AMA | April 2026

0:00:00.2 Sean Carroll: Hello everyone, and welcome to the April 2026 Ask Me Anything edition of the Mindscape Podcast. I'm your host, Sean Carroll. Back when I lived in Chicago, one of the great things about living in Chicago, the food was my favorite aspect, the restaurant scene was really good, but also good was the theater scene. I mean live theater, like plays that you would go to and see people on stage acting. It's my favorite theater scene in the United States, not that I'm a super expert, but compared to Los Angeles, where there's an actor on every street corner, but most of them want to be in TV and movies for obvious reasons, or even New York where there's a lot of live theater, but you want to be a Broadway Star or whatever, or have your one episode arc on Law & Order. Chicago doesn't have a lot of either Broadway-style theater or TV/movie production going on, so the people who were there to act in plays were really about acting in plays. And there were a lot of relatively small-scale independent theaters, some relatively big ones like Steppenwolf and the Goodman, but also a bunch of smaller ones that were really... It really added to the cultural dimension of living in the city.

0:01:09.8 SC: And I got to know some of the different places, and I even was involved in a play now and then, not as an actor, believe me, but sometimes plays, like modern movies, have science themes. And either I could help talk to the actors or the director about the science in their play, or I could give a little talk in front of a performance to the audience about the science that was going on. And I remember one play, I forget the name of it, I'm really sorry that I forget the name of it, but there's a point to the story anyway. The central protagonist of the play was a physicist, and he was a physicist who specialized in black holes. And there were various metaphors in the play about black holes and things like that. But it was very explicit in the play that a big reason why this person had become a physicist is because he had trouble dealing with human beings. [chuckle] He had trouble with interpersonal relationships because there were just too many variables. Like, he didn't know what to do, he didn't know what people were thinking, it was just too hard.

0:02:17.6 SC: Whereas in physics, there are spherical cows. They didn't put it in exactly those terms, but the idea is that there are equations, and sometimes you can solve the equations exactly, and you know exactly what's going on. You have some functions, feeling of control and mastery over it that you just can never get from human relations. And therefore, in a real sense, according to the play, physics can be an escape from the difficulties of the human realm. And I think that there's actually something to this, not for everybody, I'm not saying that all physicists or mathematicians or engineers are motivated in this way, but some of them are. And even when you're not doing that for a living, you can get some solace from thinking about science questions that we do think, some certain philosophical niceties set aside for the moment, have a right answer, right? There's some fact of the matter about what the dark matter is. We don't know it, but we can try to look for it and find out, and eventually we'll discover it. There's a fact of the matter about how to solve the Schrödinger equation, right? Or Einstein's equation, et cetera.

0:03:25.6 SC: The reason why I'm bringing all of this up is because in today's AMA, this month's AMA, there are a lot of physics questions. [chuckle] I get the impression, I have not actually scientifically looked through this, maybe there's no more physics questions in recent months than there ever have been, but there's usually a good number of physics questions and a good number of other questions, and they're all... This is not a normative, me judging the questioners kind of thing, I love all the questions, so bring them on. But I wonder if the world here in the United States where I live and where some of you live, and also in other places in the world, the world is kind of shaky in different ways. Some bad things are going on, and maybe thinking about science gives us a little bit of escape from that. That's not a bad thing, you know, that's perfectly good. You can find escape in watching basketball games or listening to good music or watching TV or going for long walks. Escape is important. Escape is part of the balance of a well-lived life. You want some engagement, you want some struggle in a good life, but then you also want some relaxation, some escape, some... Maybe you want to not completely relax, you want to engage your brain, but you want to do it in a way that is not quite as fraught with consequences for the world as we often get.

0:04:48.1 SC: We do have a world in which the problems around us... There have always been problems around us over the years, but now we have technology that brings those problems to us in a much more vivid way. So maybe that's providing a little bit extra motivation for people to think about the measurement problem of quantum mechanics or what the dark matter is. I don't know. That's a speculation. Let me know what you think. So nevertheless, in the today's AMA, we're going to get lots of questions about physics, lots of questions about other things. They'll all mix together in the usual fun way. Let me, as usual, give a huge thank you to the Patreon supporters of Mindscape who make the AMAs possible. For those of you who are new here, you can subscribe on Patreon for a nominal fee. I used to be able to say that the fee was $1 per episode, but because of capitalism, [chuckle] that is changing. I don't mean capitalism in the sense of me wanting to make more money, I'm perfectly happy with the system. But Apple, where many apps are sold and many people are pointed to, won't let Patreon use their old model, and they're changing the way that things get charged. So we're going to have to change from a charging-per-episode model to a charging-per-month model. Not a big deal, it's just that since we haven't done it yet, I can't tell you what the charge is actually going to be.

0:06:09.8 SC: Anyway, you can go to patreon.com/seanmcarroll, you can sign up. It's a feel-good thing. You'll feel better about yourself and the world for doing that. And it is the support we get from Patreon that really enables and encourages me to do these AMAs. It's the Patreon listeners who get ad-free versions of the podcast, and they also get to ask the AMA questions. So many, many thanks to everyone who's been supporting on Patreon all these years. Who knew that we'd be going on for this many years when I started this whole thing before the pandemic even came along? So thanks to them, and I think that's it. So let's go.

[music]

0:07:07.6 SC: Brandon Wheeler says, "What are the biggest flaws in the Many-Worlds version of quantum mechanics? You vaguely mention them here and there, but if it's not too hard to explain, can we have all of them?" Yeah, that's a perfectly good question. I think that I have said what they are, but maybe not laid them out in perfect detail. And in fact, I'm not going to give a very long answer to this one because I'm toying with the idea that this would make a good solo episode. Not so much the very idea of the flaws in Many-Worlds, but a particular challenge that I'm going to mention in a second. I basically think that there's two, not flaws, but important things that need to be better understood before we can say that Many-Worlds is a perfectly successful theory of reconciling quantum mechanics with our experience. The first is the probability problem. Let me actually, before even mentioning that, let me mention that there's a bunch of things that are claimed to be flaws that just aren't. The ontological extravagance of the theory, the question of what happens to me in the future, which branch will I end up on, where does the energy come from to make all these worlds? None of those is really a problem. Okay? These are problems... Sometimes you ask questions and they're good questions, but there are answers. There are answers to all of those questions. I'm not going to go into them right now because that's not the point of the question.

0:08:22.2 SC: The real questions outstanding are the probability question, which again, I think is more or less solved, but the short version of the problem is Many-Worlds is deterministic. We know that for any allowed measurement outcome, there will be a world in which any measurement outcome appears. So why are we allowed to say that predicting the future, we have a certain probability of getting some answers rather than the others? Again, I think that's more or less, we understand the answer to that question, but not everyone agrees, and I'm sympathetic with the fact that our understanding there is not absolutely airtight. So I would say it's at the 90-95% level, but still there's some lingering question there. The other one, which I think is a real question that I think is fascinating and I'm devoting my research energies to trying to study, is the problem of structure. And what I mean by that is many of you will have heard me say before that quantum mechanics describes the state of the world as what is called a vector in Hilbert space. That's a fancy way of saying that there's a certain mathematical object that is the quantum state. That mathematical object by itself doesn't have any way of saying, "Well, this part of the universe is a planet, and this part is a star, and this part is a puppy dog," or whatever. It's just a vector pointing in some direction in some giant dimensional space. So how do you decide to divide up that giant-dimensional space to be subsystems that look like the classical parts of the world that we know and love?

0:09:59.2 SC: Again, I think that there are answers to this question, but the answers here are much fuzzier. This is something where I think a lot of people don't pay any attention to it. This is kind of my favorite kind of problem now that I'm in my old age. Problems where no one thinks they're problems because they think they know the answer, but nevertheless, the reasoning that gets them to the answer, I think, is highly sketchy. So I think there's a lot of work to be done there. Everettian quantum mechanics, frankly, is almost too simple to explain the world. I think that you have to really think carefully about where the ordinary everyday picture of the world comes from in Many-Worlds, and I think that's a good open question.

0:10:38.8 SC: Emmett Francis... Oh, so yeah, so by the way, therefore, chime in, especially Patreon supporters, but everyone: Do you think that would be a sufficiently good topic for a solo episode? It would be more technical than average. You know, we just had the Daniel Harlow episode not too long ago, so there's some technicalities of quantum mechanics floating around. But I would try to make it as understandable as possible. I'd have more time to explain what is meant by a vector in Hilbert space and things like that. We did have the episode a little while ago about whether time is emergent, so space and matter and stuff is all that emergent in a similar way.

0:11:17.5 SC: Anyway, back to Emmett Francis, who says, "I find myself going to Dan Brown books as a guilty pleasure of sorts, and I can't help truly feeling a bit guilty since he often sensationalizes the science and everything else. And he definitely has had a strange fictional take on previous Mindscape guest Jeremy England." Jeremy England appeared as a character, sort of an unauthorized appearance, in one of Dan Brown's books. "You've written on the scientific mistakes from Brown before, but I'm curious if you have any updated thoughts, especially given his most recent book, which heavily focuses on the science of consciousness."

0:11:50.4 SC: I have no special thoughts on his most recent book because I didn't read it. But, you know, I don't... I've said this before once again, but I don't get too upset about bad science in fiction, okay? Bad science, by which I mean science that doesn't exactly map onto the real-world science as we know about it. You know, I was an advisor on Thor, [chuckle] who is an Asgardian who comes over on the Rainbow Bridge from Asgard to Earth. There's no accurate science there or anything like that. It's you telling a story, and that's okay. Now, a book like a Dan Brown book that sort of pretends to be or purports to be closer to reality than a Marvel comic book, you might have different questions about, and that would be fine. But it depends. If it bugs you so much that you can't enjoy it, then that's okay. If it doesn't bug you and you get pleasure out of reading the book, that's also okay. You should not feel guilty about it. I'm brought to mind, we had a conference at the Santa Fe Institute a little while ago. I forget what topic it was, we have all these crazy topics here at SFI. But Anthony Doerr, D-O-E-R-R, who's a very accomplished author, novelist himself, he wrote 'Cloud Cuckoo Land' and some other things, I highly recommend him, he gave a talk, and he opened his talk by... I think it was something about reality, the topic of the symposium, but he opened his talk by reading, like, the first page of a novel. And he didn't tell anyone what novel it was, but we all... I guessed because I knew it was 'The Da Vinci Code' by Dan Brown. And what he was doing by reading it out loud was pointing out how very little sense it made, okay? [chuckle] So in many ways, in multiple ways, if you actually paid attention to the text of just the first page of 'The Da Vinci Code,' you'd be like, "No, that makes no sense." I actually looked it up so I can read it to you so you know what Tony, Tony Doerr was talking about. I'll just read you a couple sentences from page one of 'The Da Vinci Code.' The scene is that a curator of a museum pulls down a painting and falls down, and he hears a noise and there's someone threatening him.

0:14:16.5 SC: So here's the text, "On his hands and knees, the curator froze, turning his head slowly. Only fifteen feet away, outside the sealed gate, the mountainous silhouette of his attacker stared through the iron bars. He was broad and tall, with ghost-pale skin and thinning white hair." Okay, there I read three sentences, and they might seem perfectly fine, but look at them carefully. So the very first sentence, "The curator froze, turning his head slowly." You don't get to do both of those things. [chuckle] You freeze, or you turn your head slowly. If you're turning your head slowly, then you're not frozen, okay? That's a tiny mismatch there, but it's there. The next one, "The mountainous silhouette of his attacker stared through the iron bars." Fine. What does that mean? A silhouette means you're seeing a dark outline, right? You're seeing nothing but darkness with bright light behind. But then the very next sentence says that the attacker had ghost-pale skin and thinning white hair. "His irises were pink with dark red pupils." But you don't know that if you're just seeing a silhouette! How do you know the color of someone's eyes if all you're seeing is their silhouette? Again, it completely contradicts the very sense of sentence before it. And part of the point is, nobody cares. Like, maybe you care. Maybe for the kind of reader who reads very, very carefully and this stuff is very noticeable to you, very important, then you're going to care. But clearly there are plenty of people who don't care because Dan Brown sells billions of books.

0:15:46.2 SC: And so I had the question, and I was asking both Ted Chiang, who was also there for the workshop, and Tony, I said, "Look, aside from the fact that this is sort of sloppy writing or whatever, people like it. So why? What is it that Dan Brown does? If he's making all these mistakes, there's clearly some skill, some talent, some craft that he has because he is giving some people what they want. It's enjoyable." And Ted told me that in his opinion, the explanation was that Dan Brown sacrifices everything for narrative velocity, the speed with which you're pulled through the text. Famously, Dan Brown has these very short chapters, they end on cliffhangers, you want to keep reading. But even at the level of sentences and paragraphs, you want to keep reading. He somehow puts you in a mental state where you are pulled or pushed, or however you want to put it, to figure out what happens next. And that's a skill, okay? I mean, that's not sloppiness. Like, you can try to do that. Other people have tried to do it, and they don't quite succeed.

0:16:54.6 SC: So the point is that you can read things for different purposes. You can read things just for guilty pleasures. You don't have to feel guilty about it, honestly. Pleasure is a good thing. You don't need to feel guilty about it. You can read Dan Brown books or whatever, Harlequin Romances if that's your thing, the Twilight books, I don't know. But if you get pleasure out of reading it, that's fine. If you don't get pleasure out of reading it because you read very carefully and you want everything to make sense, that's also fine. And that's just a different thing that you're trying to get out or a different kind of reading that you're doing. And when I say it's fine, it's not just narrative consistency or inconsistency, the same thing is true for scientific mistakes or whatever. If you don't care that Back to the Future has a theory of time travel that is entirely incoherent, that's fine. Back to the Future is a great movie. But if it does bug you, like it bugs me, then you're going to want to look for something that is a little bit more consistent. And that's also perfectly fine.

0:17:55.2 SC: Thomas Anderson says, "Max Tegmark," former Mindscape guest, it's my duty, I need to point out to you when anyone mentioned is a former Mindscape guest, "has described the pattern of a human through space-time as like a tube or braid of trajectories, a much messier thing than, say, a planet. How would a poetic naturalist who knows quite a bit about quantum field theory describe the pattern of a human being in quantum fields? Would I shimmer and swirl?" So I'm going to try to answer this, but I'm not exactly sure what either Thomas or Max has in mind when they're saying these things, so there's going to be some guesswork involved on my part. But let's back up. You know, in Newtonian mechanics, there's the idea that you have a particle. A particle is an idealization; a point-like particle has no extent, okay? But you can mathematically talk about it. And in Newtonian mechanics, you would say a particle has a position, it has a location. There it is, it's a dot. But it will evolve over time. At different times, the position of the particle will be different. That's fine. It's not until relativity comes along and space and time are unified into space-time that the dot as a particle at one moment of time becomes less fundamental than the worldline of that particle, the line that stretches through space-time going from the past to the future, telling you at each time where the particle is. And maybe there's a beginning and an ending to the worldline if the particle's created and destroyed or something like that.

0:19:24.4 SC: If you have an extended body like a human being or a planet that is made of many, many particles, then a traditional thing to talk about is the world tube of the extended thing. It's the volume at any one moment of time stretching through all of space-time. If you're a little bit knowledgeable about relativity, you might might get worried that we're talking about the volume at any one moment of time. But the whole point of talking about the worldline or the world tube is that it doesn't matter what reference frame you're in or how you divide up space-time into space and time, the whole four-dimensional thing, the worldline or the world tube, is completely invariant that way. I think that what Tegmark is getting at is that if you have a planet or something like that, it will have a world tube stretching through time, but it'll be fairly simple, right? The planet doesn't do anything very complicated. It goes around the sun, it obeys Kepler's laws, it rotates on its axis. Whereas a human being, which will also have a world tube, does all sorts of weird things, standing up, running around, blinking their eyelashes, waving their arms, right? So a lot more information is involved in tracking the human being over time. I think that's what Max is getting at here.

