# 2023-06-26 - Everett@50 - Apart from Universes [YouTube](https://www.youtube.com/watch?v=g3Q67dAgzwo) Duration: 00:53:59 ## Transcript ### David Deutsch 00:00:00 - 00:04:26 Right, well I'll start with a simple fact. Because it's 50 years on, at this moment, in this room, in some nearby universes, Hugh Everett is here with us, celebrating. Perhaps he's there, in that seat. Therefore, in those universes, Simon is sitting somewhere else. Okay, a fact. Now, it's customary to say how astonishing a fact that is. But that we know of other universes and can reason about them and have evidence of their attributes. But actually, by this time, the only astonishing thing is that that's controversial. After all, we know that no single universe theory can explain even the EPR experiment, let alone something like quantum computation, because we know that any process, hidden variables or whatever, that accounts for those phenomena, must be exponentially more complex than everything we see. And it must also contain many autonomous streams of information, each of which describes something resembling the universe as described by classical physics. But I don't want to address that controversy today, for reasons that I'll explain, except for one aspect, which is just terminological. Well, actually, it's more than just terminological. Whenever I say interpretation, I'm going to mean that with the scare quotes. I'm not going to refer to Everett's discovery as an interpretation, at least if I can help it. Or even as the interpretation of quantum theory, even though that's what it is, that no other is known, because the term interpretation has come to have connotations that misrepresent not just quantum theory, but science in general. For instance, we never speak of the existence of dinosaurs millions of years ago as an interpretation of our best theory of fossils. We say, or we claim, that it is the explanation of fossils. And what's more, the theory isn't primarily about fossils, it's about dinosaurs. We say that the dinosaurs really were there, even though there is an infinity of rival interpretations of the same data that make all the same predictions and yet say that the dinosaurs weren't there. For instance, there's the interpretation that fossils only come into existence when they are consciously observed. And therefore, fossils are no older than humans, so they can't be as old as claimed. And what's more, in that interpretation, they aren't evidence of dinosaurs. They're just evidence of those acts of observation. Or there's the interpretation that dinosaurs were such weird animals that conventional logic doesn't apply to them. Or there's the interpretation that it's meaningless to ask whether dinosaurs were real or just a useful fiction to explain fossils. None of those interpretations is empirically distinguishable from the rational theory of dinosaurs. But they are ruled out because all of them are general purpose means of denying anything. ### David Deutsch 00:04:26 - 00:08:43 You could even use them to deny that quantum theory is true. So insisting that parallel universes are only an interpretation and not a scientifically established theory or something, as if there were such a thing, has exactly the same logic as those stickers that they put in some American biology textbooks saying that evolution is just a theory by which they mean exactly that it's just an interpretation. Or in terms of the analogy that Everett himself used in his famous exchange of letters with Bryce DeWitt, it's like claiming that the motion of the Earth is only an interpretation which we place on our observations of the sky. Now, a prime mistake there in all those interpretations is to conceive of scientific theories as being composed of separate formalism, predictions and interpretations. Because that necessarily makes it look as though the interpretation, which then by definition doesn't make predictions, can't be tested and therefore isn't scientific. And as I've just explained, then interpretations are also infinitely variable. So are formalisms then, in that case, because on that view, the content of a scientific theory entirely consists of its testable predictions. But that is equivalent to regarding observation as an irreducible primitive in science. And that is inconsistent with developing a proper theory of measurement that is itself scientific. So you can't go that way. And this is one of several reasons why science can only be explanatory, asserting what is there in reality. For discussion of other reasons that science can only be explanation, see my book, The Fabric of Reality. The only purpose of formalism, predictions and interpretation is to express explanatory theories in such a way that they can be efficiently criticized. And the explanatory theories have to be about what is there in reality, not just about human perception. That is an intolerably parochial view. And in that regard, all the so-called interpretations of quantum theory are either versions of Everett, sometimes in heavy disguise and with a heavy equivocation like Bohm's theory. Or they claim in effect that no explanation in my sense of quantum phenomena exists. That is to say, no objective reality which alone accounts for the observations on the one hand