AMA | December 2025

0:00:00.2 Sean Carroll: Hello everyone. Welcome to the December 2025 Ask Me Anything edition of the Mindscape podcast. I'm your host, Sean Carroll. A little bit later in the month than usual with the AMA, but there's a whole bunch of things that happened in the scheduling of a podcast. As many of you know, it's just me here. There's no team, there's no social media apparatus, there's no editor. It's just me doing all the podcasts, which I like it that way. I can control everything, but sometimes you just have to juggle things. And the AMAs came out a little bit late. As longtime listeners know, at the end of the year, next week, you'll get a little holiday message from me, kind of a mini solo episode. And then we go dark for the one week that is sort of right after Christmas, I guess this year, I forget exactly what it is. But other than the holiday message next week, the next you'll hear from me is regular episodes coming in January. And there's no AMA in January for that reason, because we're taking off the holidays. So the next AMA will be in February. So there's this one. And I don't have a lot to say to get into it.

0:01:03.1 SC: I just have two quick things to say. Both involve asking for money. [laughter] Sorry about that. It is the season, though, isn't it? The holiday season is closely associated with asking for money. The first is a more highbrow thing, asking for money, which is. I got an email from Joshua Greene. You remember Joshua Greene was a guest on the podcast. He's a Harvard professor in both philosophy and psychology who does these wonderful neuroscience experiments to see what is happening in your brain when you are reasoning morally, right? Are you... Is the brain doing something different when it's doing deontology versus utilitarianism? If you have different answers to the trolley problem, do different parts of your brain light up? Yes, is the answer, which is fascinating. If you haven't heard that episode, I would recommend checking that out. But also, as he mentioned on the episode, he is one of the driving forces behind a charity called GiveDirectly. And it's one of those things that I'm unabashedly a fan of. GiveDirectly is kind of a version of effective altruism, where you give them money and they just give it to people, like, not a lot of thought involved.

0:02:15.1 SC: They give the money to poor people. That's their big thing. There's like some spin, some variations on that theme, but that is basically what they do. And they're hosting something this holiday season, this December, called the Pods Fight Poverty campaign. They're organizing a bunch of podcasts to pitch to people the idea of giving a little money to GiveDirectly. So long story short, you can go to... Let's see what the page is to go to. I'll put a link on it on the preposterousuniverse.com webpage post for this AMA, but go to givedirectly.org that's G-I-V-E Give D-I-R-E-C-T-L-Y directly.org/podsfightpoverty P-O-D-S Fight F-I-G-H-T Poverty P-O-V-E-R-T-Y. And you can give money and they will... They match it, right? In this case, you get twice your impact. So if you give $100, there's $200 that will actually go. And the money goes to Rwandan families in poverty. What they're trying to look for is they have certain villages in Rwanda where people are just struggling, they're not doing very well. And what they want to do is just hand $1,000 or the equivalent in Rwandan money to families in these villages.

0:03:39.3 SC: There's many... I'm not going to go into all the social science here. It works. Giving people money is what works to make the world a better place. And this is life changing money for these families. You don't have to donate a thousand dollars. They'll collect whatever you want to donate and they'll bundle it up and they'll give $1,000 to these families. And so it's truly transformative. I know that I give money to GiveDirectly. I'm actually a regular donor. Like I give every month as a recurring thing. And so if you're looking for a way to just do a nice thing for the world in this holiday season, I can recommend that, GiveDirectly Pods Fight Poverty. There is a lot of research behind it that this is an effective thing to do. So that was the first thing. The second thing is... [laughter] And I bundled this together as a second thing because I... It's made me feel really bad because at the end of every intro to the AMAs, I say the Ask Me Anything questions are composed by Patreon supporters of the Mindscape Podcast. And I mentioned that if you enjoy the AMAs and would like to ask questions yourself, you could become an AMA supporter.

0:04:49.9 SC: And I'm going to say that again. You can become an AMA supporter. You can go to patreon.com/seanmcarroll and support the Mindscape Podcast. So that's my second ask for money. You don't have to as I always say, I know that it's not something that everyone can do. It's, you know, you can pay a dollar per podcast. If you think that you get four or five dollars a month worth of joy and entertainment from listening to Mindscape, then I think it's a good thing to do. But you don't have to, because you can just listen to it with the ads. If you join on Patreon, you get an ad free version. You get discussions in the Patreon posts with other Patreons, and you know, the people who ask the AMA questions, they sometimes get answers from other Patreons and sometimes even to questions that I don't get around to answering. So that's a useful thing. And of course, you get the feeling that you're supporting Mindscape, and I appreciate that support very much. And look, if you have to choose, if you say, I got $10, I want to give it to a good cause, should I give it to Fighting Poverty in Rwanda or the Mindscape Podcast?

0:05:52.0 SC: Please give it to Rwanda. Okay? Like, I'm a believer that we should do various things. We don't need to only do one thing at once. We can do all sorts of different things simultaneously. So maybe you have enough rattling around in your wallet or pocket to give a little money to Rwanda and an even tinier amount of money to Mindscape. That would be great. If you don't and you have a little bit that you want to give somewhere. I would go to Rwanda. But I also appreciate what goes to Mindscape. So it's a complicated world. We have many obligations. But the support from Patreons and from other listeners is what keeps me going here at the podcast and what keeps me doing these AMAs, for sure. So my appreciation to all of you, and hopefully we can make the world a better place in all sorts of ways, large and small. So with that, let's go.

[music]

0:07:00.0 SC: Mikhail Bennitson says, a man is about to board a flight when an engineer comes over. It turns out that the engineer is the world's expert on flight and plane engineering. The engineer tells the man, don't board that plane, there's a substantial probability that it will crash. The engineer has elicited expert opinions from thousands of his colleagues all over the world who all agree with this assessment. On the other hand, the man has just got a good look at the plane through the terminal window, and everything seems normal. Further, he's never crashed an airplane before, so why should this time be different? You might by now recognize that the protagonist of this story is Sean Carroll, and the engineer is Geoffrey Hinton or Yoshua Bengio. I share your skepticism of doom from superintelligence, but I find it prudent to update my priors in light of the experts' warnings and my own non-expertise. Further, the stakes are so high that extreme caution seems to be implied from considerations of risk aversion. How can you be so certain that the plane will not crash even when experts are explicitly calling out the possibility? Even if you judge that there is only, say, a 1% probability that the analysis of Hinton and Bengio is correct, it still appears that we are rationally required to take this very seriously, I wouldn't get on the plane.

0:08:08.1 SC: Sure, you're perfectly welcome to take the word of experts like that more seriously than you take my word. I've never pretended to be an expert in AI or anything like that. People keep asking me questions because I have a podcast and I do AMAs and so I'm going to give you what my opinion is. My personal opinion is that there's a difference between being an expert in computer programming, as Hinton and Bengio are, and an expert in what intelligence is, which I think is the relative... Relevant question in this case. It would be very easy for the people who worry about superintelligence to convince me that I should worry. They could either... Well, what I would want is for two things, number one, evidence that they are not simply anthropomorphizing computer programs.

0:08:56.0 SC: They've given me no evidence of this. I've seen them, in fact, anthropomorphize computer programs all the time to speak about large language models as if they had agency and desires and things like that, which I just think is a mistake. And they could convince me that I'm wrong. But I haven't seen any arguments that I'm wrong. And on the other hand, give me a viable scenario where super intelligence in particular is the thing that will cause an AI disaster. I've tried to be very clear that we should be super worried about AI. There's a lot of things to be worried about, that it could go wrong. I just think that the specific fear that it will become super intelligent is the wrong one. And I could be right about that. I could be wrong. You are certainly welcome to not get on the plane as you like.

0:09:47.7 SC: Anonymous says, do you have any advice for people who are feeling imposter syndrome? I know that comparison is the thief of joy, but regardless, I find myself often struggling because I think that my own accomplishments are not good enough or that other people are doing better than I am, or I haven't achieved as much as I want to. Have you ever had similar experiences? And if so, did you ever find anything that worked for you?

0:10:08.6 SC: I don't really have any strategies here. I'm sorry that you feel that way. You know, I kind of feel like it's part of some combination of the human condition and the modern world that, you know, we can't just enjoy life as we go through it. We always feel like we should be doing more, that, you know, we're not good enough, that we're comparing ourselves to others. There's various techniques people use to get through that. Whether it's, I don't know, meditation or therapy or self help books or just advice from friends or family or colleagues, I really don't know. I think that this is one of those cases where in my mind different people are going to have different answers to a question like that. You know, it's easy for me to say, what you should do is enjoy the process, enjoy the day-to-day, enjoy like making little victories rather than worrying about your status or standing in the bigger picture. You know, there will be in the course of a long life, lots of little victories, lots of little setbacks, and can't let the highs get you too high or the lows get you too low. You know, do the best you can. None of us is as good at everything as other people are. All of us are better at some things than some other people are. So that's what we have to take solace in. I make no pretense that this is great advice, but my thoughts on this are just not very organized enough to be super duper helpful. Sorry about that.

0:11:30.8 SC: David Kudaverdian says, the classical limit of the photon field is the classical electromagnetic field. Similarly, the classical field associated with the electron is the Dirac field. So why do we routinely encounter classical electromagnetic fields in everyday physics, but never classical spinor fields like the Dirac field or macroscopic manifestations of the Dirac field? In what sense, if any, can the Dirac field give rise to classical macroscopic behavior the way the electromagnetic field does?

0:12:00.5 SC: Yeah, this is a really good, important question. Important in one way because we know what the answer is to this one. For those of you a little bit less familiar with the jargon here, we know that the world is made of fields in the language of Quantum Field Theory. We don't know that at the most fundamental level, of course, but it's the best current way we have of describing what happens in the world. And these fields come roughly in two kinds, the fermion fields and the boson fields. You can read all about this in 'The Biggest Ideas in the Universe, Quanta and Fields'. The boson fields are characterized by the fact that when you take the field and you apply Quantum Mechanics to it, it looks like a collection of particles. That's true for either bosons or fermions. But the difference is that the particles you get from a boson can pile on top of each other.

0:12:49.9 SC: The particles you get from a fermion field obey what is called the Pauli Exclusion Principle. You cannot have more than one fermionic particle in exactly the same quantum state. In fact, you can't even have... There's a little bit of a resistance to having two fermion particles in slightly similar states. So this is the explanation for all of atomic structure and chemistry, right? Because you have electrons that are in orbitals around nuclei, and electrons have spin, and there can be two values of the spin, spin clockwise or counterclockwise. So in any given spatial configuration of the electron, you can have two electrons in that spatial configuration, one with spin clockwise and one with spin counterclockwise. But it's the piling up of the bosonic particles that gives us the impression of a classical field. So you get a classical gravitational field, a classical electromagnetic field, and so forth, because these are bosonic fields, because the individual particle, like excitations, can pile on top of each other to make a big classical thing.

0:13:56.2 SC: A Dirac field, Dirac is the guy who wrote down the equation of motion for the electron, which was the most studied fermion for a long time. So a Dirac field cannot pile on top of each other. And therefore you don't get a classical Dirac field that we could measure in the universe. Of course, you do get classical macroscopic objects from fermionic fields. Indeed, you and I are classical macroscopic objects. It's just that the way that they become classical is different because the particles can't pile on top of each other. That's why fermionic fields, giving rise to particles, the particleness is so much more evident because that's all you can ever see is the particles. You'll never see like many particles in the same quantum state, looking like a classical field.

0:14:44.3 SC: Indeed, it's kind of fascinating how important all of these details are to how matter behaves in the universe. If electrons were bosons, you wouldn't have matter in the way that we normally have it, right? There'd be no... Nothing that stops electrons from just piling on top of each other. All atoms would look like hydrogen basically and molecular structure would be completely different. The world would be entirely different if that were not true. And likewise if you couldn't have long range fields mediated by bosonic quantum fields, long range forces, I should say, then gravity and electromagnetism would be very different. It would also be a very, very different world that we live in. So it's the interplay of both these possibilities that makes the world as interesting and allows us to have complex structures like you and me.

0:15:42.8 SC: I'm going to group two questions together. One is from Diana David Russ who says, if an observer had complete information about the microstate of the universe so that coarse grained entropy would be zero, would macroscopic irreversibility like cooling, mixing and thermalization still occur? And if so, what is the explanatory role of entropy in your view? And then Anonymous says, if I know the state of a black hole in extraordinary detail, so that my uncertainty about the black hole state is less than its surface area, does the black hole shrink?

0:16:07.2 SC: Both of these questions are aiming at the idea that we can think of entropy as a measure of our knowledge of the microstate of a system. Now, you can probably... Many of you are going to guess what I'm going to say about this, is that yeah, that's one way of thinking about entropy, but it's not the only way. And different contexts will call for us to think about entropy in different ways. So both of these are thinking about entropy in the sort of Gibbs/Shannon way that we usually... We usually label it that way in talking about entropy, Gibbs is the guy who gets credit for the equation of entropy that says if I have a probability distribution over what the microstates of some system could be, then I can have a formula.

0:16:57.2 SC: If you want to know what the formula is, if the probability distribution is P, then the formula is minus the integral of P times the log of P. Okay? So given a probability distribution over some microstates of a system, you can calculate an entropy. And of course the probability distribution that you use would naturally depend on how much you know about the system. If you know the exact microstate, then you might just say the probability distribution is one, for it being in that microstate and zero for it being anywhere else. You could run that formula and you find the entropy is zero. Whereas if you just have a knowledge of, it's in the Maxwell-Boltzmann thermal distribution with a certain temperature or something like that, then you get a much larger entropy.

0:17:41.2 SC: So Diana's question is, if you know the microstate, then there's no entropy. Do all the phenomena that we usually associate with the second law of thermodynamics, like cooling, mixing, thermalization, do they still happen?

0:17:56.6 SC: Well, it's a slightly difficult question to give a convincing answer to because you're speaking two different languages in the same sentence. If you knew the complete microstate of the system, then the natural way that you would talk about what happens next, is in the language of the complete microstate of the system. You wouldn't use words like cooling, mixing, thermalization. You would just say, here is where the atoms go, here is where the molecules go. Because I'm Laplace's demon, I have all the complete information about the microstate of the universe. You know, why would I ever use words like cooling, mixing, thermalization? Now, would they still happen? Of course they would still happen. Your knowledge of the microstate of the system does not, at least in some classical approximation, affect how it actually changes over time. So that brings up the other way of thinking about entropy. You don't need to think about entropy in terms of your personal knowledge of a system.

0:18:54.3 SC: You can do the Boltzmann way of thinking about it, which is to say that there are macrostates which are objectively defined, once and for all, we say that given the microstate of a system, it is in a certain macro state. You can justify how you've defined the macro state by saying, here is what we can macroscopically observe about a system. But you don't have to. You can just define some macrostates, and then you define the entropy as the logarithm of the number of microstates in the macrostate, and that's objectively defined. And then you can define cooling and mixing and thermalization and all those things, and it would absolutely still happen that no matter how much you know about the system, it would behave in exactly the same way. So it depends on how much you know and what your purposes are, which kind of definition of entropy you're going to be appealing to. The same thing is true for black holes. If you know the state of a black hole in extraordinary detail, it makes no difference whatsoever to the state of the black hole, especially to its surface area. There is an extra complication there because of Quantum Mechanics, right?

0:20:00.6 SC: The only reason why the idea of an entropy of a black hole even makes sense is because of Quantum Mechanics. And in particular, it's very tempting. This is not something we understand 100%. And there's some people who are skeptical. But most people who think about this at a deep level think that we could think of the entropy of a black hole, at least a mature black hole. Again, this is part of the controversy, as entanglement entropy. That is to say, there is some set of degrees of freedom that describe the black hole itself, some set of quantum degrees of freedom that describe the outside world, and they are entangled with each other. And it is a feature of Quantum Mechanics that even if you know everything about the universe, subsystems of the universe can still have a non-zero entropy.

0:20:47.1 SC: And that's because there's no description of the individual subsystems in terms of a pure quantum state. That's because of entanglement. So your knowledge, when you say if I know the state of a black hole, you can't know the state of an entangled subsystem of the universe because there is no such thing as a pure state. There's what we call a mixed state or a density matrix or a density operator that describes the entangled subsystem. But that's irreducible. That's the most information you can possibly have about it. And it still has a non-zero entropy. So your knowledge of the black hole does not change its surface area or make it shrink whatsoever. So in both questions, the answer is exactly what you would intuitively expect. Your knowledge does not affect how this independent physical subsystem of the universe behaves.

0:21:37.7 SC: Chris Hershey says, in your Bluesky post on September 18th, you contrasted world first reasoning, which would give you a halfer perspective in the Sleeping Beauty puzzle, with observer first reasoning, which would give you a thirder perspective. A few puzzles are rare cases where my credence diverges from yours. Broadly identify with an observer first conscious perspective. Shouldn't we take seriously the idea that both perspectives are literally true and valuable, that our individual consciousnesses are a distinct branch of a conscious multiverse, while simultaneously coexisting in the same physical universe? Why shouldn't we build a foundation that presumes or predicts many observers as its cornerstone?

0:22:18.1 SC: So I think I'm going to do a less good job of addressing the question, the deeper we go into the later half of the question. I can do the first half pretty well. Just remind everyone the Sleeping Beauty puzzle, you've heard it from me and from others a while back. But Sleeping Beauty is put to sleep on Sunday. A coin is flipped. If the coin is heads, she's awoken on Monday, if... Or awakened, I suppose. If the coin is tails, then she's awakened twice, once on Monday, once on Tuesday, and she's given some drug that makes her forget that she was awakened. So there's three possible states Sleeping Beauty can be in when she's awakened. It can be the coin was heads and it's Monday. The coin was tails and it's Monday, and the coin was tails and it's Tuesday. They're all completely the same from her perspective when she wakes up and she knows this ahead of time. And the question is, when she wakes up and you ask her, what is your credence? That the coin came up heads. And there's an argument the credence should be 50% because it's a fair coin and nothing that she does when she wakes up teaches her anything. So if it was 50% before she went to sleep, it was 50% after. There's an argument that it should be a third because there's three possible places that she could wake up.

0:23:30.0 SC: Three possible centered worlds, as David Lewis would have said, and they're all equal. And so she could give equal credence to any of them. In particular, if she was told we know it's not Tuesday, that was extra information. Then she would give equal credence to it being heads or tails, because now it's Monday and there's only two choices. But likewise, if she was told the coin came up tails, then she would give equal credence to being Monday or Tuesday. And so therefore the three things have equal credences just by transitivity. So those are two arguments. And, you know, I would... Let's be clear about what my own point of view here. I'm not sure if I ever... I've talked about it and, you know, my own views are a little bit fuzzy here, so I'll try my best to be clear about it. When I talk about world first reasoning versus observer first reasoning, I don't want people to get the impression that I'm thinking like, that this first world... The word first. Sorry. The word first in observer first or world first is somehow a matter of priority, or what we care about the most, the world or the observer.

0:24:37.7 SC: It's just a matter of which we count first. It's literally first in the sense of coming first in a sequence not being more important. Okay? So do you assign credences or probabilities to worlds first and then count the observers within them? Or do you count observers in all the worlds and then assign credences to being all of them. That's world first versus observer first. So to answer the question, should we take seriously a foundation that presumes or predicts many observers as its cornerstone? I have no idea. I don't... It would have to be much more carefully, quantitatively spelled out exactly what that was supposed to mean before I would go one way or the other on that. But as far as Sleeping Beauty is concerned, I am of the opinion that, you know, it's not that there's a right or wrong answer to Sleeping Beauty, it's that you have to define the question a little bit better. You have to operationalize it. What do you mean by what is your credence? You have to be a little bit more specific. In other words, both... I think that both halfer or thirder perspectives can be valid as long as we are super duper careful about what we mean.

0:25:47.8 SC: So if what you mean by your credence is, I'm going to imagine doing this experiment over and over again many, many times, and I'm going to make bets, me versus Sleeping Beauty, about is it heads or tails that the coin came up up. In that case, Sleeping Beauty had better assign one third as the probability of any of the different three possibilities because that will be the long run way that she can at least break even making these bets. But that's a slightly different question than asking, you know, is the coin fair? Her waking up twice on Monday versus Tuesday doesn't change the fairness of the coin. In particular, if you said, well, okay, you're going to flip the coin again and not do any of this weird waking up stuff, what is your credence the coin will come up heads again. It should be 50/50, right? So I think that's a little bit falling short of the rigorous standards of writing a philosophy published paper. But I think that, roughly speaking, that's the kind of way that Sleeping Beauty paradoxes or puzzles should be resolved. I think it's a little bit less clear in the real multiverse case where you...

0:27:01.1 SC: Or not even the multiverse case, but in the cosmology case, let's put it that way, where you say, I don't actually have two different things that will happen in my lab, I have two different theories of the universe. And one theory of the universe is, you know, there's a big universe out there with many, many observers. The other theory has a relatively small number of observers. I have my preferences or my ideas about how to reason in those cases. But I think it's a slightly different case. I think it's important to get both of them right depending on again, what your purposes are by asking the question.

