236 | Thomas Hertog on Quantum Cosmology and Hawking’s Final Theory

0:00:00.1 Sean Carroll: Hello everyone. Welcome to the Mindscape Podcast. I'm your host, Sean Carroll. Stephen Hawking is known for any number of revolutionary advances in theoretical physics, the singularity theorems that he did with Roger Penrose and others in the late '60s, the evaporation and radiation from black holes in the mid '70s and in the early '80s with Jim Hardell, he calculated the wave function of the universe to try to explain the creation of the universe from nothing. But in 1988, Hawking revolutionized not theoretical physics, but the scientific publishing industry with the appearance of A Brief History of Time. His surprise, runaway bestseller.

0:00:38.6 SC: I was a little bit too young to take advantage of this, but I'm told that in the late '80s, after A Brief History of Time came out, if you were a theoretical physicist with a book to write, you could get a million dollar advance. No problem. Not like that anymore, but those were the days. Andre Linde is a well-known cosmologist whose name will appear again in this episode. He'll... Also a mischievous guy. He likes to tell the story. Back in the late '80s, he would be riding an airplane sitting next to someone who was reading A Brief History of Time. And Linde would inevitably say, I like the book, but I didn't really understand it. And the person reading it would go, oh, yeah, it's really not that hard. You just have to really concentrate while you're reading it.

0:01:19.2 SC: But Hawking never gave up doing science. He wrote more books, but he also wrote a lot of technical papers in the published research literature. And his views continued to evolve about how to do quantum cosmology, how to think about the nature of the quantum universe. Today's guest, Thomas Hertog, was one of Hawking's most frequent collaborators in those years. He was a PhD student with Hawking, and then continued to write papers with him, and has now come out with his own book called On the Origin of Time, Stephen Hawking's Final Theory. And in fact it's a joint theory that he's described between himself and Stephen. So we'll talk about that theory, but we'll talk about the genesis, the evolution of what we mean by quantum cosmology.

0:02:07.8 SC: How we go about saying, okay, you have the whole universe. We're gonna apply the rules of quantum mechanics to this universe. And I think you will correctly get the impression that there's a lot that we know about how to do that, and a lot that we don't know. So our views on how best to do it are continually evolving, and it brings in both philosophical ideas about the role of the observer in defining what you mean by a universe and calculating the probability of the universe looking different ways, but also very modern cutting edge physics ideas like holography and the emergence of time from the quantum wave function. So, as I apologize, to Thomas in the middle of the podcast, this is a tough one for me, not because I don't understand it, but because I'm too close to the issues here.

0:03:01.1 SC: I think about, about these issues all the time. And so it's harder for me to put myself in the seat of the audience member who is not a super expert. I hope that I didn't interject my own views or interpretations too much here. I tried to reel myself in, but I don't think I was very successful. I think that you'll find my own views all over the place. So hopefully Thomas's views, shine through because he has a different point of view. That is a very interesting message. I think it's worth taking very seriously, especially because we don't know the final answers. We're still working on this, we're still moving forward. So let's go.

[music]

0:03:47.0 SC: Thomas Hertog, welcome to The Mindscape Podcast.

0:03:53.0 Thomas Hertog: Hey, hi, Sean.

0:03:55.1 SC: Normally, and I'm sure will happen in this episode also, here at The Mindscape podcast, we focus like a laser beam on the substantive intellectual content. And we don't dig that much into the personal fun stories of people's histories and so forth. But in your case, you were Stephen Hawking's most frequent collaborator in the last years of his life, and that collaboration forms a lot of the basis of what you're gonna tell us about in the podcast and in the book that you've written, how does one become Stephen Hawking's collaborator? I'm sure that there's a story there.

0:04:31.2 TH: Yeah, but it's a typical science story. There was a folklore, there was sort of a lore at a well known sort of story at the Department of Applied Mathematics and Theoretical Physics in Cambridge, which was, whoever got top scores, in their famous part three course would get an invitation to go talk to Steven. And so that's essentially what happened and what happened to many other students in different years, so that's how I first entered into his office. The real surprise, of course, was the experience of that first conversation, which was anything but normal. It was not normal because it was interspersed with various, journalists walking in and out. And the second thing, which I thought was very exceptional was that Steven went just straight in and started talking about, how he found that whole idea of the multiverse so paradoxical, and how his colleague, Andre Linde, had these outrageous theories. And so there I was, how could I possibly have an opinion on the multiverse and Andre Linde, as a 22-year-old student? But that was fun. That was really fun.

0:05:54.1 SC: And again, we're not going to spend most of time talking about this stuff but, how did it work, your collaboration? Again, later in life, Stephen had a tougher and tougher time, banging out the sentences, right?

0:06:06.4 TH: Right. Yes, I think I was lucky in a sense for two reasons, the timing, late '90s, this was Stephen and I met '98 really. Yeah, I think it was really a coincidence why it worked so well. First of all on your point in terms of communication so Stephen was already using his computer voice at the time, but the whole system worked really well. He had his sort of... He was used to using a mouse to select words and he sort of... My impression was that by then he sort of instinctively knew when to click to select certain words. And so the whole system was working very well in the late '90s. And of course I had a notion of time. So we would sit hours and hours in that department, shoulder to shoulder, and he would type out sentence and sentence and by by by okay, if you spend so much time by at some point you begin to understand what he's talking about and you get going. So that was important because those years really were foundation for when it became very difficult later on to communicate. I think at that point, these first few years we developed some sort of intuition.

0:07:29.9 SC: Good.

0:07:30.3 TH: Common language. The second point I think, which was equally relevant is that the late '90s were a great time in cosmology. Stephen's famous book, A Brief History of Time had been out for like a decade. So the frenzy around that book had sort of died down. He was back to research and he was back to research because cosmology was... It was a golden era, right? You had these mystifying observations about the acceleration of the universe, the CMB fluctuations, which were pointing to an early phase of acceleration, which we now call inflation. And then you had these paradoxes to do with the multiverse, which were essentially going to the core of cosmological theory. So this was a good time. Stephen was grounded in research again, and still being able to communicate. And that's what we built on, I would say.