and obeys the equations, the formalism on the other. If you deny there's an objective reality, then yeah, you can do that. But then science stops being about anything apart from your own thoughts. So the idea that science is about interpreting mathematical formalism is horribly misleading. Which is why I won't say Everett's interpretation, I'll say Everett's theory, which is quantum theory. And I'll use those terms more or less interchangeably. So much for the terminological issue. ### David Deutsch 00:08:43 - 00:13:44 Now, if Everett were here today, which is another way of saying in the universes in which Everett is here today, what would he make of the progress that has been made with his theory since his discovery? I think it's fair to say that averaged over the whole period, progress was disappointingly slow. However, thanks in no small part to some of the people here today, that rate of progress has been increasing. And I'll come to that in a moment. But first, why was progress slow? Well, a related issue, which we usually discuss at the same time, is that take-up of Everett's theory was and continues to be disappointingly slow. But I think that that is an effect, not a cause. After all, 10%, 1% of a vigorous research community like the theoretical physics community is plenty to pursue any one topic. There are plenty of causes in physics that are pursued by only 1% of the community and in which progress is made. The real impediment to progress in understanding quantum physics during those decades was quite different, I believe. It was that... well, let me digress for a moment to historical speculation. That was a bit of hyperbole by Dennis Sciama, who was my boss when I came to Oxford as a graduate student. I quote that from memory, but said it several times in different ways. That phenomenon has been diminishing, but it was once an intense effect, and it's tempting to blame it. But no, you can't blame a theory's opponents for not developing it, for not getting it. By the way, Schrödinger also had the basic idea of parallel universes shortly before Everett, but he didn't publish it. He mentioned it in a lecture in Dublin, in which he predicted that the audience would think he was crazy. Isn't that a strange assertion coming from a Nobel Prize winner, that he feared being considered crazy for claiming that his equation, the one that he won the Nobel Prize for, might be true? So, Schrödinger and Everett and DeWitt and Sciama and everyone else who realized that Everett was right, encountered that perplexing phenomenon. And as a result, the proponents of Everett's theory, to the extent that they worked on it at all, focused on defending it against criticism of rather poor quality. Initially, against its overtly irrational predecessor, the Copenhagen Interpretation, and also then against the Bohm Theory. By the way, the Bohm Theory, future historians may regard the Bohm Theory as just another of the predecessors of Everett that didn't quite get it, but almost got there. And from hundreds of years from now, people won't be able to tell much difference there. Bohm could easily have anticipated Everett entirely if he had chosen not to equivocate about which of the things in his theory were real and which weren't. And then later, they were defending against the other disguises and denials of quantum theory that were proposed for the purpose of escaping from the plain physics of the theory. So they, we, defended it, but that was no way to solve that problem. In fact, it's no way to solve any problem because defending a theory against its predecessor is inherently backwards looking. Of course, some such defense is necessary, but preoccupation with looking backwards to refute bad theory makes it hard to move forward with a good one. ### David Deutsch 00:13:44 - 00:17:58 It's as if molecular biologists were to spend all their time refuting creationism. Then, more sophisticated opponents arrived with better criticism and Everett's theory was indeed improved by meeting that. But still, even then, that's the kind of progress which is like putting better tires on your car. What is the point of having better tires on your car if you're not going to use it to drive somewhere? So, that I think wasn't the reason. This was the reason. Looking backwards instead of forwards. That's my historical speculation. And that has palpably changed, but not yet enough. So, I want, today, I want to advocate developing the sophisticated and far-reaching consequences of Everett's theory rather than endlessly defending and re-defending just the basic idea of it. Let's address the many perplexing problems to which quantum theory gives us access, each of which, presumably, promises new insights into reality. And in most cases, it's only Everett's theory that does that because most of those problems are fatally obscured in other approaches by their vagueness and crudeness and, in many cases, by their denial that there's anything real there to theorize about in the first place. You know, if someone opposes, say, realism or insists on inherently vague or equivocal language or relies on what Richard Dawkins calls the argument from personal incredulity, which, by the way, I think you will find in virtually every critique of Everett and