Episode 23: Many Worlds Quantum Mechanics

• Links to this episode: Spotify / Apple Podcasts

Unofficial, reviewed transcript. It may still contain mistakes; check the podcast when wording matters.

Speakers are labeled by first name where identified. Unidentified-speaker labels may refer to different people. Bracketed question marks indicate uncertain wording.

Transcript

[00:00:07]  Bruce: Welcome to The Theory of Anything podcast. We've got a new co-host today. This is Tracy, who is a friend of mine. Actually, we've been friends since high school. And interesting story. She was actually the one I originally started this podcast with, and it was just hard to find time that she could do it. And so just by chance, Cameo asked me to do a podcast with her too, and so we ended up starting with Cameo. But I've been trying to get Tracy on the show for forever. Tracy, say hello.

[00:00:38]  Tracy: Hello, I'm just a basic working person who's interested in all sorts of topics. I guess I'd classify myself as somewhat of a nerd. So I like to read books such as David Deutsch's books, Fabric of Reality, and anything that's interesting about the world and reality and science and learning, sign me up, I'm all about learning about it.

[00:01:08]  Bruce: And that's certainly true. You ask a lot of really good questions.

[00:01:11]  Tracy: I appreciate that.

[00:01:15]  Bruce: And then we have our guest today, Sam Kuypers. Did I say that right, Sam?

[00:01:20]  Sam: Yes.

[00:01:21]  Bruce: He and David Deutsch recently did a paper called Everettian Relative States in the Heisenberg Picture. But so Sam and David Deutsch did this paper and I had noticed it. And that was why I asked him on the show was to ask him questions about his paper and get him to kind of describe it. It is a bit of a rough go mathematically, but Sam and I have been working on also some YouTube videos that describe the mathematics for someone who's interested, layman, who's willing to put a little bit of effort in to understand the mathematics. And by the time you hear this episode, those should be up on—edited and up on YouTube. The first one anyhow, we may do some additional ones. And the first one was about explaining Q numbers, which is a big part of this paper. So Sam, can you introduce yourself? Tell us a little bit about yourself.

[00:02:15]  Sam: Yeah, so I am a DPhil student at Oxford, where I research the foundations of quantum theory, and that usually means, well, in my case, it means I'm interested in the Many Worlds interpretation of quantum theory. And that's the topic of the paper. So the paper is about kind of reformulating Everett's relative state construction, which is the foundations of the Many Worlds interpretation of quantum theory.

[00:02:44]  Bruce: Yes. So Sam, I got lots of questions about this, but I think my first question for you, though, is how many people actually accept Many Worlds interpretation of quantum physics? As someone who's in the field and really doing this in real life, how much of a minority are you?

[00:03:04]  Sam: I'm not clear on how big of a minority we are at the moment. I remember reading a blog post by Sean Carroll, in which, if I remember correctly, it's roughly 30% of the physics community who are studying these topics, they're studying kind of foundations of quantum theory, believe that the many worlds interpretation is the way to go. Of course, in Oxford, there's kind of the gang of Everettians reside here. Usually, if you read, for example, the book by Peter Byrne about Everett's life, then there's a section on the Oxfordian Everettians. And so here it's a big deal.

[00:03:51]  Bruce: Yes.

[00:03:52]  Sam: That's something.

[00:03:52]  Bruce: In fact, you got a conference coming up that you guys are doing, right? Is it Oxford that's putting that on or is it just a conference in general, the virtual conference?

[00:04:01]  Sam: That is just a virtual conference that I'm putting up with a friend of mine, a friend and colleague, Charles Bédard. Yeah, it isn't really connected to the university in any direct way other than that I'm a DPhil student here and I wanted to organize a conference. And luckily, doing that online means that there's not a lot of administrative work. So you're very free in just inviting people and having them speak. And yeah, so it's something that we didn't organize with the university specifically.

[00:04:34]  Bruce: But you've got like a lot of really big names coming to the conference from what I can see. So I actually think people should check it out. What is the name of the conference?

[00:04:44]  Sam: The name of the conference is On the Shoulders of Everett. And yeah, we have a bunch of really interesting speakers. We have Chiara Marletto, David Wallace. We have Carlo Rovelli. Yeah, Vlatko Vedral, my supervisor, will also be speaking. And I think [unclear] Zurek is another speaker that I am very much looking forward to hearing present at the conference. Yeah, we have a nice list of people attending. So it should be fun.

[00:05:16]  Bruce: Yeah, absolutely. So you just mentioned that like 30% or so of those looking into the foundations of quantum physics, accept many worlds at this point. But I'll bet you if we went back a decade or two, even just maybe even just a decade, it would be nowhere near that large. Would I be right or am I misinformed here?

[00:05:36]  Unidentified speaker: I'm not actually sure.

[00:05:38]  Sam: So I think that the poll that Carroll writes about on the blog is quite old. OK, but I could also be misremembering the statistics here. I think it is still something like that. I imagine. Well, so it's very strange because I think many students are becoming ever more interested in the topic, but that doesn't necessarily translate to there being more people in high positions and universities who actually research it. And I think there's a distinction, but I'm hopeful that it is a growing community of people. I guess the reason I don't have the statistic in mind is because I tend to just be convinced that it is the right way of doing things. And if it's the right way of doing things, then if you make the best case where you can, you will hopefully convince others and it will increase in popularity over time.

[00:06:28]  Bruce: Yes.

[00:06:28]  Sam: And also it's the thing worth working on if it's true.

[00:06:32]  Bruce: Give us a little bit of history about Many Worlds. How did that interpretation come about? And like, where did it come from and what was the, why was it necessary?

[00:06:43]  Sam: So it came from Everett's thesis under the supervision of Wheeler. The main problem that Everett tried to address was this idea that there's two ways in which quantum systems appear to evolve. One of them is the way systems evolve in general, which is kind of, if it's not being observed, then there's a set of rules. It's like, for example, the Schrodinger equation that determines how systems evolve over time. And whenever they're being measured, they apparently collapse, what they call collapse. And it's a different kind of dynamical law that is used when systems are observed. And the two can't really be unified in any nice way. And the situations in which we apply the collapse rule are rather parochial. They're when, for example, some kind of what you think of as a classical measure or a person or someone who's conscious or something, measures a system.