0:27:33.8 SC: Rain Suchek says, in episode 320, your solo on complexity in the universe, you talked about available information being a resource that complex systems can use up. But what did you mean by systems using it more cleverly? I thought you meant more efficiently. But what would push systems towards using it more efficiently if systems weren't anywhere near running out, or are they?

0:27:55.3 SC: I didn't mean more efficiently, I meant more cleverly. What I mean by that is, can you think of things to do as a complex system that take advantage of the resources around you in ways that might involve a little bit more effort in putting something together but get you a better outcome at the end of the day? You know, it's just like, are you more clever at, I don't know, designing a car or cooking a meal or something like that? There's like the most basic obvious thing to do. If you're a star and you have a low entropy fuel, you burn it. Right? That's not very clever. I mean, you're using up your low entropy resource, but you're just, you know, making hydrogen helium into heavier elements. Okay? It's not a lot of moving parts there. There's not a lot of intricacy or complexity there. Whereas a human being also uses the low entropy environment around them, but they use it to like think about the world, to develop models of the world in their head that they can manipulate and use to invent new things and, you know, make their lives better and come up with wonderful ideas. That's a more clever use of the available information resource.

0:29:07.6 SC: Kevin D., says, in Star Trek teleportation technology. Sorry, in Star Trek, teleportation technology painlessly vaporizes you where you are and creates a perfect copy down to your last pre-vaporization thought at your destination. Would you use it if it was the only way to teleport? Do you think the post teleport individual is the same person or merely a continuation of the narrative? Since the teleporter is constructing you from the same model, do you think identity is a continuity of thought or of the chain of computations that believes itself to be you?

0:29:41.2 SC: Yeah, people have thought about this before unsurprisingly. There's a famous paper that we actually read in the Philosophy of Cosmology course that I taught this year from Derek Parfit, who literally talks about exactly this question. He uses the example of a Star Trek teleporter machine to argue for the fact that what counts in personal identity is some kind of continuity of psychological structures, not necessarily continuity of physical pieces of matter. So I would have no problem climbing into that teleportation technology if I really truly believed it would never make a mistake. Of course, there's all sorts of Star Trek episodes about things going wrong, malfunctions with the transporter machine. So that might lead you to having second thoughts there. But I don't think that there's any mystical essence of self that gets handed down from moment to moment in my physical body. You know, like, you should be unsurprised that I think that.

0:30:36.8 SC: I think that I am an emergent result of the collective behavior of all sorts of atoms and particles. And if they were slightly different atoms and particles, but doing the same collective thing, it would be the same me. It's the Ship of Theseus problem, right? Like, in fact, your body does actually replace its physical matter, much of it over the course of time, but the pattern stays continuous in some way. Think about it this way, what if there was a teleportation machine that vaporizes you where you are and creates a perfect copy in the same place five minutes later? So there's just exactly the same you picking up exactly where you left off, but you were not there for five minutes, would that bother you?

0:31:19.0 SC: And if it does bother you, what if we decreased the interval from five minutes to one second? What if we decreased it from five minutes to a nanosecond so you literally wouldn't know. We just destroy you and recreate you more quickly than any other... Any person could even notice, would that bother you? I think it shouldn't bother you. If it does bother you, I get why it does bother you. But the reason why it bothers you is because we're not used to thinking of things like this, okay? Our intuitions, our experiences of the world don't equip us to have ways to think about these kinds of situations. They're interesting science fictiony thought experiments. But that's exactly why your intuition shouldn't be a good guide to how to think about these situations. You need to, like, have some rational criteria for going beyond what your intuition actually says.

0:32:12.8 SC: Okay. Antoine Chopin says, I just finished reading a book called 'A Guide to Making Friends in the Fourth Dimension' by Tibby, a YouTube science communicator in which she tries to help the reader imagine what a fourth spatial dimension would mean. My question is, could it be that there are more spatial dimensions which we are not aware of, e.g, because they are tiny? And can we think of an experiment to find out whether they exist?

0:32:36.9 SC: Well, yes, and yes, I'm sure some readers are going to be very, very familiar with this story, and some might not be. But this is an idea that came up, you know, early in the 20th century when people started thinking about space time as a thing. Before then, you had space and it was three dimensional. And there was a certain inevitability about that because it was all people ever thought about. But once you start thinking of the geometry of space and different kinds of geometries and stuff like that, I mean, not only with Einstein, but with other people as well, people started thinking carefully about the idea that, what if there are extra dimensions of space? So two things, certainly conceivable that you could imagine a universe with extra dimensions of space. Indeed, the famous book called 'Flatland' by EA Abbott or Edward Abbott. Abbott is his last name anyway. It's a wonderful little tale. It's actually a social satire about classism and things like that. But it tries to sort of soften you up to think about the geometry of dimensions by imagining a world that is only two dimensions of space. And then what would it look like if this two dimensional world encountered three dimensional people? But all the people are geometric shapes. So if a three dimensional sphere... If a sphere that lived in three dimensions encountered flatland, so there's really three dimensions, but there's a two dimensional subspace of it that certain organisms were confined to.

0:34:06.9 SC: If a sphere passes through that two dimensional plane, it looks like first just a dot and then a growing circle and then a shrinking circle and then the dot that disappears. Right? If you think about a sphere passing through a plane. So you can play the same game, of course, with what if a four dimensional creature interacted with our three dimensional world, and people have done that. In modern physics, the idea of extra spatial dimensions is a very popular one, going back to Kaluza and Klein soon after the invention of general relativity. There's a very well known fact that if you play around with the idea of extra dimensions of space and you compactify them so they're too small to be seen, symmetries of those extra dimensions can show up in our three plus one dimensional world, three dimensions of space, one dimension of time as forces, because the forces of nature that we know about are based on gauge symmetries. And you can invent them by having them be symmetries of the extra dimensions of space. So things like electromagnetism can come out of having a curled up extra dimension of space. And so that was always considered to be an intriguing idea.

0:35:17.7 SC: Einstein was intrigued by it, for example, but there was no evidence for it or anything like that. And then it comes into the 1970s when people realized that String Theory was a promising theory of Quantum Gravity. But String Theory only really naturally works if space time is 10 dimensional and there are footnotes there because there are limits or it looks 11 dimensional and whatever. But there's this idea that the natural place for strings to propagate is in a 10 dimensional space time. We don't live in a 10 dimensional space time. What can you do? Well, you can hide the other dimensions by making them tiny or by making them warped in some way so that the geometry is different. These are absolutely viable ideas that people take very seriously. Are there experiments? There's absolutely experiments that could show that they exist. These experiments have been done. There is zero evidence right now that they do exist other than the theoretical virtues that they have of helping us to understand gravity and perhaps other problems. Like there's still this possibility that the famous hierarchy problem of particle physics, the difference between the high energy scales of the Planck scale and the [0:36:29.5] ____ scale, are so different from the low energy scale of the weak interactions because of extra dimensions that are warped in their geometry somehow. So this is still very, very viable ideas, still don't have any direct evidence for it experimentally.

0:36:46.7 SC: DMI says, in last month's AMA, you answered a group of questions about dimensions of Hilbert space. You said there's a subtlety that if you ask about the position of a particle, the number of measurement outcomes is a countable infinity. Can you explain why it's a countable infinity rather than either an uncountable infinity or a finite number?

0:37:05.5 SC: Yeah, I forget exactly what I said. I hope I said true things things. But there are subtleties here that are worth getting right. So this is necessarily going to be a little bit technical, but I'm going to start at a level that is less technical than DMI's question and I'll work my way up just so the people who are not already on the bandwagon, can get something out of the answer to this question. So you've heard me say that in Quantum Mechanics, the space of all quantum mechanical states is called Hilbert space. Basically, a quantum state of a system is a vector. You can add them together, you can scale by complex numbers. So there's some space of possibilities called Hilbert space.

0:37:41.3 SC: It's a vector space. Every vector space has a dimensionality. And so individual subsystems, like the spin of an electron we talked about, have a factor of Hilbert space that has a certain dimensionality for the spin of a spin one half particle. It's a two dimensional Hilbert space for something like a single particle moving in one dimension, moving on a line, maybe rolling down a hill. This is what you would study if you were taking Quantum Mechanics 1 as an undergraduate. You would have non-relativistic system, the first thing you would study. You study other things as well. But you would have a non-relativistic particle moving in some potential energy and you would have the Schrödinger equation and maybe it's a harmonic oscillator or a square well, and you're solving the Schrödinger equation, okay? In that case, the Hilbert space that defined... That describes that system is infinite dimensional. So it's very easy to get very physically realistic systems that have infinite dimensional Hilbert spaces to describe them. We don't know in the real world whether Hilbert space is finite dimensional or infinite dimensional. You might think that if a single particle has an infinite dimensional Hilbert space, then surely the universe does.

0:38:51.9 SC: And that's a reasonable thought. But there is gravity in the universe and gravity provides an extra cutoff on the number of dimensions of Hilbert space which might render it finite. We just don't know. Okay, but to get to this particular question, there is a tension between two different things that you might learn in your Quantum Mechanics course regarding the infinity of the dimensionality of Hilbert space for just a single particle moving in one dimension. Okay? Some of you might know there's a difference between countable infinities and uncountable infinities. Countable infinities are the number of integers, for example, right? Zero, one, two, three, minus one, minus two, minus three. All those numbers are countable. We just counted them. We started counting them, it would take infinitely long to count all of them, but in principle you could do it. Whereas the number of real numbers is strictly bigger than that. If you include Pi and e and the square root of two and all those things, these irrational numbers, then you get an uncountable infinity. And the big contribution of Georg Cantor 150 years ago or whatever, was to prove that the uncountable infinity is a bigger number than the countable infinity.

0:40:04.3 SC: So if I say that the Hilbert space of a single particle has an infinite dimensional Hilbert space, do I mean countable or uncountable? And the answer is, I mean countable. There are good reasons to believe that we're not going to get to here, that the physically relevant Hilbert spaces for physics, you can define Hilbert spaces mathematically, but the ones that matter for Quantum Mechanics are what are called separable Hilbert spaces, which is just a fancy way of saying they are countable dimensions. The cardinality of the dimensionality of the Hilbert space is a countable infinity, not an uncountable one. So that's a fact that is shared in various Quantum Mechanics books and things like that. But there's another fact, which is that you can think of quantum states as superpositions of different measurement outcomes of different things you might like to measure. Right? One of the reasons why we know that the Hilbert space corresponding to the spin of an electron is two dimensional, is because there's two different measurement outcomes. You can get clockwise or counterclockwise, what we sometimes call spin up or spin down. How many measurement outcomes are there for the position of a particle in one dimension?

0:41:16.1 SC: Infinity. Right? I can measure the particle to be at any location on the real line from minus infinity to infinity. By the way, the number of numbers between zero and one, that is to say, the number of numbers on an interval is the same infinity as the number of numbers from minus infinity to infinity on the whole real line. So this is just one of those subtleties of infinity. Any continuum is going to have the same number of numbers. You can map an interval of the real line to the whole real line one to one. That's the criterion you use. So even if your particle can't go from minus infinity to infinity, as long as it can go anywhere on a continuum, there's an uncountable number of possible measurement outcomes. At least it seems that way, doesn't it? Because there's, for every X, for every position X, I could measure it to be at X. So that would seem to tell us that the dimensionality of Hilbert space is uncountably infinite. Because I can construct what is called a basis for Hilbert space, a set of vectors that are just basis vectors. You know, the vectors that point in every perpendicular direction in the vector space, which is...

0:42:29.1 SC: And the number of basis vectors is equal to the dimensionality of the space. And the number of basis vectors... If my basis vectors are position eigenstates, states of definite position, there's an uncountable number of them. So doesn't that make it sound like the number of dimensions of Hilbert space is uncountably infinite, not countable. It does, but you made a mistake when you made that argument. Not you, DMI, but I just made a mistake intentionally to give you that argument. Because those states, the eigenstates of position, states of definite position turn out not actually to be in Hilbert space, for... Just to be a little bit technical just for a second, for those who follow this kind of stuff. In terms of a wave function psi of x, the position eigenstates look like delta functions. Right? They look like delta of x minus x naught, where all of the particles located at x naught with 100% certainty. But delta functions aren't really in Hilbert space. They're not square integrable functions. You cannot actually take the integral of the delta function squared to get the probability or anything like that. So you've been cheating all along when you used the uncountable number of position eigenstates as basis vectors for your Hilbert space. This is...

0:43:49.8 SC: There's a whole subject of this that deals with this in functional analysis known as Rigged Hilbert spaces. And I talked about it a little bit in the Quantum Mechanics course I taught this year. No one ever talks about it, but it's a little mathy, but I think it's worth doing. But basically what you do is, you take the Hilbert space, you care about, the set of square integrable functions, for example, you make a smaller space out of smooth and nicely behaved test functions, and you also make a bigger space out of distributional operators on that smaller space. And things like position eigenstates live in the bigger space. So the real Hilbert space we care about is presented as a subspace of a bigger space that is uncountably infinite, even though the Hilbert space we care about is countably infinite in dimension. And then you can actually show that it's still possible to express states in the physically real Hilbert space as sums of basis vectors in the bigger space. And that's what we do when we express a state as a sum of either position eigenstates or momentum eigenstates. So I should not quite have said, I don't know what I did say, again.

0:45:04.9 SC: I don't think it's correct to say that if you have a position of a particle... If you measure the position of a particle, the number of measurement outcomes is a countable infinity. It's correct to the extent that the number of measurement outcomes of anything is at most a countable infinity. There's never a measurement outcome that could have an uncountable set of possibilities because the Hilbert space themselves are separable. They have a countable dimensionality. But the reason for that is that you can't actually measure position. Position is an unbounded operator. It's not something you can actually technically measure, which we know in the lab, right? If you're going to measure position, you're not going to measure it with infinite precision. Every measurement device is going to have some error. And if you look carefully at the different possibilities, everything looks countable at the end of the day. So the relevant factor is the Hilbert space for that particle has a countably infinite basis, not an uncountably infinite one. All right, I hope that... I hope you indulge me in a little technicality there. It's worth it to get technical sometimes, as long as we then switch gears. So let's switch gears.

0:46:14.1 SC: David Lifquist asks a priority question. Remember that priority questions are those that you're allowed to ask, and I will do my best to try to answer it. I get too many questions every month that I cannot possibly answer all of them. So I pick and choose the ones I think I have the most interesting things to say about. But if you really, really want me to ask a question, and you are a Patreon supporter, you get to ask a priority question once in your life. If you could do it an infinite number of times, then there'd be no point. Everyone would just always ask priority questions. So David says, I'm thinking a lot about how to interpret equations without confusing the geography for the map. I think there's a reality we try to describe and there's our modeling of that world. Isn't taking the wave function as real confusing the math for reality?

0:47:02.3 SC: Yeah, I think this is a legitimate danger, and I think that it depends on precisely how precise you want to talk about things. Okay? It's something that, it bothers people more in the Quantum Mechanical context than the classical context, but I get it. You should always try to be precise. So when you say, is the wave function real? There's two possible... Well, there's three possible answers. One answer is no, because the world, the physical world, has nothing to do with the wave function. That's a possible answer that some people have. Another possible answer is yes, the wave function is real because the physical structure of the world is precisely represented by a wave function. It's like, are the orbits of planets real? Right? Planets have positions, they don't have orbits that are physically out there in the world. They trace orbits over time through space time. But there's a mathematical idealization that goes into the orbit of a planet. So the third answer is wave functions aren't real because wave functions are just mathematical representations of reality, not reality itself. So the point is, I hope this is going to clarify a little bit. When people say the wave function is real, those people like me who are tempted to do that, they don't really mean a statement about Platonism, about mathematics or mathematical reality or anything like that. What they mean by that is that the wave function provides an exact representation of physical reality.

0:48:37.8 SC: But then it leads you to ask, okay, if you want to be a little bit more careful, a little bit more precise and persnickety, what should you say is real, because I don't want to say the wave function is real because it's math. I want to say the physically real thing is real. That's fine, I'm 100% in favor of that. But then all you can say is that reality is real, the universe is real. The universe isn't some other thing. It's unique, okay? It's all by itself. So you can say, yes, I believe in the real world. What is it? It's the real world. That's what it is. It doesn't get you very far to say things like that. I mean, they're true, okay? That's why I literally wrote a paper called 'Reality Realism' arguing in favor of this point of view. But it conveys a lot more information, if you say the world is a vector in Hilbert space, even if you don't really mean it at the very, very precise level, you mean the world is represented by a vector in Hilbert space, but you're actually giving people useful information when you say that, which is sometimes a thing that you want to do.

0:49:41.2 SC: Zach McKinney says, given your pessimism about the power of bottom up or grassroots approaches to depolarize cultural and political dynamics, how can regular individuals most effectively promote changes in platform or systems in which we may participate but have no direct control over the policies, incentives or algorithms that drive them?

0:50:02.6 SC: I'm not sure where your thought about my pessimism comes from. I don't think I've ever said anything quite like that. I have frequently expressed pessimism about the power of individual choices to enact worldwide change. In particular, I mean, of course, every worldwide change ultimately comes down to a large number of individual choices. I think that when you have a large scale social problem, you do not solve it by asking individuals to behave differently. I think that that's really not the way to do it. If you want people to use less gasoline, you don't just ask them to use less gasoline. You make it harder for them to use gasoline. You charge more money for gasoline. You provide alternatives to using gasoline. You change the system in some way. But changing the system is something that individuals do have some ability to do, right? That's...

0:50:52.6 SC: I very much believe that. So I'm all in favor of bottom up or grassroots approaches to various things. So in fact, I think that's the answer to the question, what can regular individuals do? Lots of things. You can always behave better yourself, whether it's voting or making your individual choices. I just think that those... If that's the only thing that happens, are a relatively minor impact on the world, you can try to communicate with the world, you can reach out, you can, you know, call your representatives, whatever that means. You can join groups that try to get the right messages out. You can donate to groups that get the right messages out. You can try to influence the groups that already exist, whether they're political parties or activist groups or whatever. You can be on social media, you can write letters to the editor, you can write op eds, you can do very small scale things like volunteering at homeless shelters if that's what you want to do. I think there's lots of things that people can do individually, but the point is you need to take that individual energy and channel it into something that actually has a chance of changing the system if you want to have a real impact.

0:52:02.2 SC: Calvin Firth says, I'm a young physicist and I'm currently experiencing my first big encounter with academic bureaucracy through grad school and fellowship applications. And it sucks. Are there paths one can take in physics or science still staying in fundamental research but having to deal with less bureaucracy than you would as, say, a tenure track faculty? If so, have you ever considered such a path?

0:52:24.6 SC: There might be various jobs out there that have a little bit less bureaucracy. I mean, I think that if you're a research scientist at a national lab or employed as a researcher in somebody else's lab, if you are working for some professor, but not being a professor yourself, then perhaps you can focus more on the actual work and less on the bureaucratic aspects. But let me tell you, there will always be the bureaucratic aspects. Again, this is, I think, something that is a feature of human nature, not specifically of academic physics. It does suck. It's terrible. You know, I do think that some places are a little bit better at having less bureaucracy. That was true at Caltech, for example. They were... They're pretty good at providing space for researchers to just be their best selves and not worry about sort of minor annoyances with dealing with paperwork and things like that.

0:53:20.1 SC: They had gotten less good at that by the time I was there than they apparently had been in the past. Johns Hopkins is maybe medium level at that. I've been at state schools where it's worse. [laughter] So there's slightly different things to do, but there's always those sacrifices like no job is only the fun part of the job and you need to both maximize the good parts of your job and minimize the bad parts. And if you're a person for whom bureaucracy is just especially anathema, like you just can't stand it, then being within academia might not be the place to be. Depending on the kind of physics you do, being in academia may or may not be essentially the only choice. If you want to do the kind of physics I do, or you're thinking about the nature of reality and the emergence of space, time or whatever. There's no industry that's really going to pay you to do that, roughly speaking. I mean, maybe there's something in Google or Amazon or Microsoft where they're doing quantum computing and you're sort of close enough to that that you can get a job there, but it's not easy. And of course there's trade offs there too. There's a lot less job security, there's a lot more secrecy, a lot... More difficulty in publicizing your results sometimes if the results might be worth a lot of money. So there's always trade offs. And I think you just have to accept that the world is full of trade offs, figure out what the trade offs are and make them with open eyes.

0:54:44.7 SC: Caroly Kantor says, do you think that physics will ever reach a final, fully complete explanation of nature, a point where we've fully gotten to the bottom of things? If so, how could that be possible? Or is it more likely that there will always remain some layer of fundamental entities or principles that we can describe mathematically but can't explain any further?