virtually every defense of it as well, then you can't prove them wrong. But what you can do with a bit of luck is make progress. And I think that most physicists and philosophers actually want to understand the world, no matter how loudly they may deny this. So it's going to be progress that's difficult or impossible to achieve in other approaches that will ultimately be the best and only possible effective defense of the rational quantum theory, just as progress in general is the only really effective defense of reason and of science in general. So I want to tell you what I think some of those unsolved problems are. They're all in one way or another about what is really out there in reality apart from universes, because the multiverse isn't just a collection of universes. In fact, universes, the things that historically nearly all the fuss about Everett has been about, are really just classical physics. An ensemble of classical universes is still classical. Even if they slightly interact, it's still classical. It doesn't have entanglements, it doesn't have phases to its amplitudes, and it doesn't have continuous motion of discrete observables or quantum, doesn't support quantum computation. It doesn't even represent observables that aren't close to being diagonal in some preferred basis. So these are all properties that the real multiverse has in addition to having universes in it, and that's not an exhaustive list. ### David Deutsch 00:17:58 - 00:22:29 Okay, so let's start with a free particle in empty space. You might think that that, at least, is something that happens in each universe independently of the others, so that we can forget about the multiverse when we're talking about just a free particle in empty space, but nothing of the sort is true. So first of all, a particle is described by wave packets, thanks to the uncertainty principle, which means that as far as universes go, it's at different positions in different universes, but then also, because of the uncertainty principle and because of the linearity of quantum mechanics, there is no region of the multiverse in which the position and also the velocity are behaving independently of what is happening elsewhere in the multiverse. That means that there are no autonomous information flows in sufficiently fine detail in the multiverse at all, anywhere. So in fine detail, even a free particle is an irreducibly multiversal object, not just a parallel universes object. And furthermore, at some later time, the particle will, the wave packet will have changed, and the instances of the particle in the multiverse will be at different positions. None of them individually will have moved to where it is. That doesn't happen, because there is no such thing as any of them individually. When the universe approximation breaks down, the autonomy of the instances of a single particle in the multiverse breaks down too. They are then fungible. In fact, all this is implicit in the equations, but we lack an explicit mathematical description of the multiverse at any level lower than that of the universe, which is an emergent level, as we now all know. And this is one of the most telling omissions, in my opinion, corroborating my historical speculation, from multiverse research so far. No one has yet done the spadework to construct a mathematical description of what the multiverse even is, to fill in that gap. In general relativity, for instance, Einstein's field equations are understood to apply to a physical object, space-time, which has a non-trivial structure. It's got a geometry and topology and singularities, all of which took a lot of careful mathematical theoretical thinking to discover and to express. Whatever possesses people to think that, for instance, quantum gravity can be cracked before we've developed a similar understanding of the multiverse of quantum theory. So where should we look for that understanding? Well, even in general relativity, much of the theory of space-time is the theory of information flow. The Lorentzian metric, event horizons, Cauchy surfaces, causality conditions, and so on. And at the more basic level, there's just the plain relativity of simultaneity as well. In the structure of the multiverse, we have good reason to believe that information flow is even more important. So it might even be the whole theory. And it's studied in the quantum theory of computation. So that's a good place to look. In fact, that's a good place to look for most important things, I've found. So at the crudest level, quantum computation is just quantum parallelism, many classical computations in parallel, which is the parallel universe approximation. But we need to do a lot better than that. ### David Deutsch 00:22:29 - 00:25:50 We know that the physical attributes of systems determine whether and how they can be used for information processing. To do Turing computations, you need discrete observables that can interact in various ways. You need stable information storage. The next step, say quantum cryptography, requires certain operations on qubits rather than bits, but only individual ones. To build general quantum gates, you need to be able to do more. And for universal quantum computation, you need to be able to do even more, such as error