0:20:39.2 SC: So how would you update that to quantum field theory, is Thomas' question. And I feel bad saying this, but I kind of feel you shouldn't update that to quantum field theory, because this is part of the poetic naturalist credo, it's that there are many different ways of talking about the world, but they should all be respected for their own sake, right? For their own sakes. You should either talk a quantum description or a classical description where they're appropriate, and they need to be consistent where they overlap. But all of this talk about worldlines and world tubes, that's classical talk. That's as if classical Newtonian mechanics or relativity were true. And in the quantum world, as you know, as we were just talking about, particles don't live in space, they live in Hilbert space, which is a whole different kind of thing. And a world tube is not what it would be in Hilbert space. You can kind of vaguely approximate what's going on by imagining a classical particle as being a bit smeared out because there's some uncertainty in where you would measure its position to be, but you're already giving up on some of the accuracy of your description. You might as well just talk classically at that point.

0:21:49.7 SC: The other thing to mention is that this shimmering and swirling thing worries me a little bit. I want to emphasize that if you have, let's say, an electron in an atom, okay? So the electron is orbiting a proton in a hydrogen atom, let's say. It is not moving in the quantum mechanical description. An electron in its lowest energy state, that lowest energy state is smeared out. And sometimes we can't help but speak as if there are quantum fluctuations, right? What we mean by that is if somehow we were to measure the position of the electron, write it down, and then reset so that once again the electron is in its ground state in the hydrogen atom, and measure it again and measure it again, et cetera, we would get different answers each time because there's a probability that comes into the Born rule. That makes us think inevitably that somehow, when we are not doing the measurements, the electron is moving and jittering back and forth. But we know that's not true. That's just excess baggage from our classical intuition. The actual quantum wave function is perfectly 100% static. So you might be a little bit smeared out in that description, but there's no shimmering and swirling going on. Your quantum wave function would just do what it does, which is actually, other than the center of mass of your wave functions when you move your arms around or whatever, other than that, it's a pretty stationary kind of description.

0:23:18.4 SC: Mark Kumeri says, "A few months ago, there was an excellent interview of Edward Witten by Brian Greene, former Mindscape guest, as part of his World Science Festival program. At around minute 49 of the YouTube video, Witten talks about the Many-Worlds approach to quantum mechanics, and he gave a critique I had never heard before. He starts talking about the Copenhagen interpretation. He states that Niels Bohr would say that when a measurement is made, the measuring device records the outcome and the observer learns the answer. But Bohr didn't say in what sense the observer knows. And it can't be that it means measuring the measuring device, as that would be an infinite regress. Witten then says that Everett shifts this logic one step further than Bohr, with the observer's memory replacing the measuring device, but claims that he didn't also solve the issue, as Everett leaves undefined what it means to access the observer's memory, as it can also create this infinite regress. Is this a valid criticism? He seems to imply some mystery about what it means to to know something. I listened to this several times, but I'm still confused. I really hope you watch this and can elaborate."

0:24:21.2 SC: So I did not watch the particular video that Mark is talking about, but I watched a seminar that Witten gave that is also on YouTube where he mentions exactly the same ideas, where he's talking about these very worries. It's clearly on his mind, and we know why it's on his mind. It's on his mind for exactly the reasons that Daniel Harlow was talking about in the recent episode there, thinking about quantum gravity in de Sitter space leads to these weird issues where you just have a one-dimensional Hilbert space, and you just start thinking about what it means to be an observer and all these questions in the foundations of quantum mechanics. It's great. And Witten has been thinking about that, along with Maldacena and a bunch of other people, Raphael Bousso, former Mindscape guest, et cetera. So... But I don't think that this particular worry that Witten has has anything to do with Copenhagen versus Everett. It's a semi-respectable worry I have, I guess I would say, concerning epistemology. Like, what does it mean to know something? Okay, that's a perfectly good question that philosophers have engaged with, but I don't think there's any special physics or quantum mechanics worry there.

0:25:30.4 SC: As a physicalist, what you would say is that what it means to know something is that somehow there are configurations of neurons or your synapses between neurons in your brain that represent a certain piece of knowledge. What does it mean to know that you know something? A different particular set of neurons and their firings and synapses in your brain, the strength of the connections between them. So I don't know what particular configurations of synapses in our connectomes correspond to knowledge, but that's a job for the neuroscientists, okay? It's not really a job for quantum mechanics. Whatever the answer to that is, it's going to be the same answer in Copenhagen or Everett or Bohm or anything else. And I don't think that there's any sense in which the answer is deeply mysterious; it just seems to be technically hard to know what is going on in the brain. So I do think that Witten has been worrying about this stuff, but I don't think that particular worry is especially salient.

0:26:29.4 SC: I will close with telling one little story, which is when I was at Caltech, Edward Witten came to give a seminar, and I was briefly chatting with him before the seminar, and I mentioned that I was thinking about Everett and quantum mechanics and things like that. And he goes, "Oh yes, I read Everett's paper a while ago. I didn't really understand what the point of it was. I'm not sure that he really solved any problems." And I wanted to keep talking, but he had to go give a seminar. So he went up and gave a seminar, and he didn't have any notes. He just stood at the blackboard and wrote for an hour and a half all this high-level math stuff about brains and gauge theories and stuff like that. And the funny thing to me was, immediately when it was done, when the last question was answered, he came back up to me and said, "So here's why I didn't like Everett's paper." So the amazing thing was, not only did he give an off-the-cuff high-level math talk for an hour and a half, but clearly there was a little subroutine running the whole time that was thinking about Everett and the foundations of quantum mechanics. That's what makes Edward Witten Edward Witten. So the rest of us just enjoy talking to him and learning things.

0:27:33.4 SC: Okay, then Ed says, not Ed Witten, well, I don't know, maybe it's Ed Witten, hi there, Ed, if you're writing questions here on the Patreon, our Ed says, "When a society is heading toward authoritarianism, do you think there is a case for deciding in advance the criteria under which you push back or leave? I.e., should you set a 'this far and no further' point to avoid being the metaphorical frog in boiling water? E.g., I've often wondered if you could show the German Jewish population in 1933 what would happen ten years later, how many would have left before it was too late?" So roughly, the answer is no, I do not think that there are criteria for deciding that you would decide in advance, okay? Partly this is... You know, people sometimes ask, "Do I learn things by doing the Mindscape podcast?" I learned something from talking to Elizabeth Anderson, the famous philosopher who was a Mindscape guest. And Anderson is a big critique of the idea of ideal theory in political philosophy. Ideal theory is basically the idea that what we should do is decide what would the perfect society would be. That's the first step. And then the second step is we try to make our real society closer to the perfect society.

0:28:50.4 SC: And her point was, if instead you start with the real society and just say, of all the ways we could change it, what would make it better? A little bit better. You might, after making it better, realize something that you hadn't realized when you were in your previous state. You might learn something that teaches you more about what a perfect society would be like. And therefore, rather than starting by thinking of the ideal society and trying to move toward it, we should start where we are and try to make things a little bit better. I thought that was actually super convincing and super plausible given the nature of complex systems, right? You can't always predict ahead of time what it is you will actually want, whether or not what you think you want will actually work or anything like that. So likewise, I don't think you decide in advance the criteria under which you would leave a society or push back against creeping authoritarianism. Furthermore, there's just a whole bunch of complicating factors here that really make it extra clear that it would be weird to decide ahead of time.

0:29:52.4 SC: I think it'd be perfectly sensible if you were a part of the German Jewish population in 1933 and somehow you had a crystal-clear vision of what was going to happen over the next ten years, that you would want to leave. That doesn't mean you can leave, right? It might be hard. I mean, it might be hard for very practical reasons, maybe they wouldn't let you leave, but also for semi-practical reasons: Do you have the money to leave? Do you have the wherewithal to sort of give up your house and your home and your friends and things like that? Can you bring along all of your family and everyone you know, or would it be okay to not do that? All of these really very difficult real-world questions are very, very hard to decide ahead of time. Finally, I think it depends a lot on who you are and who you are in that society. Like, everyone in Germany in 1933 should have been appalled by what was about to happen over the next ten years. Not everyone was, clearly. But some people have more power to resist what would be happening, and some people, just because of the way the society is organized, are just going to be the victims here. And I think that the people who are just much more likely to be the victims have a much better case for picking up and leaving rather than staying and resisting. I don't judge people who make either choice. I think that in different circumstances, either choice is perfectly viable. But I think that the right choice depends just as much on who you are and what your position is as it does on what is happening in that society.

0:31:25.9 SC: Okay, Eric Olav Chen says, "Does the existence of observers within branches require that the pointer basis be approximately the position eigenbasis? More generally, are there constraints on which pointer bases give rise to branches capable of supporting observers?" You know, I always wonder whether I should answer questions like this because they're somewhat technical. There's a lot of words in there that maybe some of you are not familiar with, and I will of course try to explain them a little bit, but maybe I should just skip the question entirely. Every time I mention that uncertainty, most of the comments say, "No, no, no, you should answer those. Even if we don't understand them, we find them fun."

0:32:05.4 SC: Okay, [chuckle] so I'm going to keep answering them occasionally, all right? But I do want the podcast to be accessible to everyone. So, you know, people get fun in different ways, as we've already discussed. The idea of pointer states and branches, this is something that we talked a lot with Daniel about. In the wave function of the universe, if you're an Everettian, you know, you say, like, "Let's do the Schrödinger's cat thing." You put the cat into a superposition of awake and asleep. And the point is, when you open the box and look, you never see the cat in a superposition; you either see an awake cat or an asleep sleep cat. Now, part of that is a motivation for moving on to a discussion about wave function collapse and realism and the role of observers and all this stuff. But there's also a very down-to-earth technical issue: Why do I specifically see either the awake cat or the asleep cat, even though quantum mechanically I can describe a state that is 1 over the square root of 2 awake + asleep and its perpendicular state, 1 over square root of 2 awake - asleep. In other words, the states in quantum mechanics that are superpositions of definite states of sleepiness for the cat are just as real, just as valid, just as honest within Hilbert space as the states that we actually see, either the cat awake or asleep.

0:33:32.4 SC: So this is called the preferred basis problem in quantum mechanics, and it's definitely a problem for Everett. It's less of a problem if you have hidden variables or whatever, because they just pick out what the right states are, or even if you have objective collapses or things like that. But this is a problem within Everett that has been solved, okay? We essentially know the solution to this one through the work of people like Zurek and others who have been talking about decoherence for a very long time. The idea being that the environment of the world interacts with these different states in different ways and becomes entangled. And when you have a superposition of the cat awake and the cat asleep, the environment, all the photons and all the air molecules, interacts differently with the awake cat and the asleep cat. Whereas if you just had the awake cat, the environment would interact with it, the light and the air molecules would hit the cat, but it doesn't interact with it differently. There's only one thing there; there's the awake cat. So what happens is you branch, and you branch in a very specific way onto very preferred states where things look more or less coherent. Things look like they have a shape, like they have a classical existence. That's why classical mechanics is such a good approximation to quantum mechanics.

0:34:47.4 SC: So what Eric is asking is, is there some relationship between this idea of being a pointer state, a pointer state is a state in which the pointer of a measuring apparatus is pointing in a definite place. It's not the best nomenclature in the world, but the idea is, instead of you looking, let's imagine you had a cat observation device looking and a big arrow, a pointer, that could either point to cat awake or cat asleep. The pointer ends up in either pointing cat away awake or cat asleep, never a superposition of both. So a pointer state is one where the pointer of your measuring apparatus has a definite position, or whatever you want to call it, a definite notion of where it's pointing. And that happens to coincide with what we call position, right? Things that are more or less pointer states have definite coherence in space. Is that an accident? Does it have something to do with, with observers or something like that?

0:35:46.4 SC: So the short answer is I don't know. The long answer is yes, they're certainly related in very, very definite ways. I know some of the arguments here. I wrote a paper with Ashmeet Singh called, "Quantum Mereology," where we talked about exactly this issue, and I'm following it up now with another graduate student. We're still thinking about this problem of how and why do you do pointer states, which are defined by being robust under being monitored by the environment, happen to coincide with states where things have definite positions or pretty close to it? Okay, those are two different statements; robust under environmental monitoring, having a more or less coherent configuration in space. There's no a priori reason why those two things have to coincide, but they do in the real world. So is that purely dynamical? Does it have something to do with observers and the existence of people? Is it anthropic in some way? I think that all of those things are possible and even plausible and even probably true, but I don't think that they have been completely worked out. I think the logic is something like: In order to get observers as we understand them, you need something like classical behavior. And in order to get something like classical behavior, you need something like locality in space. And once you have locality in space and you have a past hypothesis and increasing entropy and things like that, decoherence will pick out states that look approximately static and coherent in space as their pointer states. But all of those statements I just made are up for grabs and need a lot better job of actually showing that they are true.

0:37:30.6 SC: Henry Jacobs asks a two-word question simply, "Jürgen Habermas." You may know Jürgen Habermas was a famous German philosopher who just died very recently. He was quite prolific and quite active even into deep into the later years of his life. He died at age 96. Habermas was a big, big name in social and political philosophy, and I did know a little bit about him. I presume that I'm supposed to interpret Henry's question as, "What do you think? Do you have any thoughts about Jürgen Habermas?" So I do actually have some. They're not especially educated or anything like that, I'm certainly not an expert. But when I was an undergraduate, being a philosophy minor among other things at Villanova, my philosophy mentor, Jack Doody, he has Jürgen Habermas as his favorite philosopher. Jack was actually, got his PhD from Notre Dame in Philosophy of Science under Ernan McMullin, but then later switched to thinking about social and political philosophy himself. So I think a couple of different classes that we had talked about Habermas quite a bit. And Thomas McCarthy, who is currently a Philosophy professor at Notre Dame, is sort of the biggest proponent of Habermas's thought and bringing it to the English-speaking world. And he came to give a guest lecture, translated Habermas, things like that. So yes, I have some exposure there, and I am a fan of Habermas in general. But I'll tell you, this was many years ago, and I haven't done a lot of reading even then when I was supposed to be reading Habermas. I forget what it was we read, I think something from The Theory of Communicative Action. But Habermas is not an easy read, let me tell you. And he's not a sparkling, sprightly writer, which, you know, which is okay. He's, like I said, a prolific writer. He just finished, I don't know, a three-volume history of philosophy, thousands of pages long, in his 90s. So he could turn them out, but it was hard to figure out sometimes what he was talking about.

0:39:37.2 SC: So I came away with, I think, two ideas that have stuck with me over this large number of years, and who knows how accurate my current understanding of those ideas are. But just for those of you who have zero idea about Habermas but may be interested, I can tell you what I remember. The two big ideas: One of them is communicative action. Both words are important there; communication and action. Habermas was a big believer in the power of reasoning. And this is a non-trivial thing. You would think most philosophers are indeed big believers in the power of reasoning. But Habermas was sort of the last great name in what is called the Frankfurt School. The Frankfurt School, it still exists, but it had a heyday starting in the 1930s. So this was people like Max Horkheimer, Theodor Adorno, Erich Fromm, Herbert Marcuse, a bunch of other people. And they basically founded what is known as critical theory. Critical theory being a way of looking at the world in which you try to figure out that what the world tells you about itself, and we're talking about the social world here more than the physical world, so this is not sort of a science statement, this is a social science statement, what the world tells you that it's doing might not always be what it's doing. So you have to sort of be careful to be critical about what is actually going on versus the story you're being told. And just so we're clear, this school of critical theory does lead directly to more contemporary ideas like critical race theory and critical legal theory.