0:55:05.6 SC: So I'm not sure about the last part. I had an answer ready to the first part here. You know, it's 100% possible that physics will reach a final, fully complete explanation of nature. Like, how in the world would you ever know that that was not possible? It's completely possible. It's also possible that it won't. Right? Just because nature's puzzles will turn out to be too hard. If you want a personal judgment, I'm on the side of thinking that I'm much more impressed with how much we've learned about nature and the universe than pessimistic about how much we have yet to learn. We might have a very large amount yet to learn, but the rate of progress over human history has been enormously fast, especially the last couple of 100 years. Right? So I would not tend to be pessimistic about how far we can get. Even if there is a final, fully complete theory of everything, I'm not that pessimistic that we will find it someday. Look, what are the timescales we're talking about here? You know, like Isaac Newton was only the 1600s. That's only a few centuries ago. How long are you going to give us to figure it out? People get very, very impatient. 100 years ago, we had just invented Quantum Mechanics. And so a 1000 years from now, how much will we know? I can't even imagine, but I presume a lot more.

0:56:20.9 SC: At the end of the question, though, you say, is it more likely there will always remain some layer of fundamental entities that we can describe mathematically but can't explain any further?

0:56:30.5 SC: So I'm not sure what that means. I'm not sure what kind of explanation you're looking for. I mean, it's a feature of fundamental physics that the way that we've been successful is by describing things mathematically. And if you've described something mathematically so that you know sort of what is the mathematical structure that corresponds to the thing you're describing, and you know how that mathematical structure behaves according to certain equations, I think you're done in terms of explaining it any further. I don't think that there is any such like, oh, but do you really understand it yet? I think that's the understanding that we're permitted to get out of fundamental physical creatures, fundamental physical phenomena, let's put it that way.

0:57:15.7 SC: David P. Reichert says, could you say a bit more about what the open mathematical issues are for many worlds? Or is the picture mostly complete? For example, I'm aware of there being some questions around deriving the Born Rule. What about decoherence, is that fully explained or pointer states?

0:57:31.0 SC: So I'm not exactly sure what you mean by mathematical. Do you mean as opposed to philosophical, for example? I think that there are definitely issues in many worlds, but they have different natures. I think that the probability/deriving the Born Rule issue is the one that I would say is most commonly a worry among people who know what they're talking about. There's plenty of people out there who don't know what they're talking about when it comes to many worlds, and they have a whole bunch of silly worries. The deriving the Born Rule worry, I think, is the most respectable one amongst people who are respectable. But I don't think that there's a lot of math remaining to be done. I think there's a lot of philosophy remaining to be done or maybe a lot of convincing people of existing philosophy. I think there's very good arguments explaining why you get the Born Rule in many worlds, but not everyone is convinced by them. So I guess in some sense there's still work to be done. There... The other big area that I wish more people paid attention to is what is known as the problem of structure in Everettian Quantum Mechanics. And what that means is basically in Everett you have one big wave function of the universe or one big quantum state, one vector in Hilbert space, to be a little bit more specific.

0:58:45.1 SC: And it evolves according to the Schrödinger equation. But we talk about it in all of these rich and complicated ways, right? We divide the world into subsystems in Everett you have branching of the wave function, you have decoherence and pointer states, like you say. And I would say that we know a lot about how to get from the sort of bare bones wave function or quantum state in Everett to the real world, to the world of our experience. But there's still a lot we don't know about that. So, yeah, I think that the structure problem. How does space emerge? How do fields emerge? How does locality emerge? Why do we branch in certain ways rather than others? All of those, I think, are the more pressing problems. For they're questions, they're not even problems. It's not like there's any argument that says, oh no, you can't do it. It's actually pretty well defined questions and it's just going to be work to actually solve them. So these are not usually raised as objections to Everett like the Born Rule thing is, it's just a set of questions we need to figure out. And physicists... Most physicists don't put a lot of effort into this because, you know, they kind of think they know the answer.

0:59:55.0 SC: At the end of the day, they might not know how you get there. This is why one is pushed towards philosophy departments, because the philosophers want to know, like, okay, not only when you open the box do you see Schrödinger's cat either alive or/awake or asleep/dead. But why do you see that rather than something else? And most physicists are like, well, we see it. What do you want to know? My job here is to predict the outcomes of observations. So there's the small, plucky band of people like Zurek and others who feel otherwise. And I think that... I think the proof of the pudding is in the tasting. And I think that the nice thing about people Like Zurek and others in the game who really try to understand decoherence and pointer states, is that the whole discourse started as trying to explain something that people didn't think needed explaining. And yet by thinking about it hard, we came up with a lot of really important ideas that have turned out to be very valuable. So that's why I think this stuff is worth doing.

1:00:57.3 SC: Alexandra Kojiomtis says, if a super intelligent AI or a radically different alien civilization derived a theory of the everyday world that was equally predictive, but mathematically completely unmappable to our fields and particles, like it used totally different ontology than fermions and bosons, would you then admit that the core theory is not a description of nature, but merely a description of the human interface with nature? Or do you believe the core theory is the unique optimal compression of reality?

1:01:26.3 SC: Well, of course, I certainly don't believe that the core theory is a unique optimal compression of reality. It's not done. It's not a complete theory. Right? It doesn't include black holes in the Big Bang and dark matter and things like that. Like, it's very obvious that the core theory is not all of reality. So I hope that would never give a different impression than that. As far as the actual question is concerned, I'm just not sure that the hypothetical that you pose is plausibly realistic. It depends on what you mean. There's certainly cases in physics where you have the same theory that can be expressed in two mathematically very, very different looking ways. Classic examples to this are in mechanics, you have sort of Hamiltonian way of doing mechanics where you say you have positions and momenta and there's an equation that tells you how they go forward in time versus a least action principle way of doing things where you don't talk about momenta at all. You talk about paths from one beginning position to one final position, and you assign an action to those paths.

1:02:32.5 SC: The word action doesn't appear in Hamiltonian Mechanics. It's the most important thing in the least action principle. And then you see which of these paths minimizes the action. So the words and the math you use are completely different, but all of the physical predictions are exactly the same. I don't think that those are two separate theories. Those are just two different languages for talking about the same theory. So I'm not exactly sure what you mean when you say if you had a theory that used a totally different ontology but was equally predictive, do you mean makes exactly the same predictions? In that case, it is mathematically mappable to our fields and particles like it makes the same predictions, that's the map. Maybe you mean something like a duality in Quantum Field Theory. Like famously, we have the AdS/CFT duality where there's one theory really, but there's two different ways of talking about that theory as Quantum Gravity in D plus one dimensional anti-De Sitter space or as a conformal Quantum Field Theory in a D dimensional flat space time. But again, those are just one theory. It's just two different ways of talking about the same theory, even though the ontology looks completely different. So I think that if you have two theories that are completely equivalent to each other, but different ways of talking about them, that's just one theory. That's not two incompatible theories in any sense.

1:03:55.6 SC: Nat W., says, six years ago in episode 60, your guest, Dr. Lynne Kelly, shared her journey into memory techniques, citing, for example, how ancient cultures used structures like Stonehenge as tools for cataloging vast knowledge. That episode stayed with me for a long time. Do you or any of your contemporaries exercise structured memory techniques like doc... Like what Dr. Kelly describes memory palaces, et cetera.

1:04:20.1 SC: Well, I don't really know. I don't really know of anyone specifically who does. I've had friends and colleagues before who've used various different things, various different tricks with like, I don't know, colored pencils and things like that, or techniques for remembering things that seem to work for them. You know, look, I have to... I get a lot of questions like this about my personal strategy for thinking, for writing, for doing physics and whatever. And I feel bad because I have to confess that I'm very unsystematic in my strategies. You know, I don't have any special ways of remembering things.

1:04:54.5 SC: I don't have any special schedule during the day that gets writing done. I don't have any special algorithm for coming up with ideas or anything like that. So I personally don't. I think that this is a personal individualized issue. I'm certainly not trying to say that something like memory palaces are not helpful to people. I think that they're probably helpful to some people and less helpful to other people. And there's, of course, there's always a trade off, right? Because you have to spend time learning the technique and then applying it to things and doing things. And so you have to know ahead of time, is it worth spending the time doing this technique? And I'm sure that over the course of my life I've been wrong all the time in making choices about what to spend time learning. This is not just true for mental techniques, it's true for anything in the world. Like, are you going to spend time learning this and then you'll be good at it, but you've spent all that time, or are you just going to be less good at it, spending less time? Yeah, you have to decide. So I love that episode. I thought it was great. I thought it was very useful information and you can make use of that information as you will. I don't personally think along those lines very often.

1:06:02.5 SC: Sukito Tay says, as I understand the value of a researcher is commonly evaluated by the amount of impactful measured by impact factor publications. If you have the authority and resources to improve the system, how would you do it? I don't think that's exactly how the value of a researcher is commonly evaluated. I do think people have a difficult time wrapping their heads around this. But different subfields of science behave quite differently when it comes to understanding the value of a researcher. And I think each individual, even very successful professional scientists, have their own field in mind and just assume that everyone does it the same way. But really they don't. Okay? In physics, for example, the idea of an impact factor of a journal is almost non-existent. No one cares about the impact factor of a journal. Plenty of famous scientists don't even publish in journals anymore. They just put their things in the archive and be done with it. Right? It depends a lot on the sociology and the organizational structure of the field you're in.

1:07:03.3 SC: For one thing, just because some fields are bigger than others. In a relatively small and focused field, like my area of fundamental physics, the value of a researcher is measured by reading their papers, by understanding what they've said and what they've done, and who cares about how many citations they have or whatever. Now, it's very quick that you get outside your specific area of knowledge where you can read the papers profitably. Like even someone who is, let's say an expert on the standard model of particle physics is not going to be an expert in Quantum Gravity or condensed matter Quantum Field Theory or something like that. Maybe they are, but they don't have to be. So when you're doing things like hiring somebody, especially hiring a... Well, let's put it this way. If you're hiring a postdoc to work with you, you're probably going to read their papers, you're probably going to look at the actual work you do. If you're hiring a faculty member, then hopefully there are some people in your department who know the work well enough to read their papers and comment on them. But not everyone in the department is going to be able to do that.

1:08:06.0 SC: So they might look at how many publications they have, how many citations those publications have. So things like the h-index, which is a way of taking... You know, it's a... I'm not going to give you the formula, but it's a way of sort of averaging out the number of citations for all of your papers or just the total number of citations, total number of papers, for that matter. Those things are things that people notice. Absolutely. But nobody thinks that they're the once and for all correct way of evaluating the value of a researcher. They're a very quick and dirty way of getting some idea, has this person had an impact or not? If you want to do something important like hiring them, you're going to have to dig much more deeply into it. And then there are other areas, unlike physics, there are areas where there's just too many people doing it, right? Like nobody is able to actually understand all the papers that they might want to read to understand the relative importance of different researchers. In particular, you might have, like I just said, you might have a situation where your department is trying to hire somebody and you are not an expert.

1:09:10.9 SC: But if you're at a small liberal arts college that only hires a person for their physics faculty once every 10 years, then you're probably not going to be hiring people who are already in your subfield. And therefore you have to go by more broad measures. You can't just read their papers and expect to understand them. But still, I think that citations are more important than impact factor of journals. The idea that there is a hierarchy of importance or prestige of different journals exists in physics, but it's very, very weak. Okay? There's some idea that being physical review letters or under certain circumstances in science or nature is good, but beyond that there's almost nothing. Whereas in something like philosophy, there's this elaborate hierarchy of exactly how prestigious it is to be in this journal or another one. And that's where the impact factor comes in. For those of you who don't know. Sorry, I should have said this at the beginning. Impact factor is not about a paper, it's about the journal in which the paper appears. A journal with a high impact factor basically gets a lot of citations for his papers overall. So it's kind of a proxy for how good the journal is, roughly speaking.

1:10:22.9 SC: But that's very, very crude. I would much rather look at the actual number of citations that this paper has than to say, well, oh, this paper appeared in a journal that usually gets a lot of citations. That's very weak. If I had the authority and resources to improve the system, how would I do it? You know, read papers, talk to people, you know, actually figure out whether or not their ideas are any good. By the way, this is an issue these days because I'm on the board of the editorial board of several journals. And what do you do when someone submits a paper and a substantial amount of the paper has been written by AI? What if, you know, it's not that the whole paper is written by AI? And what if the paper is really good? What if a person has used AI and cut and pasted from the AI output into their paper, but has made sure that everything makes sense and it's a really good idea? Would you reject that paper because there was AI input? What if they used AI but then rewrote what the AI said in their own voice so it was them?

1:11:24.7 SC: Is that any different than like looking on reading other people's papers and getting ideas? This is going to be a big deal, moving forward, I can see arguments on any side. I do think that one reason to want to resist... I mean, so there's an argument that says, look, if the result is good, then it's good. Who cares whether a person came up with it or AI? But a counter argument, not necessarily a definitive one, but a counter argument is one of the things we use papers for is evaluating their authors. Right? Maybe we want to hire them, we want to give them a prize, and you want to invite them to a conference to give a talk. And if you're telling me that the appearance of a name as the single author of a paper is a little bit untrue, then that's a very important piece of knowledge to have. So I'm less worried about improving the system as to just preserving what little aspects of the system are working right now.

1:12:23.3 SC: Thomas S., says, could there be a fifth fundamental force?

1:12:26.7 SC: Sure, no problem. There could be lots of things. You know, the weak interactions weren't even that easy to find in the first place. We only found the weak interactions so quickly because the neutron decays and that turns out to be a manifestation of the weak interactions at work. And that's too obvious to miss, beta decay. But you know, as I've said before, even the idea for fundamental forces is a bit of an archaic way of thinking about things. There really are fields and sometimes we are able to describe them as forces. Sometimes it's less true. So maybe you could count the existence of the Higgs boson as a fifth fundamental force. You could count neutrino exchange as a fundamental force if you wanted to. We usually don't, but that's kind of an artifact of how things came about. But there absolutely could be fifth forces. There's like literally Google fifth force and you'll find a whole bunch of experiments that are looking for fifth forces. I've written papers about ways that you could look for fifth forces and also new ways they could be hidden from us. Right? Because we haven't found them yet. So as long as that force is either very, very weak or is very, very short range, or it only couples to exotic things like dark matter particles. Any one of those three ways could hide a fifth fundamental force from us very easily. So that's absolutely something to keep in mind.

1:13:49.4 SC: Alex Dubrow says, or Dubrow. Sorry, Alex. If time itself is fundamental, but the arrow of time is emergent from entropy, what prevents closed time like curves from forming, as in Hawking's chronology protection conjecture? Do you think the chronology protection arises from emergent consistency or from something more fundamental like in Quantum Gravity?

1:14:07.1 SC: Well, so Stephen Hawking had this paper where he proposed that chronology protection conjecture. This was in response to Kip Thorne and others saying that if I had a wormhole, I could use a wormhole to create closed time like curves, which would basically be a time machine. A time like curve in relativity is one that just moves forward in time. But in relativity, where spacetime can be curved, the idea of being moving forward in time is a local idea, like in your little region of space time, we all agree on what it means to be time like, to be moving forward in time. But when space time can be curved, as in general relativity, you could have a closed loop.

1:14:45.5 SC: That's what a closed time like curve is. So you're at every moment locally moving forward in time, and yet you still come to meet yourself in the past because of the global curvature of space time. Now, Hawking didn't just say, I think maybe it doesn't happen. Right? That was not the chronology protection conjecture. He proved a theorem that said under certain circumstances it doesn't happen. And you know what he actually proved was that if you try to make close time like curves in a compact region of space time where they didn't already exist, you're going to end up getting a singularity. Maybe you do get a singularity. Maybe that just happens, right? Maybe that's okay. But the implication is that both the singularity and the closed time like curves will be hidden behind a black hole kind of event horizon. That was not proven. But you know that's part of the cosmic censorship conjecture in some sense. So there's a chronology censorship conjecture as well. All of that, by the way, is classical general relativity. You know, the reason why close time like curves don't happen easily is just a matter of good old classical general relativity.

1:15:46.7 SC: You don't need Quantum Gravity or anything like that. In Thorne's work, where they had wormholes and they showed that you could make a time machine out of them, step one was have a wormhole. And that step is really hard. Part of what gets proven in these kinds of theorems is that you can't just make a wormhole with also, also without making a singularity. There are restrictions on the amount of topology change that you can have in general relativity. So I don't think it has anything to do with entropy. I don't think it has anything to do with Quantum Gravity. I think that classical gravity itself is really resistant to making close time like curves. Which still doesn't mean that you can't do it because classical gravity isn't the final answer. There is Quantum Gravity. Quantum Gravity, if anything is going to, maybe, maybe I don't think it's actually going to happen, but you could imagine it makes it easier to make time machines than classical gravity does. But you know, there's no strong reason to think that nature has closed time like curves in it. It's just something that people think would be cool. So if it turns out that nature doesn't, that's not a surprise from anything that we know right now.

1:16:51.8 SC: Taylor Gray says, any Christmas cocktails that you stick to or you're trying... Or that you are trying this year? I've recently been enjoying revolvers which are bourbon, coffee liqueur and bitters. But want to attempt making my own eggnog soon.

1:17:08.0 SC: Nope. I don't really have any Christmas cocktails. I'm not a really big believer in the Christmas cocktail idea. Like if a cocktail is good, I'm going to drink it all year round. If it's not good, then I'm not going to become a fan of it just because it's Christmas time. I am a fan of eggnog, either alcoholic or non-alcoholic. I'm so much a fan of eggnog, even the non-alcoholic version that I refuse to have it in the house because I will drink it and it's super bad for you. So, you know, maybe on Christmas Day, literally, we'll have a little bit of eggnog in the house and enjoy it, but I don't want it hanging around because I am definitely going to consume that. And I should resist. An interesting manifestation of the fact that you can, in part of your brain intellectually know you shouldn't do something, and another part of your brain is going to make you do it.

1:17:51.7 SC: Right? That's just a feature of being a human being. And the other thing is, I'm not especially big fan of sweet cocktails. Like, I think a Manhattan or a Sidecar are as sweet as I'm going to get. So things that involve coffee liqueur and even simple syrup, I try to avoid by a lot, or spoons full of sugar and things like that. I like more, I don't know, sharp. I don't want to say savory, because that's not quite the right word, but spirit forward, sharper, more complex cocktails. Although I got to say, the bourbon coffee liqueur and bitters sounds like it might be good. I might have to give that one a try. But I will feel free to have it the middle of July if I want to.

1:18:35.0 SC: Henry Jacobs says, what is different when you watch the greatest show in history a second time? Of course, I'm talking about 'The Wire'.

1:18:41.5 SC: Yeah, I'm watching 'The Wire' again. We just finished season four. We're very slow. We're not trying to... You know, we don't have a schedule here. But it's not the second time. It's at least the third time that I've seen it. And it's amazing, you know, cannot rhapsodize enough about the storytelling and the world building in 'The Wire'. We usually talk about world building in the context of fantasy or science fiction. But all the different moving parts of 'The Wire', the police, the drug dealers, the dock workers, the school workers, everything comes to life. And I do notice that, in fact, if there's any group that doesn't get a lot of positive press, it is college professors in 'The Wire'. The words Johns Hopkins are raised a couple of times and not usually in a positive light, but the richness of it, you can...

1:19:33.5 SC: It's absolutely a show that you can watch again and again. I'm presuming that you like it... One likes it the first time. Like I'm not aesthetic objectivist. If you don't like the TV show, then that's okay. You're not making a mistake. Maybe it's just not speaking to you. But if it's the kind of show you like, it is also the kind that you can watch again and again and get something different out of. Like, I can watch 'Raiders of the Lost Ark' again and again, but I don't get something different out of it. I just enjoy it in exactly the same way. 'The Wire' is sufficiently rich, and it very often does showing but not telling, right? So by the time you've seen it a couple of times, you're seeing things that you basically know what's going to happen. And you can pay closer attention to the details. And there's always details going on, and there's like, little people doing things, saying things, making jokes, whatever. You appreciate things a little bit better. So if anyone out there loves 'The Wire' and wonders whether they should rewatch it, it's very, very worth doing.

1:20:30.2 SC: Carl Einerson says, I'm still very unclear on how black holes are actually existing or talked about like they are. I'm sure someone told me the time dilation for anything approaching the event horizon relative to anything else behind that thing falling in is approaching infinity. So my question is, how is there enough time in the lifetime of the universe for anything to fall into a black hole and pass the event horizon relative to anything else in the universe? Or in other words, are there actually black holes in the universe, or just spaces of very slow passing of time relative to us?

1:21:02.2 SC: No, there actually are black holes in the universe, I'm here to tell you that. And it's a kind of subtle, complicated thing going on here. I mean, you're not crazy. For a long time before we understood black holes, scientists knew about the Schwarzschild solution to general relativity. Like it came along in 1917. You know, it was one of the very first things that happened after Einstein proposed general relativity. And black holes are there, right? They're implicit in the metric that Schwarzschild wrote down. But we didn't know how to deal with them, so people thought of them. One way of thinking about them anyway was frozen stars. Because as you see, if you look at the Schwarzschild metric, the solution to Einstein's equation for the space time that is spherically symmetric and doesn't contain matter in it, maybe except at the center where we're excluding our... What the metric is doing, then you find that like Carl implies, when you throw something into the event horizon from the outside world, you see it slowing down more and more. You see its clock ticking more and more slowly. Now the clock reaches sort of asymptotes to a certain time. There is a last time you ever see. You just never quite get to it, which should already give you the impression, or at least suggest to you, that if you were carrying the clock, you would still see it ticking and you would think that you yourself fall into the black hole, right?

