So in a sense-

Roger Penrose 1:23:31 ... the most magical thing I've seen in mathematics or physics, and it's in quantum mechanics.

Lex Fridman 1:23:35 And in quantum mechanics, right.

Roger Penrose 1:23:36 You see, it's there already. You might think, "What's it doing there? Okay, just a nice beautiful piece of mathematics." And then suddenly we see, nope, it's the very crucial basis of quantum mechanics. It's there-

Lex Fridman 1:23:47 So on the-

Roger Penrose 1:23:47 ... in the way the world works.

Lex Fridman 1:23:49 So on the question of whether math is discovered or invented, it sounds like you may be suggesting that partially it's possible that math is indeed discovered.

Roger Penrose 1:23:57 Oh, absolutely. Yes. No, it's more like archeology than you might think.

Lex Fridman 1:24:02 (laughs)

Roger Penrose 1:24:03 Yes. Yes.

Lex Fridman 1:24:04 So let me ask the most ridiculous, but maybe the most important question. What is the meaning of life?

Roger Penrose 1:24:12 (laughs)

Lex Fridman 1:24:12 What gives your life-

Roger Penrose 1:24:13 Oh, God.

Lex Fridman 1:24:13 ... fulfillment, purpose, happiness, and meaning? Why do you think we're here on this, given all the Big Bang and the infinities of photons that we've talked about?

Roger Penrose 1:24:21 All I would say, I think it's not a stupid question. (laughs)

Lex Fridman 1:24:24 (laughs)

Roger Penrose 1:24:26 I mean, there are some people, you know, many of my colleagues who are scientists and they say, "Well, that's a stupid question." Meaning, you know, well, we're just here