0:41:19.4 SC: You might have heard sort of cartoonish, crazy versions of critical race theory from modern provocateurs like Christopher Rufo, but the original idea of critical race theory is there can be varieties of racism that are not that explicit, that are not that blatant, right? Systemic racism. The very idea of wokeness, the motto, "stay woke," was an exhortation to be on the lookout for systemic racism in society, for ways in which the system systematically discriminated against one group or another without necessarily just coming out and saying, "Oh yes, this group is bad." You sort of hide the racism or the discrimination in the system, and that's why you need critical theory to bring it out. And so the early critical theorists of the Frankfurt School, they lived in a world, so you think about Germany in the 1930s, right? There was a lot going on, both politically and intellectually. They were still in the aftermath of the ideas of people like Marx and Freud and Nietzsche, for that matter. And politically, there were was real communism going on, right? They just had the Russian Revolution, so there was Marxism-Leninism. There was certainly real capitalism going on, and there was a rise of fascism going on. So trying to figure out how to negotiate all these things in an intellectually respectable way was one of the things that critical theory tried to do.

0:42:46.0 SC: And there's a version of thinking about Karl Marx, Sigmund Freud, Nietzsche, and people like that, which says that these people are a little bit too anti-reason, or let's put it this way: They don't give enough credit to the power of reasoning and rationality in human action. They reduce it to either economics or the unconscious or some various cultural factors that are so deeply hidden in our way of living that we don't even really know what they are. And part of the critical theory pushback was to say, "Yes, there are these sort of non-rational things going on, but there is also culture and ideas that are, that are important to us, and we should make a space for them." And I think that that's very much the world that Habermas was still living in, trying to, to balance an understanding of how rationality and reason work with an understanding that sometimes the things that happen to us are not because of rationality and reason. So communicative action was his idea that, you know, when things are going well, human beings, agents, really do try to come to mutual understanding back and forth. It's action, but it's communicative. And all of this is very, very applied to making the world a better place. This is not ivory tower kind of philosophy. Habermas and his predecessors in the Frankfurt School were very active in politics and trying to critique the world that they saw around him. So that push for an ideal of communicative action that analyzes how it works and also tries to boost it is one of Habermas's big themes.

0:44:35.1 SC: The other one that I still remember, again, I'm sure this is a tiny, slightly imperfect way of talking about Habermas's big ideas, but there was what he called the system-lifeworld distinction. So these are two things that are all around us, basically the structures that are all around us that are shaping us in society and countries and other institutions. Habermas sort of divided up into system-kind of things and lifeworld-kind of things. And the system is all of those structures that sort of kind of are self-perpetuating via structures of power, right? Not by talking to each other and trying to reason things out, but by other factors, like the economy is governed by money and how it flows and people's desires for money, or the government, or, you know, it might be described as being organized in terms of power and lawgiving. And it's not all necessarily bad, but it's not communicative rationality, right?

0:45:38.4 SC: So the system is all those things. Whereas the lifeworld is more the side of our world that has to do with communication, meaning, ideas, knowledge, things like that. So the intellectual world, the cultural world, the values we have, all the ways that we communicate, and so forth. So they're both there, they're both necessary; the system and the lifeworld. But the motto that comes out of Habermas is that the system tends to colonize the lifeworld, so that we start off with ideas and we talk to each other and we reason together, but there's a tendency, and this is sort of a very German sociological kind of way of thinking, there's a tendency of things to systematize, for the system to colonize the lifeworld by turning ideas and talk and communication into structures that are sort of stuck there without necessarily relying anymore on some particular rational underpinning. And of course, Habermas is in favor of resisting the colonization of the lifeworld by the system. And anyway, that's what I remember from all that.

0:46:52.0 SC: And it's all very... It makes a lot more sense, it has a lot more impact in the context of all the different discussions that were going on among philosophers, social theorists, political thinkers, and social thinkers of all various sorts. Habermas has, basically, there's Wikipedia entries for Habermas and his dialogue with Thinker X for a large number of X's. You know, Habermas and Gadamer, Habermas and Chomsky, Habermas and Rawls, Habermas and Derrida, Habermas and Foucault, right? They're all talking to each other in different ways. Apparently, I don't really know a lot about this, and I'm not in favor, as you know, of hero heroism, turning people into heroes. I think we should try to understand people's ideas and even admire them for their actions and their ideas, but not trick ourselves into thinking that therefore they're good people if we don't know them personally. But all the news stories, all the reports are that Habermas was a really good person as well as a brilliant philosopher. One story that was bouncing around BlueSky was by this woman, and I'm not going to remember it exactly because I wasn't planning to tell it, but it's apropos here, this woman was helping to organize a public lecture by Habermas. And she was either, you know, at the door where they were letting people in with their tickets, or she was standing around there, and there's a long line, you know, going to see Habermas was a very exciting thing in Germany.

0:48:17.8 SC: And then, you know, she's taking the tickets or whatever, and she looks up, and there's Habermas. He was standing in line for his own lecture. And she's like, "Oh my goodness, Professor Habermas, you do not need to stand in line. You're giving the lecture. You can just come in." So that's a nice thing to hear about the people who you think, it's... You know, we're in a world where sometimes we have people who have done intellectually impressive things, and therefore... And later we learn that they've been unimpressive in other ways or even bad in other ways. So it's nice just to know that most of the reports you get about a person were very positive in nature at the human level. So yeah, Habermas obviously led a big, long, productive life. He had his controversies. He's well worth understanding. I'll confess, I did toy with the idea a while back of, like, inviting him on the podcast, right? Why not? But you know, he's in his 90s, German is his native language, he's writing a several-thousand-page history of philosophy. I figure he has other things to do than appear on the Mindscape podcast, so I never did ask him. So you can't have everybody. That's okay. But maybe we should have someone to talk about that kind of thing, because I'm clearly not the world's expert on it. But for those of you who are not familiar with Habermas's work, it's the kind of thing that is worth studying up on just a little bit.

0:49:40.4 SC: Owe, O-W-E, says, "At the end of February's AMA, you talked about how it's still financially worth going to college. I think that's true on average still, but the variance has gotten way wider with the increase in loans and reduction in economic opportunity they provide. That's not to say that I disagree; normatively, I think education is a positive. Rather, your comment struck me in how hard it is to talk about outcomes and choices in terms of their ranges and probabilities without reducing to binary classifications like 'worth it' or 'not worth it.' Especially as a physicist, I'm curious if you think there's a better way we can have such discussion."

0:50:14.7 SC: I don't know if there's a better way we can have such a discussion. I get it that it's frustrating to have these very complicated multi-dimensional questions reduced to a "worth it/not worth it" kind of distinction. But at the end of the day, there's a lot of factors that go into it, but any individual person is either going to go to college or not. They will decide whether it is worth it or not worth it for them to go to college. My major point in the AMA, and the solo episode I did, or the holiday message on the romance of the university, is that it's worth it not because it gets you a higher income, but because it makes you a better human being. That's not necessarily true, you can go into college and come out a worse human being than you went in. But I'm thinking about those people who would like to become a better human being, that is part of their goals, and I do think that there's a lot of tools and resources that are there in the four years of undergraduate college experience that are almost irreplaceable elsewhere and that are super-duper useful for becoming a better person, if that's what you want to do. Now, how do you value that against the amount of money it will cost and the student loans you will accrue and things like that? That has to be an individual choice. I can't provide you an algorithm for that. I do think that a lot of people... I don't know if it's everybody or not, but a lot of people end up spending too much on college because they want to go to the fanciest one that they can get into.

0:51:52.4 SC: I think this is maybe less true than it used to be just because all colleges are pretty expensive these days. But I'm one of the people who believe that almost any good college or university has way more resources than any individual undergraduate can possibly absorb in five years. So the difference for an undergraduate in going to the world's best and fanciest university versus their local state university is not that big in terms of what they can learn and what they can experience. There are big differences, especially in the other people you get to know. If you're the kind of person who cares about having a good Rolodex, as we used to say, of influential and important people in other areas of life, then going to Harvard or Princeton or Yale is definitely valuable. If you're the kind of person who really, as a matter of socializing and your personal life, prefers to be surrounded by nerdy people who are doing all technological things, then going to Caltech or MIT might be for you. But in general, to get the worthiness of the college experience, I think that there's plenty of different kinds of places you can do it. And one simple way of saving money is to go to the least expensive of the ones that you think would serve that purpose.

0:53:11.4 SC: I guess the final thing to say is, look, it's a system versus lifeworld kind of thing going on here, right? Going to college and getting that university experience should be a lifeworld question. It should be about thoughts and ideas and talking to each other about them and learning. And so often it gets squeezed into a question of acquiring skills and getting a better job, okay? Purely economic considerations. And I think that I can't tell you how to balance all the considerations because they're all real and I respect all of them, but don't go too far in just thinking about one rather than the other. That's the best I can do. Very far from a perfect answer to your question, but that's what you're going to get from me right now.

0:53:53.4 SC: Michael Bright says, "The James Webb Space Telescope has seen things that no other space telescope has seen. For example, it has helped us deepen our understanding of the accelerated expansion of the universe. But I believe the JWST is built to focus on small regions of space, whereas the Nancy Grace Roman Space Telescope is a wider view. Can you help help those of us who want to support science research understand the importance of supporting Nancy Grace Roman and other science exploration missions that have paid off very well for NASA so that we can advocate for their continued support?"

0:54:24.9 SC: Sure. And I think this is actually a really good question because it's not always clear when we hear about these different telescopes, these different satellites, why we need more than one. Why don't you just build the best satellite out there that does all the telescoping you need? Sometimes it's kind of a little bit obvious or clear. For example, a satellite that is measuring the cosmic microwave background is going to be very different than one that measures optical, visual light, or even infrared, which is what JWST does. But among those, the Roman Space Telescope is different than JWST, even though they're both sort of optical, approximately visible light wavelength telescopes. And the biggest single difference, as Michael says, is that JWST has a narrow field of view and Nancy Grace Roman has a wide field of view. And this is with mirrors, the collecting area of the mirror in the telescope. So anyway, what happens in these telescopes ever since, I don't know who invented the reflecting telescope as opposed to the refracting telescope. In the refracting telescope telescope, you send light through glass lenses to focus it so you can collect a lot of light onto a small region. But going through glass inevitably messes you up a little bit. So a reflecting telescope where you use mirrors to collect the light is actually more efficient for precision astronomy. It's also just easier to make these mirrors than to make perfect lenses. So the mirror collecting area, the size of the mirror, basically tells you how many photons the telescope can bring in. The more photons, the better. Okay, so far so good.

0:56:06.4 SC: But the difference here being that you can still affect the field of view. So you can collect a lot of photons in a very, very narrow area, or you can collect a lot of photons in a very wide area. And what is the difference? Well, obviously, if you're collecting more or less the same number of photons, I know the real astronomers here are going to be be mad at me because it's not really the same number of photons, it's the same collecting area, but you know what I mean, you get to see deeper if you focus in on a very tiny area. If what you're interested in is things that are very far away and very dim, then something like James Webb, which really focuses in on a narrow area, is what you want. And that's what you have; that's what the telescope actually does. And the reason why they're interested in that is they want to look at the very first galaxies, the formation of things in the early universe. And then also JWST turns out to be very good at looking at exoplanets, which are not very far away at all, but they're very, very faint and very, very specifically located at a place where you can point the telescope.

0:57:12.4 SC: So the advantage of that is that you can get a lot of image of something very faint. The disadvantage is you better know where you're looking. If you just point the JWST at a random part of the sky... You can do tricks so you get, like, an ultra-deep field kind of thing, but if you focus it as much as you can on the tiniest possible area, most of those parts of the sky are going to be empty. Nothing's going on there. Now, if you do the same thing, the same collecting area in the mirror, but you have a very wide field of view, then what you can do, you're not going to see very deep. You're not going to see things in the very, very early universe. What you're going to see are many, many things in the relatively nearby universe. When we say relatively nearby, we don't still mean very nearby; it's still billions of light years away, or at least a billion light years away. But you have a huge advantage now if you're trying to survey things, if you're trying to look at many things at once.

0:58:09.4 SC: So if you're trying to measure the large-scale structure of the universe, or if you're trying to search for things like supernovae that come and go unpredictably, the only way to do that, you're not going to focus in on one galaxy and look for a supernova, that's going to happen once per century. You have to look at many, many, many galaxies at the same time to have a good number of supernovae collected. So that's why the Vera Rubin Observatory here on the ground has a very wide field of view. It is actually, actually going to be taking multiple pictures per night and scan as much of the sky as it can. And looking for supernovae is a big part of its mission, also looking for asteroids and things like that here in the solar system. So Roman is like a compromise, it's not looking at the whole sky, but it's not focusing in on a small patch either. And it has a much clearer view, of course, because there's no atmosphere out there in space. So you have different missions. The JWST will teach us about early galaxies and exoplanets. Roman will teach us about the overall structure of the universe, the acceleration of the universe, the Hubble tension, all of those things. So it's just different science that you're doing with different missions. Both of them are very worth supporting.

0:59:23.4 SC: Jan Drugalia says, "Priority question." Remember that every Patreon supporter gets to ask a priority question once in their life, and I will do my best to try to answer that question. So Jan says, "I've noticed a trend where some of my friends with STEM backgrounds are falling into strong support for the UAP, the Unidentified Aerial Phenomena, what we used to just call UFOs until people thought that that didn't sound respectable enough. They point to the congressional hearings and the anomalies of the "ATLAS interstellar object as evidence that there is something out there. While I love and respect my friends, Congress, and the world witnesses, I struggle with the fact that we still have zero peer-reviewed physical evidence for UAP. Until that evidence exists, I choose not to invest more time in the topic, and that feels almost impolite toward my friends. Did the congressional hearings or the discovery of ATLAS update your own Bayesian priors for UAPs at all?"

1:00:18.4 SC: No, they did not. And my priors are very, very low. Now, of course UAPs exist, or UFOs exist. There are things in the sky we haven't explained yet, okay? That's never a surprise. There will always be such things. The question is, do these have anything to do with aliens or extraterrestrial technologically advanced civilizations? And the credence on that remains hilariously low, low enough that you don't really need to worry about it. It's especially a tell if they think that the ATLAS interstellar object has anything to do with this at all. We have interstellar objects entering the solar systems. They're in the solar system. They're basically comets. They're basically icy things that come into the solar system and fly by at high rates of speed. You can be a little bit intrigued because sometimes these things are seen to accelerate, and you go like, "Oh, how can a dead rock just accelerate?" But comets are not rocks. They're actually very icy, and they have all this material on them that heats up when they come close to the sun and pushes them, so they are seen to accelerate. But it's not because they have an engine, it's just because they're warming up and sort of outgassing. So there's zero reason to think that the ATLAS interstellar object has anything to do with aliens or technology or anything like that.