[00:07:44]  Sam: And that brings into fundamental physics all kinds of things that I don't think belong in fundamental physics in the way that they're being presented as. So, for example, I don't think that consciousness is part of quantum theory. That is a separate interesting topic that should be treated differently. And it's very unsatisfactory to have as a rule in quantum theory that something evolves differently when it's being consciously observed. So that's kind of one criticism of the collapse postulate.

[00:08:19]  Bruce: So when I looked into quantum physics, I was immediately struck by the fact that there were many things that counted as an observation that caused this collapse. And it didn't require anything like consciousness. It could be a detector. It could be a photo plate. So it's interesting that, I mean, I agree with you when you bring this up, but even just with the collapse function, even with the way they were teaching quantum physics, it's not immediately obvious at all that you have to insert consciousness. That seems almost forced that people started inserting it in. It wasn't part of even that interpretation.

[00:08:55]  Sam: Yeah, well, that kind of brings us to a paradox known as Wigner’s friend, in which you have two observers and one of the observers is isolated in a room, and their task is to look at a quantum system and to note down what kinds of measurement outcomes they detected. And there's a second person who is usually known as Bob, and the first person is usually known as Alice, who is outside of the room that Alice finds herself in, and the room is completely isolated, and no one knows what's going on inside the room, except for Alice. And the strange thing is that if you apply the rules, according to how they're prescribed, then you would say, okay, well, Alice is performing a measurement, and therefore something collapses, but there is a larger system, which consists of Alice and the measurement device that are contained in this room, and they're isolated, and therefore they evolve according to the Schrodinger equation.

[00:10:08]  Sam: And there's this person outside of the room called Bob. He concludes, well, as long as I haven't measured this room that contains Alice, she doesn't actually have anything like objective existence, because she's in a superposition. And I think you could say the same thing of a measurement device. You could say, well, there's a measurement device in the room. Forget Alice, this is just a measurement device. It has recorded something, but it doesn't actually mean anything for it to have recorded anything until we see it, because then that's when we apply the collapse postulate.

[00:10:39]  Bruce: Sure.

[00:10:40]  Sam: So that kind of... I think people were forced to start talking about consciousness because of this nonsensical rule. Well, it's not completely nonsensical. I think it's too harsh to say it's entirely nonsensical because systems do appear to collapse, but the logic of it is unclear. There's no explanation for why the collapse happens.

[00:11:08]  Bruce: So this is the measurement problem.

[00:11:09]  Sam: Yes, yes.

[00:11:11]  Bruce: So the idea that Alice has made a measurement, and from a certain point of view, you could say the system has now collapsed, the wave function has now collapsed, and from another point of view, you could say, well, she's now caught up in the wave function, and so for Bob, she is in a superposition herself at this point.

[00:11:31]  Sam: Yeah, so in the usual interpretation, you would say something like, well, she doesn't have objective existence or something. Just like a particle in a superposition doesn't have or isn't assigned anything like objective existence until it's measured, you could then say the same thing about Alice, and then all kinds of interesting questions come up from that. Like, well, if we are in a larger room in the sense of, if we are part of the larger system that's not yet being observed by some other observer, Charlie, then how are we to know that our existence is objective? I think this is actually one of the questions that Everett asked in his thesis, where he raised the possibility that, you know, Alice, of course, has a history before Bob measures the room that she's in, and for her nothing changes when Bob walks into the room. And when Bob starts talking about how he gave her objective existence because he measured her in the room, he can then be countered with the argument that, well, you know, from her point of view, nothing changed.

[00:12:51]  Sam: How about if they're still in a larger room that is yet to be measured by some other observer, Charlie, how are they to know that they already have objective existence or not? And so all that gets you into trouble very quickly.

[00:13:06]  Bruce: Yeah, for that matter, why are we going in one direction? I mean, couldn't Alice also claim that Bob had no objective existence until he walked in the room?

[00:13:14]  Sam: Yeah, very true.

[00:13:15]  Bruce: The other thing that I found interesting when I was studying this years ago was the EPR paradox. If I recall, I don't remember much. I remember working it out once on paper so that I make sure I understood it, but then it like started slipping out of my mind almost immediately. But it was a paper that Einstein wrote with two other people, the P and the R. He was trying to demonstrate that there was a problem with quantum physics because it violated non-locality of his theory of general relativity. Is that correct? Do I understand that correctly?

[00:13:54]  Sam: Yes. So it violated locality.

[00:13:57]  Bruce: Locality, sorry.

[00:13:58]  Sam: Yes. So Einstein has what is known as Einstein's criterion of locality, which states that whenever you have two subsystems that are spatially separated from one another, like, for example, if you have the Earth and the Sun spatially separated in the sense that there's some distance between them and you do something on the Earth, then that operation that you perform on Earth, that has no immediate effect on the Sun. It's only when objects come into contact with one another that they can exert some kind of effect on one another. So operations that you perform on Earth don't necessarily affect distant stars, the Sun, other systems, etc. Maybe an even nicer example is that, you know, if the light from the Sun has to travel to the Earth, and before it arrives on the Earth, it can't have any physical effect on us. That's part of the kind of theory of relativity that the speed of light is the speed limit of the universe.

[00:15:04]  Sam: And that means that before light reaches the Earth, the Sun cannot have any effect on us. And that is another way of viewing locality that systems need to be in contact with one another before they can physically affect one another.

[00:15:19]  Bruce: Yeah. So Einstein's purpose then for coming up with the EPR paradox was to show that quantum physics must be wrong because if it were right, then there would be this non-local aspect to it. That was his point, right? The problem was that the experiment he laid out actually works in real life, thereby leading to this idea that quantum physics is non-local. And how many... To what degree do physicists currently accept the idea that quantum physics is non-local?

[00:15:55]  Sam: That's a very controversial topic. Most people, most physicists think that quantum theory is non-local. In other words, most physicists think that if you have quantum systems, then they can exert effect, they can affect one another even when they're spatially separated. So if you have a quantum system on the Earth and one on the Sun, then they can affect each other faster than the speed of light in some sense. And this is why Einstein thought of that thought experiment, because he was convinced that that is not how physics is supposed to, or how physical systems are supposed to behave. And it seems from the usual way of formulating quantum theory, that that is in fact what's going on, that these systems have effects on one another, even though they aren't supposed to, even though they're very far apart, and should not in any way be able to change each other's behavior.