0:08:36.7 SC: And in particular, the research that you did together, I think it's fair to say as always, correct me if I'm wrong here is sort of downstream from the wave function of the universe work that he did with Jim Hartle in the early 1980s. So quantum cosmology in some sense. So why don't you explain to us what that is? What's about that? What is the wave function of the universe? Thomas.

[laughter]

0:09:05.8 TH: Okay, good. Well, so in a way, the whole wave function thinking, the whole sort of idea of let's think about the universe in a quantum mechanical way as a quantum system must have been sort of the moral lesson that Stephen took out of his PhD work, his own PhD work in the 1960s when he essentially showed using Penrose's techniques that the Big Bang classically the origin of the universe, the Big Bang in Einstein's theory is a singularity where Einstein's theory breaks down, it's the origin of time, is this gonna be... And Einstein's theory, if you would take it at face value, you'd almost be driven to the statement, okay, this is not science, this lies outside science. But of course, there's another lesson. The one Stephen and most of our colleagues took, well, wait a minute. It just quantum, the quantum nature of gravity becomes important.

0:10:08.6 TH: But then how do you go about doing something about that? That's when... That's I think when Jim and Stephen pioneering work came about, well, if the universe is a quantum system, then it must have a quantum state somehow a very abstract, super abstract description of reality. And the ingenuity of Stephen's work, which featured so much in a brief history time, was that he came up with the first fairly explicit model of how you would go about giving a quantum description of the Big Bang of the creation of the universe. And their trick was really to sort of, in a way bent the time dimension of Einstein's theory into a space dimension. And if your reality is pure space dimensions, you know what to do to close it, you can just round it off like a sphere. Hence and so Stephen's famous line, of course was what is the Big Bang? It's a bit like the South Pole and what was there before. While it's like asking what's south of the South Pole, it's a meaningless question.

0:11:22.1 TH: So that was of course the typical oracular Hawking-nean kind of phase, right? But by the late '90s, Stephen and many others had realized that the creation theory, so to speak, of Brief History of Time had a fundamental problem, which is that taken at face value, you'll be led to the conclusion that the universe should be empty. That the universe should be, yeah, that there should be no stars, no galaxies, no life. And so while their original theory was beautiful in a way from a theoretical perspective, it's almost like you and I think he felt like that, that he sort of had cracked the enigma of creation, so to speak, by giving a mathematical description of how you can make a universe. It was very much, it was not the kind of universe we inhabit.

0:12:23.0 SC: Yeah.

0:12:23.1 TH: So there was something missing.

0:12:24.3 SC: Right. I actually, I do wanna... That's gonna be a heart and soul I think of this conversation 'cause it's a really, what your book builds up to. But I wanna linger in the '80s for a little while to get the setup so that everyone comes in on the same page here. So when we say the quantum state of a system, if it's an electron or something like that, something that we are very used to treating as a quantum mechanical object, it's a wave function for every position that we could measure it in. It tells us the probability, etcetera. So it's a function of every possible location we could measure it. What do you mean when you say the wave function of the universe? Is it supposed to be, it sounds hard to write down a possible quantum amplitude for every particle in the universe.

0:13:12.9 TH: Right? And it is worse than that if you treat... I mean, what we really mean and certainly what Steven meant in the '80s by a wave function of the universe is very much a wave function, not just of a particle describing various positions of a particle like an electron, but really a sort of abstract description that describes a superposition of various possible universes, including all the matter and the space and time. So it's almost like you go from one universe to a zoo of possible universes. And so you really go up a level in abstraction and a level, [chuckle] and in confusion, right?

0:14:00.6 SC: Yeah.

0:14:01.4 TH: And frankly, I think the question what we mean by a wave function may well be at the heart of these more recent developments with Steven and what we worked on.

0:14:16.1 SC: Good.

0:14:17.3 TH: Because of course, if the wave function predicts an empty universe, if the empty universe is the, by far the dominant wave crest, so to speak, yeah, then you know something's missing, right there.

[laughter]

0:14:30.3 SC: Yeah. Good but...

0:14:31.8 TH: This was Linde's complaint, right? Steven would be saying, "Yeah, with your multiverse, you have infinitely many observers and you don't know where we are." And then Linde would say, "By your wave function has no observers...

[laughter]

0:14:46.6 TH: That's equally bad."

0:14:48.6 SC: Maybe that is even worse, honestly.

[laughter]

0:14:50.4 TH: Yes, yes, I can... I didn't want to say it.

0:14:54.5 SC: But okay, I mean, I wanna... To give the listeners a feeling for how we operationally go about this. I mean, clearly you're gonna to have to make some simplifications if you're going to think about the wave function of the universe.

0:15:09.8 TH: Yes, yes. One, what is the goal here? It is really just like we do in ordinary physics problems. We try thought experiments, we try to simplify the situation, but of course, in such a way that you think or that you hope to capture the essence of the problem. And my impression is that, that this has worked pretty well in this quantum cosmology program. Of course, it is not an exact wave function, it is not a precise formulation, but somehow and a little bit miraculously, the general framework of quantum cosmology, it seems to me, has been able to capture a few key foundational features of how we go about thinking about the quantum universe, which has been very difficult to discover by other means.

0:16:20.2 SC: It obviously runs into the question that the person on the street has been told every day of their life that we don't understand quantum gravity.

[laughter]

0:16:28.6 SC: So it sounds like you're doing quantum gravity even though we don't understand it. How do you get away with that?

0:16:34.5 TH: Yeah, so somehow we get away with that. I think we understand some... I think we understand more than we sometimes admit.

0:16:45.3 SC: Good.

0:16:45.5 TH: I do think we understand, we have learned a lot about sort of the conceptual framework. Maybe we don't have a precise mathematical picture, but... And you can see where this goes, right? These toy models do capture certain essential features.