correction. And then it stops. Then we have something that is universal. So different kinds of information processing are possible in different kinds of multiverse regions. So the understanding that's currently being developed about these different kinds of computational resource is telling us about the structure of the multiverse. And many of the people doing it don't even know that. But the sophistication of these new results about computation contrasts very sharply with the wooly and archaic explanations that still accompany them. In quantum teleportation, for instance, we are still routinely told that information passes from A to B without passing through the space in between. Or that it travels backwards in time to the source of the entanglement and then forwards again to its destination. Or that it jumps instantaneously across the gap, whatever that means. Again, there's no substance accompanying any of those descriptions. No mathematical object that is even purported to represent those alleged weird processes of jumping and so on. Only the observed outcomes are described mathematically. Because again, the quantum formalism is being used purely instrumentally together with the same old single universe apologetics. The only way of understanding what's really happening in quantum teleportation and in the newly discovered important processes of cluster or graph quantum computation is to apply Everett's theory. Patrick Hayden and I analyzed information flow in general entangled systems using quantum theory of computation and showed that there's quantum information inside decoherent observables. Now there's no way even to express that fact other than in Everett's theory. Because it means that there is no classical level. Even physical variables such as those in an observer's brain which behave autonomously in each universe also carry this hidden information that's not confined to universes and which can later participate in quantum computations. By the way, in that work we used the Heisenberg picture which I don't have time to go into today but suffice it to say that working in the Heisenberg picture is the key to eliminating 50% of all misconceptions in quantum theory. Including ones about non-locality in things like quantum teleportation. ### David Deutsch 00:25:50 - 00:29:48 Okay, now there's a related and even worse spanner in the works of elementary particle physics. Particles or strings or whatever are supposed to be fully quantum mechanical entities in nature but the people who work on them only ever construct classical single universe theories. Why? Because they think that the quantum part of the theory by definition has to be trivial. It's assumed that in order to discover the true quantum mechanical equations of the world you have to enact a certain ritual. First you have to invent a theory that you know to be false using a traditional formalism and laws that were refuted almost a century ago. In some senses more than a century ago. Then you subject this theory to a formal process known as quantization which for these purposes includes renormalization and that's supposed to be a quantum theory. A classical ghost in a sort of tacked on quantum shell. In other words the true explanation of the world is supposed to be obtained by a mechanical transformation of a false theory without any additional explanation being added. That is close to being magical thinking. How far could Newtonian physics have been developed if everyone had taken for granted that there has to be a ghost of Kepler in every Newtonian theory of anything? That the only valid solutions to Newton type equations were those based on conic sections because that's what Kepler had. And because the early successes of Newton's theory also had them just as the quantization recipe did successfully discover new quantum systems what three times four times. So it was worth trying for one or two more times as a heuristic. Lots of things are worth trying. String theory was worth trying. Well worth trying. But there was no justification for relying exclusively on nature being this classical ghost in a quantum shell. And I think it's not a coincidence that the decades of not taking seriously what quantum theory says in its own right by following up on Everett was also the time when discoveries of new kinds of quantum systems frustratingly slowed down as well. But it's surely a mistake to stick to the historical single universe worldview even while using quantum mechanical equations. We should be seeking explanations not in the confines of one space time or even multiple space times but in the multiverse. Such theories are possible. I even constructed one a while ago for unrelated reasons called qubit field theory. Anyway another problem is the nature of time. Now in 1982 Don Page and Bill Wooters discovered that times are Everett universes. They were addressing again an apparently unrelated theoretical problem, namely since total energy is conserved how can we tell that the world isn't in an eigenstate of it. And they discovered that we can't tell. We can assume without loss of generality that the world is in an eigenstate of its Hamiltonian. ### David Deutsch 00:29:48 - 00:34:02 A stationary state. But then what's all this motion that we see? Well