Transcription
What is it that you're most misunderstood about? I would think the consciousness thing, probably. The trouble is that's a murky subject, and probably I'm not sure I should have got into it at all. It's a good question. I think I'm glad I did.
But it came about mainly from talks. You see, this was one of the three courses I went to which were nothing to do when I was a graduate student in Cambridge. And I went to three courses that were nothing to do with my subject. One was by Herman Bondi on general relativity, one was by Paul Dirac on quantum mechanics—big influence on me clearly—and the third one was by a man called Steen on mathematical logic. And I learnt about the Gödel theorems, and I found it absolutely stunning because I had sort of vaguely heard that Gödel's theorem showed that there were things in mathematics you couldn't prove. I didn't like the idea very much, but then when I heard what it actually said, I found this stunning because what it says—you make a sentence, and this sentence says, you have to be clever to make it do it, but you make this sentence say in effect, “I am not provable by those rules,” and that's what it says. So you say, well, maybe it's false, and if it's false, it is provable by the rules, and therefore it's true, and not provable by the rules. Amazing.
So how do you make it do that? Well, that's the trick, of course. But what I found remarkable is that how do you know it's true? You know it's true not because of the rules that you're using, but because you believe that those rules only give you truths. And what's the point of using these rules as methods of proof if you don't trust the rules? Why is it proof anyway? Why is it something which you believe to be true? Only because you trust the rules. But if you trust the rules, then you can prove this other statement which is based on your belief that the rules only give you truths, but you can't obtain that using the rules. In fact, it wouldn't be much good if you did use the rules because you have to know that the rules only give you truths in order to trust the conclusion. So you have to understand why the rules only give you truths.
So then I'm going to think, well, what does understanding mean? What are you doing? Well, it seemed to me whatever it is, it seems to involve consciousness. I mean, the very term understanding, you wouldn't really say an entity understands something unless it's aware of it. That's normal usage of the language. So to understand something, you have to be aware of it. So for some reason, the awareness of things gives you something beyond what you could achieve with computers. So that is a view which I held then and still hold, which is more relevant than ever now, of course, because people talk about AI, which is already a misnomer as far as I'm concerned. And it's not artificial intelligence. It may be artificial cleverness or something, but it doesn't involve any understanding. You're just following the rules. And the argument that I'd convinced myself, learning from Mr. Steen about the mathematical logic, was to say that the understanding of why the rules work is something which is not a computational thing. What is it if it's not computational? Well, it's something to do with consciousness.
Now, you see, I don't know anything about what I... I wrote my book, The Emperor's New Mind, to try and express this point of view. And I learned a bit of neurophysiology, thinking that if I learned enough neurophysiology, I would find out somewhere what it is that, you see, I formed a view. Yes, I have to backtrack a little bit. I formed a view thinking of all the laws of physics. You see, I'm a physicalist, so I believe whatever is going on in our heads is following the laws of physics. But what are the laws of physics? Well, the laws of physics, Newtonian mechanics does pretty well, but then you have to go further than that, you have to have Maxwell's equations, you have to talk about special relativity, general relativity, and quantum mechanics. But all these things, although there's a little bit of incompleteness in the argument, you see, all these theories depend on the continuum. And it meant there could be a catch there, that something is not based on the continuum. But the continuum seems to be how you can approximate. I mean, you can approximate the continuum and calculate very accurately. Even at that time, there weren't good calculations on general relativity, but I realized that was still a computational problem. And nowadays, people do great things with calculators on computing what happens with black holes spiraling into each other and things like that, which is pretty impressive. So sure, that's things you could put on a computer. Then relativity, as well as Newtonian mechanics, Maxwell's equations, all these things you can put on a computer. Quantum mechanics, well, the Schrödinger equation, that's you can put that on a computer. Maybe lots of variables you have to worry about and so on and so forth. But still, it's still the same sort of thing.