1:01:36.4 SC: And again, you know, ask yourself the Bayesian question: If it really were technology from aliens, why would it just fly by? Why would it look so much like a comet? Why wouldn't it just stop and say hi? None of these are remotely plausible, these scenarios that you have to cook up to make this be related to aliens in any way. As far as the congressional hearings go, I mean, look at the history of this. For at least the 20 years that I have been talking about this stuff online, every year I say something like, "No, UFOs are not aliens," and I get people saying, "Oh, there's going to be evidence coming out six months from now. You're going to be sorry, you're going to change your tune, you're going to regret saying this." 20 years later, I'm still waiting to do that. The idea that aliens would come by, be good enough to have technology that can travel across interstellar distances and visit us, but not have the technology to remain hidden from us, the idea that they would just crash and get little fuzzy photographs or, you know, eyewitness testimony from people in airplanes or whatever, is just crazy. That's just very, very unlikely to happen. Whereas the idea that human beings would make the mistake of thinking that and keep tantalizing you with evidence that is just beyond actually being concrete, that's 100% believable to me. So having congressional hearings or whatever has not changed my credences at all.

1:03:06.4 SC: Gag Halfron says, "What do you consider to be the emergent long-term geopolitical effects of the US-Israeli excursion in Iran?" Well, of course, this is just terrible and tragic. I mean, the whole thing is tragic. The ongoing wars for a long time in the Middle East have been tragic. The Hamas attack on Israel was tragic. Israel's total decimation of the Gaza Strip has been tragic. The US and Israel invading... Not invading yet, but maybe that will happen, but attacking Iran has been tragic. Israel going into Lebanon has been tragic. It's all just terrible. Like, it's not just annoying, there's people there who are living there and are really suffering because of this. The Iran debacle is especially bad because it serves literally no one's interests. It's only happening because we... I'm only going to talk for the United States here because I don't follow Israeli politics like that, but here in the United States, I can speak with complete confidence that the people who are making the decisions are idiots. They do not know what they're doing. They're very, very stupid. They have very bad motivations. If you listen to Pete Hegseth, our Secretary of Defense, and other people talk about their bloodlust for war and to commit war crimes and to spread Christianity across the globe, et cetera, it's horrifying. It really just makes you shake your head.

1:04:33.4 SC: And of course, it was entire... There's no purpose being served here. The purpose that is trotted out, which, by the way, let me... I keep backing up because the whole thing is just so terrible, ordinarily, if you were going to do something major like this, like, Iran is not a small country, okay? They're pretty tough. And you would at least, to the general public in the United States, make a case for getting involved in something this dramatic. But the administration did not do that. And the case that they're making ex post facto has to do with Iran's nuclear program and the possibility they would make weapons of mass destruction. We already had a treaty in place that was preventing Iran from doing that, and Donald Trump in his first administration left the treaty unilaterally because he just was annoyed that it had been negotiated by the Obama administration. And so we're doing nothing that has absolutely any purpose whatsoever. And oil prices are going up. Many people have died in Iran. Apparently, Iran is going to come out of the situation geopolitically stronger than it went in. Right now they're saying that the Strait of Hormuz, which is a little tiny strait, I suppose, where many, many tankers go through to bring oil to the rest of the world, not just from Iran but from other Persian Gulf states, is going to go completely under Iranian control, whereas it was not before. And this affects not only the United States and Iran and Israel, but every country in the world because all of our supply lines are interconnected in very important ways.

1:06:10.4 SC: And so it's especially heartbreaking to me because I know that the government of Iran is terrible. I mean, there's just nothing good you can say about it. It's an autocratic, repressive theocracy, and it deserves to be gone. But it's also not very popular in Iran, as far as I can understand it. Iran is not, by its nature, an oppressive theocratic regime. There's a lot of strongly liberal currents in the Iranian people. Education and... Science education in particular in Iran is very high level, especially compared to other countries in that part of the world. We have great physicists who I know personally who've come from Iran, including Cumrun Vafa, former Mindscape guest. And a lot of the people in Iran don't want to be led by a repressive theocratic regime. The United States has a lot of responsibility for the fact that they are, because United States historically, it's not just Donald Trump, it's just the United States, likes to throw its weight around and think that it should decide who runs other countries. And it instituted the Shah of Iran many, many years ago. You know, I'm old enough to have been barely... I sort of became interested in news and politics and things like that around the time of the Iranian Revolution and the hostage crisis and Jimmy Carter versus Ronald Reagan and things like that.

1:07:31.4 SC: So this is us reaping what we've sown a long time ago. And if anything, there was a possibility that the Iranian people would have been able to overthrow their regime, there were demonstrations against them, et cetera, and the United States, of course, has blown it because, as mentioned already, we are ruled by idiots. You know, they can't get it through their heads that if you drop bombs on a country, that country is not going to like you as much. It goes back to Dick Cheney in the Iraq War in the George W. Bush administration saying that he thought that we would be greeted as liberators when we invaded these other countries. And it's become a joke because if you invade other countries, even if the people of those countries were against their own regimes, they're not going to like you invading them. They're going to change... They're going to rally around the flag, as it were. And so now we may have even lost any possibility that we might have had for a regime change in Iran. So anyway, I'm not an expert on any of these things. You shouldn't take my view as especially super well-informed compared to the people who are well-informed. But it just makes me shake my head in sadness. So I did answer that one, but, you know, it's not something I talk about very much just because there are smarter and more informed people out there who you can listen to, and I encourage you to do that.

1:08:53.4 SC: Kevin's Disobedience says, "Say one good thing about loop quantum gravity and one bad. What's it got going for it, and what are its major shortcomings as you see it?" I think it's easy to say both good things and bad things about loop quantum gravity. For those of you who don't know, loop quantum gravity is basically a clever way of trying to quantize general relativity. You've all heard that gravity and general relativity... Sorry, general relativity and quantum mechanics are hard to reconcile. So maybe that's because the version of general relativity, or the way that we're writing down general relativity, is not the most convenient for quantization purposes. That's basically the philosophy behind loop quantum gravity. It started with Abhay Ashtekar, who figured out a way to rewrite the fundamental dynamical degrees of freedom of general relativity in a different way. And then people like Lee Smolin, Carlo Rovelli, both former Mindscape guests, worked out how to rewrite those variables in yet a different way based on loops. Basically, the loops in question: You take a vector or something like that, and you transport it around a loop in space-time, and you ask, "How does it get rotated by the curvature of space-time?" So you can try to quantize that representation of general relativity rather than just the usual one where you have a metric and so on.

1:10:14.4 SC: The good thing to say about it is it's a perfectly natural, sensible thing to try, right? You should try to quantize general relativity, see if the reasons why it hasn't been working are just because you didn't use the right variables or something like that. The problem is there's almost no chance it's going to work. And I think it kind of hasn't worked, in fact. I mean, there's not a big understanding that we now have of quantum gravity because of loop quantum gravity that we didn't have before. Or big understanding of quantum gravity, for that matter. And I think we kind of understand why it doesn't work. Unlike other forces of nature, not only is gravity different because it's the metric of space-time and whatever, but from a quantum field theory point of view, and general relativity is a classical field theory, after all, from a quantum field theory point of view, you know that if you want to get an exact theory, not just an effective theory, but if you want to get a theory that is valid at all energy scales, then the trouble is going to be at very, very short energy scales, right? Energy scales where there's wild fluctuations in all the fields, the space-time metric itself, the energies are very high, we have no experimental traffic, data, things like that.

1:11:29.4 SC: String theory specifically solves those problems by smoothing everything out at high energies because instead of point particles, you have strings. They're sort of floppier and looser, and you can show mathematically that all the worrisome infinities that you might have worried about go away in string theory. In loop quantum gravity, there's nothing to guarantee that anything like that is going to happen. And furthermore, loop quantum gravity treats general relativity as separate. It treats the curvature of space-time as one thing, and then whatever matter fields you have as completely other things. But the infinities and the worries that you would get at short distances in gravity depend not only on gravity, but everything gravity is talking to. And the whole thing about gravity is that it talks to everything. So there's no reason to suspect that somehow gravity will be quantizable until you also understand all of the other fields in the universe. Again, in string theory, that naturally happens. All the fields in the universe come from the propagating strings; it's a complete unification. In loop quantum gravity, all the other fields are completely separate. To imagine that there's some conspiracy that makes everything smooth and finite in loop quantum gravity despite not knowing what the matter fields are doing seems implausible to most of us thinking about quantum gravity. That's why, honestly, it's not that popular among people who do quantum gravity. It's maybe the second most popular after string theory, but that's not really saying very much.

1:13:00.4 SC: Iminfinity Category says, "In your 2020 podcast episode featuring Sean B. Carroll, you mentioned that you've received looks from peers when they find out that you're active in public science discourse, just as your evil twin did when he took up the writing of biographical material as opposed to just research. You've also mentioned often that doing anything other than research isn't conducive for tenureship for junior researchers. Can you say something about when and how you started breaking away from the research-only mold? As a junior research fellow interested in writing for the public, should I just wait for tenure before doing anything?"

1:13:33.2 SC: You know, I can't give anyone precise advice here because, as I always say, everyone's situation is different. In particular, it depends on, where you are, if you're a junior faculty member, what kind of university you're at, what they're interested in, how your research has been going. Like if you're... As I also like to say, if you're a genius, if the whole genius thing is working out for you and everyone recognizes that you're a genius, you can do whatever you want, you'll be fine. It's a matter of shifting probabilities and likelihoods. It's not a matter of absolutely yes or no, should you do this or not. So you have to judge that. It's certainly the safest to wait for tenure before doing anything. Once you have tenure, you can be much more confident that you can mix in research with other kinds of activities, and that will be okay. They will not fire you. Indeed, this is one of the reasons why it's dangerous to do that before getting tenure, because the universities know that once you get tenure, they have a much less leverage over you. And so if you're the kind of person who likes doing other things, they will worry that you will stop doing research. And so that's something that's very hard to prevent. But it can be done. People do do it. People are successful starting with public outreach things at a very young age. I think you have to judge from the department you're in, the university you're in, and things like that, how likely you are to get away with something like that.

1:14:57.4 SC: Igor Kopolov says, "The story I've heard about the history of quantum mechanics is that after Schrödinger wrote down his equation, he hoped that the wave would naturally coalesce toward a point over time. That wasn't the case at all, and only later did Max Born propose that the wave should be thought of as a probability distribution. The part I'm missing is why, without the later insight, does the Schrödinger equation appear to be on the right track? What problem does it solve even if you don't know about the Born rule?"

1:15:22.8 SC: That's a good one. It's a good question, because I know the answer, the answer is it solves the spectrum of radiation from hydrogen and other atoms. Remember the big thing that physicists were focusing on, physicists always like to have a specific, potentially solvable problem to look at and think about, and in the case of quantum mechanics in the 1920s, it was electrons in atoms. And they would both absorb radiation and give off radiation, and that was quantized, right? They would jump up and down in energies by certain discrete amounts. Niels Bohr had come up with a kind of ad hoc rule for what kinds of energy levels were allowed that worked very well for hydrogen, an atom with just one proton and one electron in it. Once you added another electron, and so the electrons are talking to each other, there was no guidance from Bohr's theory about what to do next.

1:16:18.4 SC: So the Schrödinger equation, forget about observations, forget about measurements, forget about probabilities and collapses and things like that, just solve the Schrödinger equation for what electrons do in atoms, and you turn out to get exactly the right answer. It's very, very beautiful. You can find the energy levels. In fact, if you look at the title of Schrödinger's paper where he proposes the Schrödinger equation, the title is "Quantum Mechanics as an Eigenvalue Problem." Eigenvalues are just basically the energies of these specific states that the electron can be in, these specific solutions to the Schrödinger equation. So that's what he was worried about. And the thing you're measuring there is the energy of a photon that is emitted by the electrons as they shift from one energy level to another. So you're not directly measuring the location of the electrons. Indeed, what Heisenberg was saying, contra Schrödinger, is that there's no such thing as where the electrons are or what the wave function of the electron is doing; there is only what you measure at the end of the day. That was the birth of the Copenhagen interpretation. So there was still good quantitative reason to think that Schrödinger was on the right track even before the Born rule came along.

1:17:31.4 SC: Kyle Cabosaris says, "Do you have any writing tips for a first-time author who wants to write a book about a topic that already has many of other popular and technical books in the market, for example, quantum mechanics? How do you personally find the courage or motivation to write on things that have been explained by others ad nauseam?"

1:17:49.1 SC: This is a really good question. And this is unlike the previous one where it's good because I know the answer, this is good because I don't know the answer. It's very, very hard. It's very common for a first-time author to have a topic where they know perfectly well someone else has written about it already, or many people have written about it already, but they feel that no one has done a good job. And so their sales pitch to agents or publishers or whatever is, "Yeah, I know that's been written about, but those books are all bad. Mine will be good." And I can tell you that's not a very convincing sales pitch because everyone thinks their book is going to be good. You're not actually conveying any new information to the publishers or agents when you tell them that.

1:18:30.4 SC: So I think you not only need to think that you're going to write a book that is good, but you have to have an angle. You have to have... If you're writing about something that's been written about many times before, you have to be doing something new and different, at least usually. You know, there certainly are, if you look at the set of books that get published, there absolutely are books that say nothing new [chuckle] but are just, "Well, it's been a while since there's been a book about this topic, and we need to sort of freshen it up a little bit, reach a different audience," which is a entirely fine. But for the most part, you want to say something either substantively different or in a different way than it's been said before. Maybe write your book in the form of a dialogue or something like that. Have more pictures. You know, when I wrote my quantum mechanics book, in 'Something Deeply Hidden,' I talk a lot about quantum mechanics as quantum mechanics, but the main thrust of it was talking about Many-Worlds and emergent space-time. So that was something a little bit different. How you actually go about doing that, that's your job. That's what you're going to get paid the big bucks to do, figuring out a way to put a new spin on an old topic.

1:19:36.4 SC: Mikhail Sirotenko says, "I have a question about the timeless universe and Boltzmann brains. If time is not fundamental and the universe is just a static wave function with different moments in time just components of this wave function, then the amplitude of those components would depend on the chosen basis. And assuming that we cannot choose this basis arbitrarily because we need a basis that allows for observers and laws of physics, does the selection of such a consistent basis mathematically suppress the amplitudes of Boltzmann brain states relative to the classical-looking states?"

1:20:07.0 SC: No, probably not. [chuckle] I think that the question is a little bit difficult to answer directly because I'm not quite sure I agree with what is going on here. So to back up a little bit, you've heard me talk about how in quantum mechanics, the state of the quantum system, whether it's the universe or an electron in a hydrogen atom, is a vector in Hilbert space, this gigantic vector space. And therefore, there's this very special property that quantum mechanics has; that different physical states of the system can be literally added together. You can add vectors together. And that's a big part of hoping to explain how emergent time works, because you could take what you would ordinarily think of as the state of the universe at different moments in time and just add them together. And you could actually, if you know what basis to use in the vector space, the basis we're talking about here is literally the set of basis vectors. So if you have a plane that... You know, you have x-axis and y-axis, there is a basis vector pointing along the x-direction and a different basis vector pointing along the y-direction. Informally, what you think is that nothing should care about what your basis is. Your basis is something you choose; it's a convenient language in which to express the values of your different vectors. But it's not exactly true that all bases are the same. There are physical states that you will see as an observer and other states that you won't. Schrödinger's cat has the famous property that you will see the cat awake or the cat asleep, not 1 over the square root of 2 (one + the other one). And that's because of physics. Physics explains why certain states are more likely to be observed than others. There's no rule that says you have to use those states as a basis in Hilbert space, but you can, and it's often very convenient to do so.