0:17:04.1 TH: The universe... The fact that the universe inflates at early times. And also, yeah this idea that, well, as we all know from quantum mechanics, the act of observation plays a crucial role. An electron doesn't really have a position as long as we don't ask for it. But that is a fundamental different thing from a classical system, which of course has a position and a momentum. So imagine now thinking about the universe as a wave function, as a description of all possible universes. Maybe it isn't quite real until we bring in the observer. And so that has been a whole fruitful area, I think, to study the kind of questions, to study ultimately the relation between our existence and the nature of the universe in a quantum mechanical setting, something which classically you cannot begin to ask really and so I am a bit more optimistic; we keep saying we don't understand quantum gravity, but I think we're somewhere along around these...

[laughter]

0:18:16.3 TH: And that's... And then we haven't even talked about holography.

0:18:22.9 SC: Right. We will. Don't you worry. But I do want to... I do want to, again, give a flavor of some of the issues that one faces here. You already mentioned turning time into something that looks like space. I mean, this was infamously the part in A Brief History of Time where most people are like, "Okay, I give up because he started talking about imaginary time."

0:18:48.5 TH: Yeah. Yeah.

0:18:49.0 SC: So you're welcome to say no, but could you explain what imaginary time is and why it mattered? Why you had to do that? Time is real to you and me. Why do you have to make it imaginary?

0:19:02.0 TH: Well, yeah. Okay, time is real to you and me here and that's all fine. But as we discussed already, when we go back in the history of the universe to the earliest stages, the Einsteinian way of thinking about the expanding universe, we run it backwards and time, time stops. So you could already have said in the '60s or even earlier, in fact, because this idea that time had an origin and that that was the Big Bang was been around for 90 years. So the discovery of the Big Bang to me, the fact that the Big Bang is the origin of time already shows that there must be something emergent about time. If you're gonna understand the Big Bang, we better put in time as a prior assumption because it's all about how the dimension or perception of time as we know it and as we experience it comes about. So I would say that the dimension of time has been a problem all along in modern relativistic cosmology. And in that sense Stephen's trick to sort of turn time into space is in a way exactly what the doctor ordered.

[laughter]

0:20:19.4 TH: It's a bit radical, but then the big bang is a very radical phenomenon. Right?

0:20:25.9 SC: Yeah.

0:20:27.1 TH: And in fact, that was later, much later. We shouldn't probably go too deep into this, but now almost 40 years on really from Stephen's time into, time goes into space business. Now we understand how that trick, so to speak. Is much less random as it looks, but in fact emerges from our new holographic way of thinking about the universe much more as an effective description. So of course, this was Stephen's bold, sort of characteristic way of doing physics back in the days. He had an intuition that he could do all of physics without time, essentially [chuckle] Everything could be just spatial Euclidean geometries. And I must say that since he died, that kind of physics, that kind of approach to quantum gravity, both in terms of black holes and in terms of the Big Bang, has regained importancy.

0:21:39.1 SC: Right. We did have a podcast episode with Netta Engelhardt, who was one of the people working on getting information out of black holes. And the idea of Euclidean quantum worm holes loomed large.

0:21:52.0 TH: Yeah. And so Hawking would've liked that. I think.

0:21:54.8 SC: That's right. Okay, good. But I'm still stuck in the '80s, because, look I'm older than you. My formative years were back in the '80s, and the big thing at that time, you already mentioned Andrei Linde and Hawking had a little bit of a disagreement about the wave function of the universe. It grew into this disagreement about the multiverse, etcetera. But back in the day it was just about inflation and can we get inflation out of our theory of quantum cosmology? So. Why don't you explain to the listeners what inflation is and why it matters to us.

0:22:29.5 TH: Okay. Yeah. So Inflation indeed came along in the early '80s, somewhat independent, I think, of Stephen's quantum creation model, as a way of more of inflation. What is inflation? Inflation is a very rapid phase of expansion in the earliest stages of the universe's evolution, which creates a big universe in a fraction of a sec. And so it sort of interconnects our entire observable universe. And to me the big bonus of inflation is that because it's such a rapid phase of expansion, it sort of generates with it a pattern of fluctuations, a pattern of variations in the universe. Purely from quantum uncertainty essentially. There are particles that are being sort of teared out of the vacuum and set you up with a big universe that is not exactly the same everywhere it comes with some sort of roughness.

0:23:43.6 TH: And that roughness is exactly what you need to, over millions and millions of years, generate stars and galaxies and so forth. So inflation on its own, and that's, it has been I would say there's significant observation support for such an early phase of rapid expansion because the roughness that you generate during inflation is reflected in the famous cosmic microwave background images, which show that the temperature wasn't equally distributed, but nearly equally distributed. Right? So inflation stands on its own really. But the big question of course, which must have been the question, I think in the early '80s, well, okay, how does inflation start? And that's where all the disagreements came along.

0:24:35.7 SC: Yeah. I think that amazingly that was not a question that most inflationary cosmologists cared about. They just said, well, as long as it starts, it gets us what we want. But Hawking and Linde and a few others like Alex Lincoln were a plucky minority who really tried to understand why it would start. And that was part of what the wave function, the universe was supposed to be about.

0:25:01.5 TH: Right. Okay. But of course, also since then, I think we've learned that, it is an important question how inflation started. Because the pattern of variations didn't say the cosmic microwave background radiation, that the afterglow of the Big Bang, is gonna depend on precisely how inflation unfolded. So it's not an empty question. The specific mechanism that drives inflation in the early universe leaves its observational traces. So if we wanna predict the details, the fine details of those fossils, so to speak, we better understand how it started.