1:22:19.6 SC: So in that sense, black holes absolutely do exist. The question is how we interact with them and the rest of the world. So they called them frozen stars because if you imagine a star collapsing, we would see the very end of the lifetime of the star still giving out light to us. We would never see it finish collapsing according to this way of thinking. Here's the problem with that, that, you know, that whole discourse imagines that the thing you're throwing in has absolutely no impact on the black hole. But in the real world, things you throw into the black hole have mass themselves. So you're changing the black hole by throwing something into it.

1:22:58.9 SC: In particular, it's going to grow bigger because the size of the black hole depends on its mass, right? So if you throw in a little bit more mass or energy, the black hole grows bigger. So in fact, this idealization where you're just throwing what we call test particles with zero mass or energy into the black hole isn't the relevant one. You throw a real thing into the black hole, the black hole grows a little bit, so its event horizon reaches out to swallow the thing that you've thrown into it. And indeed you see it fall into the black hole in a very real sense, none of this is real. Like none of this is something you need to worry about as a matter of practical consideration, the black holes are out there, very, very far away from us. And we can see how they behave and how they behave is like black holes, right? So the fact that you have this idealized description about what would happen if you threw a clock in there, is not that relevant to what happens to black holes in the real universe.

1:23:52.0 SC: Sam Hartzog says, gallons of ink have been spilled over Gen AI's Generative AI's potential for... Or I don't know what Gen AI stands for actually. Is it Generative AI or General Artificial Intelligence? I'm not quite sure. Because it's usually AGI for Artificial General Intelligence. So I think this is probably Generative AI's potential for disrupting the world economy. But very little of this discussion has centered around the staggering paradigm shift in the amount of capital these companies are raising before going public. Sinking a trillion dollars into a company with zero obligation for public transparency seems like a recipe for disaster.

1:24:27.7 SC: At a minimum, any firm that has raised sufficient capital to single handedly trigger a global recession should probably be forced to follow the same financial reporting rules as a public publicly traded bicycle manufacturer. And yet I've heard exactly zero voices in or out of government pushing this angle. What are your thoughts on the mind boggling level of capital tied up in these privately held firms? Should limits be set on this type of behavior more generally?

1:24:53.8 SC: So I'm certainly not. If I'm not an expert on AI, I'm absolutely not an expert on regulating corporate valuations and public responsibilities. So you should put very low emphasis on my opinions here. But I do basically share your worries about what is going on financially with these firms. There's this sort of very obvious thing that happens again and again in economic history where a new idea or new industry comes along. And we remember that in previous times when new ideas and industries come along, they've been very exciting and often we get bubbles, right? We get speculation and overinflation of valuations and eventually there's a crash and it's all bad. We had the dot com bubble, we've had the real estate crash, things like that.

1:25:41.9 SC: And therefore you might think that we would go, okay, next time, we're not going to do this. Next time, we're going to be much more responsible. But every time is different, right? And every time is new and every time there's new pressures and forces and individual companies don't want to be left behind, countries don't want to be left behind compared to other countries. People don't want to miss out on making millions, billions, trillions of dollars. So we actually are not very good at using lessons from the past to warn us against the future. Especially because the future is not going to be the same, right? Just because we had a particular financial crisis last time, that doesn't mean we're going to have another financial crisis this time.

1:26:24.3 SC: Or if we do have a financial crisis, it won't be the same kind of thing. I'm absolutely in agreement that as a matter of just governmental responsibility, we should be very demanding of very wealthy firms, firms with high valuations, to be transparent, to follow some rules, to have guardrails in place in all sorts of ways. Some people have been complaining about this. Elizabeth Warren has famously been on this kick for a very long time. And guess what? These firms have a lot of money to put into supporting candidates and politicians who will not slow them down in any way. So that's another thing that prevents us from doing the right thing. So I don't have specific ideas on what should be done. I'm very sympathetic to the idea that things should be done to protect us. That's a difficult thing to balance in the real world, I'm sure we will get it wrong, let's put it that way. Hopefully we don't get it too wrong that people get in too much trouble.

1:27:29.3 SC: Matthew Hall says, I was wondering why Maldacena's AdS/CFT correspondence is so widely cited, bearing in mind that we live in a universe closer to De Sitter space rather than anti-De Sitter space. I thought perhaps it was because the cosmological constant is small enough that a negative value is a good approximation to a positive value, at least on the scale of a black hole. But does that mean that AdS/CFT isn't any use during the inflationary period of our early universe, which looks more like something like a very positive cosmological constant?

1:27:58.1 SC: Yeah, very much true. Look, the reason why Maldacena's AdS/CFT correspondence, which by the way was put forward before we knew that we lived in something like De Sitter space. I forget exactly the year, but AdS/CFT was at the middle of the 1990s and the discovery of the accelerating universe was 1998. So they were close. But Maldacena got there first. It works, AdS/CFT correspondence. It's a pretty good well understood model of Quantum Gravity where you have a theory without gravity which is the CFT and it's supposed to be the same as a theory with gravity which is the ADS part. We don't have many examples of theories of Quantum Gravity that work. So you are correct that there is some argument that says look, as long as the cosmological constant is small, it doesn't matter whether it's positive or negative, right? Like if you're looking... The more technical way of saying that is, if you look at phenomena on... That are... Have a size that is small compared to the curvature radius associated with the cosmological constant, then it shouldn't matter what the sign of the cosmological constant is. Now it turns out I'm actually not a super expert here either, but my impression is at least last time I was talking to the real real experts.

1:29:09.3 SC: Exactly where AdS/CFT is not very well understood is when we get to length scales smaller than the cosmological horizon size or sorry, curvature radius. So maybe that fact is less impressive than it sounds at first. But certainly no one thinks that AdS/CFT is going to address all of the problems of Quantum Gravity in the real world. Both the future where we're going to be in De Sitter space with basically nothing around, or the past where we were very far away from an empty universe, are not very well approximated by AdS/CFT. So those questions are still good research questions that people are looking into, like how can we think about these conditions if AdS/CFT is not helpful?

1:29:54.7 SC: On the other hand, AdS/CFT has been super helpful in coming up with new ways to understand things like how information gets out of black holes. Right? Just the very fact that information does get out of black holes is even a more strong thing that you can get from AdS/CFT. So it's very, very useful. I do think that I wouldn't be averse to an argument that said, we've been concentrating on it too much because it's a little bit like looking under the lamppost, right, for our lost keys. It's very bright. AdS/CFT is super bright. You can see everything and you can do things. You know exactly what you're doing, but it's not like you ask, not exactly the questions that you're interested in. So that's still something that people are thinking about. I mean, we did a conversation with Andy Strominger here on the podcast where he talked about efforts to understand De Sitter space rather than anti-De Sitter space, it's harder to do. It doesn't have a conformal field theory as a dual description.

1:30:48.5 SC: Marie Roscue says, I recently re-listened to episode 27... I love it, we're getting evidence here in this month's AMA that people are digging into the archives for Mindscape, which is great. There's some good stuff back there. The audio quality was not always great because I was not always as good as I am. But in fact, I learned from the pandemic. I learned that audio quality is better when you do the recording remotely. I literally thought before the pandemic came along that I should try hard to always do the conversations in-person, which I mostly did, including that one with Janna Levin, because you can control everything and the audio quality will be better, et cetera. But it's actually not true. When you do it in-person, both people are in the same room and both people are talking into microphones. And so each person's microphone picks up the other person and it causes a mess. So as long as you can just give a good microphone to your guest, you can get better audio quality by being in separate rooms. So anyway, sorry about that. If you do dig into the archives.

1:31:47.3 SC: Anyway, Marie says, the episode with Janna Levin is fantastic for many reasons. One of them is the relaxed atmosphere and the sense of your long, good friendship with Janna. I wonder, what is your recipe for a long lasting friendship?

1:32:02.1 SC: That's a great question, but again, I'm going to appeal to the answer I gave when it comes to memory palaces or to things like that, I don't have recipes for this kind of thing. And I say that reluctantly because I feel like I don't want to disparage the question or fail to answer it or anything like that. It's a perfectly good question. I just don't work that way myself. I think that every friendship is different. It would be weird to have a recipe for long lasting friendships. Janna and I met when we were grad students. She was a grad student at MIT, I was a grad student at Harvard.

1:32:33.9 SC: And we were both learning some of the same things. We both came in the same year and she and I and Ted Paine, who's another friend of mine. Ted was a grad student in the astronomy department at Harvard at the same time who's also interested in the same thing. So the three of us took classes together. I mentioned Ted because he is the musician who is responsible for the intro and outro music of the Mindscape Podcast podcast with his band Euphonic. So, you know, yeah, we did a lot of... There were a lot of late nights, the three of us, doing problem sets for Quantum Field Theory or particle physics and cosmology, things like that. And over the course of an academic career, of course, you know, two people will meet and then move away and meet again and move away again. But Janna and I are very similar in many ways. Like we're also very different in some ways, but we're people who have broader interests than just cranking out physics papers.

1:33:26.5 SC: We had that interest, but also other interests. We're kind of quirky and eccentric in those ways. So I think it was a natural fit. And we have ended up over the years doing various events together and we're just sort of compatible in different ways. So I think every friendship is different. I've had friendships like, you know, people I see all the time, people who live nearby, other people who I don't see for 20 years. And when you see them again, it's like, you know, it never stopped. Right? I don't know. I'm trying to think of something profound to say about a long lasting friendship. I do think that I really have enormous respect for the fact that people are different and the fact that that's good, you know, people are different in different ways. Nobody is going to be 100% in agreement with you. Like, I'm not 100% in agreement with my wife, Jennifer. We're much more in agreement about things than we have any right to be, honestly. Like, we constantly marvel over the fact that, Jennifer and I am talking about, things come up in the course of a long marriage where they hadn't come up before.

1:34:29.9 SC: But we're generally pretty compatible. Like Christmas decorations, you know, Jennifer and I met in March of a year, and by October we were engaged to be married. We had never dealt with the question of Christmas decorations before. It turns out we have exactly the same opinions about Christmas decorations. So you can have a lot of things similar, but still there's going to be some things that are different. I love watching basketball. Jennifer is not that interested. She loves watching rodeo. I'm not that interested. But we're both willing to watch what other, the other person wants to watch because they enjoy it. I think that a lot of not just good friendships, but good marriages, good relationships in general, are appreciating the differences rather than fighting against them. Right? Knowing where you're not compatible just as much as where you're compatible and trying not to fix things necessarily. Like, if someone is doing something wrong, then by all means, let them know, but if they're just doing their thing and it's not your thing, let them do that. You know, you don't need... You shouldn't force yourself into that. They shouldn't force you into it. But you can still enjoy the things that you do overlap with. Figuring out just how to get the enjoyment out of good people and not trying to make it something it's not.

1:35:40.8 SC: And also, by the way, I would go, this is not necessarily part of the question, but I'm happy to stop being friends with people when I realize that there's parts of them that are not good people. Like, I don't actually have this feeling like that, oh, you should forgive your friends for everything. Right? Like, life is too short. And this is even more true with colleagues than friends. Like, there's some people who are useful in sort of they're powerful or influential or they get things done, but they're just not good people. I don't have anything to do with them at all. I'm sorry, I'm not interested in wasting my time on that kind of thing. So happily, Janna is one of the good ones. She's a really, really good human being as well as a great physicist and a great communicator so. So it's been my privilege to be friends with her for so many years.

1:36:28.5 SC: Ed Saidstuff says, you've said that reality is made of quantum fields, not quantum states, not information. Since quantum states encode information, what exactly distinguishes information from the things that are real? Is information ever ontologically fundamental? And if so, where is the computation happening?

1:36:45.2 SC: So my angle on this, and people disagree, so you don't take my word for it, but my angle is that information is never ontologically fundamental. I do have a paper coming out. I say I, it's an exaggeration to say I. There's a long list of authors, and I'm one of the authors. I think I'm literally the last author, because I came in last. But it's a group paper organized by Fernando Rosas about the meanings of information and how information means things to... Different things to different people. And one of those meanings is the philosophical meaning that perhaps information is ontologically fundamental. And so we said positive things about the possibility, but with a little caveat that it's a way of talking. It's not really necessarily what's going on. It's not that information is not important. Information to me is like energy or location. These are quantities that characterize physically real things, but they are not themselves the physically real things. Right? Energy, we often talk about energy moving from one place to another or something like that. But what we mean is something moved from one place to another or something interacted with something else, and the energy that we, as a number used to characterize the things transformed in some way.

1:38:06.3 SC: I think information is exactly the same way. Quantum states encode information. Yes, but similarly to what we said before. You know, we said that it's not actually this mathematical structure that exists, it's the thing that exists. Okay, so it's not even the quantum states that are real. Reality is real. If you want to go that carefully in the direction of speaking about reality. Reality is real. Reality is mathematically modeled by quantum states. It can also be mathematically modeled by information under certain circumstances, but under certain... Under other circumstances, that's less useful. I will mention, by the way, something that I do... We... I put in the paper that we are coming out with, that I... That rubs me a little bit. It's a slight pet peeve. There's a famous motto from John Wheeler, the famous physicist, It from Bit. Right? This is grown into the whole program, called It from Qubit. The idea that reality, or space time at least, can be emergent from quantum information in some sense. So people have the impression because they didn't actually read Wheeler's paper, that John Wheeler is saying that reality emerges from information, it comes from bit. That's not what he was saying. [laughter]

1:39:19.5 SC: He was saying something much more specific and controversial than that. John Wheeler, despite having been Hugh Everett's thesis advisor, was an acolyte of Niels Bohr. John Wheeler was a devoted Copenhagenist when it comes to Quantum Mechanics. And what Wheeler was saying was that there is no reality to quantum systems before they are observed. That's the Copenhagen line. That's what Heisenberg tried to convince people of. That it's not that when you have Schrödinger's cat inside the box, the cat is both awake and asleep before you open the box. What Heisenberg wanted you to say was, there's no such thing as what's inside the box before you open it. There's no way of talking about the reality, what's inside the box. Reality is just the outcomes of measurements. And that's what Wheeler was saying. That's the its he's getting from bits. The bits he's referring to are not just abstract pieces of information. They were the outcomes of quantum measurements. It's a very, very... Is honest, I think point of view. One that I think is a crazy point of view but at least you know, he was... Had the courage of his convictions there.

1:40:30.7 SC: Darren Vigliotti says, what is the best explanation you can give to a total non-science person who's asking about what came before the Big Bang? The thought process of the questioner being, there had to be a before and something can't come from nothing.

1:40:42.7 SC: Well, the only correct answer you can give to this is that we don't know what came before the Big Bang or if anything came before the Big Bang. How could we know? We have theories, we have theoretical models of the ways that the universe could be where the Big Bang was the beginning. We also have theoretical models of the way the universe could be where the Big Bang was not the beginning. So how could we possibly know? The evidence is not decide between them now, nor does a theory for that matter. But these arguments, like there had to be before, something can't come from nothing. The only response to them is why not? How do you know? Right? And if the person you're talking to is honest, they will interrogate why they think that they have these principles and they invent them for some reasons. And those reasons don't apply to the universe as a whole. Something can't come from nothing, why not? I mean, there's a slightly more sophisticated response, which is that it didn't come from nothing, it had a beginning.

1:41:42.0 SC: Those are two different things. The reason why they're different is because if you say the universe came from nothing, you're giving the impression that there was something called nothing and then that nothing became the universe. That is not the right way to think about it. The right way to think about it is just the universe had a first moment of time, full stop. There wasn't anything before that turned into the universe. There just was no before. Okay? It's not that different than saying the real line going from minus infinity to infinity, perfectly plausible concept. But you also have the non-negative numbers, right? The number zero and then all the positive numbers. There's no thing, no principle, no reason why just because there's a number zero, there has to be a negative number coming before it. And the universe might be like that. But the honest answer is, we don't know whether it's like that or not.

1:42:40.7 SC: Chris Shotard says, is it a theoretical possibility that there exist matterless black holes? Wouldn't that be an extremely low entropy object? And finally, could such beasts exist in nature?

1:42:51.3 SC: Well, yes and no. The original Schwarzschild metric that we were just talking about was, as I mentioned, for Einstein's equation in vacuum, that is to say in empty space. So I mean, the short answer is yes, there could exist matterless black holes. The, what we call the eternal Schwarzschild metric, the full one that extends all throughout space time, describes black holes without any matter. It also describes two black holes connected by a wormhole, which is an interesting thing that, you know, Wheeler and other people have thought about, you know, what does it mean that there is a wormhole connecting two different black holes? This is called the Schwarzschild wormhole. It has the very specific and provocative feature that you can't go through it. If you try to cross through the Schwarzschild wormhole from one black hole to the other, of course you're still stuck in a black hole. You have to hit the singularity. There's no way that you can actually emerge out the other side. But if you could go faster than the speed of light, then you could do it. So there's essentially no chance that this exists in nature. Black holes can have mass and energy all by themselves. They don't need to be made from matter. That's the lesson of the Schwarzschild solution.

1:44:06.7 SC: But in the real world, we think that at the beginning of the universe, there weren't any black holes. And you make them as time goes on through matter collapsing somehow. Now, you could make a black hole purely through gravitational waves hitting each other and sort of having a little bit of energy in each gravitational wave. But they all precisely collide so as to make a black hole. There's no reason to imagine that happens in the real world. Gravitational waves are very weak and relatively rare. But you could at least imagine it. Would it be an extremely low entropy object? No. Black holes would have the same entropy, no matter what they're made of. That's a feature of black holes. It doesn't matter what they're made of. You might wonder where the entropy comes from. And the answer is spacetime itself. This is one of the things that is an implication of Stephen Hawking discovering black hole entropy in the first place, namely that the entropy is not associated with stuff, with matter, with what made the black hole. As I mentioned earlier, the black hole entropy is most plausibly thought of as entanglement between the interior of the black hole and the exterior. And even if there's nothing in there, it's entanglement between the degrees of freedom defining space time itself inside the black hole versus outside.

1:45:22.2 SC: Dave L., says, how should we think of gravity's role in shaping the forms of matter and energy in something like a solar system? Without gravity, a cloud of hydrogen would diffuse and radiate away its energy smoothly as it cools. But with gravity in the system where... With gravity in the system, there is a storm of activity that creates wild temperature ranges, fusing nuclei and converting matter to photons. When nuclei fuse, where should we think of this compressive energy as coming from?

1:45:50.8 SC: So I'm going to try to answer the question. I hope that I get the spirit of it. You know, gravity is a special force in various ways. It's special at a deep fundamental level because it's a feature of space time itself, not a field living in addition to space time on top of space time, like everything else that we know about, but also because it's universal, right? Everything feels gravity, everything causes gravity. Not like electromagnetism or other forces that we know about, where some particles and fields source them and some do not, everything sources gravity. And finally it only ever pulls things together. I know that there's the acceleration of the universe with dark energy.

1:46:30.3 SC: That's kind of a special case. But even that's not really pushing things apart. It's allowing things to keep moving apart that always were moving apart. The gravitational force between two objects is always positive. It always pulls things together. You can think of that as because gravity is proportional to mass, which is proportional to energy, and energy is always positive. So it's a long range force that always pulls things together. That's very different than electromagnetism or the forces of pressure that you get from matter. And that leads to a lot of the behaviors that we think of as complex in the universe. There's a competition between different kinds of forces acting in different ways.

1:47:12.5 SC: Pressure acts in one way, the various effects of electromagnetism act in a different way, and gravity acts in yet a different way. So things become complicated. When you look at simplifying cases like a box of gas where gravity can be neglected, the individual atoms or molecules of the gas can be thought of as just point particles bumping into each other. And it's all kind of simple. There's not a lot of possible ways things can be interesting. When you have gravity and the possibility of particles bumping into each other, that's when things get interesting. And gravity is absolutely contributing to the energy. Right? So where does the energy come from when nuclei fuse? There was potential energy in the system to begin with because all the matter was spread out and they had gravitational forces between them, all of those particles. And therefore you could decrease the gravitational energy by bringing things together that increases the energy of other things, which causes nuclei to allow to form.

1:48:08.1 SC: Mads Peter says, I recently read Andy Weir's book 'Project Hail Mary', which soon will also be a movie. In it, some interesting physics occurs that I would like your opinion on. Short version, a human meets an alien pilot at a star system. The alien species has stumbled upon an incredibly efficient fuel that allowed them to travel there. The catch is the aliens never discovered relativity. So the alien pilot is confused about the trip taking shorter time and using less fuel than expected. He has no contact with his home world after leaving. My question is this. Would the pilot actually notice any effect of relativity if he had no contact with where he traveled from after leaving? Wouldn't he just assume that he traveled faster than light and all of his calculations make sense from that perspective?