by analyzing closed physical systems that included clocks they showed that motion is an entanglement effect. Clock states are correlated with states of other systems. This important result should be the basis of everyone's conception of time. Yet it has barely and rarely been taken on board. Julian Barbour's a very honorable exception to that. Why not? Because again this is an issue that one can address and solve and understand only in Everett's theory. All other approaches just command us to talk nonsense or assume a classical conception of time by fiat. Or if they were being consistent they would say that the present is the only real time and the rest are just possibilities. I should also mention time travel. I think that more work needs to be done on information flow around closed timelike curves. Because even if it turns out that we have to rule those out, how can we do that without understanding what causality means in the multiverse? And one thing we do know is that it means something different to what it means in space-time. Then there's the arrow of time. Arrows of time, entropy is a kind of measure of the branchiness of the multiverse. And the so-called subjective or knowledge arrow of time is also related to multiverse structure because knowledge is associated with complexity that extends over a multiversal region. So it seems likely that the entropy and knowledge arrows of time are connected via the structure of the multiverse. That's another potential avenue for progress that's open only in Everett's theory or nearly only. One of the successes along the lines I've been advocating that has already been achieved is in the understanding of probability. That is answering that question with that rider. Others at this conference have already discussed and no doubt will discuss more that work. But what I want to stress now is that Everett's theory provided the solution to a philosophical puzzle that people had worried about quite independently of quantum theory. It's a problem that's a bit like deriving an ought from an is. Deriving a tends to from a does. So like you try to explain what a probability statement means, like that heads will probably come up about half the time when you toss coins. What does that mean? Well, you find yourself reducing it to other probability statements like that if you bet on some other proportion than 50-50 you're likely to lose. But then what's likely? It seems that you can never reduce it to so-and-so will happen. And so people historically have been reduced to talking nonsense about this. Like for instance the frequency or ensemble interpretations of probability. They say that something is probable if it would happen in the majority of cases if an experiment were repeated infinitely many times. But then it's not talking about reality because the real experiments aren't performed infinitely many times. And even if they were, we want to know what to do on particular occasions. And in any case, the frequency interpretation, whether with an infinite number of instances or a finite number of instances, logically doesn't reproduce probabilities unless the probability is a constant. In fact the instances are equally likely. So you get to a circular definition of likely. Or people have tried to say that probability is a measure of subjective belief. But then you have to explain why my subjective belief about a coin is a property of the coin rather than of me. And the answer has to be that it's not just any old belief. ### David Deutsch 00:34:02 - 00:37:51 It's the belief that you should have if you're rational. But then why should a rational person have that subjective belief? Because if you adopt some other belief you're likely to bet on the wrong things. But what does likely mean? So there's circularity again. David Papineau once called this state of affairs at the foundations of probability theory a scandal. Well, as I said, I think it's been solved. Everett's theory, which is a completely deterministic theory of a multiverse without stochastic processes, determines what is rational behavior for observers who bifurcate in the multiverse. Consequently, anyone who sets out to deny Everett now has the embarrassing problem that the philosophy of probability is a scandal without it. I know that some people here disagree, but some people here agree, which goes to illustrate the merit of seeking progress instead of ever more devastating demolitions of what were always bad ideas in the first place. However, there do still exist some outstanding and promising problems in regard to probability, because there's more to probability than just decision-making when observers bifurcate. There are also important situations where an observer only exists in some regions of the multiverse and not in others. One example is anthropic-principle argument, where people try to explain the values of observed quantities by calculating their expectation values given that someone asks what the values should be. In particular, they're usually concerned with ensembles of universes with different sets of laws of physics. We exist in some of them and not in others, and so we can make predictions, but one problem that arises is who are we? This isn't just the problem of personal identity in the