0:25:51.2 SC: Yeah, I agree with you. But, again, plucky minority, I think you're right that it's more common these days and, but there is a slight, I don't wanna say downside, but implication to this that you already mentioned, which is that there are these quantum fluctuations that mean that inflation is a little bit rough. It doesn't end the same everywhere and on very large scales, those fluctuations can be very big and give rise to a multiverse and different things going on in different places. And someone like Andrei Linde embraced that multiverse and said, okay, there it is. That explains why our own universe is so unusual looking. Because it's a tiny, tiny part of some gigantic ensemble. My impression is that Stephen and you did not embrace that picture quite as lovingly.

0:26:40.9 TH: That is correct. [laughter] Yes, yes. And right, right. This is exactly the moment where I entered Stephen's office at the heart of that disagreement. Somehow I think Andre, so the problem of, it's appealing in one sense, the multiverse because I would say suppose you will need to generate a huge expanding space, where different regions behave like different universes, even with different effective laws of physics, yeah, then you generate some sort of gigantic reality in which the apparent biophilic design of our universe would be just a natural fluke and that's it. So I think it appealed to some cosmologists that this would get us around a lot of the perceived fine tuning issues. The idea for the observation, that our universe is at a level of physics remarkably fit for life.

0:27:56.1 TH: Of course if there are a zillion universes out there, then once in a while you're gonna have such a universe. But there was one problem from and which was, which was clear in the '90s already, which is, okay, suppose you have a multiverse, then if you wanna turn this into a fully, a full fledged scientific hypothesis, you better tell me in which of these universes we should be, and therefore what we should observe, what kind of roughness in the CMB we should observe, or what kind of value for this or that parameter we should expect to observe. And so that's something between cosmologists, called the Measure Issue and the Measure Problem. And the Measure Problem is really how should we, in a gigantic multiverse, what weight should we associate to different kinds of universes? How important are different kind of universes in this gigantic reality?

0:29:00.8 TH: And so I think that was a crucial point. Somehow I think Stephen thought that to get a proper scientific falsifiable hypothesis out of the multiverse, would require a radical quantum thinking. Whereas other people like Linde thought okay, the measure issue, it's eventually it's gonna go away by some sort of anthropic principle or by another means. And that's of course a very interesting debate because this goes to the heart of what cosmological theory is about. How do we fit into the grand scheme? Are we, is there a giant inflating space in which the anthropic principle is gonna select our universe, or is this giant inflating space not quite there without bringing in that observer's perspective in a more fundamental way, interwoven with physical theory itself, with quantum thinking.

0:30:09.0 SC: And you're gonna be on the latter half?

0:30:10.7 TH: You bet.

0:30:14.4 SC: Well, let's linger lovingly over this distinction, because I think it's an important one, but it's also a difficult one. Cosmologists who do think about the universe, or for that matter, people who do black hole information or whatever. Anyone who talks about quantum gravity, it seems to me, is very tempted by still drawing a classical picture of space time, even though they know they're talking about quantum gravity and saying, well, there are fluctuations of some sort. But I take it the point you're making, it seems from reading the book, I cheated by reading the book. That's not really fair to a truly quantum universe isn't just a big, fluctuating classical universe. Is that fair?

0:31:00.5 TH: Right. I think that is indeed the key distinction. That you either assume that there is some sort of background out there, which can be wildly fluctuating in different regions, but the sort of big, big background in which all this is happening acts as yeah, some sort of foundation, various, but this took many years. There were, now I'm jumping around. I mean Stephen in the late '90s didn't have the, didn't have the solution. But eventually as you suggest, we came to see that this is still too classical. This is still too much of, yeah, it's not enough quantum. I can't say it differently. And so we start to try to take a fully quantum view, even though we of course didn't have a precise theory to do so, and you're led to a different picture in which we have rather a classical space around us, of course, and which can be much bigger than the observable universe, but which sort of dissolves in uncertainty on the largest scale.

0:32:15.2 TH: So it's much like what we were saying earlier about the electron. The electron doesn't have quite a position in quantum theory before we ask for its position. So if we think in the same way about the universe globally, we should be saying that the universe is a definite space, time and configuration around us. But on the larger scales it's rather uncertainty which dominates instead of a definite classical structure extending to infinity as some people would say. So it's a picture which I came to see that builds in a certain finitude. So quantum theory is interesting in that respect. It has always been interesting in that respect. In the sense that it's a theory for what we can know, but also a theory that sort of tells us what we can't know. And here, in our model, this is sort of playing out, yeah, at the level of the larger scales.

0:33:17.7 SC: By the way, this will be of interest to our listeners. Are you assuming in the background something like the many world's interpretation of quantum mechanics?

0:33:26.7 TH: I'm certainly assuming an interpretation that is like many worlds in the sense that I'm trying to work with an interpretation of quantum theory that doesn't require anything external.

0:33:42.4 SC: Yeah.

0:33:43.5 TH: Right.

0:33:44.0 SC: Or any hidden variables for that matter.

0:33:46.4 TH: Yeah, yeah, yeah, yeah. Yeah. The funny thing in cosmology though, we often think about, quantum mechanics and the many world interpretation when we think about future branchings. And so we make, we prepare an experiment the wave function splits. The observer gets correlated with one outcome and so forth. But the thing which I found striking in cosmology is that the current state of the universe around us is already the result of a giant question asked over the wave function. And so it's sort of the many world interpretation or in cosmology also acts a lot, or I think it's important when it comes to the past.

0:34:25.4 SC: Sure.

0:34:25.8 TH: In selecting this or that subset of histories. And like editing, like any branching in quantum cosmology say it comes with limitations. So in a sense you could say the multiverse, it's almost, we are behaving as if we have access to an infinite amount of information. Whereas of course, from an observer's perspective within the universe, there's the extent to which our observations distill one or another branch of the wave function is finite. And Stephen's trick to close the universe to turn time into space helps in that respect I think.