1:21:59.4 SC: So what Mikhail is asking about is, since we cannot choose the basis arbitrarily, does the selection of a consistent basis suppress the amplitude of Boltzmann brain states relative to classical-looking states? The selection of a basis does not, no. You can very easily have a classical-looking Boltzmann brain. There's a quantum fluctuation that leads to the Boltzmann brain popping into existence, and that would be a perfectly legitimate basis state, just like any of the other states we use, like the cat's awake or the cat's asleep or whatever. So I don't think that's enough to do it. There are plenty of quantum mechanical subtleties about how to deal with Boltzmann brains in quantum mechanics, and I wrote a paper that I've mentioned many times with Kim Boddy and Jason Pollack saying that in certain circumstances, the dynamics of the state can be such that Boltzmann brains never appear. But it has nothing to do with choosing the basis correctly, and I don't even think it has much to do with the emergence of time, unless I'm just missing something about the question.

1:22:58.4 SC: Wonder says, "Light cannot escape from a black hole. I understand this when I think of light as a particle. Please explain it in the context of a field." Sure. There is something called the speed of light. Remember, the speed of light was invented and talked about before we knew that light was made of particles, right? The idea of a speed of light, of course... The concept of a speed of light was measured, et cetera, long before we had electromagnetism as a theory. But once Maxwell wrote down his equations in the mid-1800s for electricity and magnetism, the speed of light suddenly became important in a way it hadn't before. It was always important because it was the speed light moved at, but now it's a constant of nature that seems to be the same to everyone and appears as a fundamental parameter in these equations called Maxwell's equations. And that's when people knew for sure that what light was was a vibration in the classical electromagnetic field. So the fact that a vibration in the classical electromagnetic field moves at the speed of light is built into Maxwell's equations from the mid-19th century. And it's much like if you have a pond of water and you throw a pebble into the pond and you see a little wave ripple out, that wave moves at some speed, right? Same exact thing is true for light considered as an electromagnetic wave.

1:24:26.4 SC: So the only other thing you have to know is that when people doing general relativity say light cannot escape from a black hole, what they mean is black holes are regions of space-time bounded by light cones. That is to say, the event horizon, the region surrounding the boundary of the inside of the black hole to the outside of the black hole, is moving outward in a very real sense at the speed of light. The reason why nothing can escape a black hole is because, literally, you need to move faster than the speed of light to escape a black hole. So light cannot travel faster than the speed of light. That's true whether it's a wave or whether it's a particle. If it's a field, then what light is is a vibration in that field, and those vibrations move at the speed of light. Therefore, they cannot escape a black hole.

1:25:16.4 SC: Nikola Ivanov says, "You've argued that if the universe settles into a stationary de Sitter quantum state, it shouldn't keep producing Boltzmann brains." Aha, that's the paper I just referred to, yes. Nikola says, "But a single de Sitter horizon patch is also said to have a finite entropy, which makes it sound as if that patch might have only finitely many possible states. And a finite-state system is usually expected to recur over long times. So why doesn't that make the Boltzmann brain problem come back?"

1:25:42.4 SC: Yeah, this is a perfectly good question. And so again, to try to give some of the background here, we just said that quantum states are vectors in some big-dimensional vector space called Hilbert space. But what does "big" mean? Does "big" mean infinitely big or finitely big? As Nicola says, the entropy of a region of space, of de Sitter space, that is to say, empty space-time with a positive cosmological constant, the entropy is finite. And that points to that region of space being described by a finite-dimensional Hilbert space. But the point is... I'm sorry, I should say one more thing, if you are in a finite-dimensional Hilbert space and you're obeying the Schrödinger equation, no matter where you start, you will eventually come back to where you left. Because the Hilbert space is finite-dimensional, there's not an infinite number of places you can go. You will eventually have have to recur, return to your starting point. And likewise, you will fluctuate forever, you never evolve into a static state and just stay there for all of eternity. Whereas if Hilbert space is infinite-dimensional, then the state can keep evolving forever and never return or never fluctuate, it can just go into a region of Hilbert space where everything looks perfectly static to an observer. So how do you reconcile that? Well, the answer is, knowing that our observable patch of de Sitter space is described by a finite-dimensional Hilbert space doesn't say anything about what's outside the observable patch.

1:27:12.4 SC: So we say in our paper very explicitly, there are two choices. One choice is there is a strict finitude for all of Hilbert space, that not only is our observable patch a finite-dimensional Hilbert space system, but the whole universe is finite-dimensional. And then we say you will get Boltzmann brains. But it's also completely allowed, given what we currently know about the universe, to say that outside there's infinitely more dimensions of Hilbert space, and really we are not a closed system. Our observable universe is connected to the rest of the universe, and therefore it can basically shake off all of its peculiarities and settle down. It can obey the cosmic no-hair theorem that says that no matter what is going on in our universe now, classically, with a positive cosmological constant, the universe settles down into a quiescent state. If Hilbert space is truly infinite-dimensional, that can happen quantum mechanically as well. And then you're not going to get any Boltzmann brains.

1:28:12.4 SC: Ryan Cobine says, "I found the following martini to be palatable but a bit off, peppery, which isn't a bad thing, and also a touch sour, which is a bad thing. What alterations would you make to bring this above the merely palatable to enjoyable?" And then the recipe he gives is 1 1/2 ounces Ford's Gin, 3/4 ounces of Dolin Vermouth, and a dash of Regan's Orange Bitters, expressed lemon oil, lemon peel garnish, stirred. "P.S. This is my first time trying a martini."

1:28:40.7 SC: Well, I think this is... I don't see anything wrong with that martini recipe per se. It'll depend on your personal tastes whether you like it or not. I would say that you're trying a little hard to put stuff in there, right, with the orange bitters and the lemon oil and things like that. Lemon peel garnish, completely 100% fine. I would say just start with a more basic martini. Just do the gin and vermouth and the lemon peel. And then little bit by little bit, add orange bitters, lemon oil. It might depend on the details of the bitters that you're adding, of the gin, et cetera. But a martini, you know, is driven by mostly the gin and the vermouth, I think, if it's done correctly. And both of those are not, smack-you-in-the-face-with-flavors kind of thing. They're kind of delicate in both cases; both the gin and the vermouth, they have some botanicals, some herbal notes and stuff like that. And so if you add extra stuff, you can easily throw the balance off. So adding extra stuff is fun and fine and good when it succeeds, but I would try with the basics first, and if it's your very first martini, and then just go and see what you might want to add from there. You might want to add nothing at all.

1:29:56.4 SC: David Levitt says, "Do you ever explain to general relativity students that Earth's surface accelerates outward at 1G?" So what David is referring to is the idea that in general relativity, there's no... What can I say? There's no fixed reference frame in which you should be evaluating anything. So there's certainly in any version of relativity, special or general, there's no fixed velocity that you can count as the correct velocity that something is moving at. All there are are relative velocities. And in fact, in general relativity, there are only relative velocities that are well-defined when two things are essentially at the same point in space-time. Since the space-time itself can be dynamical, if you have two things that are far apart from each other, there's no such thing as their relative velocity because there's no unique way to measure it. There can be some circumstances, like cosmology, where there's sort of a natural way to measure it, and we do, we talk about the velocity of receding galaxies, but it's actually not completely well-defined in the strict general relativity sense.

1:31:00.4 SC: What there is is a way of measuring acceleration. You know whether you're accelerating or not, you can feel it, okay? So there are no preferred set of trajectories in terms of whether or not they're moving, but there is a preferred set of trajectories in the sense of whether or not they're accelerating. So the inertial trajectories, or the geodesics, the free-fall trajectories, are sort of the natural well-defined ones in general relativity. And if you are sitting on the surface of the Earth, you are not on a free-fall trajectory. You can, again, feel it. You can feel it in your feet when you're standing up; you feel the Earth pushing out against you. And the sort of conventional Newtonian way of talking about that would be to say, "I am standing still, and the Earth is pulling down on me, and therefore I am feeling the force of the Earth on my feet." The slightly different general relativity way of saying that is that I would naturally be in free fall if it weren't for the Earth getting in my way. And so the reason why I'm feeling the force of the Earth on my feet is because the Earth is pushing me at an acceleration of 1G against my natural free-fall trajectory.

1:32:18.4 SC: Now, I wouldn't quite say that as the Earth's surface is accelerating outward at 1G. You can say that; it's not wrong. But certainly it conjures up an image that the Earth's surface is changing its velocity with time, and velocity is not quite well-defined. It's perfectly okay to say the Earth's surface is static, but it's absolutely correct to say that the Earth's surface is exerting a force on you, causing you to accelerate at 1G away from your natural free-fall trajectory. And to the actual question, do I ever explain that to my general relativity students? Yes, of course I do explain that to general relativity students. I might not put it in exactly those words, but the ideas are certainly there.

1:32:58.4 SC: Peter 42 says, "When we perform the double-slit experiment with electrons, we say that the electron is in a superposition of going through the left and right slit simultaneously. This results in an interference pattern on the screen. When we measure which slit the electron goes through, the superposition collapses and we get a pattern with two bands. How come the interaction of the electron with the air molecules, slits, and the light in the room does not count as a measurement/collapse of the wave function like it does in a quantum computer, which has to be shielded and cooled to maintain quantum effects?"

1:33:28.1 SC: The answer is it totally does. It absolutely would count if you let that happen. This is why doing the double-slit experiment with electrons is super hard. You know, people talked about the double-slit experiment years before because they knew what quantum mechanics predicted for it, but to actually do it with electrons is very difficult. And I think that the first semi-successful attempts were in the 1970s. But you basically do have to shield it from decoherence exactly the same way you have to do a quantum computer. So don't let the toy explanations given to you by theorists fool you here.

1:34:03.4 SC: Eugene Brevdo says, "Could a boundary conformal field theory in AdS/CFT support time-directed complexity rich enough for complex life? And if yes, would that count in favor of distributed rather than neatly localized life? And what would bulk observers look for as evidence?" So this is a question about the AdS/CFT correspondence. Many of you might have heard about it. The idea put forward by Juan Maldacena 30 years ago now is that anti-de Sitter space is a certain cosmological solution to general relativity with a negative cosmological constant. And it has the amazing property that they're at infinity, infinitely far away in a spatial direction, you can define a boundary to AdS in a way that you can't do it in Minkowski space or de Sitter space or something like that. It's special to AdS that the boundary is a space-time all by itself. You can always define a boundary, but it might not be a space-time; it might be space, or it might be null or whatever. So it's possible that you could define a theory of physics, quantum field theory, living on that boundary, and you could ask what relationship that has to what happens in the original anti-de Sitter space, the bulk or the interior, whatever you want to call it. And Maldacena's answer is, if you pick the right version of physics in anti-de Sitter space and the right version of physics on the boundary, they are the same. They are literally just copies of each other interpreted in different ways.

1:35:36.7 SC: So there's a couple of things going on here. One is, when Eugene at the end says, "What would bulk observers look for as evidence?" When you're in the bulk of AdS, to you, there's not extra things happening on the boundary of the conformal field theory, as we say, that is defined on the boundary. They're the same. There's what's happening in your universe reinterpreted in some highly weird non-local way from your point of view. So there's really nothing that you could look for that would tell you anything, you're just observing things about your local environment, and maybe you could figure out how to interpret them as what's going on in the boundary. But that would be super-duper hard since you don't really have access to what's going on in an infinitely big cosmological space-time. So that's a little bit kind of helpless.

1:36:24.4 SC: The other thing, more importantly to this kind of question, is the letter C in the abbreviation CFT stands for conformal field theory. So a conformal field theory is a kind of quantum field theory, but a kind of quantum field theory that has no scale built into it. In other words, there's no way to measure a distance or a mass compared to any elementary particles or anything like that. In ordinary physics as we know it, you have the electron, okay? The electron has a mass; you can measure it. The electron has a Compton wavelength that defines a distance, and so on. In a conformal field theory, which the Standard Model of particle physics is not, there are no mass parameters, no length parameters. So any configuration you can make, you can also make a similar configuration that would act the same but is twice the size or a billion times the size, et cetera. And therefore... This is just a conjecture on my part, a speculation, but it's a feeling that I have, life is impossible in a conformal field theory because living creatures as we know them have a different definite size. Having a size is kind of important to being a living creature. We have well-defined boundaries in space, and signals can travel in definite times from point to point in our minds and so forth, in our brains. And all that is stuff that is very, very central to what we know of as being a biological organism. Now, you can have things of definite size in a conformal field theory; it's just that they could become bigger or smaller equally well, right? There's nothing that fixes them at a size. And being fixed at a size by some physical parameter is kind of central to what we think of as biology. So I suspect, although can't prove, that the complexity that we think of when we think of biology is not going to happen in a conformal field theory.

1:38:16.4 SC: Jake Turin says, "The idea that a photon doesn't experience the passage of time has me confused. Clearly a photon doesn't experience anything, maybe that's part of my confusion. Consider a photon being ejected from the sun, traveling through space, and eventually being absorbed by a retina cell in my eye. From our perspective, the solar ejection takes place before the retinal absorption, and a measurable and finite time elapses between those events. How can we describe those events from the photon's point of view? The null trajectory description above suggests that there is no, no apparent passage of time, so are the solar ejection and the retinal absorption simultaneous? What about causality?"

1:38:52.2 SC: So I think that you... You're almost answering your own question; you're just sort of refusing to believe the answer that you're suggesting. Photons don't experience anything, okay? Photons don't have a notion of elapsed time. So whenever you're trying to talk about these weird, slightly counterintuitive ideas in quantum mechanics or relativity, you have a choice: You can either just say true things, just be completely correct, or you can try to translate the correct things into language that is more familiar and we understand some of the implications of from our everyday life experience and so forth. Translating is not ba, it can give people more insight than they might otherwise have had. You know, there's more to life than just the bare-bones equations describing the world. But the problem with the translation is that you can take it too literally, and then it can get you confused. So the correct thing to say about photon trajectories in relativity is that they have zero proper length, they have zero interval, zero proper time, whatever you want to call them, it's the same thing for a photon because they're all null. So the amount of proper time that elapses on a photon's trajectory is zero.

1:40:10.4 SC: That's the correct thing to say, and you should just stop there. You should just say that and not worry about what the photon experiences. Because as soon as you talk about experiencing, you're bringing in a whole bunch of baggage that has to do with we non-photon objects who do experience proper time. And when we experience things, we mean that there's a sequence of events happening inside ourselves that sort of accumulate memories and things like that. None of that happens for a photon, so it's just inapplicable. As far as causality and things like that, you don't have to be the photon. Like you say in the question, Jake, you can talk about what happens from the point of view of an external observer who is not moving at the speed of light. And there it's very clear that the emission happens before the absorption, and causality is perfectly normal. So it's really just more of a language question than a deep physics question going on here.

1:41:03.4 SC: I'm going to group two questions together, one from Tara Lumaghi and one from Barry Bai. Tara says, "Hoping you can explain in your excellent way: What the heck is a time crystal? How do they defy the second law of thermodynamics? What does it mean to repeat in space and time?" And then Barry Bai says, "Care to do a brief primer, or primer, depending on where you're from, on time crystals?"