How about the collapse of the wave function? Well, that seemed to be the gap that maybe is where one needs a gap in this whole discussion. And that the only gap that I know of—it could be there's something else of course that we don't know in physics—but that seems to be an important gap, and collapse of the wave function. You see, many people, including Wigner, you see, I remember talking to Wigner when I was in Princeton. And I remember I had dinner with him and discussing this question with him. Because one of the views attributed to Wigner was that it's the conscious being observing the quantum system which collapses the wave function. So it requires consciousness to make the collapse work. I talked to him a bit, and he was not so dogmatic as people tend to think about this. He said that was a view that he played with and thought it was a possibility. I didn't think it was at all likely for all sorts of reasons. It seems to me that—I had this argument about a distant planet which is an earth-like planet, and the weather on that planet, there's no life. It just never got started and there's no butterfly to flap its wings and cause the weather to do this or that. It's just a quantum mixture of all the different weathers it can possibly have. And there's no conscious observer to look at it, so it's just a quantum superposition of all these different alternatives. So the space probe which travels out takes a photograph of all this superposition of all these different weathers. And when it gets back to Earth or close to Earth, it sends a signal to the Earth and says this is the photograph I've just taken. And somebody sitting looking at a screen sees this superposition of weathers. First time a conscious being has seen it, suddenly it becomes one weather. Doesn't make any sense. So that's not what I believe. It's not that consciousness causes the collapse of the wave function. My view is almost the opposite. It's that whatever consciousness is, it depends on the collapse of the wave function. That there is the only thing which I could see which was not obviously computable. Okay, the way we do it is we just introduce a randomness, but that I would argue is an approximation, that there is something going on which is not just explained by randomness. But I don't have a solution to that question. But it was my entry into the consciousness discussion. And I wrote my book, The Emperor's New Mind, to try and explain this point of view. I had no answer. I remember getting to the end thinking I would learn about the Hodgkin-Huxley nerve transmissions, and I would learn enough about that to see what the answer might be. No, I didn't. So I just sort of tapered off at the end with something I didn't really believe, and that was the end of the book.
But later on, I learned from Stuart Hameroff about microtubules, which I thought were much more likely. Something in microtubules could easily be more likely, but the subject is not something—it involves biology. I don't understand about biology. I can't remember the names that people keep using, and they talk about different chemicals, processes, and it's all beyond me. So it's not something that I feel I could develop thinking in. So that's—I mean, I did have lots of discussions with people, but the discussion sort of went off in different directions and not anything which I could really contribute to. I think there's interesting aspects about the collapse of the wave function, most of which I didn't know at the time, but thinking about later, there's something extremely puzzling about how it works with regard to time. And this puzzlement could easily have relevance to conscious perception and things like this. So I think there's a story there which is worth exploring, but it's not really an area which I don't—I can never remember the names that people talk about, the chemicals which are involved and so on. That's not the kind of thing I can do. So I'm happy for other people to follow these lines up. Except I don't think they've got very close as yet.
What if you could create a song simply by describing it? That's what I did for the music in this video using Suno. I needed custom tracks for Theories of Everything, ones that wouldn't trigger copyright claims, and I was skeptical about AI music, but here's what surprised me. I prompted it with contemplative ambient music with subtle complexity evoking questions about reality. Yes, I know that's a complex prompt, but it's actually just a sentence about mood. Basically, I'm trying to capture the philosophical atmosphere we go for here. The output sounds fantastic, not like algorithmic noise. It's legitimately studio quality. Now what makes this useful is I can generate dozens of variations in minutes, find out what fits the episode's mood, and I have full commercial rights to use it. No licensing headaches, no hiring composers for every single video, and for someone with zero formal music training, being able to translate abstract concepts into actual audio is quite remarkable. The whole process is iterative. I describe what I want, I listen, I tweak a few words, and within seconds, I have a new version that feels like a different take on the same idea. Anyone who can describe an idea in words can start making music. If you want to turn your ideas into music, try Suno at suno.com. S-U-N-O dot com. All the music in this video was made by us using Suno.
I don't know, it's quite possible that this is not my idea because I can't really contribute. But one thing I would worry about is you have the cerebrum and the cerebellum underneath. The cerebellum is organized much more in the way that if you were a computer scientist, you would organize it. The left-hand part controls the left hand, all the left side of the body. The right-hand part to the right-hand side of the body. There's none of this crazy crossing over and all that stuff that the cerebrum is involved with. It's much more organized like a computer. It's completely unconscious. It is like a computer, maybe. But the thing that impressed me a little bit about what Stuart Hameroff was doing later on is that he considers that these main relevant structures in the brain have to do with things called pyramidal cells. And that's the first time I've heard about certain types of cell which are not found in the cerebellum. Everything else you see is the same kinds of cells and so on and so on. But these cells occupy a particular location in the brain, and Stuart seems to think that they are very important with regard to consciousness. I can't make any comment there because it's not my area. But I do find that particular suggestion as quite plausible. That's all I can say.