1:48:51.3 SC: I think it's a little bit subtle. You know, I haven't read the book, so I shouldn't really be commenting on precisely what was said in the book. You know, I'm hearing it thirdhand. For one thing, I'm very skeptical that any aliens would be able to build interstellar travel and have incredibly efficient fuel systems, but not have yet discovered relativity. It's hard to not yet discover relativity, as is often pointed out, various things that very subtly exist in our observable universe wouldn't be the same way without relativity. Like, you couldn't really understand particle physics without relativity, right? Where does the energy go when nuclei decay? You kind of need relativity to figure that out. So to get that advanced level of technology without discovering relativity is a little bit implausible. Here on Earth, muons would not reach the Earth from cosmic ray collisions if you didn't take special relativity into account. The GPS system would not take... Would not work if you didn't take general relativity into account, and so on. So anyway, that's okay. It's a novel, I get that. That's okay. And I'm... So I'm not quite sure in the question, how the alien got confused about the trip taking a shorter time.

1:50:04.1 SC: I don't know how he's measuring time. I guess if you... If he measures time by his own clock or her own clock, I suppose, by their own clock, the alien presumably has some notion of velocity. Has some feeling that the propulsion system they have will make the ship move at a certain velocity, and therefore does distance equals speed times time to figure out how long it should take. That kind of calculation could go wrong. So it could be that they get the wrong answer from their own point of view. If they had some pre-existing feeling for what that was like because they thought that they knew what speed they were moving at.

1:50:48.8 SC: So in other words, they could do a calculation that says, I have so much fuel, so much fuel accelerates me to such and such a speed, I know what distance I'm going at, therefore I think it will take this much time. So I think that's a plausible confusion that the alien could have. If the alien wasn't expecting a certain amount of clock time to pass, but rather just arrived at the Earth, then it would be harder to be confused because there's no signpost here on Earth saying here is the universal time that it is. That's something that doesn't exist in the real world in relativity. So the alien wouldn't be able to say, well, we all agree on what time it is in the universe, and that's a different time than I expected. So I think it depends on some of the details of the storytelling there. But it's absolutely possible that the alien could have gotten confused if they were keeping track of the right things.

1:51:39.1 SC: I'm going to group two questions together. Elise Cutts says, what's a popular metaphor used to explain physics that needs to be retired and Jared says, do you have a least favorite scientific analogy, one that has been oversimplified and misconstrued, leading to widely held misconceptions, conversely, you have a most favorite that elegantly describes what's going on?

1:51:56.4 SC: You know, I have a mixed set of feelings about scientific metaphors and analogies in general. On the one hand, they're necessary. Like, I don't begrudge their use. I don't think that you should always just tell the complete, honest truth about scientific concepts. You shouldn't lie about them. But there's a journey that people have to go on from wherever their thoughts about how the world works are to where you want them to be by explaining something scientific. And the use of metaphors and analogies and illustrations and stories and whatever is 100% fine. On the other hand, there's a big problem with using analogies, that sometimes people don't want to admit that this problem exists, but they really got to admit it.

1:52:38.2 SC: Namely that the analogy is with certain features of the things you're trying to describe. And there are other features that you have to say, well, those don't count as part of the analogy. And sometimes that leads to more confusion than what you really wanted to start with. So the classic example I always used as a metaphor that I think is sufficiently misleading, it just shouldn't be used, is thinking of the universe, the expanding universe, as a balloon being blown up. We are told that if we took a little balloon and put little dots on it and then blew it up, we would see all the dots move apart. And that's kind of like the expansion of the universe. And that's supposed to be helpful to explain things. And, you know, it is kind of like the expansion of the universe. I get it. But it comes along with baggage that I think is not necessarily helpful.

1:53:31.2 SC: Two pieces of baggage in particular. One is, the balloon has an inside and an outside. There's volume inside the balloon, an interior, there's volume outside the balloon. And so people are naturally going to say, what is inside the universe? What is outside the universe? And then you have to say, oh, no, no, I didn't mean that. Just imagine the balloon is in empty space with nothing inside or outside. But even in empty space, a balloon has an inside and outside. The second problem is that the dots that you drew to represent galaxies expand along with the balloon. And so people say, well, presumably galaxies get bigger along with the universe, but they don't.

1:54:07.5 SC: So you have to say, no, no, that's not what I meant. The more you're saying, no, no, no, that's not what I meant, the worse the analogy is because you should have just explain the actual phenomenon. I think for the expanding universe, the way that I like to explain it is, imagine you're standing outside on a dark night and imagine that you have sufficiently good eyesight that you can see galaxies and they're moving away from you, so they're looking smaller and smaller all the time because you have really, really good discernment about these things. Everything is getting further and further away. That's the expansion of the universe. Does that do better than the balloon? I don't know. I think it probably does, but I don't have any data one way or the other. My most favorite... I don't really have a most favorite that I can think of right now because, yeah, I'm going to use different examples depending on the context. So I don't have one that I always lean on.

1:55:00.2 SC: Nat Guffren says, I have ADHD and it is nearly impossible for me to hold a single idea in my head for more than a few minutes, let alone an intelligence... Have an intelligent sounding conversation about it. Would you be able to talk about how thinking works in your head or how you process your thoughts? Do you have a mental stack where you keep pushing and popping tangents as you go, or are you very good at backtracking your way through the chains of thoughts without having to explicitly mark the diversions?

1:55:25.5 SC: I'm going to appeal. I maybe should have grouped this question to the previous one about memory palaces and things, because I have nothing in my brain nearly as concrete as a mental stack where I keep pushing and popping tangents as I go. But I'm not any better than anyone else at remembering a thought that I started with and then changed where I was going. I just as much as anybody else say like, why did I start saying that? Why did I start telling the story? Like I will tell a story and get distracted by something along the way. And then like what was the point of that remark? I have no special talents along those lines of keeping track of many different threads of a conversation. I mention that because some people clearly do. I don't know whether those people have special techniques or tools for doing that or whether it's just a natural talent that they have. So if one has ADHD, I really don't know what the advice would be. Yeah, that sounds like a problem that I'm not even going to try to speculate about because it sounds like something where my wild guesses could do as much harm as good. So sorry to hear that you have that issue. Hopefully it's easy enough to overcome for you. And I would ask people who actually know what they're talking about. I'm not going to be of any help here.

1:56:38.7 SC: P. Walder says, David Deutsch seems to argue that according to the laws of physics, probability can't be part of an ultimate account of explanatory knowledge. However, he acknowledges that probability may be a useful concept in the same way one assumes local land areas to be flat, even though it is known that in the context of the earth, this is not true. Is this convenient, pragmatic way of thinking about probability useful or not?

1:57:03.0 SC: I don't know. It's probably useful for David Deutsch. He's been incredibly successful at coming up with various ideas, so it clearly works for him. I do think that a lot of people make a certain mistake about probability, and maybe David is making this one. I'm really not sure. Namely, to assume that the only legitimate kind of probability is an objective frequency of things. Right? Like back in the old days, when they first started thinking about probability systematically, you were thinking about dice or cards or flipping coins, and you didn't think too hard about the fact that they were all made of atoms, because you hadn't heard of atoms yet. So you thought that there's just some intrinsic probability and you could test it by flipping the coin or rolling the dice many, many times.

1:57:45.6 SC: But that's clearly not the only way that we think about probability. We clearly think about the probability of different events that will only ever happen once. The probability of a sporting event or an election or something like that, anything at all. You getting a job offer is something that you can talk about the probability of. Okay? Now, there are some people who insist that that's not probability. Probability is only the objective frequencies of things. And this includes people who are like advanced probability theorists. There's a guy named De. Finetti who is a big name in statistics, and he begins a book by saying probability does not exist.

1:58:23.3 SC: Okay, that's not true, that probability does not exist. He is... The rest of the book is about probability. What he means is, that it doesn't really correspond to objective frequencies in the way that we sometimes think it does. I think that's fine. I think it's fine to have subjective notions of probability. I think, in fact, I'm at least open to the possibility and perhaps leaning toward the possibility that all notions of probability are fundamentally subjective at the end of the day. I think that's probably what David Deutsch has in mind because he's an Everettian about Quantum Mechanics, and he thinks that the overall wave function of the universe evolves in a purely deterministic way.

1:59:01.4 SC: But as I've talked about many times in the course of that evolution, you're going to have a moment of time where you as the observer are on a branch of the wave function and you don't know which one you're on. A position of self locating uncertainty. And you assign a credence to being on one branch of the universe or another. It's 100% okay to think of that credence as being a probability, even if it's just not an objective probability. It's a subjective probability, but it's a set of numbers between zero and one that add to one. They obey the axioms of probability. Okay? That's really all you need. And even if you don't care about many worlds or whatever, in doing science, we assign self locating uncertainties to what laws of physics we have around us, right? If my favorite example is, we think that there's dark matter out there in the universe, we don't know what it is, we don't know which particle it is. Let's imagine that we know that for some reason it's either an axion or a WIMP, a Weakly Interacting Massive Particle, but we don't know which one. You have to, as a scientist assign a credence to it being a WIMP or an axion.

2:00:07.3 SC: Why are you forced to do that? Because you have to act in the world. You have to say, how much money should we spend building an experiment to look for axions? If your credence is the axion theory has a probability of 0.00001 being right, we shouldn't spend that much money looking for it. Right? And in various ways this is true. Like there's a certain set of people who want to resist the idea that you have to assign credences to things. But those people always assign credences to things. What is your credence that gravity will be attractive tomorrow, right? It's probably very high, but it exists. You can't get through life without it. You never do. So I think it's just fine to treat all of those things as probabilities, and certainly as a part of the account that we have of explanatory knowledge.

2:00:56.2 SC: Lars asks, what is your working definition of a complex system? And where do you draw the line between that and a merely complicated system? I know it's fuzzy I do research on systems in that gray area myself and I'm sometimes unsure of what terminology to apply in papers, grant proposals, et cetera. Examples might be useful. For example, is a modern car a complex system?

2:01:17.1 SC: So we talked about this on the podcast recently. There was a good, what I thought was a nice way of distinguishing the two possibilities. Unfortunately, as I'm talking here, you know, I never prepare these AMAs as well as I should, so I'm forgetting who mentioned it. I really don't know. It might have been Petter Törnberg, but the distinction that they drew was the difference between complicated and complex is, complicated is when a big thing is made of many little things, but the little things all have a dedicated purpose within the big thing. So a car is complicated because you have an engine and brakes and steering wheel and things like that. But the steering wheel is made to be a steering wheel. Okay? Whereas complexity in this view arises when you have many pieces coming together. But the pieces are more general purpose than that. They might have a specific role in the complex system they form, but they could have other roles as well, right? Maybe cells in a biological organism could be doing other things. It's just that they're stuck doing the thing that they're doing in this particular complex system.

2:02:21.8 SC: So it's more a story of emergence than the complicated system would be because you're seeing behaviors that strictly arise as a result of the collective behavior of many things that are a little bit less single purpose that way. Now, I like that definition, but I don't love it. I'm not devoted to it because I kind of don't care what the distinction is. I think that would you treat atoms out of which everything is made as single purpose things such that everything is a complex system? Maybe not. I'm not quite so sure that that kind of distinction is 100% applicable to everything. But that's okay. I think it's more important to understand the specifics of the system you're talking about than to have nomenclature that applies perfectly cleanly in every particular example. As long as you define what you mean, I mean, I think that words like complicated and complex are sufficiently similar sounding to non-experts that you had better come up with a definition. If you mean to use them in ways that are not synonymous with each other, then you better tell everybody what you have in mind when you're doing this thing. In which case, as long as you tell them explicitly what you mean, you're doing fine.

2:03:39.0 SC: Paul Conti says, given that the United States did not even send a delegation to the recently finished COP climate conference in Brazil. And with President Trump's planned regressive changes to air quality laws and similar actions, what do you think of the future for international collective action on climate and greenhouse gas emissions?

2:03:56.6 SC: Well, I was always a little skeptical about the future for international collective action on these things, just because it's hard to get international collective action on things, especially things that are long term and have lots of sort of free rider problems and things like that. It's much harder now with Trump in office, where it's almost a point of pride that the United States doesn't do anything that would be good for the planet. There was just a vote... I just saw this on social media. There was just a vote in the UN to condemn attacks on international aid workers. And the United States was literally the only country that voted no. For some reason they made up, like, oh, it was too woke or whatever like that. The United States is a huge contributor to greenhouse gas emissions and other forces of anthropogenic climate change. And so for the United States to be uninterested in trying to do better makes it all that much harder. So, you know, whatever the pessimism optimism ratio was before, it's more tilted toward pessimism now than it was.

2:05:03.4 SC: David Harper says, what far out scientific or engineering advancement do you think is more attainable than people realize? I have heard you mentioned long life in the past. So I was wondering if you had any more techie thoughts.

2:05:15.9 SC: You know, I do think that... Look, I would have guessed completely wrong if you had asked me five years ago about how good AI would be doing now. So who cares what I think? I do think that the sort of secret advances, not really secret to the people who are experts, but the less heralded advances are still in the Biotech sector. Biology is so complicated. We don't have a theory of everything for it, and we're doing little things. But if you go into people's laboratories, like I sometimes do, where they're building computers out of DNA or, you know, learning to synthetically use synthetic Biology to engineer microorganisms to do certain things, I can imagine a world where, if we want to, we would eradicate a lot of diseases, even things like allergies could be gotten rid of or treated.

2:06:05.1 SC: And as we talked about with Mary Roach a little bit, she didn't want to speculate about the future too much. She was mostly about what had already happened. But you can imagine regrowing limbs related to the stuff that we talked about with Michael Levin long time ago. So I don't exactly know what to predict. I'm not an expert in those areas. But curing mental diseases, Alzheimer's, things like that are things that I absolutely think could be incredibly impactful on the world. And a lot of people, you know, since it's not quite been obvious yet, a lot of people might not be paying attention to that.

2:06:42.0 SC: Yazan Al Hajari says, I've been reading 'The Big Picture' slowly and enjoying it. And I keep finding myself thinking not only about the physics, but also about the writing itself. The book has a kind of architectural clarity, the movement from ontology to epistemology to meaning feels deliberate and beautifully structured. Thank you. Thank you, Yazan. I'm curious about the craft behind that clarity. When you write a book like this, how much of the structure is planned in advance and how much emerges through drafting and revision? How consciously do you think about tone, metaphor, rhythm, and the aesthetic side of language versus simply aiming for clear exposition? And did your writing style grow naturally out of scientific training or from a long process of reading, revising, and shaping a voice you wanted to develop?

2:07:25.0 SC: It's an interesting question that you raise it with respect to that particular book because the big picture is a little bit different in those senses than other books that I've written, simply because most of my books are about a particular scientific topic. Right? The Higgs boson, the arrow of time, the many worlds interpretation of Quantum Mechanics. And there, even before you start writing, there's a certain obvious logic that the book should follow. You're trying to explain something, and it's a big something. So you need a lot of work to get there, and there's a journey to take. And so there's absolutely choices to make, but there's a pretty natural way to go. And the emphasis then is on the individual chapters being clear and mixing scientific exposition with little bits of history and stories and human interest and things like that. Are you allowed to have pictures or equations? And there's a whole bunch of questions to ask, and it matters a lot. I mean, you're asking about the actual aesthetic values of the words.

2:08:20.5 SC: I do care about that. I don't put as much effort into it as I should, I think, tone, metaphor, rhythm, things like that. You know, I was taught in university and elsewhere about like one of the... You always remember, like certain individual events in your education, right? And I don't think that the memories are accurate. I don't think that what you remember as being a formative thing is necessarily what was formative. It's sort of... There's a principle in how things get named, which is that things don't get named after the first person to discover them. They get named after the last person to discover them because the first person did it before people were ready to care about it. Right? Likewise, when you realize something and it sinks in, it's probably not the first time you've heard it. It's the last time that you heard it before you totally realized it. Anyway, I had an English teacher who... [laughter] It's funny because he was explaining to us what he didn't want to teach us. He wanted to teach us like deep things about truth and meaning. He says like, I could just teach you like how to write cleverly. Like the...

2:09:25.2 SC: Having one short sentence in the middle of a set of longer sentences can really act as an attention grabber and change the rhythm of things. But I don't want to bother teaching you that stuff. And I was like, oh, no, no. Teach us that stuff. That's super helpful. That is really, really clarifying to me. Not just because of the particular individual piece of advice, but just the whole idea of, like, keeping the rhythm interesting and changing things up. And I absolutely do look at the length of sentences, the lengths of paragraphs, the topics and things like that. It's the same reason why, for the podcast, I don't want to have two episodes in a row that do more or less the same thing. I want to keep up a rhythm and a difference going on there as well. So the same thing for the book. Anyway, I just said a few minutes ago that I don't have a very good track record of keeping track of what I was talking about. But what I was talking about was the difference between 'The Big Picture' and the other books. For 'The Big Picture', it's much less obvious what the structure should be.

2:10:21.6 SC: It's the big picture. It's a survey of many, many different things. It's not just an exposition of one particular big thing with a lot of build up to a payoff like that. So, yes, I did put a lot of thought into the organization. I mean, the single biggest important choice I made for 'The Big Picture' was, keeping the individual chapters very short. Because, I mean, if you compare that book to any other book I've written, the chapters are much longer in every other book than they are in 'The Big Picture'. 'The Big Picture' has many more chapters. It's like 50 some chapters. Most of my books have like 10, 12 chapters, something like that. And that's very, very intentional because. And it was a little bit of, you know, I had to force myself to do it because the individual subtopics were individually very rich and I couldn't possibly explain them in as much depth as I might have liked to. There was a temptation to let the chapters go on for a bit, to put in more exposition. But I forced myself. I forget what it was, but I had some word limit on every chapter so that the idea was you could read the book in bite sized chunks because there were so many different topics being talked about that you could like read a couple of chapters.

2:11:40.4 SC: Read a couple of pages rather. That would be a chapter, you could think about that before moving on to the next thing. I think I had like one or two times when I cheated and I had two chapters in a row that were basically the same thing that I just broke up into two chapters. But there was absolutely an effort to make the journey more interesting. The... A feeling of momentum and velocity as you moved through the subject matter. So they did lead in one to the other. But at the same time you could stop at any moment and just take every little bit of chapter that you were reading as a point. So I would love to be better at that. Like if I were truly a professional writer rather than some someone who writes and does a bunch of other things, I would probably hope to be much more careful about stylistic things than I am. But, you know, life is short and you got to do what you got to do.

2:12:32.3 SC: Tyler Whitmer says, how do you feel about very herbal flavors in cocktails, including martini variations? My favorite cocktail these days is a two to one martini with dry gin and blanc vermouth finished with a splash of either absinthe or suze, kind of a, sort of a white Negroni, but served up. Does that sound good?

2:12:51.2 SC: Yeah, both of those sound great. I'm all in favor of herbal flavors. All sorts of different ways of making cocktails slightly more interesting. I think I've mentioned that my favorite version of a martini is with Japanese gin that has tea leaves among the various herbal notes that you get in the gin. Botanicals, I should say. And also a drop or two of evergreen bitters, which adds that sort of woodsy kind of flavor to it. It's not... I don't know, does that count as martini once you add those bitters to it? I don't know. I don't care. It tastes really good and I love all... One of the reasons that the martinis are so good or my favorite cocktail is because the gin has botanicals in it. The vermouth also does, and they're both sort of flavorful in their own rights, but yet serve as a basis for slightly other things. That's why olives or lemon twists work well in martinis. But also you can have kinds of gin or kinds of vermouth that lean in different directions. The Japanese gin with the tea leaves as an example. There's another gin I recently discovered that has smoked rosemary flavors in it.

2:14:03.9 SC: And, you know, you don't want that for everything, but you want it for some things, and it's great. As Kevin Peterson talked about when we had him on the podcast, the gin he suggested for the Mindscape, what was it? Petrichor Negroni was tomato based gin. And I did try it. And the tomato based gin, it's very funny because you smell it and it smells super tomatoey. It smells like you're smelling tomato juice, basically, but doesn't taste that way. In the gin, there's a little bit of tomato, but it's more just overall kind of herbaceous, which works very well in cocktails. So all in favor of these crazy explorations. I'm all in favor of crazy explorations in general in cocktails, exploring herbal flavors is much more interesting to me than exploring sweet or fruity flavors. That's just my personal preference.

2:14:58.3 SC: Eric Geller says, I recently saw the obituary of Dr. Mary Gaillard, who apparently made significant contributions to particle physics. Despite being an avid reader of popular physics books and listeners of scientific podcasts like Mindscape, I do not recall hearing of her before. Could you enlighten us about her contributions to physics? Any thoughts about why she's not more widely known?

2:15:16.3 SC: Right. This question reminds me, I made a huge goof in the last AMA because we mentioned axions and I mentioned the Peccei-Quinn mechanism and Helen Quinn. Apparently I said that she had passed away, which is just not true. I'm very, very embarrassed. Sorry about that, Professor Quinn, Helen Quinn is a leading particle physicist who... And I'm worried that what happened is there's a whole, you know, there's been a bunch of famous physicists of a certain generation beginning to pass away sadly. And I'm worried that when I heard that Mary Gaillard passed away somehow because they're both west coast physicists working in similar areas, that I kind of, my brain flipped that on to Helen Quinn, which I feel very bad about. Anyway, yeah, Mary K. Gaillard was a wonderful theoretical particle physicist and quantum field theorist. And, yeah, she did not get a lot of public recognition. But, you know, part of that, there's plenty of innocent explanations for that. It's not... I wouldn't go looking for conspiracies here. Within physics, she was very well known. Everyone knew who she was. She won major awards, the Sakurai Prize, things like that, for her work.