1:41:24.4 SC: I don't know why time crystals are suddenly coming up. And look, I'm not a super expert on time crystals, but I can give you the very most basic idea to help you understand why it's interesting. If you think about thermodynamics and you think about, a gas in a box, traditional thing you think about when you think about thermodynamics, you know that I can imagine starting with all the gas on one side of the box and letting it evolve. And what will happen is it will smooth out, right? The equilibrium high-entropy configuration looks more or less the same everywhere. There'll be a slight difference because there's a gravitational gradient, but basically the same features everywhere. And this is very common. If you're in the highest entropy, or if you're just a single object, not a box of gas, if you're like a ball rolling down a hill, your lowest energy state, these tend to be uniform, simple, not doing anything. Okay?

1:42:21.4 SC: Now a crystal, an ordinary crystal, is a little bit of a variation of that theme. Crystals are mostly uniform, but they're periodic in some sense, right? The idea of a crystal is that the atoms get arranged in some kind of lattice. So because you're a thing that is not just a perfectly smooth fluid, because you're actually made of atoms, there's some spatial structure there to the lowest energy or highest entropy state of the crystal. Usually you consider the crystals at essentially zero temperature, so the box of gas is not a good analogy; the ball sitting at the bottom of a hill is a better analogy. All the atoms are settled into their lowest energy states, and there's some periodic lattice structure there. So I think it was Frank Wilczek, at least Frank Wilczek made it kind of famous, former Mindscape guest Frank Wilczek, but he may have been elaborating ideas that happened before, I really don't know. But he pointed out that, okay, if you have this idea of a crystal which is in its lowest energy state and periodic in space, could you imagine a system that is in its lowest energy state but moving? Moving in a very recognizable way, not just like moving with respect to some reference frame, but maybe oscillating back and forth in time.

1:43:43.4 SC: The reason why this is strange is because if you imagine a harmonic oscillator or something like that, okay, just classically, just don't make your life difficult by thinking about quantum mechanics, just think about classically, there's a way that the oscillator can rock back and forth, pendulum or whatever you have. There's a way that it can just sit still at the bottom of its potential at its minimum energy state. And the minimum energy state is going to coincide with the state that stands still. That's the traditional thing; that's the thing you would expect in a physical system. The lowest energy state is stationary because if you're moving, you have some kinetic energy, and that's a certain amount of energy. So the idea of a time crystal is to come up with a system where there is no state where nothing is moving, so that even the lowest energy state is oscillating back and forth in time. And they were able to do it, and I don't really know how they did it, but you can imagine, you know, probably taking advantage of certain quantum mechanical magic, having a system whose ground state is oscillating back and forth in time simply because there is no allowed state where everything is stationary. And because you have this sort of temporal repeating structure and you're in the lowest energy state, it's kind of like a crystal but in time rather than in space. And that's what you call a time crystal. More details than that, I don't know. So I'm not the place to go to, sorry.

1:45:10.4 SC: Ken Wolf says, "In your recent episode 'Liberal Democracy and How to Fight for It,' Adam Gorry more or less dismisses partitioning as a solution to polarization, since the partitioning often gets done using violence and leaves behind persecuted minorities." I think that what Ken means by partitioning is like dividing up the country into, you know, somehow continuous regions of either ethnicity or religion or culture or values or whatever. So he goes on to say, "But does the difficulty not diminish with the increasing granularity of partition? At extreme cases, one person leaving an abusive relationship can be as simple as moving out, but a persecuted minority leaving a hermit kingdom with closed borders is well-nigh impossible. This suggests pushing disagreements down to the smallest possible polity size. Could a case not be made for something like a voluntarist approach at the international level, libertarian at the national level, liberal democratic at the state level, social democrat at the local level, and communitarian at the family level? Or is that a step too far?"

1:46:13.4 SC: I think I have mixed feelings. I think I get the overall idea here, but I think there's a reason for pushing back a little bit. I think the general idea, if I understand what Ken is saying, is that there should be less and less specification of how you should behave as you get to larger and larger groupings of people, right? The idea, I think, being put forward is once you get to nation-states or international agreements or whatever, there should be less and less sort of paternalistic telling you how to behave, right? "Act this way, not that way," because for the very sensible reason, as you get to bigger and bigger groupings, you will have more heterogeneity amongst the values and culture of the people in the group, right? So you should have a more live-and-let-live environment. And that makes perfect sense to me. At the level of a family, the parents will tell the kids how to behave in a way that would be completely impermissible if you tried to do it for a whole society, okay? I think that's the basic idea.

1:47:15.2 SC: And that makes perfect sense. I think that that's fine. But I also think that there is an ideal that is worth striving for of people with different values coming to live together and cooperate. The thing about larger groups of people, states and nations and international agreements, is that we're going to do much better if we work together than if we go it alone. And so the whole idea of a democratic nation is that even though we have different values in some ways, we all agree on the importance of liberal democracy, and therefore we try to find room for our individual, you know, religions and cultures and tastes in food and movies and whatever, while nevertheless cooperating at the level of politics and economics and the law. And I still think that's very, very much an ideal worth pushing for. I do think that human beings work better and succeed more when they cooperate. And so the project of liberal democracy in a very, very large modern nation-state is finding ways to put up with polarization and nevertheless have a coherent country. And, you know, some people are going to say that's just not possible, that's just utopian pie-in-the-sky, that's why we see liberal democracy under threat these days. And I don't agree with that, but I get the argument, and I think that we just need to be more serious about explaining the advantages of having these liberal democratic values.

1:48:43.4 SC: Laurent Delamere says, "Have you ever been in a self-driving car like a Waymo or even a Tesla using full self-driving? I'm curious what you think of this technology and if you agree with me that the sooner self-driving cars are ubiquitous, the better, to make our roads much safer and to allow humans to do safely what they already do anyway in their cars, i.e., be distracted by their phone and other things." Well, I have not been in a full self-driving car. I've been in a car that was owned by somebody else and was driving around under close supervision but more or less by itself. I've never been in a Waymo. I was just about to go to San Francisco, but the trip got canceled, so I have not had a chance to actually try out the Waymos. I'm happy to try them out. I'm certainly not of the opinion that the sooner the better, the sooner self-driving becomes ubiquitous the better, because I really don't think the problems have been solved yet. Yes, there is a utopian view of cars that are driven completely autonomously by a system that is much, much safer than human beings. I believe in the possibility of that. But as many people have pointed out, nothing new here with me, it's exactly the tiny, tiny fraction of things that are unexpected and not in the training data that human beings are good at dealing with and autonomous systems are not good at dealing with.

1:50:09.4 SC: So I don't know as a matter of fact whether or not we could have a system right now in present technology, or even in the next hundred years, where most cars were driven autonomously and it would be safer than most cars driven by human beings. I do think that... There's certain advantages that are inescapable, especially because, like you say, people like to be distracted by their phones, people also like to drink or whatever, people are tired sometimes, people are old or young or not good drivers for various other reasons. And there's all sorts of room for improvement in that. And the number of deaths and injuries we have due to drivers in the United States of America certainly is way larger than it should be. So there's a lot of room for improvement. Maybe self-driving or autonomous driving will get us there; I just don't think that it's right around the corner.

1:50:59.4 SC: Tony Nardini says, "I wanted to ask you a pedagogical question. I'm a high school AP government teacher, and I often run into the issue of depth versus breadth when teaching the class. There's so much philosophy, history, and nuance about our government that I could teach, but there's also just the nuts and bolts they need to know as well. My question is: What do you find most important in the classes you teach? That the students are exposed to a wide range of concepts, or that they delve deeply into just a few?" Well, I think it's a little bit different for me than for you, since I'm not teaching high school. I'm teaching university-level and often graduate-level classes where the whole point is to go deeply into a relatively narrow area. At the level of high school, I'm very much in favor of breadth. I think you need to know a lot of things. I mean, I remember I was talking to a high school teacher at one point years and years ago where they were talking about, they had this new approach to teaching history where they would do what was called "post-holing." So they would dig deeply into some, like, particular event in history in their class and then skip forward 100 years and dig deeply into another event.

1:52:07.4 SC: And that kind of rubbed me the wrong way. I think that high school or secondary school is exactly where you should get the basic outlines of things. So I think there should be a balance. I don't think it should be superficial; I don't think you should just, "Here's the important dates, memorize them." I think it is important to sort of give some color and background and nuance and detail to individual things that happen, but not at the expense of not letting students know what the French Revolution was about or something like that, right? I mean, you have to make sure that the basics are also covered one way or another.

1:52:42.4 SC: S Sanders says, "You, along with David Albert and Alyssa Ney, have defended wave function realism, the position that reality is best represented by a wave function evolving in a higher-dimensional space. However, Albert and Ney argue that the best representation is a field evolving in a configuration space, whereas you've argued that the best representation is a vector evolving in a Hilbert space. What, if any, are the main metaphysical differences that you see between your version of wave function realism and theirs?"

1:53:09.0 SC: Yeah, so just as a terminological clarification, my position is not what is called wave function realism. I know why you would think that. So let me explain, because I didn't understand it myself for a while. So what is being talked about here is that in quantum mechanics, we describe the quantum state of a system as a vector in Hilbert space. At the most abstract level, as a practical matter, when we come up with how to teach quantum mechanics, et cetera, we start by saying, "Think about the classical idea of a position and a momentum. Throw away half of those, so either use position or momentum, and then describe a wave function living as a function of position or as a function of momentum." And when you have more than one particle, you take the configuration space for the whole system. So if you have N particles, you have N particles in three-dimensional space, is a single 3N-dimensional configuration space, and the wave function is defined there.

1:54:09.4 SC: So the difference between a quantum state as a vector in Hilbert space and a quantum state as a wave function in configuration space is that in the wave function in configuration space version, which is what is actually called wave function realism, you're privileging the idea of representing that vector in Hilbert space in a certain way, namely as a function on configuration space, as a complex-valued function that you then square to get the probability of observing a certain configuration of stuff. So to me... So I think that people like David Albert and Alyssa Ney and others who think of themselves as wave function realists, they think that not only is the quantum state a vector in Hilbert space, but it is correctly and properly and best and ontologically most importantly described as a function on configuration space. And I completely disagree with that latter point. For one thing, it makes no sense if you have a qubit. [chuckle] If you have a single qubit in two-dimensional Hilbert space, there is no configuration space to have a wave function on.

1:55:19.4 SC: But more importantly, it's just a choice of basis, right? We were just talking a little bit ago about the fact that when you have a vector space, usually, and I think it's certainly true in quantum mechanics, you're taught that the basis you use to express your vectors in is totally up to you. It's not fundamental, foundational, physical, important, et cetera, et cetera. It's a choice that you make, and all the physical things you calculate should be independent of that choice. We know in quantum mechanics that I can equally well represent a quantum state in momentum space as position space. And so how in the world can configuration space be so important? I just don't think it's fundamental at all. And I know why they want to say it's fundamental, it's because it makes your life easier, because you can then say, you know, "I live in the space from which configuration space is made," and you can take a shortcut to connecting the three-dimensional space that we think we live in to the configuration space and then the Hilbert space that quantum mechanics happens in. But I think that's cheating. I think that you should go the other way. You should start with the wave function, not the wave function, but the vector in Hilbert space, and you should make an argument why it is convenient to express express that as a wave function in configuration space.

1:56:36.4 SC: AJ says, "What do you think of Slavoj Žižek and other similar thinkers' use of physics concepts in their philosophy?" Well, I think it's potentially harmless, potentially harmful. It depends on how you do it. Well, I have not seen Žižek's book on quantum mechanics, which he... There's a book out that he uses quantum mechanics. I haven't actually seen it. So I'm very much in favor of people in all different areas of intellectual endeavor being inspired by not just physics, but ideas from many other areas of intellectual endeavor. That's why I think that the physics of democracy is an interesting thing to think about. It can inspire you to think about political science or economics or whatever in different ways. But of course, you have to ask whether the thing you're inspired to think makes sense. So you might say, like, "Okay, let's treat the people in a democracy like, like a box of gas," or, "Let's treat a person who is undecided about what to do as if they were a quantum vector in a superposition." Okay, then you have to be very rigorous and careful about saying, "Did I learn anything by doing that?" You know, maybe it sounds cool, but did I actually get any insight that I might not have had before? And I just don't know. I mean, the answer could very well be yes, it could very well be no. You just have to keep your wits about you.

1:57:58.4 SC: Mary Marks says, "I'll be celebrating my 83rd birthday this year." Congratulations, Mary, that's very nice. "And the closer I get to the end of my life, the faster time passes. A week seems no longer than three days. Everyone knows the cliché about life being like a roll of toilet paper: The closer you get to the end, the faster the paper runs out. But I'd like to know why that perception occurs. My life is no longer filled with tasks or deadlines or stressful to-do lists, so theoretically I should be able to savor each moment and slow the feeling of time passing, yet the opposite happens."

1:58:31.2 SC: So this is clearly a question for psychologists and neuroscientists, not for physicists. But because I have thought about the nature of time, I've talked to psychologists and neuroscientists, and they do have a kind of consistent picture. I'm not in a position to judge whether the picture is accurate or not, but it does kind of make sense, which is the following: It's absolutely true that in some measurable sense, time seems to move faster as you grow older. It doesn't actually move faster, but of course, when you say, "What is the rate at which time moves?" You have to compare it to something, and that's a tricky subject to get into. The point, according to the psychologists, is that as you're older, you're experiencing less novelty in the world, right? So you're experiencing just as many seconds per second as a child is. There's just as many photons hitting your retina or sound waves hitting your ears, et cetera. But you've kind of seen it all before, right? So there's less novelty to you. When you're a kid and you go to the beach for the first time, everything is new, right? And you're just soaking in all these, these experiences that your brain has never encountered before, and you're keeping memories of them. And the memories stick with you because they're so novel and important, and you want to be able to retain them for a long time.

1:59:47.4 SC: When you are older, that happens less often, right? You're in a routine. You've seen what happens before. Even if you travel to a different part of the world or whatever, you've traveled to different parts of the world before, so you kind of have a background to fall back on. There is a lower density of novel experiences being recorded in your brain. And that, the claim is, is the reason why time seems to go faster, because somehow your perception of the passage of time has to do with the novelty that you're experiencing. Now, there is a curious inversion here. Of course, if you're super bored, [chuckle] if you're on an airplane, a long flight, and there's nothing to do, it seems like time goes interminably, right? So there's no novelty whatsoever. And that's just because your brain is at a loss a little bit about what to do; it's getting no stimulation whatsoever. But the claim is, and again, I'm not in the position to adjudicate this, but the claim is that while you're bored and on the plane, time seems to go very slowly, but in retrospect, you don't actually feel like a lot of time passed on the plane flight because you didn't accumulate any new novelty or memories. So there's an important difference between the moment-to-moment feeling of time passing and then the slightly retrospective, you know, looking back on the last day or week or month of your life. If nothing new has happened in the last day or week or month, it will seem to you in retrospect like time has passed faster. So my advice, 83rd birthday, celebrate and go out, do something you've never done before, and time will pass relatively more slowly, if you believe this theory.

2:01:29.4 SC: David Lerfqvist says, "Grand unification theories seem to say that all the four forces were just once the one. It's described as a phase transition, like water turning to ice. I'm confused and I don't get that metaphor. Forces seem very different from matter. Is there a clear way to think about it?" So first, another terminological clarification: The technical phrase "grand unified theory" does not refer to all the four forces, just to the three, just to the strong force, weak force, and electromagnetic force. It does not include gravity. That's why people had to invent the phrase "theory of everything," which would include the known three particle physics forces plus gravity as well. But we understand the physics of grand unification. We have no evidence for it experimentally, but we understand how it would work much better than quantum gravity. So it's a perfectly sensible question to ask. It's very much a phase transition. Think about the phase transitions you know about with water going from ice to liquid to vapor, et cetera. What happens? The density changes, but more importantly, the equation of state changes when you go from one phase to another. The equation of state is the relationship between density, pressure, volume, temperature, things like that.