[00:17:01]  Bruce: So in your paper, you mentioned that Everett formulated his picture of quantum physics in the Schrödinger picture instead of the Heisenberg picture. And that's like kind of the main reason why you're writing this paper, is because you're saying that the Schrödinger picture is non-local, and therefore it seems like if we are taking those equations seriously, then we would have to say that quantum physics is non-local. Is that correct?

[00:17:31]  Sam: Yes. Yeah. So everything I just said about quantum systems affecting one another, even when they're not at the same location, it appears to be true in the Schrödinger picture. And the reason for this is that whenever you have multiple systems that you describe in the Schrödinger picture, so whenever you have, say, perhaps you have two, three or more particles that you're describing in the Schrödinger picture, the way that you describe those particles depends on how you describe the other particles. So there's no way of describing the particles individually, separate from the other particles present in the system, in the composite system. And this seems to suggest that there is no such thing as the individual particle. [There are] ways around this, what's known as the density matrix formalism, in which you can talk about the state of an individual particle, but that formalism is incomplete.

[00:18:26]  Sam: So in the Schrödinger picture, if you have, if someone gives you what's known as the state vector, then in some sense, you know everything about the system. You know all of the possible expectation values of every possible measurement on the system. And if you use this different formalism, the so-called density matrix formalism, then although you are able to describe subsystems independently of the other systems, so you're able to describe, say, a single particle in this multiparticle composite system, the density matrix formalism isn't complete, so you're not able to actually compute all of the possible expectation values. You lose some information by going to this local description. And consequently, people have concluded that quantum theory really isn't local, that it isn't really meaningful to describe subsystems individually because they always have effects on one another when they are part of a larger system.

[00:19:28]  Bruce: You mentioned in your paper that even some really big-name physicists, even ones that believe in Many Worlds, see it as non-local, like Vaidman you mentioned.

[00:19:39]  Sam: Yes. Yeah, Vaidman has a very nice entry on the Stanford Encyclopedia of Philosophy in which he writes about this. He explicitly states that his notion of a world, of the universe, is non-local. In other words, in our terminology, we would say that the universe is all of classical mechanics [wording unclear]. It's kind of the classical world that we see around us. It's one of many such universes which exists in the multiverse. The multiverse is a larger object than just the universe. And in Vaidman's terminology, he says that the universe is a non-local thing, that whenever a split happens, it's really the entire universe that splits all at once.

[00:20:34]  Bruce: Which follows from the Schrodinger picture, correct?

[00:20:36]  Sam: Yes. Yes, that's right. So if you're working in the Schrodinger picture, then I think that's indeed what's going on. Because again, there's not really a way of describing systems separately from one another. So whenever one system splits, this has to also have an effect on the other systems in the Schrodinger picture. Because there is no way of describing them separately.

[00:21:00]  Bruce: So let's take just a moment to kind of explain what we mean by non-local. Kind of the standard experiment would be like, you've got a particle that's in a spin up and a spin down superposition. And it splits into two particles, which are now in a superposition. Do I have this correct? And then you like separate them, take one light years away and leave one on Earth, as long as they don't become entangled with their environment, they should still be entangled with each other?

[00:21:30]  Sam: Yes. Yeah, so it's what's known as a Bell state, where you have... Well, Einstein and Rosen-Podolsky, they didn't actually know about Bell states, but the way in which typically it's formulated today is basically the way that you formulate it now. Namely, you have these small quantum systems, namely qubits, that are entangled with one another, and they're entangled in such a way that if you, for example, measure up in one of the qubits, then the other qubits will also be in an up state, and if you measure down on one of the qubits, then the other qubit will also be in a down state. For example, this is one of the possible ways in which they could be entangled.

[00:22:11]  Bruce: Don't they have to be reversed? I thought one had to be up and one had to be down because of the qubits.

[00:22:15]  Sam: I think both are fine. It just depends. All that matters is that they are entangled in the sense that the joint measurement gives you some certainty. So if you know that one of them is up, then you know with certainty what the other qubit will be when you measure it. It doesn't matter if it's- sorry, go ahead.

[00:22:37]  Bruce: Well, that's why it seems non-local then, right? Is because we just measured this one particle and light years away, we now know something about the other particle that wasn't known prior to that point.

[00:22:49]  Sam: Yes. Part of the reason why that isn't exactly true is, for example, that you can still perform operations on the other particle, and you know a conditional truth, which is that if, for example, so you take the state in which both particles are in superposition, in such a way that if you measure one particle in the upstate, then the other one is also in the upstate, and if you measure one particle in the downstate, then the other one is also in the downstate. Now you separate those two states, again, keep one on Earth, go and bring the other one to another galaxy for purposes of this thought experiment, and you perform a measurement on the qubit here on Earth, then you immediately know, you think you immediately know what's going on with the qubit in the other galaxy, but that's not exactly true because you know that only if nothing happened to that qubit, you're able to determine its state, but it could be that it rotated or that something happened to it that you can't know about because physical systems actually adhere to the Einstein's criterion.

[00:23:52]  Sam: So in other words, maybe some observer decided to perform an operation on its qubits, and you won't know about that operation because it turns out it has no physical effect on you.

[00:24:04]  Bruce: Right.

[00:24:05]  Sam: If you’re spacelike separated.

[00:24:07]  Bruce: Yeah. So that makes, that's a very good example. I've never actually heard it explained that well before. The reason why I guess they see it as non-local is because at least in the idealized circumstance, you had two particles in superposition of up and down, and one of them resolves, and in their minds, they're thinking that resolved the other one too. So it was light years away. It was in superposition, now it's not in superposition because you resolved one light years away.

[00:24:40]  Sam: Yeah, and that's all because of the Schrodinger picture. I think that in physical reality, the particles cannot exert any effect on one another when they're not near to one another, unless something propagates in between, that we're considering the case in which that nothing is propagating in between them. That's kind of what makes it so bizarre is that without anything physically propagating from one particle to the other, they still seem to have some effect on one another. And the resolution is basically to say, well, that's all just an attribute of the Schrodinger picture. And there's other ways of describing quantum theory. So in particular, there is the Heisenberg picture, and it has all of the properties of locality that we could desire from a physical theory. In other words, whereas in the Schrodinger picture, there's no way of describing subsystems independently of other systems.