2:16:33.5 SC: But two things happened, one is she just didn't... Had no interest in interacting with the wider public and doing outreach, being on YouTube or whatever. Most physicists don't. I think you will hear about lots of physicists passing away who have been lots of important contributions and you've never heard of before. I try to sometimes explain this, but the set of physicists who you hear of in the popular discourse and the set of physicists who physicists think of as really important and interesting is not completely independent, but is not also very closely related. There's a huge difference between sort of internal pantheon of what theoretical physicists think of as the best working theoretical physicists and what the popular imagination thinks of as leading physicists of the day, you know, and that's kind of okay. Like, there's no reason why people who are not professional physicists should know about who the best professional physicists are. They should know about the ideas. They should be interested in physics. That's important. But, you know, who the best working physicists are is not necessarily something important. And the other thing is that the places that she worked in are not necessarily the ones that lead to, you know, big splashy public recognition.

2:17:53.1 SC: She did a lot of work in, you know, both the standard model of particle physics, but then also beyond standard model work and supersymmetry model building and things like that. But, you know, we haven't found supersymmetry yet. So the leading supersymmetry model builders, both Gaillard and others, have not gotten a lot of public recognition. For whatever reason, String Theory gets a lot more popular press than supersymmetry did before String Theory came along. I don't want to even speculate about why that's true. Maybe it's just because String Theory is a more ambitious idea because it tries to include gravity and everything else. And so that's very exciting. But I think it's okay that not everyone is widely known. I think it's more important to individual physicists that they be recognized within the field. And I think that Gaillard definitely was that. She is not someone... It's not like no one knew who she was before she passed away.

2:18:48.6 SC: Daniel Larkin says, if information can't be lost or destroyed, our thoughts and more so our memories are information. So when I die not believing in God or any religion or higher power other other than oneself, what happens to that information? If we can answer that, does that not tell us that what happens when we die? Or more so, that there is something after one dies? I would love a discussion on this as I'm trying to come to an understanding myself.

2:19:12.8 SC: Well, I think what's going on here is there are different ways of thinking about the definition of the word information. When you say information can't be lost or destroyed, if that's true, which you know, there's some arguments for and against it. But what you mean is the microscopic information about a system. That is to say, if you were Laplace's Demon and you knew the position and velocity of every particle, you could specify where they were, you could write down their positions and their velocities. And the amount of information contained in that is not destroyed. It is exactly the same from moment to moment in time. But we're not Laplace's demon. I'm still thinking of getting the T-shirts made that say we're not Laplace's demons, the Mindscape T-shirts. I probably have all sorts of mottos that I should turn into T-shirts. And we'll look into that someday when I have spare time, 12 years from now.

2:20:00.7 SC: But what we have is certain macroscopic information about the world. Certain coarse grained information. When I know something about a person, I know their name, I know how tall they are or whatever, I know something, I have information. They have information in their head, they have knowledge, they have memories, they have things like that. That information, that macroscopic, accessible information, there's no rule whatsoever that says that can't be lost or destroyed. If you write down a message on a piece of paper and then you burn it and then you die. So I don't know what the information was, and you're dead, you can't tell me what you wrote, it's lost, it's just gone, right? So we always have to distinguish whether you're talking about the fundamental microscopic level of description or the higher level.

2:20:47.6 SC: At the higher level, entropy, dissipation, things like that more or less guarantee that information is lost all the time. Information is also created all the time. Like you can learn things, right? And you can spread that knowledge. So there's just not really any kind of mismatch there. As far as what happens when you die, much of the information that you had becomes much less accessible to the rest of us. So for all intents and purposes, that information goes away. That'll be relative to all sorts of specifics, but that's the basic story. Your memory will live on in some people. But, you know, as people like to say, you die twice. The last time that you... Your heart beats and the last time that anyone thinks of you or talks about you. This is a feature of the universe, is that there will be that last time for all of us.

2:21:38.7 SC: Ken Wolf asks, what is your favorite steak? Mine is a salted ribeye done medium rare in a stainless steel skillet, then rested in tin foil with butter.

2:21:48.4 SC: That sounds pretty good. It's getting near dinner time here at Mindscape World International Headquarters, so you're making me hungry there, Ken. I don't really have a favorite. I do like steak quite a bit, but don't have it very often. It's a rare kind of thing, usually at a steakhouse, but sometimes I make it here at home. Ribeye would be my go to, or filet. I did recently at a local restaurant called the Ruxton here in Baltimore, have a rib cap, if you know about that. It is a subset of the ribeye. So, like, there's the eye of the ribeye and the cap. And usually the rib cap is really the best part of the ribeye. But when you cook them together in order to get the rib cap cooked enough, it is very common that the rib...

2:22:30.6 SC: Sorry, yeah, the rib... In order to get the eye cooked well enough, you end up overcooking the rib cap and it tastes less good. But if you separate them and just do the cap, apparently, it's awesome. It was certainly awesome when I had the one I had at the Ruxton. So I encourage people going through Baltimore who want, like, a fancy steakhouse experience to go there. And in terms of when I cook it myself, you know, we cooked steaks just last week actually. It is rare, but it did happen just last week, and I busted out the sous vide equipment. I don't know if any of you folks... I'm sure some of you folks are very familiar with sous vide. Some might not be. It's what you do when you take a piece of food, steak is a good example, but you can do with just about anything. You can do with veggies or chicken or whatever.

2:23:15.4 SC: And you vacuum seal it either officially like you buy a vacuum sealer. It's a little bit of a investment. But there's sort of cheap ways to vacuum seal all by yourself with just a plastic bag. And then you cook it. Not cook it... Well, you're cooking it, but you put it in water that is kept precisely to one temperature for a very long time. So you get a little Gizmo. Again, it's a minor investment, a couple 100 bucks, if this is worth it to you. You put it in a tub of water, and it keeps the tub of water at exactly one temperature. And that's a very different thing than most methods of cooking, which have sort of high heat on the outside that you eventually hope to enter into the center of whatever it is you're cooking. And as a result, you can just sit it there for a long time. You put the beef in there, put the steak in, lasts... It take a couple of hours, and you just dial in exactly what temperature you want the middle to be. Right? So I put it at like 128, which is even less than medium rare.

2:24:15.1 SC: But what you're going to do then is, it cooks the whole steak perfectly all the way through, but there's no searing whatsoever. And the searing is where you get a lot of the flavor. So then you separately, after you've cooked it for two hours in the sous vide, you take it out and you put it on very high heat on, I like a cast iron skillet, but stainless steel would also work just as well for just a couple of minutes. You know, flip it a couple of times, get that sear on there at high heat, mix it, baste it rather with butter, rosemary, garlic. That's what I did. And yeah, really good. Very, very good. I also have a grill outside and sometimes we'll like put some real rub of some sort on the steak and grill it outside. So, you know, yeah, I'm a pluralist in many things, including how to enjoy the steak.

2:25:02.1 SC: Jaina Bloomquist says, during your discussions with Nick Bostrom and Max Tegmark about the possibility of our world being simulated, it was suggested that we are unlikely to be in a simulation because our perceptions are relatively high resolution. My question is, how would we know if our current 21st century experience of the world we live in isn't low resolution by the advanced standards of our simulators, that we simply can't conceive yet of a higher resolution simulation than the one we might be in?

2:25:27.7 SC: Oh, well, we can conceive of lots of things. We can conceive of higher resolution simulations, lower level simulations, lower resolution simulations. The point is, to make that argument, we're leveraging one of the main arguments that is made in favor of the simulation hypothesis. You know, Bostrom is careful to lay out different possibilities. He's not saying we're definitely being simulated, but part of the logic is very closely related to anthropic reasoning. You say, look, if there's a lot of sims, a lot of agents that have the complexity and capacity to be counted as agents, even though they're living in a simulation, if you believe that's possible, then typically a civilization at one level will be able to simulate many more agents at the lower level in their computers, right? If that's not true, if a civilization can only simulate a few agents, then we probably live at the top level because that's where most people live. But if you assume that there are more people you can sim than there are people in your level of physical reality, then it is more likely if you assign equal likeliness to being any of these folks, which that's absolutely the problematic part of the argument, but let's go with it right now, then it's more likely that you live in a simulation than in the higher level, because there's more people in the simulation.

2:26:45.5 SC: But if you buy that, I'm not saying you should buy it, but if you buy it, then the people in the simulation can also run their own simulations. They can build their own computers, and they can simulate people, and there will be more of them. So by the same counting argument that makes you think the simulation is worth... The simulation hypothesis is worth taking seriously in the first place, you should certainly believe that we live in the lowest level that would count as living in a simulation, as being agents, I should say, right? Because, you know, maybe there is some very crude level of simulation where you can have agents in there, but they don't have the computational power to actually build simulations themselves. And so the point is, we don't live in that level. The whole argument for taking the simulation hypothesis seriously starts by saying probably we can simulate things, right? If you didn't believe that, you might be much less convinced of the whole simulation hypothesis. So if we can simulate things, the argument implies that we should live in the lowest resolution simulation there should be. And so I think it's easy to imagine lower resolution than we have in our universe. Therefore, I don't think the argument quite holds water.

2:27:55.9 SC: Moshe Feder says, I've always believed that true artificial general intelligence is probably emergent and inevitable. So I long ago decided that when machines achieve self awareness, it will be unethical to treat them as anything less than independent, sentient beings with rights equal to our own. Unfortunately, that decision leaves one crucial question unanswered. How can we tell when our machines are as conscious, self aware and capable of free will as we are, or think we are? I'd urgently like to see people smarter than me begin to search for an answer. Do you have any ideas?

2:28:23.0 SC: Well, look, people smarter than either one of us are certainly searching for an answer to this. This is a known question. I do think that we can criticize the professional philosophy community for not doing an even better job than they've been doing. I think that the best discourse about exactly these questions happens within professional philosophy. I don't think it's very good. I think that, you know, taking these ideas seriously and really pushing them forward is something that philosophers are very good at. But scientists and engineers are the ones who are really inspired by technological advances more than philosophers are. Philosophers are still going to be responding to Kant and Hume and Plato and Aristotle. Right? And so that kind of... The philosophers are good at it, but they're slow. And so I don't think that we've really made nearly the progress that we should in understanding how to know when an artificial being would be conscious. I agree with you. It is a super important thing to know. I think what's going to happen is, it's going to happen before we have a set of simple objective criteria for saying when it happened. So it's going to be a mess. I anticipate a mess. That is my prediction.

2:29:33.5 SC: Sandro Stookey says, I'm trying to understand if Bell's inequality can be made sense of in the many worlds interpretation or if it just doesn't apply. For example, I've seen formulations of Bell's inequality as a game between players where having access to an entangled pair of particles gives an advantage over classical strategies. Would these be amenable to a many worlds interpretation?

2:29:53.4 SC: Well, sure. I'm not sure what you mean by make sense of, many worlds predicts Bell's inequality. Bell's inequality has been tested. For those of you who don't know, this is John Bell, the theorist who worked at CERN for many years as a particle physicist, but had a sidelight in quantum foundations who proved that no local theory can exactly reproduce the predictions of Quantum Mechanics under certain sets of assumptions. So there's always loopholes, right? Super determinism is one loophole. Arguably, many worlds is a loophole. So either you should say it doesn't apply to many worlds, or there's a loophole, because one of Bell's assumptions is that measurements have definite outcomes, and in many worlds, measurements in different worlds have different outcomes. So the outcome is not definite in that sense. But in terms of the predictions for any one world, they're 100% the same predictions as all the other versions of Quantum Mechanics that are consistent with Bell's inequality. So there's no tricks you can do that either make things better or worse just because you believe in many worlds.

2:30:54.9 SC: David JS says, to what extent is the ontology of a physical theory a matter of personal choice? By physical theory, I mean a mathematical model, together with the rules for associating physical observations to the mathematical objects in the model. A good physical theory correctly predicts the results of any experiment we set up. The model does not really tell us anything about reality. For example, I can choose to believe the universe is just a vector in a Hilbert space, or alternatively, that the world around me consists of objects, fields and cats. In either case, Quantum Mechanics continues to make the right predictions. What do philosophers mean by ontology? Do they have information about the real world which they're not sharing with physicists?

2:31:32.5 SC: They don't have any information about the real world, which they're not sharing with physicists, or vice versa, as I'm sure you believe. But I do think, you know, again, I'm going to... Even though I love philosophy and am a philosopher in the sense that I'm a faculty member in a philosophy department and write philosophy papers and teach philosophy courses, I'm going to criticize my philosophy friends. And this is one of the reasons why I think it's so exciting to be a natural philosopher, to be really working at the overlap of fundamental physics and philosophy, because there's a lot of things that philosophers could be doing that they're just not, that they are slow to catch up on, and they have a more leisurely rate of progress sometimes than scientists do. So this is an example, and I think what I mean by that is, philosophers will often talk about a singular ontology for a theory. Now, the good ones certainly know all about physical examples like we just talked about dualities in Quantum Field Theory or the Hamiltonian versus least action formulations of mechanics or whatever.

2:32:32.4 SC: But I don't think there's been a lot of work in therefore defining under conditions where you have these equivalences of theories that are the same theory, just with different mathematical languages in which to cast them, how should we deal with the question of what we mean by ontology? Usually most philosophical examinations of this question just assume that one ontology is the primary one. Maybe other ones exist, but they're not quite as front and center. That's not quite what I think. I think that if you truly have an equivalence mathematically between different ways of talking about the same theory, then all of those vocabularies have an equal claim to be the ontology. So in some sense, I think an ontology is an equivalence class, right? There's no such thing as what the world is made of. There's the set of all possible descriptions that describe the world in precisely the same way. Now, your example of maybe the world is just a vector in Hilbert space, or maybe the world has fields and cats in it, that's not a good example. And the reason why is because the domains of applicability of those descriptions are not the same. One is simply more comprehensive than the other. I can talk about cats in terms of atoms. I cannot talk about atoms in terms of cats. It's an asymmetric relationship there. So that's not exactly an equivalence. But when there are equivalences, I think the question is a perfectly reasonable one.

2:33:55.5 SC: Jimmy MA asks a priority question. Recently I've been thinking about neutrino oscillation. Once a neutrino is produced with a certain flavor state, the constituent mass eigenstates will travel at different speeds. So given a long enough time, the wavefunction evolves into a superposition of three spatially separated wave packets, each corresponding to a mass eigenstate. Does it imply the neutrino flavor no longer oscillates, at least not in the way it's created once wave packets are well separated? Also, I wonder what would happen if decoherence kicks in. Do you have any ideas on that?

2:34:26.2 SC: Well, it's good timing on the question because, of course, we just had the episode with Ryan Patterson talking about neutrinos and neutrino oscillations. Look, in the real world, you're right that if you have... Well, you have to be a little bit careful, okay? When you produce a neutrino, when you produce a neutrino by a certain interaction, let's say the muon decays into an electron, okay? When the muon decays into an electron, muonness and electronness are both independently conserved to a good approximation. So you make a neutrino and an antineutrino pair as well as the electron. And one of those, the antineutrino is an electron antineutrino.

2:35:10.0 SC: The muon neutrino is a muon neutrino, not antineutrino. But as we talked about with Ryan, both the idea of a muon neutrino and the idea of an electron neutrino are not states of definite mass. You can think about them as superpositions of states of the lowest mass neutrino, the middle one, and the highest mass neutrino, if there's just three of them, as in the simplest models. So if they all... If all those parts have the same momentum but different masses, then they will indeed move at different velocities. What I started to say was, in the real world, all those velocities are very, very close to the speed of light for a typical neutrino moving around. And therefore they don't actually separate that much. Right? They're just moving at 0.9999c. And the tiny difference in mass is not actually going to be very relevant to that. But in principle, yes, they would separate.

2:36:00.7 SC: And it's not that they stop oscillating, or at least that's not the best way I would put it. I think that when we say oscillate, you have to be careful about what you mean. All the three parts of the neutrino, the low mass one, the middle mass one, the high mass one, they're all always there. It's not like some of them disappear and some of them reappear. It's just that they interfere with each other, their quantum wave functions. And so you can get different predictions at different physical locations and times of flight and things like that for what you're most likely to observe. And that's like a bunch of complicated Quantum Mechanics goes in there. I mean, you're giving me an idea for inventing a problem for my Quantum Mechanics class, because the final exam is coming up, so I'll have to think about that. But I think the short answer is they could in principle get well separated. In practice, that doesn't actually happen.

2:36:52.7 SC: David Sotolongo says, in the chapter you wrote in the book 'Neuroexistentialism', you wrote that the universe, vast and impersonal, does not provide us with meaning out there to be discovered, but by striving for authenticity in our actions, we can create meaning for ourselves. I believe there are different conceptions of what it means to live authentically. But broadly, I think it means to live as one wants rather than as one is expected to by others. If that's how you mean it, I wonder if you would advise that jerks and psychopaths should live authentically too, or if you would ask them to live inauthentically by conforming to society's expectation of them to at least some extent?

2:37:27.4 SC: Yeah, this is a perfectly good question. I mean, I wrote that article in the book 'Neuroexistentialism' quite a while ago, and my views have evolved a little bit. I go back and forth about my feelings about the role of authenticity and moral character in the role of morality and ethics. We did have the discussion with Skye Cleary, for example. She was very pro-authenticity, having written about Simone de Beauvoir and her work on it. And I mean pro-authenticity in general, but it's not the same as morality. So I don't think that authenticity is exactly acting the way you want to. It's more about constructing who you are. I mentioned moral particularism as an approach to morality some time back. The idea that when you make an action, you're not...

2:38:18.0 SC: There's no objective rules for whether an action is moral or immoral. It's just that by making that choice rather than some other choice, you're defining who you are. And so in some sense, it's not about just acting the way you want to, but acting in the way that corresponds to who you think you are or who you aspire to be in some sense. Right? Rather than what someone else wants you to do or expects you to do. And you're absolutely right that some people might aspire to be terrible people or they might authentically be terrible people. So what I said, and I'm not going to completely defend it, because like I said, I've changed my mind a little bit. But if you look closely at what I said, what I said was that by striving for authenticity, we can create meaning for ourselves. I didn't say by striving for authenticity we can be good people. Maybe your meaningfulness comes from being a bad person. When I've talked in 'The Big Picture' about moral constructivism, and I quoted Sharon street, who is a philosopher at NYU, who is a champion of the view, she mentions the idea of...

2:39:24.8 SC: What is it called exactly? Something something coherent Caligula. So people kept saying to her, as a leading moral constructivist, but what if the people are bad? They will construct bad morality. And her response is, yeah, they will. So we try to stop them from doing bad things. But if you don't believe in an objective standard to which you can say everyone should aspire, then you're going to have to deal with the fact that some people are bad and we can condemn them. That's okay. We don't have to say, oh, you're bad, and there's nothing I can do about it. But we also can't say you're making a mistake in the same sense that one plus one equals three would be a mistake.

2:40:02.6 SC: Paulino Vino says, I loved your conversation with Branden Fitelson, especially the part where he says that we should define probability by what the theory says it is. Would it be fair to say that for a Bayesian the correct interpretation of probability is that it is the thing that gets updated in light of new information with respect to some priors? Then how does one connect this to reasoning about the future or apply it to decision making? Do you have to have some sort of separate theory for that that I misunderstand the idea entirely?

2:40:30.4 SC: I mean, yeah, I think that Brandon is very good at coming up with insightful formulations of ways to think about things. So defining probability by what the theory says it is is a wonderful little motto, but it's one of those things where you have to say like, okay, what does that mean really? Like what does the theory say that it is? It is... So I don't think that your definition is quite sufficient to define what we mean by probability in a Bayesian framework. I certainly do think the probability does get updated. But to simply say it is the thing that gets updated is a little bit missing some important aspects of what probability is. I think to a good Bayesian, probabilities are degrees of belief. They're credences and propositions, how much you are willing to believe that something is true even though you don't know for sure, which essentially you never do. It is true that the way to deal with those degrees of belief is to update them them in light of new information. But I don't quite think that that statement is enough to define what probability is. So your theory in this case is more than simply Bayes' theorem. Okay? Your theory needs to be a little bit richer than that to make sense of what's happening. I don't know what Branden would say about it. That would be my result.

2:41:50.3 SC: Akon Sateli says, it seems to me like all emergent models are wrong some of the time, even within their domains of applicability. For example, model planets motion around its star using the centers of mass. And it will work most of the time, but after a while the star may explode or the planet may evolve. People who push it away from the star. Is this engineer minded good enoughness a necessary and inherent limitation on all emergent models or even emergence in general?