2:02:46.4 SC: Okay, and so what that means is that things like the speed of sound changes. The speed of sound in ice and in liquid water is going to be different because the equation of state is different. So basically, the response of the system to small perturbations depends on what phase you are in in a very straightforward way, whether things are solid or liquid is one example of that. Now, this doesn't seem to be quite applicable to particle physics forces and what happens as they evolve in the early universe, but secretly it is. So think of... Forget about grand unification, that's a trickier thing, but think about the relatively well-understood example of electroweak unification, okay? We know that, and we do have experimental evidence for the idea that the electromagnetic force and the weak nuclear force came from a sort of single quasi-unified electroweak force that then underwent a phase transition. So before that phase transition, the literal thing that happened at the phase transition is that the Higgs field went from zero expectation value to non-zero expectation value. And what that means is that before the phase transition, what you would would call the different kinds of forces, electromagnetic and weak force, they were all bundled together. They all acted kind of the same. And the particles that we know and love, like electrons, neutrinos, the quarks, et cetera, the quarks were still different because they felt the strong nuclear force, but the different kinds of quarks were all the same. The up quark, down quark, top, bottom, charm, strange, they all had the same mass. They all responded to the electroweak forces the same. Likewise, the electron and its neutrino, the muon and its neutrino, the tau and its neutrino all had the same mass, namely zero, by the way; they had no mass at all. And they responded to these forces in more or less exactly the same way.

2:04:39.8 SC: And part of that is, let's make it a little bit more closely connected to the ice and water analogy, before the electroweak phase transition, the SU(2) bosons, remember that in the technical jargon, if you've read Quanta and Fields, you know all this already, but in the technical jargon, the weak part of the electroweak theory is, roughly speaking, an SU(2) gauge theory. I say roughly speaking because really there's a mixture of SU(2) and U(1) that comes into things when the phase transition happens, but that's all technicalities you don't need to worry about. The point is, in the current world where we live, there are things called the W boson and the Z boson descended from this SU(2) gauge symmetry. But they're very massive, they're very heavy. So that's why the weak force is so weak, because these bosons are so heavy, they don't travel very far before decaying, their lifetimes are very short, it's hard to make them, et cetera.

2:05:37.3 SC: Before the phase transition, they were massless. So the weak bosons, the bosons that are now the W's and the Z's, could travel all over the place and have a big effect macroscopically. So that's a different world to live in before the phase transition than after. And it's actually kind of fascinating to sit down and think about how real-world particles did behave and things like that. The whole thing, of course, is only true... A necessary thing about the phase transition is that if you're in the previous phase where electro and weak were unified, you're at a high temperature. So you also have a density and temperature to worry about, and those could be just as important as the masses of the gauge bosons. There's a whole bunch of physics going on, but it's really a different world with different responses to perturbations, different pressure, temperature, density, curves, et cetera. So it really is a phase transition in the very direct sense, I've got to say.

2:06:36.4 SC: Robert Sachs says, "Do you have any views on embodied cognition, specifically Turner's work in conceptual metaphor and Lakoff's work in the metaphor and embodiment? The strong view that cognition cannot arise in the absence of a body and sensorial and causal relationship with the external world has profound implications on the possibility of AGI. Do you have any thoughts?"

2:06:54.4 SC: I don't have many thoughts about this. In particular, let me be very honest, I don't know really anything about Turner's work or Lakoff's work except occasional popular references to them, so I can't really comment on that. I do think, in accord with things that we've heard from Ned Block and Anil Seth here on the podcast, and also you can check out their current... These guys are having a currently very interesting debate about the biological necessities, or the necessity for biology in consciousness, okay? I don't know if there's any necessity for biology in conscious. I don't know if there's any necessity that cognition cannot arise in the absence of a body. It depends on what you mean by cognition, depends on what you mean by intelligence, consciousness, et cetera. I think that a safer thing to say is that cognition and consciousness and intelligence as we know them, as we are familiar with them with ourselves and other creatures in the world, is very, very closely connected to our bodies and the way that our bodies both internally survive through metabolism and things like that and externally interact with the rest of the world. And I think that, of course, a lot of work on artificial intelligence just completely ignores that. Now, that doesn't answer the question: Is it ignorable? Is it okay to ignore it or not? Is the fact that we intelligent beings are embodied and constantly in interaction with the world and using the free energy resource that we have around us, et cetera, et cetera, is that central and crucially important to intelligence and agency and consciousness, or is it just something that happens to be there for the intelligence that we know about? That I don't know. I think this is, like, a fascinating question. We've got to dig into it, and I'm glad that people are.

2:08:48.4 SC: David Carr says, "I'm looking for guidance on a career switch to physics, both because of an interest and uncertainty in my current career. I'm a software engineer in the US in my late 20s with a bachelor's degree in Computer Science, a minor in Physics. I'm currently teaching myself physics using textbooks, as it's always been what I've been most interested in. I've also been thinking about volunteering my time and expertise in software engineering for research if that is a possibility. Could self-teaching, paired with my degrees and a good score in the physics GRE, be an effective strategy for applying to doctoral programs in the future, or is a full undergraduate degree in physics the only way?"

2:09:22.0 SC: Well, I think it's not the only way. A full undergraduate degree in physics, that's certainly not the only way. I think that you have to convince the admissions committee at some PhD department to let you in. And I've been on admissions committees, so I've seen how they talk. It's certainly... There's a straightforward road and there's less straightforward roads, but people can succeed on either of them. They might not be equally easy, right? So your question needs to be: From where you are now, what is the most direct and high-probability-of-success route to take? I think what will be crucial is...I think you already know what will be crucial because you mentioned getting a good GRE score. I think if you have a lot of background taking a lot of physics courses and get good grades in them from a good school especially, then your GRE scores are less important. But if you're doing this unconventional route, this is one of the reasons why GREs can be important. The GREs, for those of you who don't know, these are standardized tests you can take in different subject areas that help you get into graduate school. And there's been this weird de-emphasizing of them in recent years. And I think it's a little bit overly done, in fact, the de-emphasis, because I think that it's just one thing. You don't want to just accept the people with the best GRE scores. When I was at the University of Chicago, I tried to make this point, but... I've told this story before, but I had the whole faculty fill out a survey of all the students that had gone through the PhD program in the last five years and rank them by how successful they were as physicists.

2:10:55.4 SC: And there was a little cluster of people who... And then we compared them, compared to the ranks of how successful they were in grad school to their GRE scores, which of course we all had because we had let them in. And there was, like, this little cluster of like three or four people who were brilliantly good at the GREs and fantastic physicists who mostly did theory and string theory and things like that. But apart from that, there's no correlation whatsoever between how well you did in the GRE and how well you did as a physics student, as a graduate student. Now, of course, there's a huge solution selection effect, of course, because you're selected to be people who are going to be successful in graduate school. It's not that there's no correlation between GRE scores and grad school success, but that was filtered out and controlled for by the process that actually accepted certain people.

2:11:45.4 SC: But it's very, very helpful in exactly... The GRE scores are very helpful in exactly the circumstances that someone has not gone through a traditional path. Maybe they didn't get a physics degree or whatever, they went to a school you're not familiar with. That's when the GREs are very helpful, so I don't think they should be underemphasized. So getting a dynamite score on the physics GREs will be hugely helpful to you in applying to graduate school. The other thing that'll be hugely helpful is getting good letters of recommendation. That's something that a lot of undergraduates don't do a good job of. You need to be well-known enough to some real physics professors who can write you letters saying, "Yes, I know David wasn't a physics major, but trust me, he will do well in graduate school." That's what you want to get in a letter. And you can't get that from a letter by someone who just had you in a class that they taught or something like that. You really have to get to know somebody, whether you're doing research with them or communicating with them in some other way, et cetera, et cetera. That's the other thing you have to try to do.

2:12:49.4 SC: Now, something that I can't help you with is you say, like, "Thinking of volunteering your time and expertise in software engineering for research if that is a possibility." Maybe. I mean, I get a lot of people who try to volunteer to help me with my research, they might be high school students or undergrads elsewhere or whatever, and I turn them all down because if... For my kind of work, which is highly theoretical, if you're not already highly trained as a physicist, it's not helping me to have you around doing work. It's much more effort on my part to try to bring you up to speed, et cetera. Once you're a PhD student, it's still a little bit of work on my part, but you quickly get to a point where you are very helpful to me versus the other way around. For a lab or experimental physicist, that might be a very, very different story. So that's where you might want to try out that particular pitch.

2:13:44.4 SC: Fernando Curiel says, "Since Alcubierre drive was originally proposed, several modifications have been put forward that appear to make it more feasible in principle in terms of not having to use exotic mass or energy or enormous amounts of energy. The question is: Do you, as an expert in GR, think that the idea of using the drive would generate real paradoxes? I mean, like going to a nearby star and back faster than a photon doing the same traveling."

2:14:10.5 SC: So the Alcubierre warp drive was this fun idea put forward by Miguel Alcubierre back in the '90s where, if you allow yourself complete freedom to imagine different kinds of exotic energy and things like that, so in particular, exotic energy is usually a code word for name negative energy density at some region of space. And that's provocative because classically you would just think, "No, you can't do negative energy densities." But quantum mechanically, you can maybe have a little bit of negative energy density for a little bit of time if you balance it somewhere else, and it's all very ill-understood as far as I know. But anyway, Alcubierre said that if you arrange that negative energy, you can basically solve Einstein's equations for general relativity in such a way that you can basically warp the space-time metric to make it look like you're going faster than the speed of light. You're never really going faster than the speed of light.

2:15:03.4 SC: But if you think about, you know, an external space-time, let's... Think of it this way: Think of a space-time like just Minkowski space, like nothing going on, right? Just flat, special relativity applies, ignore all of the effects of gravity, et cetera. But then insert this sort of Alcubierre warp drive in there, and you would imagine like a little tube of negative energy that is moving through the universe. And from the point of view of the people outside that tube, it would look like this warp drive was going faster than the speed of light. Now, nothing there says that there's any paradoxes or anything like that, right? There's nothing paradoxical about moving faster than some other speed. There's an apparent paradox that you might worry about because you're taught in special relativity that if you can go faster than the speed of light, then you can travel backward in time, right? And then you can go on a closed timelike curve. But that's not necessarily true in general relativity because you're not really going faster than the speed of light, you're appearing to go faster from someone else's point of view, okay? So that's not quite the same thing.

2:16:12.2 SC: Now, I don't know of a well-defined answer to the question. The right question to ask in this circumstance is: Is there a well-defined, unique, deterministic solution to the initial value problem? The initial value problem says, "I give you some initial configuration of stuff, and I give you some equations that are the dynamical equations telling you that how that stuff will evolve forward in time, and I solve those equations and I find that there's a unique future," okay? That's the initial value problem, whether you're doing classical mechanics or general relativity or anything else. Maybe there is a unique solution to the initial value problem even in the presence of a warp drive. Again, I've got to emphasize over and over again, nothing is really going faster than the speed of light. It's only apparently going faster than the speed of light from the perspective of someone not inside this particular warped tube that the warp drive has created. Now, having said all of that, I think it's hilariously unrealistic, the idea of an actual warp drive. The energies you would need, the kinds of energies you need, sure, various modifications have been put forward trying to make those problems less severe than they originally were. They're still super-duper-duper severe. Like, you need astronomically large amounts of energy to warp space-time enough to make this happen. This is not something that anyone is going to do technologically in the next thousand years, let's put it that way. So, fun to think about as a science fiction thing, but not something that is realistic physics right now.

2:17:50.4 SC: Ophir Averbuk says, "You often advocate the view that the language of probability theory and Bayes's law is the right way to think about how scientists learn and how science progresses. Referencing the terminology that was used in the episode with David Deutsch, you would say that probability theory is the calculus of inductive logic in science. Is this view meant to be prescriptive or descriptive? Also, how far are you willing to take this view? Do you think that scientists or science communicators should keep track of the community's current probabilistic beliefs about different open questions? How would we go about doing that?"

2:18:22.1 SC: I do think it's mostly prescriptive. I think that there's... It's not a completely inaccurate prescription, so as a description, it's pretty good. I think that scientists do have credences and they do update them. I think that where scientists often fall short is being good about saying what their credences are before we start doing things. Like, someone will say, "We should build a telescope to measure the evolution of the dark energy because if the dark energy is evolving, it will be a super-duper cosmologically important result." And that's true. So what is your credence that the dark energy actually is evolving, right? Or, "We should build a detector to look for a certain dark matter particle because if we find it, it would be important." Yes, okay, so what is your credence that we will find it? And, you know, you better add up to all of the different exclusive possibilities adding up to one. So we don't always do it correctly. I do think that we should be better at it. I think specifically that you are right, we should be a little bit more open about what the probabilistic beliefs are. Now, many individuals just aren't very clear about what their probabilistic beliefs are, people being people. So that's why I kind of like these surveys that get done. You know, the philosophers David Chalmers and David Bourget did a very nice job in this paper, "What Do Philosophers Believe?" Right? And so they tried their best to come up with a fair sample of philosophers and ask them questions about difficult questions and figure out what their answers are. And highly enlightening what the answers are, not a lot of consensus in philosophy, as you might guess.

2:20:04.4 SC: And in physics, people try to do that sometimes. You know, I've helped out with a recent survey on trying to figure out what physicists believe about the foundations of quantum mechanics. Very, very difficult to do it well, just because when you say, "Okay, do you believe in this model or that model?" no one even agrees on what the models are. Like when you say the Copenhagen interpretation, no one is quite sure what that means. So it's difficult. And, you know, people might in their heads assign probabilities to things that don't add up to one or something like that. So is that terrible? Does that really get in the way of how science gets done? Probably not. It's not the worst sin out there, but we could be better at it, and I would be in favor of trying to be better at it.

2:20:48.4 SC: Charles Hertz says, "Ever since reading Douglas Hofstadter's 'I Am a Strange Loop,' I've been musing about the hard problem of consciousness and have formed the strong press prejudice that the development of consciousness requires the passage of some non-trivial amount of time, whether it be in some Hofstadter-esque way or through some entirely different process. If that were to prove correct, then it would seem to rule out by that fact alone the idea of a Boltzmann brain, since I assume that the lifespan of such an entity would be a few nanoseconds or less, and therefore that a Boltzmann brain would not last long enough to develop consciousness."

2:21:21.5 SC: Then I'm cutting off more of the question, but I think that's the basic question, which I can answer now because, no, Boltzmann brains do not necessarily last for just nanoseconds or less, for a couple of reasons. One is that the amount of time it takes for a brain, if it were to spontaneously fluctuate out of the random nothingness of the universe, to both appear and disappear is the same. The process of the assembly of the brain is just the time reverse of the process of the disassembly of the brain, and that can take a long time. But more importantly, you don't take the Boltzmann brain image too literally. The correct statement is: You tell me what you think is necessary to count as an observer or conscious or intelligent or whatever. There is some configuration of matter corresponding to that, okay? And that configuration of matter does not involve the entire universe with hundreds of billions of galaxies, each of which have hundreds of billions of stars. You don't need all of that to make a single conscious creature. I mean, maybe you think you need the whole solar system and the whole Earth or something like that. Whatever it is, maybe you just need a brain, maybe you need more than that, it doesn't matter matter. The statement is that whatever it is you need, in a universe that fluctuates randomly for infinity years, the overwhelming majority of appearances of that thing that you need, that configuration you need to make a conscious creature, is going to mostly be a random fluctuation surrounded by thermal equilibrium, okay?