0:35:09.7 SC: I think this there'd be a great opportunity to clarify something about, again, the person on the street who's not playing with equations, hears words and tries their best to figure out what's going on. So we hear about Feinman and his some over histories, right? Like Feinman said, consider all possible histories of the particle or the universe. And there's a certain way of adding their contributions together to get the quantum wave function. And Steven and Jim Hardell used that idea intimately when they wrote down their wave function. That's different than Everett's view of many branches of the wave function, because his individual branches are supposed to be real. They're not mathematical fictions that we add up and in some sense they're kind of classical. Right. So is that a fair distinction the way you're thinking about it?

0:36:00.6 TH: Yes, I think so. So I think it's more the Fineman kind of description that is perhaps at the heart of this theory.

0:36:14.2 SC: Well, let's get down to bras tacks. Do you believe that there really exist other universes where things are very different, other branches?

0:36:22.9 TH: That does not quite enter in our theory.

0:36:28.0 SC: Okay.

0:36:31.0 TH: And that is because in the end certainly inspired by these holographic constructions, they were very much what we call top down, so backwards in time. So they're very much anchored, so to speak, the histories that play a role in the wave function or anchors on, yeah, I would say our observational situation around us. And so in a sense, my feeling is that the new holographic wave flooding cosmology is going to at the very least trim the wave function of the universe down to, yeah. I would say a more manageable thing. [laughter]

0:37:17.0 SC: Okay. But if I observe a spin that's in a superposition of spin up and spin down and I see that it's spin up. Do you think there's another version of me that saw it spin down?

0:37:25.9 TH: Well, as an operational meaning. Sure.

[laughter]

0:37:29.8 SC: You say sure. A lot of people think that's a radical thought to think that there's a version of me that saw a spin down.

0:37:35.0 TH: Sure. But that branching once you observe it. Sure, sure. I would view that as an operational way of describing your setup.

0:37:43.7 SC: Okay.

0:37:44.0 TH: Your experiment, your observation. But once the observation has unfolded, what happens to the other you? It's lost in uncertainty again, I would say. And every branch that grows out of the other you will no longer be contributing to this universe.

0:38:04.0 SC: Okay. So let me try out the following analogy that struck me as I was thinking about your book.

0:38:10.7 TH: Okay.

0:38:11.0 SC: So Let's say we do Schrodinger's cat, right? So Schrodinger puts the cat in a quantum superposition of alive and dead. And famously if we open the box and look at the cat, we don't see the quantum superposition, we see the cat alive or the cat dead. I think that what you're saying is kind of like the following, that if I had an infinite series of cats spread out in space, I could look at one of them and it would either be alive or dead, but very, very far away. The cats could still be in a super position and it would be a mistake for me to think of this ensemble as just a random collection of cats alive and dead. It becomes more and more quantum as you go further away.

0:38:51.2 TH: Yeah. Uncertain indeed. Yeah. Just like the electron position.

0:38:54.8 SC: Yeah. Yeah. And so you're saying...

0:38:57.8 TH: Indeed.

0:38:58.3 SC: We should think of cosmology like that we can talk about the classical world that we see, but let's not extend this classical picture too far away.

0:39:06.1 TH: Let's leave it uncertain. That seems to me to be the lesson. And that's also at the heart of how this quantum way of thinking about it resolves the measure problem because it is anchored on what you just said, what we see, rather than try to get us into the picture, into the cosmos apostle theory sort to speak.

0:39:33.0 SC: Yeah.

0:39:34.2 TH: Like for instance someone with an tropic principle would do.

0:39:39.2 SC: And is this the, what do you mean by the top down approach?

0:39:43.0 TH: Yes.

0:39:44.3 SC: So say pretend that we didn't just say that. Tell us what the top down approach is.

[laughter]

0:39:50.3 TH: Right, right. So what we mean by a top-down approach is indeed that we regard the universe as we observe it around us as a kind of starting point for which of the many possible histories of the universe contribute to what we see and what we observe. And that is important because it provides you selecting those, selecting those histories, or selecting those subset of branches in the wave function that allows you to make predictions for future observations. Because that's kind of the problem with the multiverse, right? If you have many different universes and you wanna protect something for a future observation for the next satellite, yeah. You need some sort of criteria. And that was very much at the sort of... That was sort of the guideline also for Steven and me.

0:40:51.8 TH: So to get, we sort of had this intuition that a proper quantum way of thinking about the universe should somehow resolve this measure, should sort of give us a measure and give us an ambiguous criterion for future predictions. But it comes with a radical different perspective, of course, because we used to be able, we used to think that we would one day be able to predict from first principles how the universe should be, how the universe should turn out. That was the kind of attitude that Hawking took in Brief History of Time, like a sort of transcendental theory that tells us why and how the universe is the way it is. That's how he phrased it. And he totally came a... He totally turned 180 degrees on this point, which I think well was a very interesting evolution to witness in his thinking.

0:41:57.2 SC: So I wanna make sure that the listeners know what we mean when we say the measure problem, in a multiverse, in a very big multiverse that we do as you and I agree, it would be sloppy and careless to think of it as a big classical ensemble of things, but let's think of it that way. Anyway, there's a lot of observers. They see a lot of different things, different cosmological constants, different masses of the electron or whatever. And a traditional multiverse, cosmology thing to do would be to say, what is the chance that you observe the electron mass to be a certain number? And the problem is, there's an infinite number of observers in this universe who observe it to be a certain number and also an infinite number that observe it to be a different number. And it's very hard to take infinity divided by infinity to figure out what fraction of people will see a certain thing. That's the measure problem in my mind. Yeah.

0:42:53.5 TH: Yeah. I think it's one version. There's a different aspect which I think is closely related to what you say. So we asked a question faced with a multiverse, we would ask the question, what's the probability that we see this or that?

0:43:07.0 SC: Yeah.

0:43:07.0 TH: But there is this subtlety in what we mean by we.

0:43:12.2 SC: Yes.

0:43:13.8 TH: Depending if you have a different definition or a different description of what we means, what physical characteristics you associate to an observer, be it a human observer or a habitable planet, or just a galaxy, depending on how you choose to define that, you're gonna get a different answer.