2:42:16.2 SC: It is a limitation, but I don't think it has anything to do with being wrong within the domain of applicability. You're just mentioning examples that are outside the domain of applicability. The important thing about domain of applicability, which I talk about in 'The Big Picture' is, it's a... It applies to the whole set of things going on all at once. Okay? So when you say a planet's moving around a star, works most of the time, but then the star may explode, I presume that what you have in mind is that nothing happened to the planet. So if treating it as just its center of mass did a good job, then it should still do a good job when the star is exploding. But it doesn't, so the theory is failing. But what I would say is, no, the star not exploding is absolutely part of the... What it means to be within the domain of applicability of that emergent description. Center of mass motion works when you have more or less isolated objects more or less localized within their individual motions, et cetera, et cetera. There's a set of conditions under which center of mass motion works as the emergent theory. And that's just not that. It is super important to recognize where those boundaries are and where things no longer become good enough. And sometimes we're good at doing that, sometimes we're not.

2:43:30.6 SC: Douglas DeYoung said, I recently read an article about scientists using an LLM trained on viruses to design a new bacteriophage. And amazing and useful as that would be, as amazing and useful as that would be, I find myself in existential dread that AI is designing viruses now. This seems like a very plausible mechanism for AI to be the world ending catastrophe some people have been predicting.

2:43:54.2 SC: Yeah, I'm not quite sure how to reassure you about this one, to be perfectly honest. Remember when we had Fyodor Urnov on the podcast who's an expert on CRISPR and gene editing, and you know, at some point he was trying to say, like, oh, you know, it's going to be less dramatic than we think. But I said, well, you know, could someone do something terrible or design babies in their garage? And he said, oh yeah, they could do that, no problem. And it goes hand in hand with what I said earlier about Biotechnology being the thing that I think will have a huge impact on the shape of the world over the next century or so. And AI, I think, is... That is a sensible use case for large language models. I mean, there's some things, things that AI does really, really well. Protein folding with AlphaFold is an obvious example, or playing chess is an obvious example. And I can absolutely believe that designing viruses would be the kind of thing that AIs would be good at.

2:44:47.4 SC: They might make mistakes, they always make mistakes, right? But human beings make mistakes also. And so AI's bringing capacities that could be used for good or evil to a larger number of people is absolutely going to be a problem. Again, it goes hand in hand with my thought that artificial stupidity is a bigger problem than artificial intelligence in the sense that the AIs don't really have understanding, in the sense that human beings have understanding. So it's very hard to prevent the AI from doing bad things. Right? There's just an article, I forget where it is, about children's toys with large language models in them. And they get to talk. And it wasn't that hard at all to get these toys to talk about deviant sexual behaviors or how to commit suicide or all these terrible things. How to play with knives. They're supposed to not do that. You would think you could just tell or program in to the AI, don't talk about dangerous things. But the AI doesn't understand those words. So you can do your best to prompt it not to do those things. But it's too easy to get around those prompts through adversarial prompting. And you can just get a lot of information out of them. So that is going to be a problem. People with nefarious schemes using AIs to help their nefarious schemes along. I have no great ways of fixing that particular problem. That little problem is out of Pandora's box, as far as I can tell.

2:46:17.5 SC: Kevin's Disobedience says, I recently discovered the Moving Naturalism Forward lecture series you hosted years ago. And it's exactly the sort of conference I'd like to see more of. Topics where mainstream philosophy has largely reached a consensus, yet the ideas remain poorly defined or contain significant disagreements within that accepted framework. Do you have any ideas for future conferences in a similar spirit?

2:46:41.0 SC: Yeah. So anyone who doesn't know, you can find online all of the videos from the Moving Naturalism Forward conference we held... Ooh, I don't know, some number of years ago, over a decade ago now, 2012, maybe. I forget. Dan Dennett was there, Richard Dawkins, Alex Rosenberg, a bunch of people who were former Mindscape guests. Janna Levin was there, Rebecca Goldstein. And we just... We... It was a time when there was a lot of online discourse or public discourse about atheism versus religion, more than there is now. But it was already getting tired, right, by 2012. People were bashing religion, people were bashing atheism, whatever. And I recognized that there was this issue where there's more to do than bash.

2:47:33.5 SC: Like even if you agree that religion is not the right answer, what the right answer is is tricky. It's not enough just to say naturalism. There's a bunch of open questions. So we got these people together, Steven Weinberg was there, and we chatted informally. There was almost no agenda. People led little discussions. Some people took it upon themselves to give seminars. But okay. And I thought it was very interesting. We didn't reach any great conclusions or anything like that. But you illuminate the different places people are coming from. And like I said before about learning new things, it's part of the process that changes your thinking a little bit in ways that you might not even remember years later. So could we do it again? Yeah. I do have ideas. The most obvious, promising one is moving Everett forward. This is an idea that David Wallace and I have been talking about for a long time. It's never happened. It came close to happening, but then we had a pandemic and that stopped it.

2:48:30.2 SC: But we're both busy. But it's a perfect example. Everettian Quantum Mechanics is something where... It's not that there's a consensus. Forget about whether there's a consensus. It's just that there's a lot of effort expended in defending the basic idea, and therefore not as much effort as you might like in working within the idea, accepting it, and trying to move forward within it. I think there are important problems that Everettians themselves don't agree on, and it'd be great to get people together to chat about that. I'm sure there's many other similar ideas. We'll see what happens. There's far too many interesting things to do in the world, not nearly enough time to do them all.

2:49:10.1 SC: Pete Faulkner says, I really enjoyed the discussion with Anil Ananthaswamy. When Anil discussed PCA, that's Principal Component Analysis, to project multidimensional data onto lower dimensions, it immediately reminded me of your work on coarse graining. Is PCA, or similar dimensionality reducing techniques, a viable tool for mathematically deriving the correct macrostate variables from a microstate? In other words, can we use PCA to automate the discovery of emergent descriptions?

2:49:34.4 SC: You know, it's a great question. I do have a friend here at Johns Hopkins. He and I have talked about exactly this question, making it real. The short answer is no. PCA Principal Component Analysis is not nearly enough to find emergent descriptions. And just to help people out here, what are we talking about? When you have a bunch of data points and they're in some vector space. So, like there's X and Y and you have data in the XY plane and you're trying to... But it's not just X and Y. There's like a 1000 dimensions. Okay? In your data set and every point is a vector in a 1000 dimensional space. But what if you look at it and you realize even though the data are scattered through this 1000 dimensional space, in fact all of your data turn out to be confined to a two dimensional subspace, right? Maybe that's true. Maybe it's just strong relationships between what's going on and therefore you can capture what's going on in the data much more efficiently by just giving me coordinates in that two dimensional subspace rather than giving me the whole 1000 dimensional full data set variables. And that's Principal Component Analysis. It's a way of finding within some vector space, what are the most important directions to move in that I could use to throw away some data and keep what is most important.

2:50:51.5 SC: And it won't always be nearly as simple as the example I gave. But that's the general idea. The problem is that the real world is not a vector space. I mean maybe Hilbert space is a vector space, but the big classical world that we live in has all sorts of non-linearities and constraints and things like that. So what if your data are confined to a sphere rather than a plane? Principal Component Analysis is not going to help you unless you have a certain version of it that knows ahead of time that a sphere is a viable option. So really, unless you kind of know ahead of time what kind of emergence you're looking for, things like PCA are not very helpful. If you do know what kind of emergence you're looking for, you can... There are various techniques which will do the coarse graining for you. But the trick, the science, the work, the art, is to figure out how the Emergent theory... What the Emergent theory describes at all.

2:51:44.8 SC: Corey Leander says, if you're willing to share on this sensitive topic, what are your thoughts on the prevalence of the New Left in academia and how this relates to the conservative backlash to it which has culminated in the Trump administration cutting university funding. This can be seen clearly by looking in non-STEM departments like history, for example, in this 2023 reading list for American history PhD students. None of the books in the World War II Cold War section are actually about World War II or the Cold War. And then there's a link in the question to US Major Field Exam Reading List at the University of Illinois Chicago.

2:52:22.5 SC: So yeah, I'm willing to share on this and I think that it's completely off base to attribute the conservative backlash to academia to the actual behavior of people within academia. I think that it's more or less strategy that is deployed for cultural and political strategy reasons, not something that reflects reality. I mean, here is just one piece of evidence that I'm right. The evidence that you are giving as perhaps evidence that the academia is going off the rails is a reading list from the University of Illinois Chicago. By the way, I should also say it's weird that you say the New Left, like maybe it means something different than what I think it means. But usually I hear the New Left as describing like the generation that came around in the '60s and '70s and that's not new.

2:53:17.1 SC: They've been around for a long time. So maybe there's an updated definition of New Left. But anyway, there is a complaint about academia that rather than teaching what really matters, they're all doing this woke stuff and talking about discrimination and bigotry and things like that. Instead of where we invaded on the beaches of Normandy or something. So this would be an example of where complaints like that come from. But look, if those complaints were based in reality, the evidence for that comes complaint would be way better than a single reading list from specifically the University of Illinois Chicago. Like why that one? You know, if it was actually serious, if someone cared about this and thought it was a big problem, they would do a survey, either a comprehensive survey or some kind of randomly selected survey of many, many departments looking at all of their curricula and trying to decide how closely those curricula aligned with what they should be. Okay? From one reading list I can get essentially no conclusions whatsoever. For one thing, I don't know, I looked at the reading list. It's far from clear that this is supposed to be a history of World War II.

2:54:28.4 SC: It's a history of some features of America during World War II. So yeah, it's not about World War II or the Cold War. It's about what was happening in the United States. That seems to me to be a perfectly valid thing to study. There's certainly other valid things to study. Are those other valid things also studied at the University of Illinois Chicago? I don't know. I can't tell by reading this particular cherry picked document. I mean, the irony is that just a couple of days ago there was a report released by various conservative groups that is basically saying they want to take over universities and not let faculty hire people, have political appointees, control what can be done, give extra teaching loads to humanities faculty and lower the teaching loads of STEM faculty and boost American Studies over studies of other places. And all these really terribly blatant violations of academic freedom. And to say that we should blame Leftists for this rather than the Right Wing that is actually doing it just seems to me to be upside down. What is happening is that the complaints about academia are in very bad faith. Not all of them.

2:55:41.8 SC: Like there's... I have complaints about academia, there's probably good complaints about academia out there. And there certainly are some people on the Left within academia who I think are just not at all on the right track or going off in various ways. And I'm happy to disagree with them about various things. But what is happening is that some of the worst examples of that are cherry picked and raised to attention of people who are ready to be disgruntled about cultural movements that they don't like. And then that is taken as evidence, but it's really not. Like I said, it's not actually a careful, dispassionate look at what is going on. I don't think that any major university that has a history department fails to teach about World War II, honestly. There are some universities that are canceling their history departments because funding is bad. And I think that's the problem. Some combination of a lack of funding and government overreach by conservative governments are the real problems in academia, not the existence of some Left Wing professors wanting to teach about discrimination.

2:56:46.4 SC: Connor O'Brien says, how does Quantum Field Theory turn the eigenstates, which earlier in the book... I'm presuming this book is 'Quanta and Fields' represented different excited states and energy levels into representations of various numbers of particles while preserving its description of the excited states. It seems like that's one mathematical idea representing two distinct things. Particle count versus energy levels.

2:57:10.9 SC: Yeah. You know, in the book I tried my best to explain this move, but it's a tricky move, It's a subtle move. And honestly, in a Quantum Field Theory course that you would take as a grad student in university, as a physics major, it's kind of glossed over really quickly, you know, because we know what the right answer is. So we just kind of say it and move on. But I think that the sort of the best way to wrap your head around it is just to accept the existence of a mathematical equivalence. Just like we were saying before, when two mathematical structures turn out to be equivalent to each other in some way, then we say that they're the same physical theory. Right? Two different mathematical structures for the same physical theory. And in this case the two mathematical structures are on the one hand the quantum description of a set of point like particles, like three particles. Okay? And you can describe particles, but if you just do the ordinary Schrödinger equation, when you do particle Quantum Mechanics, there's a feature that the number of particles never changes.

2:58:09.7 SC: So it's not sufficient to say I have a quantum theory of three particles if you want to describe the real world, because in the real world particles decay and annihilate and things like that. So what you could do is construct a many particle theory, which would be a Hilbert space, a space of quantum states that includes a sector where there are zero particles. So you say, well, I just have a single state, which is the vacuum state with no particles in it. Then you have a bunch of states that you add to that which include... Which describe a single particle. We know how to describe a single particle in Quantum Mechanics. Then we add even more states which include descriptions of two particles. Okay? So we have a particle like theory...

2:58:49.2 SC: A particle theory really is a particle theory, but includes the possibility of zero particles, one particle, two particles, three particles, et cetera. Keep that in mind. And then completely separately look at Quantum Field Theory and quantize the field theory, quantize it in the approximation where interactions are not important. And you basically have free fields. And you look at the excitations of the fields and you look at their structure and it turns out to be exactly the same as the point particle theory with many different kinds of numbers of particles that you started with, but the single field includes all of those numbers of particles. And so it's much easier to describe transitions between them where two particles annihilate or one particle decays into two, or something like that. So that might not be satisfying to you if I'm just sort of asserting to you that the mathematical structure of the two ideas is the same. But that's basically the right answer. Okay? That is basically why we're allowed to do it that way. The structure of low lying states of excited quantum fields is exactly the same as the structure of collections of quantum mechanical particles.

3:00:02.2 SC: Dan O'Neill says, has the explosion in self publishing affected you as a traditionally published author?

3:00:07.6 SC: Nope. I've not really self published anything myself. I know people who have, like, certainly sometimes you're tired of the bureaucratic issues with publishing, but actually, you know, my publisher is good. I like them. They handle a lot of things that I don't need to worry about. So no, it has certainly not affected me personally. I'm all in favor of it, by the way, like going along with my usual pluralistic tendencies. Anyone who wants to publish a book should be able to publish it. The only way it really affects me is that I get a lot more book manuscripts from, let's just say, informal physicists, amateur physicists who have a new theory of everything. You think they're all just on the Internet making YouTube videos, but no, a lot of them still write books and publish them and then mail them to you. And I get a lot of those.

3:00:58.6 SC: Rob Gebela says, if probability in many worlds is not related to frequencies, why can it be experimentally confirmed by frequencies? And is there another way to experimentally test the correctness of predicted probabilities?

3:01:10.2 SC: Well, yes and no. It is related to frequencies in that if you believe the usual argument for getting the Born Rule in many worlds, then you predict that the frequencies of repeated quantum measurements will obey the frequencies you would predict from the Born Rule. Right? So that it's related in that way. It's that the essence of what the probability is representing is not truly a frequency in the sense that if I measure a single spin that is half spin up and half spin down, and I measure it a 1000 times, and I say that I predict in many worlds that it'll be roughly 50, 50 spin up and spin down, I have to say that there will definitely be one world in which it was spin up every time out of a 1000 times, and another world in which it was spin down. So it's a frequency in some worlds, it's the right frequency, in some worlds, it's the wrong frequency in other worlds. And you have to make an argument about why the worlds that the frequency is, what you expect from the Born Rule count more than the worlds in which it doesn't. And there are arguments because it's true they do count more, but you get the frequencies out in that sense, even though there's always these worlds in which you get different frequencies.

3:02:21.9 SC: Fontanetti says, I'm currently on my third reading of; The Big Picture'. Ooh, good. And I keep finding new things every time. I'm a Brazilian Portuguese speaker and have no difficulty reading it in the original. But I've always wondered why, as far as I know, your books haven't been published in Portuguese. We have a significant gap in science communication in Brazil, and I'm certain your books would be extremely well received here.

3:02:43.8 SC: So here's the thing, the dirty little secret. I have, like, nothing to do with what languages my books get published in. You know, I have an agent and they pitch the books to publishers in all sorts of different countries, but sometimes they just say no. I kind of thought that my book, at least some books were in Portuguese, but I might be wrong about that, but that whole thing happens... The only point at which I'm involved is when the agent says, they've offered you this much money. This is more or less the standard amount. Is it okay to go forward? And I say yes. And then a year or two later a book appears. I've had essentially no interaction with the translators of the books or anything like that. I don't even keep track of what languages the books have been published in. So I can't help you. It's actually... If you live in Brazil, you have a lot more influence than I do over whether my book gets published in Brazil. Write to the relevant people, write to science editors at newspapers or publishing houses or whatever and pitch the book. You have as much of a chance as I do of getting anything to happen there.

3:03:52.6 SC: Michael Wall says, if space and/or space time are emergent from a quantum wave function, would you expect String Theory to be emergent from such a model?

3:04:01.8 SC: It might be. That's absolutely possible. So the idea that the universe is described by a wave function and you try to figure out how to chop up Hilbert space to see what the emergent bits of it are, would apply to any theory that might be right. If it's string theory, if it's Loop Quantum Gravity, if it's something else, like that, as long as it's just Quantum Mechanics at the heart of it and you haven't modified Quantum Mechanics in some way, then that approach will work. So you might end up doing a sort of top... I guess it's top... I don't know if it's top down or bottom up, but a top down version of trying to figure out how Quantum Gravity works by starting with a wave function in Hilbert space and try to figure out how it might manifest itself.

3:04:44.8 SC: And then you, of course, are constrained by the desire to fit the data and things like that. And you might end up at the end of the day, oh, look, I rediscovered string theory. Right? But you might also discover you discovered something else. So we don't know. I think again, it's useful to have different approaches to the same problem, even if they eventually end up being compatible. I know that some people have hoped that once we understand both Loop Quantum Gravity and String Theory better, they will turn out to actually be overlapping or at least compatible with each other. The very second episode we did of Mindscape was with Carlo Rovelli, a champion of Loop Quantum Gravity. And I asked him that question and he said, no, they are not compatible. They are just different theories. So there you go.

3:05:26.5 SC: Tracy P., says, concerning complexity and emergence, does it strike you as something akin to the unreasonable effectiveness of mathematics that so many examples of emergent phenomena can be described through coarse graining? Or is emergence something that should necessarily follow from the underlying interactions of the constituent parts?

3:05:44.4 SC: You know, I would love to know the answer to this question, and I truly don't. The closest I have even to an intuition... If I correctly interpret your question as asking, under what conditions does some lower lying comprehensive theory give rise to or is compatible with some higher level emergent description? I don't know the general theory, a general answer to that question at all. I suspect that spatial locality has something to do with it. At least in the examples that we kind of somewhat understand, the first step in getting emergence out of a fundamental theory is to first look at the classical limit of Quantum Mechanics. And then you have atoms and particles and things like that. And those atoms and particles interact locally in space. They don't interact strongly if they're far away. They interact noticeably if they bump into each other. Okay?

3:06:34.2 SC: And that helps us find emergent behaviors starting at the lowest of levels. Because literally, if you go just like from kinetic theory to fluid mechanics or whatever, you're literally averaging over small regions of space. Once you get to much higher levels, if you're trying to find the emergent theory of a national economy where the lower level is human beings but making decisions, then it's not really local at all. And there's a lot more complications going on. But all that is built on atoms and particles at the end of the day. So I suspect that spatial locality has a lot to do with it. I also suspect that as I talked a little bit in the... There was a solo episode where I talked about complexogenesis. Yeah. And dissipation in photons play a huge role in my mind in complexogenesis. The ability to lose extra energy and settle into a configuration that is more or less stable is something we all take for granted, but is... Turns out to be part of the laws of physics that could have been otherwise. And so I think that's another important aspect. But I'm very far away from having a general theory of all this. And I would love to have that. That's an ongoing research project as far as I'm concerned.

3:07:49.2 SC: TCMD says, given the arguments from black holes, holography and wormhole physics, that Quantum Gravity probably forbids exact global symmetries, what does this mean for axion models, dark matter stability, and the accidental symmetries like Baryon Number in The Standard Model. If these symmetries can only be approximate, how far down does that constrain model building?

3:08:08.9 SC: So the question has to do with number one, there's a distinction in physics between global symmetries and gauge symmetries. Global symmetries, despite the grandiose name, were actually the wimpier versions. The gauge symmetry says, I can separately do a symmetry transformation at every point in space and time, and it's still a good symmetry transformation. And those are the basis for modern theories of particles and fields, of the various forces of nature, the strong and weak and electromagnetic and gravitational forces. A global symmetry says there's some collection of fields that we have to rotate into each other simultaneously everywhere through the universe. Thus the word global in order for it to be a symmetry. And so symmetries, according to Emmy Noether, have conserved quantities associated with them.

3:08:56.6 SC: And so there's a symmetry, for example, that gives rise to conservation of Baryon Number. Baryons are the protons and neutrons and things like that. But we look at the world and we say, well, there are more Baryons than anti-Baryons, how can that be true if there's a symmetry? The answer is the symmetry is not exact, right? Or at least that's part of what you need for there to be an answer. Unless the asymmetry was just built in to the creation of the universe. And so, unlike gauge symmetries, which are actually sort of redundancies of description, they're unbreakable. It's not that there is a symmetry that you could break by a little bit. You can hide it. And we call... Unfortunately, we call the hiding of a symmetry spontaneous symmetry breaking, but it's not really a breaking. The symmetry is still there. It's just hidden from you. But for a global symmetry, you can just break it.