[00:25:39]  Sam: So there's no way of describing the two-qubit system as the wave function of two separate qubits. You always have to consider the wave function of the combined system for it to give you full information about the state that the system is in. In the Heisenberg picture, you can have a complete description in which it's perfectly possible to talk about subsystems which evolve locally in the sense that if you do something to the one particle, it doesn't have any effect on the description of the other particle. And it's also a complete description in the sense that anything you could possibly want to compute is computable in the Heisenberg picture. So you have complete information about the state.

[00:26:25]  Bruce: So it can compute the same things as the as the Schrödinger picture, but it does so in a way that makes that is transparent about where the locality is.

[00:26:36]  Unidentified speaker: Is that correct?

[00:26:37]  Sam: Yes. And this is a discovery made by Deutsch and Hayden. And the way that they kind of summarize it is by saying, well, in a sense, the Schrodinger picture is like a highly condensed way of formulating quantum theory. It's like having a compressed file on your computer. And you can kind of see that you're losing out on information because of the compression. And this is a result that sometimes you can't describe systems as being independent from one another. For example, when they're entangled, you cannot write down their state as the state of two independent wave functions, but they have to be somehow the state of the combined wave function of both particles or both systems. But that's an artifact of the Schrodinger picture being a very condensed description, and the Heisenberg picture does not have this feature.

[00:27:33]  Bruce: So is this the motivation for your paper, then, is to make this clear? Are there other motivations for why you are writing this paper, and you're trying to do this in the Heisenberg picture?

[00:27:43]  Sam: Well, I think this is the main one. I mean, the way that I kind of view this is that the Heisenberg picture is the more fundamental description of quantum theory, that the Heisenberg picture has the property of locality because it is superior in some sense. Like, it should be stressed that both the Heisenberg picture and the Schrodinger picture are mathematically equivalent, but they're nonetheless very different descriptions, and I suspect that the Heisenberg picture is the fundamental description of quantum theory. Maybe, I mean, these are all unknown, but perhaps the Heisenberg picture, perhaps in a future theory of quantum mechanics, or a successive theory, rather, there will be no Schrodinger picture. So Dirac actually wrote a nice paper about this, kind of arguing that, well, maybe there's field theories which allow solutions in the Heisenberg picture, but which don't have solutions in the Schrodinger picture, so maybe there's an asymmetry between the two.

[00:28:42]  Sam: And that's all kind of the motivation for thinking that, well, maybe the Heisenberg picture is just the way to go in general, and we should leave behind the Schrodinger picture as just a useful tool for calculations. Because of that, and because of the fact that before the paper there wasn't really a good work on how to formulate Everett's construction in the Heisenberg picture—sorry, in the Heisenberg picture, that it seems useful to us to provide that, to give a description of of Everett's construction, which I think is the solution to the measurement problem in the Heisenberg picture, which is the solution to the problem of locality in a sense. I guess it's not usually called that, but I guess we could call it that for the podcast, the problem of locality that Einstein noticed in his EPR paper.

[00:29:31]  Bruce: Okay, why did, two questions here, why did Everett choose to formulate this in the Schrodinger picture instead of the Heisenberg picture? And is there any advantages to the Schrodinger picture? Like, is it like easier to compute in some cases or something like that?

[00:29:49]  Sam: Yes, well, the answer, the first question is rather simple. It's just that it's the most widely used description of quantum theory, and the reason for that is that the Schrodinger picture is just, as you put it, easier to compute with. Yeah, so the Heisenberg picture isn't the picture that people traditionally use. It's invented around the same time as Schrodinger, but I think that researchers very quickly adapted the Schrodinger picture because it's just so much easier to calculate with, and also at that time it wasn't known that there was this benefit with the Heisenberg picture, because those issues with locality were still being kind of uncovered.

[00:30:38]  Bruce: I see.

[00:30:38]  Sam: And I think, well, it's only been since Deutsch and Hayden, which is in the late 90s, that we have discovered these attributes of the Heisenberg picture that I think make it preferable over the Schrodinger picture.

[00:30:56]  Bruce: So as you have Everett come up with Many Worlds to try to explain the measurement problem. It's my understanding that initially he was largely just dismissed as crackpot science, with the exception of his advisor. Who was his advisor again?

[00:31:15]  Sam: So his advisor, his supervisor was Wheeler, and Wheeler had a very strange reaction to Everett's thesis. I think initially he was rather optimistic about it, and he felt like Everett's solution to the measurement problem could be used to gain a deeper insight into quantum gravity, because there could be such a thing as, say, the wave function for the universe. But because Wheeler was educated by Bohr, I think in turn that Wheeler's supervisor was Bohr, whom he greatly admired. And Bohr was very unfavorable to the Many Worlds interpretation. And consequently, I think that Wheeler changed his mind and was of the opinion that the Many Worlds interpretation was the wrong way of viewing quantum theory. There's a nice interview with Wheeler on YouTube in which he summarizes Everett's interpretation as you couldn't get quantum theory more wrong.

[00:32:19]  Sam: So that was his opinion about the theory for most of his life, I think, after it had been presented to Bohr. And it went to the extent where Wheeler really found it necessary to edit, to severely edit Everett's thesis. So they sat together over the course of several evenings after Everett had presented his work to other researchers. And they edited out many of the words and much of the core of the theory that Everett presented in his, what's now known as his longer thesis, which is available online for free. And that kind of shows that he was a bit hostile towards the whole idea of Many Worlds.

[00:33:06]  Bruce: Is there a connection between Wheeler and David Deutsch?

[00:33:11]  Sam: Yes, so Wheeler also supervised David, Dave Deutsch. Yes, there's an interesting history where Deutsch was supervised by Wheeler at the time when Everett's work was kind of being rediscovered by other researchers like Bryce DeWitt. There's actually a nice documentary in which they talk about this. I think it's called Parallel Lives. And it's about the son of Everett, who is now a famous rock star of the Eels, the band the Eels. And they kind of follow him. They follow this, the son of Everett. Well, he discovers about his dad's work and about kind of notoriety that he has acquired in the physics community. And there's a nice segment where David talks about meeting Everett and how that meeting came about. I think it was due to Bryce DeWitt and Wheeler having invited him back for a conference and a series of talks on the topics of Many Worlds.