Now, if you're a physicalist and you believe that the collapse of the wave function somehow produces consciousness, then I don't see that as inconsistent with a computer being conscious, because a computer is a physical system after all. Well, no, it would have to be harnessed, and you would have to make use of it in some way. You see, it's, I mean, it's physical in the sense that the collapse of the wave function is part of physics. It's not to do with an observer looking at it, as I was saying earlier. It has to do with the system getting too big in some sense. And that's another line of thinking which I've developed to some extent independently. I had an argument which I put forward around about the turn of the century, I can't remember exactly when, and it came up, this argument produced a lifetime for a superposition. So if you have a body, take this cup. Forget about the coffee in it. Take this cup. And imagine it's in superposition of being here and here. Well, I can tell you that that will have a lifetime of becoming one or the other. What is that lifetime? Well, I work out, if I take two copies of the cup right on top of each other, I move one of them away from the other, I ask how much energy that would cost me if I consider only gravity. So the gravitational force between the particles as I move this away, it's tiny but still huge in another respect. That energy is a fundamental uncertainty whose reciprocal gives you the lifetime. So the lifetime for this cup would be absolutely instantaneous. If it's a molecule, maybe not. Maybe take a long time. If it's a big enough molecule, or maybe the collapse would happen quite quickly. Depends how big it is. But that calculation tells you the lifetime. I should say that that lifetime was independently discovered independently about two years before me by Diósi. So you can call it the Diósi lifetime. I sometimes call it that. However, his scheme is in a way different from mine. And there was, he still had an argument for a lifetime, but it has other implications in his scheme which have been refuted. There was an experiment done a few years ago, about three or four years ago, I can't remember now, in a mountain, because his point of view would predict that the collapsing is happening all the time, and this gives you an effective heating. And so the system would gradually heat up. There would be a slight, slight spontaneous heating, and that heating is measured not to take place. So some people say this disproves the Penrose theory, but it's not the Penrose theory, because my scheme does not involve the heating. It would be a nuisance because I think it would cause real problems. With cosmology, you have neutron stars which probably would already be heating up spontaneously in a way that they don't. So I don't believe that. But in order to avoid the heating problem, you have to have something else which is very curious. And this curious thing depends on certain things being retrocausal, as though the cause goes back into the past. But you can only make sense of this if you have what I call two kinds of reality. One is classical reality, and the other is quantum reality. And classical reality, well, the cup here has classical reality. I can say, “Hello, cup, what is your shape?” And it can say, “Well, I've got more or less a circle up there, and I have this particular curve there, and it's actually symmetric apart from the handle,” and so on, you see. That's classical reality. I can ask it what its shape is. Quantum reality is better described by, say, the spin of an electron. The spin of electron, the spin state could be in any direction in space, but it's the quantum superposition of only two directions. If you want to ask it, you see, Einstein had a criterion. See, I think people were worrying about whether the quantum state was real or not. When I say real, I don't mean real in mathematical sense as opposed to complex, because it's complex numbers. But I mean real, is it physically real? And Einstein had the following criteria. He says, if you have a state that you can perform a measurement on the state such that it gives you the answer, yes, you've got the right answer, yes with certainty, that gives it a reality without disturbing the system. That's right. If you can make a measurement without disturbing the system to say are you, like a spin half particle say, is your spin in that direction? And if you've got it right, it will say yes with certainty. That's Einstein's criterion for the reality. I'm slightly changing the terminology. That is, in my mind, the criterion for quantum reality. You can't say, “Hello particle, which way are you spinning?” You can say that, but it won't tell you. It looks at you blankly and says, “I don't answer questions like that.” Suggests a direction, you see. Then you can suggest a direction. If you got it right, it will say, “Yes, you got it right.” If you've got it wrong, it may say yes, it may say no, with various probabilities depending on where. If you've got it completely wrong, opposite right, it will say no with certainty. So Einstein's criterion is satisfied that you can't ask the particle which way it's spinning. That's quantum reality. So you have to distinguish between the quantum and classical realities. And the quantum reality has very curious causal behavior. It looks as though it goes into the past and things like that. So you have to be very careful about it, that it doesn't give you a causal anomaly. Somehow you could predict the past or something, kind of retrodict. So it doesn't do that because it's only quantum reality. But it's a curious kind of situation which I have written about but not probably sufficiently detailed, and I'm thinking about writing a paper which would explain this more completely. It's actually in my book, Fashion, Faith, and Fantasy. So it's in the book, but perhaps it needs a little bit more sharpening up. Yeah.
Hi everyone, hope you're enjoying today's episode. If you're hungry for deeper dives into physics, AI, consciousness, philosophy, along with my personal reflections, you'll find it all on my Substack. Subscribers get first access to new episodes, new posts as well, behind the scenes insights, and the chance to be a part of a thriving community of like-minded pilgrimers. By joining, you'll directly be supporting my work and helping keep these conversations at the cutting edge. So click the link on screen here, hit subscribe, and let's keep pushing the boundaries of knowledge together. Thank you and enjoy the show. Just so you know, if you're listening, it's C-U-R-T-J-A-I-M-U-N-G-A-L dot org.