0:43:39.2 SC: Yeah. [laughter]

0:43:40.7 TH: And so there is, it's almost not settable by rational arguments because you could turn all, you could turn a negative outcome into a positive outcome by changing what you mean by we. And so that's another version of I think the measure problem, a version which points very clearly to the underlying problem with the multiverse, that it is a construction, a kind of platonic construction that is out there independently of whether it is observed or not. It's out there with an independent existence from us. And that frankly, it took many years, by the time I sort of fully realized the depth of the problem.

0:44:32.2 SC: The God's eye view in other words.

0:44:34.3 TH: Yeah. It's what Stephen Coles called indeed and many others, I think. Yeah. The God's eye view, which in indeed and by 2005 or so, we were absolutely convinced that we had to construct cosmology in a different way from what Stephen called a worm's eye view. Not a very good term I think.

0:44:58.5 SC: That's okay.

0:45:00.0 TH: You get the idea right.

[laughter]

0:45:02.0 SC: Well, I do think, I actually really like the philosophy behind it. And I think it's kind of a shame that Stephen famously went to rhetorical war against the philosophers. 'Cause I think that there's useful...

0:45:14.8 TH: I agree. I agree. No, that's a good point. And one can wonder why that was.

0:45:23.2 SC: Oh, to sell books. [chuckle],

0:45:26.2 TH: You think so?

0:45:28.5 SC: You know, he...

0:45:28.9 TH: Little bit more than I do?

0:45:31.0 SC: Well, I can say that people who knew about, how he constructed his famous sentences about philosophy being dead and so forth in The Grand Design, it was very clearly to sell books.

0:45:46.3 TH: Oh. But... Right. That is probably true. But my feeling is... But I'm not sure. I'm not a biographer. Right?

0:45:54.4 SC: Yeah.

0:45:54.5 TH: But my feeling is that the whole philosophy is that thing of Hawking predates the Grand Design.

0:46:01.5 SC: He was never a fan of philosophy. No. That's true. But that doesn't distinguish him from plenty of other physicists. Right. [laughter] like plenty of physicists. Well.

0:46:10.5 TH: Right. But wrongly so, I think because...

0:46:13.2 SC: I think so too.

0:46:14.0 TH: If you now look back on our just on the conversation we had this issue, God's eye versus let's call it worm's eye, is foundational.

0:46:25.2 SC: Yes.

0:46:25.7 TH: Because it is really about what is it ultimately that physical theory finds out about the world? Is it some sort of eternal transcendental truth, or is physical theory once you get the observer fully incorporated in there, a different beast from what we thought it was contingent on our existence within the universe.

0:46:50.7 SC: Right.

0:46:51.0 TH: And that's... For this I think we must admire Steven for the... For just for the simple fact that he was able to change his mind on this.

0:47:03.1 SC: Oh yeah, sure.

0:47:04.0 TH: And so in... At the end... Towards the end of my life, of his life, he said literally but okay, with top down with that new approach to cosmology, somehow we put human kind back in the center. That is a very different Steven from the one we could read in A Brief History of Time.

0:47:22.6 SC: Very much. And I will... I do... I should apologize to you because all of what you do and talk about in the book is too close to things that I care about. So instead of asking you questions, I keep saying, what about this? But I... Hopefully you can deal with my question asking style. So let me do it again, let me say what about this, 'cause I think that this question of predicting what we should be like if this certain multiverse were true, is exactly wrongheaded, that's the point on which I completely agree with you. What do you mean, what we should be like, we're us, we are what we are like, I do think it's possible and you could probably say this even classically in a big fluctuating ensemble, you could ask, what is the probability that your theory predicts the existence of anybody like you? And if that probability is one who cares? If there's many more people not like you, you're gonna be there in the multiverse or in the theory.

0:48:23.2 TH: That's right. That's right. Yeah. Yeah, yeah, yeah, yeah. Certain probabilities we just don't care about.

0:48:28.1 SC: Yeah. And Jim Hartel and Mark Srednicki made a point about this with the example of Jovians. Do you know their example about the Jovians?

0:48:36.6 TH: Sure, sure.

0:48:37.7 SC: Can you tell it to us? To the readers? To the listeners?

0:48:40.8 TH: Oh, I don't remember the details, but they... You can correct me, but their point was that should be expect in some giant ensemble of inhabitants say, should we expect to be typical in one or another sense, given that all we know is that we exist.

0:49:07.5 SC: Yeah. [laughter]

0:49:08.3 TH: And so they made a big point in explaining that the mere observation of the fact that we exist is very, very different, if you don't have access to other civilizations or planets or extraterrestrials, from saying that we should be typical. And the reason is one's and is the same as what we were discussing earlier. Typicality in the end always boils down to treating certain features of our living systems or biosphere or planet or galaxy as preferred as the most probable and... But that is falacial thinking really.

0:49:54.5 SC: So good. So I think that very much on the same wavelength there, I take it you would agree that there's no reason to think that you or I as individuals or human beings as a species are typical in the universe.

0:50:10.1 TH: And even that this is the key point, and even that nature of the physical laws that we observe is the typical outcome of some grand cosmic evolution, exactly.

0:50:22.4 SC: Good.

0:50:22.8 TH: So neither... So the neither is the... Neither is the tree of life on earth as sketched first by Darwin, the typical outcome of that evolution in a sense, Steven and I pushed that same kind of thinking further down and we are saying, well, wait a minute, maybe the physical laws as we have them are also not a typical outcome. But... And this is the crucial point, just like Darwin... Just like Darwin didn't need a zillion other planets to do biological evolution on this planet, we claim we don't need a zillion other universes, to study the evolution of this universe. But it comes crucially... As you point out, it comes crucially with the caveat that there is no assumption of typicality and there is no... It could have turned out differently. There's... And that said and to me the big surprise and this is really when the moment that Steven sort of told me, look, now it's time for a new book, [laughter] the real surprise, this is as a matter of giving some homework, right?

0:51:26.9 SC: Yeah.