[00:34:39]  Bruce: Interesting. And David Deutsch, I guess I perceive him as being kind of the premier bearer of the Everett Torch today. I know there's some other famous names like Max Tegmark, who believes in Many Worlds also. But there seems to be a bit of a difference. Max Tegmark is just interested in interesting, weird ideas, and he latches on to a lot of them, and he's very open-minded, which is good. I don't mean that in a negative way. Where Deutsch seems to have come to it based on epistemology and the realization that this is the only actual explanation of quantum physics that's available to us, and then was compelled to believe it based on that. Is that correct?

[00:35:26]  Sam: Well, I don't know about how Tegmark came to discover Many Worlds. I think he's pretty sound on the topic. And yeah, he seems sound enough, but I'm not sure. I imagine that if you're interested in Many Worlds, you have to be a realist. You have to think that quantum theory really describes the physical world. It's a theory that isn't just a parochial description of very small things, but it applies to all physical systems in some sense. And those are attributes of Everettians in general. I think those are the kinds of things that lead people to being an Everettian in general. So I imagine that Tegmark is much the same, that he was interested in similar issues of just describing the real world. But yeah, I'm not sure how he came to the theory.

[00:36:25]  Bruce: I've read some of his books, and I think you're correct that he buys into it quite strongly. But he also has advanced other types of many worlds theories that are a lot more, there's no evidence for. And he'll tell you that. He's not like trying to hide that, but he's the one who came up with like the four different kinds of multiverses and things like that. Are you familiar with those?

[00:36:52]  Sam: Right, yes. I'm only somewhat familiar with those ideas. It's like you have multiverses that are really bubbles in space, right? They're, yes.

[00:37:03]  Bruce: Like type one and two, right?

[00:37:05]  Sam: Yes. And yeah, I'm not too familiar with the topic. I basically just about know what the different universes are according to Tegmark, or what the different types of multiverses are. But yeah, other than that, I'm not very familiar.

[00:37:22]  Bruce: The ultimate multiverse in his mind, which is clearly not something there's any evidence for, would be that reality is mathematical. And so any mathematical system that's consistent exists. So he has this kind of concept of a mathematical multiverse, which clearly is very speculative, very different than the quantum multiverse, where it's a requirement to understand the explanation of quantum physics. He's definitely more willing to dip into really speculative things.

[00:37:56]  Sam: Right. Yes. Well, that's definitely different from Everett's construction. And I think it's solving a different problem. It sounds like that's trying to solve the issue of why does mathematics describe the real world so well?

[00:38:11]  Bruce: Yep. That's the problem he's trying to solve. He mentions that.

[00:38:14]  Unidentified speaker: Yeah.

[00:38:15]  Sam: So that's a different issue. It's a related one in some sense because it's about why the universe is Turing complete, which has something to do with quantum computation. Well, I say the universe is Turing complete, but what I mean with that is a rather inexact statement. What I mean is that the universe allows for physical systems to be simulated with arbitrary accuracy with finite means. So if you have a computer running on finite physical means, then it can simulate any other physical system. But that's a related thing in the sense that we have a theory of computation, which is the quantum theory of computation, and that kind of gets you back to quantum theory. But the motivations for Tegmark's multiverse seem to be very different in that Everett was solely concerned with an issue in quantum theory.

[00:39:17]  Bruce: So maybe let's describe what the, in layman's terms, without using math, and I know with quantum physics, it's really hard to describe things without the math. But let's describe what the Heisenberg picture looks like and what the multiverse looks like and what does a local version of the multiverse look like? And what is the Heisenberg picture telling us about it?

[00:39:40]  Sam: There's perhaps there's this thing that we should discuss before we get into this, which is entanglement, which we already mentioned.

[00:39:46]  Bruce: Yes.

[00:39:47]  Sam: And it's kind of at the heart of what quantum theory is about. So there's this property of quantum systems known as entanglement where, as we said, if you, say, have two particles or two qubits, I think qubits are probably the easier example. If you have two qubits, then each qubit individually can be in a superposition, meaning that they can have value, say, 1 or minus 1. But they can also be in both of those states simultaneously. They can be in the state 1 and minus 1 simultaneously.

[00:40:30]  Bruce: Just to clarify, a qubit is a generalization of the concept of a bit, just like with a computer. So this would be a quantum bit.

[00:40:38]  Sam: Yes. It's a thing that has two definite states. It can be in a state 1 or minus 1, or as it turns out, in the superposition of 1 and minus 1.

[00:40:51]  Bruce: This is something that always bothered me a bit about the way people describe quantum computation is they'll say it like the media just butchers science a lot I found. But they'll often say that you can do more with a quantum computer because you can have this bit as a 1 or as, they usually use 1 and 0 because of bits instead of 1 and negative 1, but just a matter of how you're interpreting them. So it can be 1 or 0 or it can be both and therefore it's more powerful. And the first time I ever read that, I thought that would be so easy to emulate on a classical computer. So that can't possibly be by itself the full explanation. So entanglement’s the real answer. Entanglement is more than that.

[00:41:35]  Sam: Yeah, I think that's a nice point. It's because there's loads of superpositions that exist in nature. The superposition is this idea that you can have that if you have a bunch of solutions for a set of equations, then a sum of those solutions is also a solution. And perhaps the simplest case of that is just if you have light being a superposition, you can imagine that you have a ray of light pointing to the west and a ray of light pointing to the north. And those are separately allowed rays of light. And also if you add them together, so if there are two light rays traveling north and west or whatever directions I chose just now, then those are also allowed. So you can have rays of light traveling in different directions at the same time. And that's because light is a kind of a wave in the electromagnetic field and light rays. Each individual light ray is a solution to to Maxwell's equations, and Maxwell's equations allow you to superpose rays of light.

[00:42:40]  Sam: So rays of light traveling in different directions are allowed, provided that each light ray individually is an allowed solution. And in the same way, quantum theory says that, well, if you have two solutions, then the sum of their solutions is also a solution to the Schrodinger equation. I don't know if that clarifies it. I feel like I just dropped a whole lot of terminology on you.

[00:43:07]  Bruce: I think I can actually explain this. In quantum computation, it's true that you have the bits in a superposition, but they also have a relationship to each other. And if you were to try to emulate that on a classical computer, that relationship that they have to each other, the amount of memory you need grows exponentially. And even after, like, I started actually programming a quantum computation simulator on a computer, I couldn't even get to nine qubits. Before, the amount of memory I needed to be able to track all the relationships that exist between each of the qubits was more than anything I was going to be able to afford. And in fact, I doubt, I mean, like, the reason why they, you couldn't even get to, like, 30 qubits. You would have to use actual quantum computer to be able to get that far. I know some of the quantum computers that exist use, like, 30 qubits, or equivalent to 30 qubits.