philosophy of mind, though that too is an issue that one can't hope to understand without the full quantum context, as Michael Lockwood and others have pointed out. For instance, the universe may be spatially infinite, in which case again there may be infinitely many instances of the whole universe that's observable from here. And if we identify all identical instances of ourselves as being the same person, which surely we have to do if we're a physicalist, then should we not also be identifying identical observable universes across space as being the same universe? But in that case, space would really be finite after all, even in the cosmologies where space is infinite in the classical general relativity approximation. Then there's the so-called quantum suicide argument to the effect that if you want to win the lottery, all you have to do is buy yourself a ticket and set up a machine that will kill you automatically in your sleep if you lose. And then in the universes in which you do wake up, in which you exist to make the observation at all, you're bound to be a winner. And then there are various types of what's called a simulation argument, for instance by Nick Vostrom, such as that in the distant future the whole universe as we know it is going to be simulated in computers many times, and therefore the overwhelming majority of instances of whatever we have observed will be observed in those simulations, and therefore we probably are in one of them. So the argument goes. ### David Deutsch 00:37:51 - 00:41:39 Now at present, little if any progress can be made in any of those controversies. Ingenious arguments have been devised for all of them, but at present they all ultimately reduce to hand-waving because at the root of all of them there is some assumption about probability, and all of them are guilty of using the frequency interpretation or the subjective belief interpretation or both. And they are therefore all in a state of sin at present. What is needed is to express any such argument in the framework of a theory of a multiverse. Sometimes it has to be a bigger multiverse than the multiverse of Everett. And then one has to derive the premises of one's argument from laws of physics, purported laws of physics, as Everett did. One has to determine then, as was done with the Everett multiverse, whether that theory forces a unique meaning on the numbers that those arguments want to regard as probabilities. A unique meaning such that those numbers can be used to make predictions and place rational bets and infer, make Bayesian arguments about inference about the state of the universe and so on. And all the other uses of probability. And if it doesn't force that, then no progress can be made with it and the argument in question isn't really valid, isn't really scientific. My guess is that when that's done, we'll find that the quantum suicide argument and most versions of the simulation argument are actually invalid. And the anthropic principle arguments, I suspect, are actually right as far as they go, but insufficient in themselves to explain anything. But that's my opinion. It's just hand-waving. Hand-waving, as I said, can't resolve those issues. Only future development in multiverse theory can. Okay, so here's a quick summary of the open problems that I've mentioned. Really just look at the length and diversity of that rather than the content. Everett's theory is pretty far-reaching. That's the point I'm trying to make. And this is just the tip of the iceberg, I'm sure, because quantum theory is deeper, much deeper, than all the other theories in physics that are considered to be fundamental. So the implication of that for physicists is, well, we've got to understand this thing, the multiverse, because otherwise in every fundamental branch of physics, it's as if we were planning an expedition to the moon while still thinking that the Earth is flat. We may not go wrong, but we probably will, and we probably won't have the wherewithal to proceed. To philosophers, the implication is some of our basic intuitions about the physical world are plain wrong, not just in regard to how many universes there are. As I said, that is the least of it. That's just classical. It's what is there apart from the universe. And hence, since philosophy is about understanding things like probability and time and the nature of existence and non-existence and the self and causation and laws of nature and the relationship of mathematics to physical reality, for all of those, for any of those, you've got to understand the multiverse first. ### David Deutsch 00:41:40 - 00:42:18 Learn to think in terms of it. Build on that understanding. Leave behind the older misconceptions and the older conceptions of the multiverse. And apply that to learning about everything else and discover its successor. As I said, there are people here already doing those things, but to the rest of you and to the physics and philosophy communities at large, don't let it be another 50 years before you too become serious about building on Everett's theory. Thanks. ### Oliver Pooley 00:42:18 - 00:42:33 OK, well, we've got about three to five minutes for questions, but we may get a little over that. ### Lev Vaidman 00:42:33 - 00:43:23 Certainly, very