0:51:28.4 TH: Was that holographic way of thinking about cosmology builds in much of that top-down reasoning, because the holography in a cosmological context, really flows backwards in time from data, from an observational situation in the present, the time, the past, the time dimension is in a sense, the emergent dimension and it's contingent on the kind of questions you ask. And that, for me was sort of the key transition point, because previously much of that top-down reasoning we were preaching, so to speak, remained controversial because it felt like a choice. It felt to so many people like David Gross, who would tell us, "Ah, but wait a minute, you are putting in the answer."

0:52:22.8 TH: So that's this typicality reasoning again, you were putting in the answer, I'm trying to explain, I'm trying to predict the answer. And, you kind of feel like maybe he's right, he has a Nobel Prize and all that you see. But then holography sort of solidified the top-down reasoning, precisely because it flows backward. It works backwards in time and I was very surprised by that and I think Steven was too. And so, okay, that's when it all sort of began to click together our picture.

0:53:00.5 SC: Well, if you want us to believe that, which is a good thing to do, you're gonna have to tell us more about holography and how it goes backward in time. I don't know where you wanna start, but what do you mean when you're saying holography in this context?

0:53:14.8 TH: Okay, so we wanna talk a bit about holography.

0:53:18.7 SC: Yeah. You wrote a book, it's your job [laughter]

0:53:24.5 TH: So holography has been a fact. Let's face it, holography has been the talk of the town in theoretical physics for 25 years, right?

0:53:32.4 SC: Yeah.

0:53:32.9 TH: But of course it's true, it's been mostly practiced in highly idealized, abstract, non-realistic, mathematical situations. Universes that have nothing to do with ours. But there's a general lesson behind holography, which is, I think it's been the way which we're finding out in which quantum theory and gravity can finally work together more or less harmoniously. And the way this works is that one appears to be the hologram of the other. The clearest example perhaps is the case of a black hole. We think about a black hole. We've seen images of a black hole, that's all very nice. And a black hole is something very gravitational, right? It, space time is curved, highly curved.

0:54:30.3 TH: Einstein says there is a surface, there is a horizon. And inside the horizon, inside the black hole, space time really crumbles [0:54:40.5] ____. So that's the gravitational description of a black hole. But then when you start thinking about a black hole from a quantum perspective, you begin to discover, going back to the work of Bekenstein and Hawking and many others, that well, maybe all there is to know about a black hole is in fact located in bits of information that are living on the horizon surface, that are living on the surface. So if you start reasoning about a black hole that way, you might arrive at the conclusion that the inside of the black hole doesn't really exist, or isn't, or is in a sense, yeah, no, not quite there. Is some sort of emergent phenomenon which you may not need.

0:55:33.2 TH: If you wanna ask physical questions, you could ask physical questions from a quantum perspective and just only talk about the horizon or [0:55:41.6] ____ horizon. So I think that's much of the more motivation or inspiration for maybe there is a fully quantum way of thinking about the universe about space and time is in a sense holographic in the sense that there is one dimension in the case of a black hole, the interior dimension that is emergent, that is not quite phenomenon. And now you begin to think about, wait a minute, wait a minute. The Big Bang is another problematic thing. Just like black holes, space-time crumbles. What dimension in cosmology could be the one that is holographically projected, that is sort of encoded in a lower dimensional screen like thing, just like a hologram. Well, as we discussed in cosmology, it is very much the dimension of time, which is the problematic one.

0:56:34.0 TH: It's the one that has an origin. It's the one that disappears with the big bang. It's the one that causes us a headache. And development of those holographic ideas in theoretical physics, indeed suggests that it is the dimension of time in a cosmological context that can be holographically encoded in yeah, a hologram.

0:56:57.6 SC: So we start with a moment of time or some spatial description of the universe, and then we kind of do holographic tricks to understand how that could be projected into time evolution. I'm just stringing words together like ChatGPT here. You can fix that.

[laughter]

0:57:21.4 TH: Right, right. The way I see it is that there is, okay, I wanna say two things here. First of all, this holographic way of thinking about reality is completely useless in normal circumstances. Today, here around you, around me, around everywhere, there is time and there is space and we can work with that.

0:57:45.4 SC: Yeah.

0:57:46.0 TH: But where holography becomes important I think, is where Einstein's theory, where the description of reality in terms of space and time that we experience, where that description doesn't hold. So inside black holes and at the big bang, my feeling is that in those extreme regions of the universe, the more fundamental holographic quantum nature rises to the forefront. And so what I mean by that is that in those extreme regions of the universe, one of the familiar dimensions disappears. So in the case of the black hole, it's the interior of the black hole. In the case of the universe, if we go... If we trace the history of the universe backwards, it's all fine. But at some point, the bending of time becomes so strong. And you think what holography is telling us is that well in fact, the dimension doesn't reach further. The holographic way of saying the same thing would be that the hologram doesn't quite encode the information to push history further backwards. And so the Big Bang in holographic way of thinking about the universe becomes almost like, it becomes almost like an epistemic horizon. A region where you can't, yeah, you'll run out of bits almost literally. That's kind of where it stands. Of course it has to be, this is a brand new hypothesis. It has to be developed in so many ways but you sort of get the gist, right?

0:59:25.3 SC: Yeah, no, I do. And so maybe a motto might be, classically we would say if we kept going backward in the history of the universe, time would end because we hit the big bang singularity. And what you're saying is time kind of ceases to be a thing. It's not that it ends, but it ceases to be a useful way of talking about the universe.

0:59:46.6 TH: Yeah. I think indeed. Right. Right.

0:59:52.0 SC: Gradually maybe.

0:59:53.3 TH: In a way, what we have been trying to do in cosmology ever since the discovery of the Big Bang is to let time, when we go backwards, disappear in a controlled fashion.

1:00:07.0 SC: Sure.