[00:44:07]  Bruce: That's probably well beyond, like, to try to emulate that on a computer. I would, I don't have the math worked out in my head, but I suspect it would require a computer the size of the universe or something, to be able to track 30 qubits. So it very quickly becomes completely intractable for a classical computer to be able to even just keep track of all the entanglements that are going on. And at least that was how I came to understand, oh, now I can see why a quantum computer would be more powerful than a classical computer, is because it's actually capable of computing things that would be physically impossible for a classical computer without a universe-sized computer.

[00:44:55]  Sam: Yes, that's definitely part of it. It's the possible states that quantum systems can be in, is just so much larger than physical, than classical systems, that one of them scales exponentially and the other one doesn't, meaning that the quantum systems scale exponentially in the state space.

[00:45:23]  Bruce: Yeah. And that's also why it can sometimes do these, a quantum computer can sometimes do these exponentially faster computations, is because it can actually compute across all the qubits, which would be like doing a computation that is exponentially large, but in just a few steps.

[00:45:47]  Sam: Yeah, so that's the argument that Deutsch usually gives for why quantum computation is like a reason to believe in the multiverse. It's because you are performing these computations somewhere in physical reality, but they're not in our universe. They're happening somewhere else, and they're happening simultaneously in different universes. That isn't the whole answer, because also you need to be able to kind of obtain the right result from those parallel computations happening in different universes. That is where quantum interference comes in, and you have to use clever methods of selecting the right results.

[00:46:28]  Bruce: It's almost like you're tricking the system. You set up the system so that you're doing this quantum experiment. You want to do a computation, and so you set up almost like a quantum experiment, and you say, okay, if I get this result, that's going to tell me that maybe I'm doing a search problem, that my answer is in the top half of the search, and then you can continue to do that, and it gives you a piece of information that in physical reality, in classical physical reality, shouldn't be available to you. That allows you to run at a much faster, much faster in one step than a classical Turing machine would be able to.

[00:47:10]  Sam: Yes, but I feel like we, so we haven't really addressed yet what entanglement is. I think we went down an interesting path.

[00:47:17]  Bruce: Yeah.

[00:47:17]  Sam: Yes. So to kind of come back to the example of the qubits. So first of all, there's this notion of superposition, which is that something can be in two states at the same time. So the qubit can be in state one and zero at the same time. And then there's an interesting feature where the qubits can also be in these indefinite states at the same time in such a way that they are what we call entangled. So you have, say, a particle like we described earlier, or a qubit where if you measure the qubit to be in the up position or in the up state, then you know the other qubit is also in the up state. And if it's in the down state, then the other qubit is also in the down state. Even though separate measurements on the qubits, or individual measurements on them, are uncertain. So an individual measurement on the qubit will not tell you with certainty whether you're going to see up or down.

[00:48:28]  Sam: But you know that if you're going to see up on one of the qubits, then you're also going to see up on the other qubit. So a joint measurement on them gives you some certainty about the results, about the joint results. But the individual measurements are uncertain. And that's what we call entanglements.

[00:48:53]  Bruce: Yeah. So, and my original question was how do we relate this to locality? So maybe let's try to work that in with locality now, because entanglement certainly feels non-local.

[00:49:08]  Sam: So usually in the Schrodinger picture, it seems like when things become entangled, they can't be described separately from one another, because your description of the different systems doesn't allow you to describe each system individually without also losing information about what the state of the system is in. And that seems to suggest that if you perform these measurements on the qubits individually, you're also affecting the other qubit, because again, you can't really separate them from one another if they're entangled. And that in turn suggests that if in Everett's picture, some branching event happens where the universe suddenly has two histories, then that must affect the entirety of the universe, because there is no way of really describing what's happening in the universe separately from what's happening to everything else in the universe.

[00:50:09]  Sam: So if one system branches, then everything branches, almost necessarily, but as it turns out, that is a feature of the Schrodinger picture. That is not a feature of physical reality. It's a feature of a way of describing physical reality, and it's an attribute of the Schrodinger picture that makes it less fundamental than the Heisenberg picture. And if you describe things in the Heisenberg picture, then it turns out that you have these local branching events, where, for example, you imagine you're in a lab, you perform a measurement on a qubit. Well, when you perform a measurement on a qubit, then you branch into two versions of yourself, one of whom sees one particular measurement outcome, and the other one sees another measurement outcome. And that is actually a completely isolated event, in the sense that it's just you and the particle or the qubit that branched.

[00:51:12]  Sam: And before the environment starts to obtain information about the branching event that happened, it is not yet in any one of those universes. So in some sense, there's this bubble that contains you and the qubit, where within that bubble you can talk about two different histories resulting from the measurement that occurred. But outside of the bubble, there's not yet any information about which branch you are in. So that kind of spreads out to the environment. So maybe you have this bet with a friend where if you measure a qubit in the upstate, then you will buy your friend a beer. And in the other state, if you measure the other state, then he will buy you a beer. Okay, well, that means that you measure the particle and it's an up. You go to the bar and buy a beer. Well, that's a very different history from you and your friend going to the bar and him buying you a beer.

[00:52:15]  Sam: And systems behave very differently in those two universes, but it has to kind of... So that's one way in which the measurement can have effects on the environment. And now by knowing who is buying the beer, you know what the outcome of the measurement was. And so information about the measurement outcome has leaked into the environment. And many things in these different universes will be different as a result of the measurements. And you can kind of imagine a sphere of influence that's growing over time, where the measurement result has had some effect on the environment. And this is the view of the many worlds in which interactions happen locally. So it's not that the whole universe splits all at once. It's that systems kind of split or they branch. I think branching is a term that I prefer. They branch only because they have interacted with something that has interacted with the measurement result.

[00:53:22]  Sam: So again, you holding the beer rather than your friend is a sign of what the measurement result was. And that means that the bar that you're at has also split in as part of this history that you are in. So that is a local picture of what's going on in the Many Worlds.