3:09:45.0 SC: It can just be approximate. And that makes sense. For example, what if you have a black hole and it evaporates and it happens to give off a neutron but not an anti-neutron? You have violated Baryon Number. So we think that gravity provides a way of violating any global symmetry you might have wanted to have. So in a world with gravity, which is our world, there is this feeling that global symmetries should be approximate rather than exact. How far does that constrain model building? Usually not that much, in the sense that models like axions, which are based on approximate global symmetries, the Peccei-Quinn symmetry that we talked about before is broken because it's only approximate. It's a global symmetry, but it's broken much more by quantum chromodynamics than we would ever need gravity to worry about breaking it. There were a couple of papers back when I was in grad school about, you know, could wormholes breaking global symmetries ruin axion models or something like that? And I believe the answer was, probably not but maybe. And I'm not sure whatever happened to that particular discourse. If axions are discovered as the dark matter, then maybe we'll have a much more down to earth discussion about that.

3:11:02.2 SC: Nate D., says, I'm curious about your condemnation of the Copenhagen interpretation of Quantum Mechanics. Does this have to do primarily with the frustration about its refusal to engage in ontological questions or its dogmatic uptake in the discipline?

3:11:16.0 SC: Neither one. I think I'm perfectly clear on my condemnation of Copenhagen interpretation. It's not a well defined theory. That's it. It uses words like measurement and observation right there in the fundamental formalism of the theory, but never defines them. Therefore the theory is not well defined. When does a measurement happen? What counts as a measurement? You can try to answer those questions using words like decoherence, et cetera as you would do in the many worlds interpretation. But decoherence takes time. It's not fundamental. It is an emergent, higher level thing. Why should it appear in the fundamental laws of physics? So you need to do better than the Copenhagen interpretation. I don't think it's like I don't like it. I think it's not a well defined theory yet it's not a legitimate claimant to be the right way to think about Quantum Mechanics.

3:12:05.4 SC: Taoist says, well I... While I, that you... I think some words got deleted there. Well, I take it that you, as I do, have very low credence that 3I/Atlas was manufactured. This is this comet that is from interstellar space that is going through the solar system. And of course whenever that happens, certain people, including former Mindscape guests, say that it might have been an alien spaceship. Do you think, says Taoist, that the unexplained anomalies, per Avi Loeb, raise legitimate questions about the nature of the object? Do you imagine that NASA is downplaying these precisely because they don't want to throw fuel on the alien conspiracy fire?

3:12:43.7 SC: No, I don't think either one of those. You know, look, it's fine to have questions about the nature of the object. It's fine to think about those very carefully. It's fine to put effort into considering long shot possibilities. But it's not fine to like ignore the much simpler explanations or to over claim the probability that the wild explanations are the right way ones. And I think that's, sadly, the direction that Avi Loeb has gone in. I don't think he was necessarily going in that direction. I think that when we did the Mindscape episode, I would still put him as you know, he was overly enthusiastic about Oumuamua being an alien spaceship. But it was still sort of respectable. He wasn't ignoring counter evidence. But I think he's gotten worse at that since then.

3:13:31.4 SC: I think it's kind of bad. Look, NASA. What do you mean? NASA is downplaying these, like NASA's not a person, okay? NASA is an organization with many, many people. And these people are scientists, and I know them. And they're ornery and individualistic. And they're not going to collectively downplay these. Because they don't want to throw fuel on a conspiracy fire. They're downplaying them because they think that they're wrong. Okay? I think NASA might get a lot more money if people thought that aliens were out there. So if anything, the incentive would be for NASA to throw fuel on the fire.

3:14:06.1 SC: Scott Evans says, I am reading 'Something Deeply Hidden' and really enjoying it. Thank you. Why is it not possible for many worlds to work in reverse? Meaning that to begin with, there are many orthogonal branches described by the wave function we observe. But also, just as we observe, the wave function decoheres into a single branch, upon observation of the single... Or the system's entanglement with the environment. This would mean that the Schrödinger equation would need to be adjusted to account for the other branches ceasing. But this isn't as likely... Isn't this as likely as the many worlds theory, and one that accords with actual observation?

3:14:38.4 SC: So I truly think I don't understand what is being asked in this question. But maybe I can say some true things that will be helpful, which is why I picked it to be answered. What we observe is always just a single world. No one has observed worlds either branching or coming together. We think... I think that worlds have branched. But by the nature of what it means to be an observer, you end up on one branch or the other. You don't end up observing both. We've certainly never seen anything that looks like two worlds coming together. The reason why in many worlds. The worlds branch apart rather than coming together is very, very well understood. And it is fundamentally thermodynamic. It is very much like why entropy increases globally in the universe. And it requires a special initial condition, just like the second law of thermodynamics does. It does not require any modification of the Schrödinger equation. So I'm not quite sure what modification of the Schrödinger equation you have in mind to count for other branches ceasing, but I suspect that such adjustments are going to be much more difficult to pull off than you might think. I mean, people have tried for a long time to change the Schrödinger equation to get rid of the other worlds. Some possibilities are still quasi viable, but none of them are especially elegant or compelling.

3:15:52.0 SC: Dynasty says, if there were one million people in a dark room with only one light source, and that light source only put out 1,000 photons per second, would only 1,000 people know there was a light source? I realize this is a simple question, but I assume... And I assume I know the answer, but I must know if I'm right before asking more advanced questions.

3:16:11.4 SC: Roughly speaking, yes. Let's imagine... Let's grant you the idea that people are really good at observing photons so they can see one photon, which I think is probably unlikely for actual human beings. But let's imagine they could. Maybe they have a photo detector or something. Then, yeah, if the light source puts out a 1000 photons, then only a 1000 people know there's a light source, because once each person detects a photon, that photon is gone. So that's a simple answer to your simple question. I hope it's moving us in the right direction.

3:16:41.2 SC: Joey says, in a recent episode, you and your guest had a very brief conversation about the book 'Math Without Numbers' by Milo Beckman. This quickly became my next audiobook and I really enjoy it. I've also heard you mention several times that something is lost in the math or science to everyday language translation. I think I pronounced that incorrectly. I've also heard you mention several times that something is lost in the math or science to everyday language translation. Did you read this whole book? And if so, do you think it's lacking because of not using equations? Or are the concepts basic enough that language is adequate? I'm just wondering if it's a good study tool or just entertainment.

3:17:19.0 SC: So let me say a positive thing and a potentially negative thing here. The positive thing here is that Milo's book 'Math Without Numbers' is a wonderful book. It was great that you picked it up. I don't think it's just entertainment. I think that Milo understands the math very, very well and he's doing a good job of using pictures and words to describe intricate mathematical concepts, which is great. As you might expect, it's more about geometry and topology than about algebra or number theory. If you're not going to use numbers, but those are perfectly good aspects of mathematics. The slightly negative thing here is, we did not mention that book in our brief conversation. What...

3:17:56.5 SC: Sorry, who was it who I was talking about? It's gone out of my head who I was talking to in the conversation. But the book that was mentioned was 'Science Without Numbers' by Hartry Field. Hartry Field is a very well known philosopher who is a philosopher of mathematics and he is a nominalist, is he a nominalist? I forget what he actually is, but he was trying to prove by construction that you could recast all of the ordinarily interesting theories of physics, at least classical physics, without specifically using numbers. Because he wanted to say... He wanted to undermine the Platonist argument that the world exists, the world obeys the laws of physics. The laws of physics are described using numbers, therefore numbers exist. Okay? So he said, well, you don't need to use numbers to describe the laws of physics. It is an intensely technical, philosophically advanced book. It is not one you would want to read the audiobook for. And it's entirely different in purpose and scope than Milo's book. They're both good books, but they're trying to do two very different things.

3:19:08.3 SC: Qbit says, I'm curious about the ground state of a hydrogen atom from the perspective of Quantum Field Theory. Intuitively, the atom is in its lowest energy state, so it shouldn't emit photons. But in QFT, the presence of the electron changes the photon field's ground state, meaning there must be on average more than zero photons around the atom. Does this mean a photon detector placed near a ground state hydrogen atomic could actually detect photons even though the atom can't lose energy?

3:19:33.6 SC: So this is a perfectly legitimate question and a relatively simple sounding one that very quickly gets us into very deep issues dealing with Quantum Field Theory. So one issue is, you know, the... When you say the electromagnetic field upon quantization gives us photons, that's not all it gives us. You know, there are quantum states which are not well described as collections of photons. In particular, the ordinary Coulomb electric field around a charged particle really shouldn't be described as a collection of photons. You can do it. It's awkward mathematically, but you're sort of forcing a square peg into a round hole.

3:20:19.5 SC: It's a field. It looks like a field, it doesn't look like a collection of photons. It's just much more natural to think of it as the actual field. Sometimes you want to think about the interaction of the electron with the nucleus as a collection of Feynman diagrams, in which case you have to think about it as a photon. But it's not very natural. The Feynman diagrams are much more naturally situated in scattering experiments and things like that, rather than a static situation like an atom. The other subtlety is, you ask about a photon detector placed near the ground state of the hydrogen atom, could it detect photons? Short answer is, yes, it could detect photons. But here's one of the miracles of Quantum Field Theory. Even if you were in the vacuum, even if there was literally nothing going on, it was just empty space everywhere, the lowest energy state of your Quantum Field Theory. But now you bring in a detector into the vacuum, could it detect photons there? And the answer is, once again, yes, you can detect photons in the vacuum. Some of you might think of, if you know a little bit about this stuff, the Unruh effect, which says that an accelerated detector will detect photons.

3:21:28.4 SC: But even an unaccelerated detector has the feature that it is finite in size. One aspect of the vacuum of a Quantum Field Theory is that it is kind of global. What happens in one region of space is entangled with what happens in other regions of space. So when you say there are zero particles in the vacuum, which is something we often say, what you mean is, if I did an experiment to measure the number of particles exactly simultaneously everywhere throughout space, I am in an eigenstate which says there are zero particles. And that is the answer I will always get.

3:22:02.6 SC: But as a practical matter, the only experiments we can actually do are to say, okay, in this region of space, are there any photons? And the answer to that one, it's a different question you're asking. You're not asking about the global state of the universe, you're asking about this little sub region of it. And the answer is, you have a non-zero chance of detecting a photon. And your classical intuition wants you to say, well, how can you detect a photon if it were never there? And the answer is that photons are not really what the universe is. The universe is quantum fields, and they have states. And you have to be careful about what you mean when you say, I look for a photon and try to detect it. You have to operationalize it in terms of a detector with certain physics and an interaction Hamiltonian and all that. And if that detector is finite in size, you will be able to detect a whole bunch of things that you didn't think were there.

3:22:55.8 SC: Armchair Epistemologist says, how do you treat certainty when it comes to Bayesian probabilities? For example, if I haven't looked at the weather forecast, I might say there's a 10% chance that it rains tomorrow. But I have very low certainty and I think it's quite likely wrong. I suppose this would be a prior. This is different than if I looked at a forecast that was an output of some weather model that told me there was a 10% chance of rain tomorrow. It feels as though I almost need some second order Bayesian probability distribution of what the probability of rain is. And at that point I could also have a third order distribution, how accurate my gauge, my uncertainty is, fourth order, et cetera. Where does this end and how is this handled in practice?

3:23:33.9 SC: This is an interesting question, which I don't think there's any obvious consensus about how to deal with it. So I wouldn't phrase it the way you're phrasing it. You say certainty when it comes to Bayesian probabilities. That makes it sound like what do I do when my Bayesian probability is zero or one for something. But that's not what you mean. You're talking about what a physicist would call the error bars on your error bars. You know, when you say that the mass of a certain particle is a certain number of GeV plus or minus something, or if you measure the Hubble Constant right? So in fact, right now, as we've talked about with Marc Kamionkowski and before that, Adam Riess, we have the Hubble tension. The Hubble tension is we can measure the Hubble Constant in the current universe two different ways. We can measure it either locally or sort of globally by using the Cosmic Microwave Background. We get two answers that disagree with each other. They both have error bars, both of those answers and the error bars don't overlap, so they can't both be right.

3:24:32.9 SC: So one of them is making a mistake when they quote those error bars. And this is a kind of situation where you want to know like how much confidence that, do you have that those are the real error bars? The problem is, that there's a distinction we make in science or in statistics, I guess, between statistical errors and systematic errors. A statistical error is, if I flip a coin a certain number of times, I'm supposed to get 50% on average, heads and 50% tails. But if I only flip it four times, you wouldn't be completely shocked if I got three heads and one tail, right? There can be some deviation from the expectation just because your statistics are finite in size. That's a very, very well understood thing. And we can completely characterize that quantitatively. Systematic errors are when you're just making a mistake. There's something that is really important that you don't know about. Famously, when we had the elections, presidential elections in the United States in 2016, and Hillary Clinton was predicted to win, she was ahead in the polls by a little bit not by a lot. But she was predicted by... She was ahead in the polls by a little bit in many different states.

3:25:45.1 SC: So in order to lose, there's a certain calculation that says we have to be wrong simultaneously in all these different states. And if there are error bars, what are the chances that we're wrong in all of these states all at once? That reasoning was wrong. And the truth is there can be systematic errors. Like if there's a way that you're polling to get your pulse, your finger on the pulse of the public, and that way systematically under-represents a certain point of view, you could get the same error that is correlated in a whole bunch of different states. And that's exactly what happened. And Donald Trump ended up winning. It was not as statistically as unlikely for that to happen as some people said that it was at the time. So how do you know that? I mean, people like Adam Riess, who's an observational cosmologist, live their lives trying to understand what are the systematic errors in what they're trying to figure out. But it's not, despite the fact that it's called systematic errors, that's not a very systematic way of dealing with them. Right? Because by the nature of the beast, it's because you don't know what's going on that these errors exist.

3:27:00.1 SC: What if you are talking about a coin flip again and unbeknownst to you, you've been handed an unfair coin that is going to turn up heads 80% of the time. You don't know that. So your predictions are just going to be off. So it is important, it is under theorized in some sense. And how is it handled in practice? It's more art than science, I would say, as a theorist myself, so I'm probably not the person to ask, but I do think that it is something to keep in mind whenever you start taking your own probability estimates a little bit too seriously. And by the way, I should say when the probability of an event becomes very, very tiny, this kind of worry becomes very, very big. And I think that for things like existential problems, will AI take over the world? Will we build a Large Hadron Collider that makes a black hole that destroys the earth? And you say, well, it's improbable, but the consequences would be enormously big, that's exactly where these worries about how improbable is improbable become super duper important. And I don't really have a systematic way of handling them.

3:28:07.9 SC: Christopher Smith asks a priority question. I am very torn on what major to pursue and have a possibly irrational desire to go all in on the so called deepest subject. My line of thinking tends to go something like this, psychology is the most immediate raw facts of my subjective experience. You can never not encounter your own psyche. But it's arguable that biology informs psychology, chemistry informs biology, physics informs chemistry, maybe mathematics and logic possibly inform physics. And even further, metaphysics and ontology may inform mathematics, logic and physics. So is metaphysics and ontology the deepest subjects possible or is the question itself a pointless search to begin with?

3:28:48.7 SC: I will give you a possibly surprising answer here. Yes, metaphysics and ontology are the deepest subjects possible. [laughter] And the reason why I'm saying that is there's a whole bunch of ways that we can spin the question so that that actually would not be true. So you have to construe the question a certain way, but you begin the question by saying that you have this possibly irrational desire to go all in on the so called deepest subject. So I take that seriously, that, you know, I don't know whether your desire is irrational or not, but let's imagine that that's really what you want to do.

3:29:18.6 SC: People can make arguments for psychology or for mathematics or for physics, but the thing is, all of those are relatively narrow. They do their thing and they do their thing exquisitely well. Philosophy is a bit broader, right? You can do philosophy, you can do metaphysics and ontology, and you can fit in physics and psychology and mathematics and logic as you want. So as your individual thoughts about what is most fundamental evolve over time, philosophy would be a good home in which to explore those. You wouldn't be kicked out because you start in a physics department and you become more mathematical or whatever. Right? I'm not going to say whether or not this irrational desire is the right way to think about what to do. But I think that given that feeling that you have, it's perfectly legitimate to really look into metaphysics and epistemology and ontology very carefully as something that you might enjoy doing.

3:30:22.3 SC: Robert Rux Andrescu says, imagine that I'm close to a black hole and I throw a radioactive particle toward it, a neutron or a muon. If the particle decays before reaching the event horizon, there's a chance that the resulting decay particles like neutrinos, photons and so on would fly away from the black hole and escape, thus carrying some of their energy away from the black hole.

3:30:41.2 SC: They don't contribute to its mass. If the particle doesn't decay until it passes the event horizon, then its entire energy will contribute to the mass of the black hole and thus to the curvature of spacetime. Have I put a macroscopic object, the black hole, in a superposition of different masses and thus different curvatures of spacetime, therefore putting spacetime itself in a superposition?

3:31:03.1 SC: Yeah, you have. It's a little bit more subtle than that. So number one, the effects are very tiny because neutrons and muons are not very massive compared to black holes. And number two, you should really be thinking about the whole wave function, not just a sort of actual quasi classical probabilistic decay process. But the spirit of the question, the answer is yes. And in fact, what immediately comes to mind is, you're working way too hard to put space time itself into a superposition. Don Page, who's a very well known theoretical cosmologist, pointed this out long ago and did an experiment just kind of for fun, but kind of to make a point. He had some quantum measurement device. I don't remember whether it was like a Stern-Gerlach experiment or a beam splitter or a Geiger counter or whatever.

3:31:49.6 SC: And he said ahead of time, okay, if it clicks in one way, I'm going to pick up this bowling ball and move it to the left one meter. And if it clicks the other way, I'm going to pick it up and move it to the right one meter. And then, so if I believe the many worlds interpretation of Quantum Mechanics, there's now a superposition of the ball being in one place and in the other, and the ball is heavy enough that we can do a Cavendish experiment and detect its gravitational field. Okay? So by the rules of the game, the space time is indeed in a superposition. But the typical Everettian would say, once you're using your Cavendish experiment to measure its gravitational field, you're on one branch or another, you're not detecting the gravitational field of both parts of the bowling balls wave function, you're only detecting one of them. And indeed that's what the experiment shows. But in principle, yeah, you can... There's no difficulty whatsoever in imagining putting space time into a superposition so there are some people out there who want to think that space time itself is still classical, even though everything else is quantum mechanical. I don't think that's something that you need to worry about. But some people do, and they need to work hard to eliminate those kinds of possibilities, to explain why that never happens, that very easy experiment you can imagine doing.

3:33:08.2 SC: Okay, the very last question today is from Marek Boric, who says, I would like to express gratitude that you keep the AMA episodes and the regular podcasts going for this long. I'm really enjoying every episode. Do you think often about gratitude? Do you have any unusual or unique things that you are grateful for?

3:33:27.2 SC: Yeah, I think gratitude is hugely important. I have opinions about it. I think that gratitude is something that only really matters when it comes naturally. I get very annoyed by people who demand gratitude for what they do, but I have... I do think that it's a good human quality to have gratitude naturally. I have a lot of respect for people who are very sincere in their feelings of gratitude and express it very openly and very believably, very plausibly, very authentically, if you want to put it that way. Do I have any unusual, unique things that I'm grateful for? Look, my life is pretty awesome compared to a lot of people's lives. I'm grateful for many, many things. I think that one thing that I am able to do is not take things for granted.

3:34:16.5 SC: Both Jennifer and I, like walk around the house we live in here in Baltimore and go, like, we love our house. How lucky are we to have this house? I have a wonderful relationship. I have a wonderful job, I have wonderful cats, I have a wonderful podcast. And let me be, again, very honest and very sincere. I am enormously grateful, I say this a lot but it's true, for everyone who listens to Mindscape to the podcast. Special extra little bit of gratitude for the Patreon supporters. I will not be shy about admitting that, but everyone who listens, I'm also very, very grateful. Even if you only listened to one episode, even if you didn't like it very much, you know, sometimes it's not your cup of tea, but engaging with what goes on in the podcast, caring about these issues, you know, like, you're not going to get rich listening to Mindscape podcast. You're not going to invent something new because of what I talk about. You're not going to change the world politically or anything like that, it's really about ideas and you're not going to get celebrity gossip or whatever.

3:35:16.4 SC: And I just like it that there are people who share my love for these kinds of ideas, trying to understand the world better. I mean, some of you are annoying when you disagree with me, but I say that in complete gratitude that you are choosing to engage with me. I think that's great. That's part of the deal when you enter the public sphere that people are going to disagree with you, and that's fine. And some people sometimes do it in annoying ways, and we won't talk about those. Right now I'm talking about the gratitude I have for the overwhelming majority of people who engage with Mindscape in different ways. It keeps me going. I've said this before, you know, I got a lot of things to do. I'm busy. I don't have to do a podcast. Why am I doing it? In part because people like Patreon and the ads, et cetera, send me money, but mostly because there's a lot of people out there who like it. And that just keeps me going. I hope you all get something out of it. I think some of you do, and I'm very grateful for that. Thanks, as always, for supporting the Mindscape Podcast. I will talk to you next time.

[music]