1:00:08.0 TH: And that's essentially, that has essentially been the goal. And of course it's kind of interesting to look back on this because this is what the singularity theorems in the '60s tell us to do. Find a better way to let time disappear into the Big Bang so that physics doesn't break down. And of course, these ideas about the multiverse or about pre-Big Bang cosmologies, they're sort of ideas that all go in the direction while maybe the Big Bang wasn't really the origin. Maybe there is something, maybe we can just push through. Do physics as we normally do it. But it's kind of interesting that this hypothesis that I developed with Hawking is very different. It's taking the idea of an origin very seriously. In fact even more seriously than the early Hawking would've done it, in the sense that if you let the time dimension disappear, it's as if the laws of physics disappear. And so it's really sort of placing that notion of an origin very central in our thinking about the early universe. And in that sense, I think we can now begin to see clearer the difference between this hypothesis and other hypotheses that evoke an evolution before the Big Bang and all that.

1:01:38.6 SC: Well, let me consider two different cosmological scenarios. One is one much like we think is real. In other words, we have observers like us today and we trace back 14 billion years and there was a Big Bang. And that Big Bang by the way, was a very low entropy special condition as Roger Penrose and others have pointed out. Another one might be, there's sort of a galaxy, kind of like the Milky Way and people like us, but the whole background space time is otherwise empty. So there's no Big Bang. There's just a weird random fluctuation in which all the particles came together to make our galaxy and then they'll disperse in the future. And there's no beginning to end of time. Does your theory explain why our universe looks like the former rather than the latter?

1:02:30.1 TH: Right. I'm writing a paper on that.

1:02:31.8 SC: Good. Hurry up. [laughter] What are you doing being on podcasts when you should be writing your paper, Thomas? [laughter]

1:02:39.7 TH: Okay. My claim is the following that if you specify in sufficient detail the local galactic configuration that you sketched by which I mean really the actual configuration, so you specify enough data. Then you will see a switch. I'm revealing really the latest research here you'll feel a switch in from your second scenario to your first scenario. So if you would only sort of loosely say, well, I've got some sort of Milky Way, I'm not very much interested in its precise description, then you might as well favor an empty universe without anything else. But if you begin to describe that observation situation, the fact that, act in your insufficient detail, then at some point it'll become, you will see a phase transition. You will see a shift towards the universe like we actually observe it.

1:03:46.1 SC: Okay. I will go on the record as saying that would be great. And I don't believe you [laughter], but we can talk about that.

1:03:52.3 TH: You should be skeptical.

1:03:54.1 SC: I am skeptical. I think that it even I would claim that you can specify to whatever level of detail you want the world around us to, a 100,000 parsecs in every direction surrounded by vacuum and everything is perfectly fine.

1:04:13.8 TH: Okay. Wait, what is the statement then?

1:04:18.2 SC: The statement is that it doesn't matter how carefully I specified my current observations here in the Big Bang, I can always embed them in a universe... Sorry. Here in the Milky Ways. My mistake. I can always embed them very easily in the universe without a Big Bang at all. And I suspect strongly, and though I don't know, maybe I'm wrong about this. Maybe this is what you can convince me of, but I suspect strongly that in most known principled ways of comparing the likelihood of those two possibilities, the empty universe is gonna come out more likely.

1:04:48.8 TH: Yeah. So indeed what I'm going to reveal is indeed, of course a different way to compare these likelihoods.

1:04:57.5 SC: Good. I look forward to seeing that.

1:04:58.0 TH: Okay. Good.

1:05:01.1 SC: But okay. I guess the last loose end here, this has been excellent. Thank you very much for explaining a lot of this modern research to us. I just wanna get straight one last time the comparison to the multiverse story. So if I understand what you're saying, which I think I mostly do, I can kind of conditionalize on here I am, here's the big bang, here's what I observe, and then I use your theory to reconstruct the past and the future of the universe.

1:05:27.9 TH: Yeah.

1:05:29.3 SC: Could I have conditionalized on completely different kinds of people and completely different laws of physics and things like that and told a similar story all still within your framework?

1:05:43.6 TH: Oh, yes, yes. Sure, sure. You could do a thought experiment and conditionalize... In fact, we do many of those thought experiments in theory and conditionalize or start, so to speak, from an entirely different configuration. And you'd get a different past and future.

1:06:03.6 SC: Good.

1:06:04.5 TH: All of these past and futures are limited, just like we discussed this.

1:06:12.3 SC: So good. So just to... I'm just trying to get it right. So I keep repeating. So what you're saying is that once I say who I am, the classical world around me is finite, it's limited. Because it sort of dissolves into quantum uncertainty if I go too far. But I can think of it as an ensemble of many different patchwork classical realities, all of which are there in the wave function of the universe.

1:06:35.3 TH: Good. This is his last point, I am no longer convinced of.

1:06:38.3 SC: Okay, good.

1:06:39.1 TH: All these different, at all these different classical worlds fit in one grand wave function. That's indeed the heart of that top-down approach taken, taken fully. And the evidence we have for this, and this is a crucial point, I think yes. The evidence we have for this is that they're not all there in one grand wave function, when we think holographically about this. Holography is really sort of, yeah. It's kind of interesting. It brings in this observational perspective with, in the theory, but at the same time, it then also limits the range of that wave function that we've been talking about. It limits the range of different realities that the wave function encompasses. Now of course, we are talking really cutting edge stuff. But you're absolutely right. Is there a grand overarching wave function and comprising all possible holographic theories. Or is there a limitation on the reach of physical theory that is contingent on, say, a boundary configuration or an observational configuration?

1:07:55.0 SC: Yeah, and we don't know yet still work to be done.

1:07:57.6 TH: That remains to be seen. Yes. Sure. Sure.

1:08:00.6 SC: It's good that not all the questions are yet answered, because that leaves something for you to say in your next book. [laughter], Which I predict is gonna come out eventually. So Thomas Hertog thanks very much for being on the Mindscape podcast.

1:08:13.6 TH: Thank you so much Sean lovely.

[music]