[00:53:47]  Bruce: So relating it back to the Alice and Bob example, it solves the measurement problem because first you have the instrument that gets entangled with the particle. So you've got one instrument in one universe that measures up and one that measures down. Then Alice looks at it and she becomes entangled into it. And so now her brain contains information about it either being an up particle or down particle. Then Bob enters the room and she tells him what it was. And so now that he becomes entangled in into this, then they decide that they're going to get the beer and because it was up, she's buying. So they go out to the bar. Now that bar becomes entangled into it and it just continues to kind of move out from there. Is that correct?

[00:54:33]  Sam: Yes, that's how it works. And that's how you can kind of imagine there being a sphere of influence where the effects that the measurement can have had on its environment have to grow with time. So if you're very far away from where the measurement happened, then the information about what the measurement outcome was has to travel some distance before it can reach you and before you can become part of that history. And until then, you cannot be in any one of these histories yet.

[00:55:08]  Bruce: So I've got a weird question for you that I've always wondered about. In Beginning of Infinity, David Deutsch talks about how an article in a newspaper could be part fiction, have a fictional history and a true history at the same time. And I never really understood what he was getting at. But I later read a book by Frank Tipler where he explained it a little differently. And he was saying that once universes, there is nothing different between them, then you can think of them as the same universe again. And it could in theory be that that history of say, you know, the Pharaoh or something, that there was a fictional version that created exactly the same state as the real version and that those would both have some feed into our history. Is that correct? That always seems so weird to me.

[00:56:05]  Sam: Oh, I'm not sure. I guess what's being said there is that certain certain fictional, well, certain works of fiction have counterparts in the multiverse that are very close to to that fiction. So maybe there's a story about two people falling in love or I don't know what's what's a good fiction book. Maybe it can't be too outrageous.

[00:56:32]  Bruce: It still has to follow physical reality, right?

[00:56:34]  Sam: Yes. Yeah, it still has to obey our laws of physics. But maybe somewhere in the multiverse, something like Breaking Bad happened. And like not just a little bit like Breaking Bad, very much like Breaking Bad. Like maybe there really was a guy called Walter White who lived in Albuquerque, who was dealing drugs as a high school teacher and who almost won a Nobel Prize in chemistry. And that will be true somewhere in the multiverse. Yeah. But here in our part of the multiverse, it's a fictional story because it's the television show.

[00:57:07]  Bruce: It's related to that, but it takes it just a little bit further where it says, okay, let's see, let's use King Arthur as an example. We have these stories of King Arthur and there's this debate over whether King Arthur actually existed or not. So we can imagine one set of universes where King Arthur actually existed and he was the inspiration for the stories that we tell today. But everything else about him has been lost. There's no history of him other than these stories. Then you have another universe where you have the exact same stories happen to come into being, but there was never a King Arthur. He was a totally fictional character. Both these universes in their current state today are identical. They both produced the exact same stories about King Arthur. So you have this history, one's fictional and one's not, and both would have fed into the universe that we see today.

[00:58:05]  Bruce: And so there would be an open question. The question, did King Arthur exist, may not even be a valid question in that universe. That was really what Frank Tipler was getting at, that you could actually, maybe that one's fed far more into it—into the other [referent unclear]. Perhaps the one where King Arthur did exist is a much more realistic, more probable circumstance. So you end up with that one feeding into the present much stronger than the fictional version. But he was suggesting that Many Worlds suggested that you would actually have both as part of your history.

[00:58:43]  Sam: Yeah, that seems, I mean, that's true in principle, but unlikely in practice, because the ways in which those universes differ from one another is not just about whether King Arthur really existed in those universes. So, for example, if King Arthur really existed, then there will be certain historical artifacts that are, that will be here today, and certain people's lives will have been different in relevant ways. And it is true that if all of that information was lost, then perhaps we could re-merge with that universe in which King Arthur was alive.

[00:59:30]  Bruce: The way you're describing this actually is exactly what Tipler was getting at, that they would have to be identical. It would have to be that there's no other differences between these universes except... There are no differences between these universes. They both have exactly the same stories. For whatever reason, they happen to get into the exact same state, but one King Arthur was real, one he wasn't. And that's our universe today. We don't know if King Arthur is real or not today. There's people who say he was, people that say he wasn't. They're trying to look at the, quote, evidence to determine if he's real or not. But the idea would be that, in theory, because we don't have anything that really shows for certain that he did or didn't exist, that it could be that both those histories are valid histories of our universe today. I don't know how realistic that is.

[01:00:28]  Bruce: Obviously, they're trying to give a kind of out-there example of how you could have a history that's both fictional and isn't, and it would require some really very special things to happen where all the information got completely deleted, so there was no other differences.

[01:00:47]  Sam: Yes, and also the stories are completely the same.

[01:00:49]  Bruce: The stories happen to be completely the same, correct.

[01:00:52]  Sam: Yes, so those things are very unlikely, because even if there's some historical evidence that we have yet to uncover that could point to there having been King Arthur, that is a distinction between the universe where King Arthur was a real historical figure and the one in which he wasn't.

[01:01:11]  Bruce: That's right.

[01:01:12]  Sam: And there's just stories about him.

[01:01:13]  Bruce: That's right. So even if this idea were true, it seems like it would be some little infinitesimal part of the multiverse for one of them, and the other one would be a much stronger feed into the history.

[01:01:27]  Sam: Yes, but the principle is true. The principle that universes rejoin or can rejoin because they lose some information is true. So, for example, if you have a particle that goes along two paths because it's hit a beam splitter and then you make it re-merge again, then the particle at the end of the experiment, which is in a definite sharp state, there's not really an answer to the question, which trajectory did it take, because it took both at the same time and then it re-merged. So the information about which trajectory it took is completely lost to everyone. All that we know is that it took, well, if we know enough, then we know that it took both at the same time. But there's no way of saying, well, those instances of the particle took the one path and the others took the other path. Because those instances that have taken different paths have become indistinguishable from one another, again, upon re-merging.

[01:02:31]  Bruce: That makes sense. So they're taking that idea to an extreme when they suggest something like this to show how far in theory it could happen. But like in practice, it doesn't seem like even slightly probable.

[01:02:46]  Sam: Yes.

[01:02:48]  Bruce: Makes sense. Could you maybe just briefly, if possible, describe what decoherence is? Your paper talks about a couple of things that I don't know, I have kind of a vague notion of. Decoherence, which is, you know, why is it, it is somehow related to why is it that we can't see both universes at the same time? And then you also talk about quasi-classical systems. Is that related to the same question?