much with you, Everett, by far the best interpretation is correct and it's a miracle, but it's not understandable till now. But when I tried to read it, I think why, I think some of the things, excuse me, you may be somewhat responsible. I think the first one is the myth, which says that there are many worlds and the two states that are really separate. And it looks like your multiverse, which before I thought it could be with different names, looks like you consider multiverse a theory about multiverse. And I think Everett, me and especially Schrödinger, which is what we went for, it's a single physical universe. I don't think the multiverse, it's not a theory, it's multiverse. The main important thing is that there is no collapse and this is the main point. ### David Deutsch 00:43:23 - 00:43:25 I agree with that. Yes, go ahead. ### Lev Vaidman 00:43:25 - 00:44:11 I have to be brief, so I just speak. And the other thing, which you made big contribution, which would put it forward, but you also ask things that you believe can be achieved, which might be not. And I think Everett was the first to say that you can get a probability out of the many worlds. If he didn't say so, I think now it would be accepted, because this was not proven at that time, it's not proven until today. And when people didn't show that this doesn't work, they killed the whole theory. But the main thing is that there is no collapse and no contradiction of experience. This is Everett's theory. If you would not add some other things which may be not achievable, then it would be accepted. ### David Deutsch 00:44:11 - 00:45:22 Okay. Two things there. First of all, I think what you say about universe, multiverse, I don't care about the terminology. I'm quite happy to call the multiverse the universe and the universe a branch. In fact, I did that in one paper. I don't care. Certainly the point of Everett's theory is that we don't need collapse and that quantum theory can be regarded as a complete theory of the universe. As a probability, I agree with you that the objection to Everett that it didn't describe probability properly was always a mistake. What was happening there was that there is a problem at the foundations of probability theory which was obscured by silly interpretations. But in Everett's interpretation, you could actually see that problem. It wasn't a problem with Everett's interpretation. It's something that you could see through Everett's interpretation and then, we believe, solve. If we hadn't solved it, it would be no argument against Everett. A sort of friendly amendment, I think. ### Simon Saunders 00:45:22 - 00:45:56 You had your single particle in space, classical multiverse approximation breakdown. It seems to me that being the case, it's a retrograde move to describe that thing as multidimensional in the sense of describing it as living in different universes at different points. Those sorts of position-based distinctions are there in the quantum mechanics, by all means, but calling them universes doesn't seem to me to be necessarily that helpful. They're not dynamically autonomous. They're not that clunky thing. They're just those bits of quantum space. ### David Deutsch 00:45:56 - 00:46:12 I entirely agree. If only we had a proper mathematical description of the multiverse, then I'd be very happy to say what's actually happening with a single particle is so-and-so. You can see that you can't then make the universe approximation and have it be accurate. Yes, I agree. ### Oliver Pooley 00:46:12 - 00:46:16 Simon, do you want to respond? ### Simon Saunders 00:46:16 - 00:46:30 When you say we don't have a proper understanding of the multiverse, what sort of understanding do you have in mind? Is it mathematical? ### David Deutsch 00:46:30 - 00:48:15 There's a mathematical issue and a mathematical physics issue and also a philosophical issue. The mathematical physics issue is that we don't know what mathematical object the equations of quantum theory are supposed to apply to. We know how to get answers out of them and we know how to describe something about this object in emergent approximation. But we don't have an exact theory of it. We don't have a theory that's in principle exact. For instance, in that work that I did on time travel, going back in time and coming back into a different universe, I have a diagram there where there's a space-time and the loop in space-time gets unfolded into a... I don't think any of you are familiar with that word. But anyway, the point is that that manifold, which is a hybrid conception of a relativity thing with a quantum thing, there's no mathematical object that I can say that that's an approximation to. There ought to be. And in the philosophical... So that's the mathematical physics part of it. I think we need that in order to make progress in certain directions. There's also the philosophical thing that we don't have a vocabulary and a language. We have a vocabulary to talk about situations where an observable is sharp and we have vocabulary to talk about parallel universes. But most of the multiverse isn't like that and we don't have a good way of talking about what that is like, even in the case of a single particle. So those are the two things that are lacking. ### Oliver