[01:03:20]  Sam: So, okay, if you're a many-worldser, and you think that everything is described by quantum theory, then a very natural question to ask is, well, why do things even seem classical in that case? Like, I think I live in a classical universe where things have definite values and chairs and tables and other things are not in superpositions as far as I can tell. And they behave roughly as though Newton's equations are true. So why is that? Why is it the case that this quantum world somehow behaves classically at all? And it turns out that when quantum systems evolve generically, then due to the interactions between all of the quantum systems, they start to behave somewhat classically. So if you have some environment that's constantly interacting with, I'm just going to take a kind of random object, imagine you have a chair, you have a quantum chair, and the quantum chair is in the superposition of being in all kinds of different locations.

[01:04:28]  Sam: Well, okay, that chair isn't isolated. It's in a room and it's interacting with the light in that room, interacting with several trillions of photons that are hitting it at every point in time, or every couple of seconds, I think, I should say. And so that chair is providing information to the photons about what position it's at. The photons are getting entangled with the chair, and it's not just the photons, it's everything else in the room that's getting entangled with it. So the photons are hitting off the chair, they're then bouncing off the room or the door or the walls in the room, and then the photon hits your eye, and you see where the chair is. And in that process, the photons, the chair, the wall and you have all become entangled with one another, and that's happening very, very often. So it seems like everything has a definite position, because everything is constantly interacting with almost everything else, at least within the near vicinity.

[01:05:39]  Sam: So everything in the room is interacting with everything else in the room. And the result of that is that things behave roughly classically, so that kind of suppresses the typical quantum effect. So the chair can't interfere with the other versions of itself, because of all these different interactions that are occurring. And that's roughly why the chair is like a classical system. It's not because it has only a single history, it's that the different histories of the chair can't interfere with one another, because of all this noise, because of all the different interactions that the chair is partaking in.

[01:06:16]  Bruce: And that's what decoherence is?

[01:06:21]  Sam: Yes, so decoherence is what occurs because of all these different interactions with the environment. So you could say like the chair is some physical system, it's interacting with an environment in this case, say the air molecules in the room, the photons that are hitting it, and basically anything that interacts with the chair at all. And these are decohering the chair, so the chair roughly has a single position, and it will, or rather, every chair in the multiverse, the chair has multiple positions, but these different instances of the chair will never interfere with one another because of the interactions that are taking place with all the other systems.

[01:07:05]  Bruce: And what is a quasi-classical system then? Does that just mean that effectively we can treat it as if it's classical, or is quasi-classical something that's actually more in between?

[01:07:21]  Sam: Well, so the reason why we use the term quasi-classical is because the chair is never really a classical system. I see. For example, if you were to say that the chair is a classical system, it would only have one history. And many things about the chair would be false. Many attributes that the chair has in reality, it could not have if it was a classical system. Like the very fact that it is stable is an attribute of it being a quantum system, because only quantum theory allows us to explain why atoms are stable. And so these attributes of the chair are attributes that it couldn't have if it was a classical system. And yet it behaves roughly as if it's a classical system, in the sense that chairs appear to only have a single definite position, somewhere in space, and they don't interfere with other instances of itself. They're always, unless we interact with them, wherever we leave them.

[01:08:31]  Bruce: Interesting.

[01:08:32]  Sam: So it's a way of saying, things seem to behave as if they, as we thought they would before discovering quantum theory, and we have to explain the behavior of particles, of chairs, of everyday objects, as a result of quantum theory, but those things never really are classical systems.

[01:08:55]  Bruce: Okay. So that's what a quasi-classical system is. It's the explanation for why things seem classical to us, but they never really are classical.

[01:09:03]  Sam: Yes.

[01:09:04]  Bruce: Okay.

[01:09:05]  Sam: So my favorite example of this is that, say, refrigerator magnets are, they can’t be classical systems. There’s this nice proof by Bohr, where he showed that magnets, as in materials with magnetic properties, ferromagnets, cannot exist according to classical physics. In classical physics, any material that has some magnetization can only have that magnetization because it's in a magnetic field, but it can't have any intrinsic magnetization. And we actually, quantum theory is part of the explanation for why things like magnets, refrigerator magnets, ferromagnets, can exist. And in that sense, those magnets are not classical objects. They're quantum objects, just like everything else in the world is a quantum object. But they behave roughly according to our classical intuitions. So we never see these magnets interfere with other instances of the magnet in daily life.

[01:10:12]  Bruce: Right. Interesting. I think you've given us all a lot to think about, Sam. This is a very interesting paper. I hope that this overview will kind of help people either gain interest in it or understand it a little bit better. And we will continue to do some YouTube videos. I'll put those up that will also help explain the mathematical side of it. So, Sam, thank you very much for coming on the show. I really appreciate you doing that. I think this has just been fascinating. Cool.

[01:10:48]  Sam: Thanks so much for having me. It was great fun chatting with you.

[01:10:51]  Bruce: Yes. Thank you. And Tracy, thank you for joining us also.

[01:10:55]  Tracy: Thank you both. That’s great.

[01:10:58]  Unidentified speaker: Glad you liked it.

[01:11:00]  Sam: Thanks so much for having me again and talk to you soon.

[01:11:03]  Bruce: Thank you. The Theory of Anything podcast could use your help. We have a small but loyal audience, and we'd like to get the word out about the podcast to others so others can enjoy it as well. To the best of our knowledge, we're the only podcast that covers all four strands of David Deutsch's philosophy as well as other interesting subjects. If you're enjoying this podcast, please give us a five-star rating on Apple Podcasts. This can usually be done right inside your podcast player. Or you can Google The Theory of Anything Podcast Apple or something like that. Some players have their own rating system and giving us a five-star rating on any rating system would be helpful. If you enjoy a particular episode, please consider tweeting about us or linking to us on Facebook or other social media to help get the word out. If you are interested in financially supporting the podcast, we have two ways to do that.

[01:11:53]  Bruce: The first is via our podcast host site, Anchor. Just go to anchor.fm slash four dash strands, F-O-U-R dash S-T-R-A-N-D-S. There's a support button available that allows you to do reoccurring donations. If you want to make a one-time donation, go to our blog, which is four strands dot org. There is a donation button there that uses Paypal. Thank you.