Pooley 00:48:18 - 00:48:28 I think we should get underway by just after 5:20. There are three other people who wanted to ask questions. James, Meir and Julian. So, James. ### James Ladyman 00:48:28 - 00:49:28 I thought you overstated the argument against the anti-realism and also that you ran together two distinct things. So an anti-realist about science doesn't have to deny that there's any objective reality. They just deny that we're finding out about it. So you shouldn't run those things together because obviously one is much less plausible than the other. And I think that anti-realism is perfectly not anti-realist but it is an intelligible view to take about science. That is exemplified by the attitude of lots of people who think about Newtonian mechanics. You can think it's a wonderful theory, use it every day, predict lots of stuff with it, don't believe that there really are Newtonian action-at-a-distance forces. That's not a crazy view to have of the world. So someone who thinks about quantum mechanics, someone might think about quantum mechanics just like that. I think it's a good theory to use but I don't see why I should believe it to be telling me the truth about it. ### David Deutsch 00:49:28 - 00:50:43 Okay. First of all, I'm glad you're not an anti-realist because it's very hard to argue with someone who denies that they exist. So about whether I was too hard on anti-realism, all I said about it was that you say it could be that there is a reality but that science doesn't have access to it. All I said about it was that if you take that view then science isn't about anything apart from your own mind. And I think that would be true of a person who took the view that you outlined as well. So that's... Now as for explanation, the trouble is if you regard a theory as being purely a set of observable predictions, then you must logically be thinking of observations as unanalyzable primitive things. And that is incompatible with having universal scientific theory. Also, there's the fact that explanation... There are infinitely many interpretations of those things which make different predictions for the future. And then you'll run into the problem of induction and so on. I think... Well, I have to refer you to the fabric of reality for why science has to be explanatory. ### Meir Hemmo 00:50:44 - 00:51:29 Okay, I just wanted to... We had a discussion this morning about the meaning of probability. Sorry, I want to ask that. Yeah. And many people in this room, for example, feel that your derivation of the Born rule is not based solely on rationality. But you do have some assumptions which are not part of the rational decision theory. And second, it might even be not possible to justify this assumption in the context of theory. ### David Deutsch 00:51:30 - 00:51:50 Well, at the moment, all I can say... I think they are justified. But all I can say about that at the moment is the same as my reply to Lev's question. If it turns out that there is a hidden assumption there, and we have to admit that there has been... There was an assumption in it. Okay, well if... ### Meir Hemmo 00:51:50 - 00:51:56 It was your negativity in your proof and what David wanted in the other... ### David Deutsch 00:51:56 - 00:52:25 If this really is an assumption over and above rationality, then it's no argument against Everett's theory. It simply says that there is a remaining puzzle about what probability really is. And I say we have to admit that all previous attempts for whatever it is, 300 years, to solve this problem of deriving a tends-to from a does, have contained hidden assumptions. And it could be that we're wrong as well. We're not. ### Julian Barbour 00:52:27 - 00:52:41 David, you've completely persuaded me that the central task is to find what the multiverse is like. Would you agree with me in Popperian lines that ultimately someone's got to conjecture what it's like and then we've got to... ### David Deutsch 00:52:41 - 00:52:57 Absolutely. I agreed entirely with what you said about that yesterday. Yes. Yes. It's got to be... They've got to conjecture an explanation, that is an assertion about what the reality is like, observed and unobserved, and then that has to be tested. ### Julian Barbour 00:52:57 - 00:53:19 As a follow-up to that, there is an argument for saying that science actually progresses in small steps, so I wouldn't be quite so quick in saying dismissing the ghost within the thing, because that's really actually not... The ghost within the... The classical ghost within quantum is a bit like not making a giant leap rather than a series of small leaps. ### David Deutsch 00:53:19 - 00:53:49 Absolutely. And to some extent I have made a conjecture about what the multiverse is like. I think it's probably almost certainly too naive and we'll be wrong. But that is essentially what [unclear name] conjectured 40 years ago. So I wouldn't totally dismiss that as a possibility. I said it was worth trying heuristically, and it has been tried, but it seems to be running out of steam. So there seems to be an obvious explanation for why it's running out of steam. ### Oliver Pooley 00:53:49 - 00:53:56 Okay, I think we should call this part of the session to a close, and thank David again for a wonderful speech.