2:23:03.4 SC: Even if you claimed for some reason reason that you need a million years of evolution to make a conscious creature, okay, then take a million-light-year part of the universe and let that fluctuate into existence, you know? Then the things will last for millions of years. That's fine. The point is that you certainly don't need the whole Big Bang 14 billion years ago giving us hundreds of billions of galaxies. So whatever your criteria are for consciousness or for intelligence or whatever, they do not get you out of the Boltzmann brain problem. It's really a Boltzmann fluctuation problem, as I do say every time I write about this stuff.

2:23:45.4 SC: Todd Pellman says, "Can you share the actual software that you use for Mindscape, and has it changed over the years?" Yeah, it hasn't changed much over the years. By the way, if you go to preposterousuniverse.com/podcast, the homepage for Mindscape, there is a thing on the sidebar where you can click, it says "About Mindscape", and I tell you both the hardware and software that I use. So you're welcome to check that out. But the rough picture is, I have... Hardware-wise, I have a nice microphone, an Electro-Voice RE320, and a Sound Devices MixPre-3, which is both a recording device and a mixer that takes the right kind of cable. If you have a nice microphone, the really nice microphones don't plug in via USB to your computer, they use what is called an XLR cable. So you need to plug in the microphone to something. And so for a long time, I've been using this MixPre-3, and then that connects via USB into the computer. And so I can both record directly on the MixPre-3, but then the software I use on the computer is mostly Audacity. Audacity is a free digital audio workstation, so you can both do recording and do a little bit of editing in there.

2:24:58.4 SC: There's... I forget exactly the name of it, there's something called iZotope RX 7, which is a way to do a little bit more sophisticated editing. Somehow I used to be able to embed that inside Audacity, but I can't do that anymore, I don't know, something broke, something happened, so things like noise reduction and so forth I can do through that. Sometimes the audio from the guest is just not that good, and there is a service you can get from Adobe that uses AI... Or it's really not AI, of course, it's machine learning, but it cleans up the audio from the podcast. And I forget exactly what it's called, it's something like Enhance Speech from Adobe or something like that. So you can upload an audio file, and it will try its best to sort of pick out the actual words that are being spoken by a person. The problem with that is there's a slider, you can go from 0% enhancement to 100% enhancement. And 100% enhancement sounds really good when it sounds good, but sometimes it interprets actual words as noise and eliminates them, and so that's bad.

2:26:02.4 SC: But basically, Audacity is the only thing that I really need. I upload the audio files to Scribie, which makes a transcript. Costs money to make a transcript, that's what was the original motivation for getting Patreon support was, was to pay for the transcripts of the podcast. So I still use that. I am shifting a little bit because carrying around the... I mean, the MixPre-3 is a little bit expensive and a little bit bulky, not very bulky, but I got this little Scarlett Focusrite Solo, which is an alternative to the MixPre-3. It also lets me plug in my computer, but it's a little bit lighter weight, easier to use. There's two downsides, one is that it doesn't record; that's a big downside. So I have to record on the computer, on the software. And the other is it's a little bit trickier to sort of combine the signal I'm getting from the guest with my own signal. So I'm learning how to use other software like Loopback and Audio Hijack to learn how to do that correctly. But that's a very recent development; I don't know if it's going to make any difference or be very common. But Audacity is the main thing that I'm using. Oh, for the actual recording, I use Zencastr. Zencastr is something like Riverside that you might have heard of before. It provides a connection on the web between you and the guest. And the nice thing about Zencastr is it records locally and uploads your files very efficiently. So even if the internet connection is not that good, the audio file that I get from the guest can be good because then it sends me the files. And of course, for publishing, now I'm on Libsyn. I moved from Wondery to Libsyn because Wondery has shifted its emphasis, they just want to make TV shows, really. So Libsyn has been great as a host for the podcast.

2:27:58.4 SC: David Declutt says, "Is my understanding correct that if someone believes in the Copenhagen interpretation, the Schrödinger equation still holds when there are no observations? And so long as there are no observations, you can still speak of decoherence, and you can still split the wave equation into separate orthogonal components, which we would call worlds. It seems to me that Copenhagen doesn't exclude multiple worlds, it just reduces the number of worlds when observations occur, but still needs many worlds when no observations occur."

2:28:25.0 SC: Well, I don't think that's... I think that's maybe being a little bit too fair to Copenhagen, or not fair enough. I'm not sure which direction it's going in. But I think that you're sort of using an Everettian language to talk about Copenhagen, which isn't a very good fit. Copenhagen, if you take it seriously, the wave function doesn't represent reality. So yes, the Schrödinger equation still holds when there are no observations, and you can talk about decoherence and all that stuff, but none of that is real in the Copenhagen interpretation. The only thing that's real are the measurement outcomes. And to make sense of Copenhagen, you really need to have the idea of a measurement outcome play an important ontological role. Okay? You need to define what you mean by a measurement and an outcome, and no one does. That's why Copenhagen is not very well-defined.

2:29:15.7 SC: Now, in Many-Worlds, there's no such thing as measurement or observation or anything like that as a separate ontological category. There's just the wave function evolving according to the Schrödinger equation. You treat the wave function, or whatever it represents, the vector in Hilbert space, as real, as describing reality. And then what you thought was a measurement outcome is actually, in Many-Worlds, just decoherence and branching. So you can talk about decoherence and things like that in Copenhagen, it just doesn't have any meaning. It doesn't do anything. And a lot of physicists are kind of very confused about this. They want to say they believe in the Copenhagen interpretation, they also want to say that observations happen when you have decoherence, and it doesn't actually quite fit together in a sensible way.

2:30:02.4 SC: The Great Deceiver says, "I really enjoyed the Andrew Guthrie Ferguson podcast and have a question which dovetails nicely. Recently here in Canada, we had our second-worst mass school shooting. Obviously it was horrible, with eight dead in the end. But something interesting came up very quickly afterwards, the shooter had their ChatGPT account flagged and banned just weeks prior to the event because of violent scenarios. This prompted a meeting between our AI Minister and Sam Altman, which resulted in promises to ensure better cooperation with law enforcement and better surveillance of potentially dangerous users of the platform. Many people think this was a missed opportunity for Canada to force companies like OpenAI and AI in general to make more systemic changes which prevent these kinds of tragedies. In response to this event, which frankly could have been prevented, what responsibility to its users and the public do you think these companies should/could have going forward?"

2:30:55.4 SC: Well, I think two things. I think on the one hand, at the level of the public, I don't think that the government should have the right to snoop on people's conversations with the AIs. It's not... I mean, it seems like different, like this new thing, AI, et cetera, et cetera, but it's just saying, "Does the government have the right to snoop in on your phone conversations or just put microphones in your living room?" Or things like that. There are certain things that should be in the private sphere. Even if those things include planning for a school shooting, that's the price you pay. You try to get rid of those not by doing ultra-surveillance and making sure you know what every person in the country is thinking, but by trying to create a society where school shootings are not even tempting for most people and there's safeguards to prevent them, like maybe making it hard to get guns unless you can show that you're a responsible person, and so on and so forth.

2:31:45.4 SC: The companies, on the other hand, forgetting about the public sphere, the companies might very sensibly be thought to have some responsibility. You don't want to be a company that helps a school shooter along, and therefore if you are OpenAI or Anthropic or Google or anyone else, you might very well want to put in safeguards that sort of collapse or turn off the AI when you think that it is getting input from a user which is giving an indication that something terribly dangerous might happen. Now, that's easier said than done; I completely understand that. I don't know how to do that; that would be very difficult work. It's somewhat counter to the economic incentives of a lot of these companies to worry about things like that until a tragedy happens and then they look bad. And maybe that there is also some indirect room for government intervention here in, you know, not that government is actually surveilling users of AI, but the government can say, "Look, if someone uses your AI product and because of that does something really terrible, you will be liable. So deal with it one way or the other." I think that's a perfectly fair thing for a government to say.

2:32:58.4 SC: Dennis Banks says, "Do you think that there is any way that information could be causal in its own right, enabling some sort of top-down causation, for example, in biology?" Roughly speaking, no, I don't think that information is like that. I think that people are a little bit overly tempted by the language that we use. You know, again and again and again, ordinary everyday language is just not meant to talk about fundamental physics or ontology or anything like that. So we talk about information, and I often analogize it to energy, right? I don't think energy has any causal influence either, even though we talk about it as it does all the time. If you really want to be careful, if you want to be casual, then go ahead; having information, having energy certainly plays a causal role at some level of discussion, but if you really want to be really careful about your fundamental ontology, things like energy and information are properties that matter has, that certain configurations of stuff. The actual causal influence comes from the stuff. Energy is not a fluid that flows into a ball rolling down a hill, it is a property of the ball rolling down a hill. And information is exactly the same way. Information is the way that we talk about configurations of stuff, and it's the stuff that is ultimately responsible for things happening

2:34:15.4 SC: Jamie says, "I remember you once said to David Albert about Many-Worlds that self-locating uncertainty was like betting, and that the Born rule gives you odds, which if you don't like... If you don't follow, you'll get burned. When it comes to the Sleeping Beauty problem, what is the betting situation? It seems to me that the disagreement between thirders and halfers comes down to how you define the payout. You bet on your credence of heads or tails. Would you bet each time you wake up and count your money at the end, or would you bet each count on its own?" I think that's almost exactly right. I mean, I think that... As I've said before, I think that the actual answer to slightly Sleeping Beauty is you've got to define the problem better. You can sort of operationalize what you mean by having a credence that you have awakened a certain number of times. So if you promise to do the Sleeping Beauty experiment over and over again, and every time Sleeping Beauty wakes up, she has the chance to bet that it is either heads or tails, then it's perfectly clear, she should place on odds that it is 1/3 heads and 2/3 tails, if the heads is the option that wakes you up once and tails wakes you up twice. But maybe that's not what you mean, right? There's different ways to operationalize the Sleeping Beauty problem, and you could easily justify getting a halfer approach to it. So I do think that it's just a matter of being very careful about what you mean by the problem. Again, even though the language sounds very clear, there's details hidden in there until you go and specify exactly what it means to have a credence on something.

2:35:46.4 SC: DA says, "If you could restructure the usual undergraduate physics curriculum in the United States, how would you go about doing it? Would you introduce a few extra classes geared toward students who would like a more descriptive introduction while they work on building math skills? Would you introduce some problem-solving classes that take a slightly different approach, e.g., greater focus on dimensional analysis than in regular classes?" You know, I don't have a strong feeling about this. I've never really sat down and carefully tried to think about what the undergraduate physics curriculum should be. I mean, part of your question slides into how we should teach undergraduates who are not physics majors versus how we should teach the ones who are physics majors. I think we do a pretty good job with the physics majors. I do think that there's always a lag between how we teach our undergraduates and the current state of the field, right? I mean, in modern physics, there's things going on in quantum field theory and cosmology and condensed matter and certainly the intersection of AI and physics that those are a little bit more contemporary and new to actually have bubbled down into the undergraduate curriculum.

2:36:53.4 SC: I do think it would be nice for undergrads, and many places do this, by the way, to have a class or two that was just "Topics in Modern Physics" or "Modern Research Seminar" or something like that, right? To give even the undergraduates a taste of what is going on at the cutting edge of research. Now, for non-physics majors, I think that there it's more clear that we do a bad job. And I think that maybe, arguably, we do a better job for humanities made majors than for, like, engineering majors or math majors or whatever. Because people who are able and encouraged to take a quantitative version of a physics course but aren't going to become physics majors are often taught the watered-down version of the introductory course for the physics majors, right? You learn about Newtonian mechanics, maybe a little bit of special relativity or field theory or something like that, but not very much. You certainly don't hear about quantum mechanics, really, cosmology, or anything like that. I think that if you know that a student is quantitatively able to do problem sets and things like that, but they will only ever take either one semester or two semesters of physics, you don't need to just do inclined planes and pushes and pulls and things like that, you can teach them something more modern. But overall, I don't want to complain too much. I do think think that the general undergraduate curriculum is pretty good. Everyone has their complaints, but I think that the complaints on the whole average out, so I don't think that there's any obvious way to improve it. But again, I have not actually thought about it too hard.

2:38:33.4 SC: Elijah Massey says, "Can moral constructivism provide someone with the strength necessary to make significant sacrifices for one's ideals, such as giving up some something one strongly cares about, living an ascetic life, or being willing to die if that is required? It seems like it would be harder to have such strength if one does not believe morality is certain and based on something external, but believes that we construct it the best we can and other moralities are no less valid." Nope, I think it's completely the opposite of that, in fact. I think that there's no problem whatsoever... Well, look, there's always a problem with people, people as a general rule, having the strength necessary to make significant sacrifices for one's ideals. That's true for moral constructivists, for relativists, for objectivists, realists, et cetera. That's just a human nature thing. I see absolutely nothing in the fact that one takes a clear-eyed view of the origin of morality getting in the way of taking that morality seriously. I think that the fact that you think that morality is something that human beings invent doesn't stop you from taking it super-duper seriously any more than other things that human beings invent are taken very seriously.

2:39:40.4 SC: The real problem is with moral objectivists or relativists who are either nudged towards doing something that they don't want to do by a pretend opinion that they're talking about objective morality, or for that matter, justifying bad behavior on the basis that it's objective morality. Just on social media this morning before I was recording this, there's this controversy about the Pope, you know, Pope Francis, the Pope from Chicago and Villanova, the "woke Pope" people were calling him, or the "based Pope," I guess. He's very liberal-leaning overall and has gotten a lot of plaudits from certain progressive circles for saying things that they want to hear. But at the end, he also says, he's very anti-transgender and things like that, very anti-homosexuality. Why? Because he's the head of the Catholic Church. [chuckle] It's part of what it's like to be the head of the Catholic Church. If in fact morality is not objective, then people who act as if morality is objective will use that as an excuse to do bad things. And that's the real worry, not that the people who are correctly realizing that morality is not objective don't somehow have the strength to do right things.

2:40:59.4 SC: Ben Lloyd says, "Do you lean more toward eternalism or presentism when thinking about time? In discussions of eternalism, I often hear that past events still exist in the same sense that the present exists. I trust this is true but also find it confusing. For example, if we consider a specific event like my time, 10th birthday, in what sense is that moment still real even though it's not happening now?" So I'm an eternalist, absolutely. I've said that many times. But I think that, again, people fall into this linguistic trap. Eternalism says all moments of time are equally real, okay? That's not exactly what you said in your question, Ben. You said things like, "Past events still exist in the same sense that the present exists." The word "still" should not be there. The word "still" is temporally laden. The word "still" means that the present and past exist both right now. "Right now" picks out the present. It's not that... Eternalism doesn't say the past exists right now, it says the past exists. It's literally exactly like space, right? If I say I'm a spatial realist, I'm a spatial eternalist, all points in space exist. That doesn't mean all points in space are here. Points in space are at different points in space. If I say all moments of time exist, that doesn't mean all moments of time are now. So when you say again, "In what sense is that moment still real even though it's not happening now?" it's in exactly the same sense as a point far away, you know, a kilometer away, is real even though it's not happening here. There you go.

2:42:42.4 SC: Gauta Einaval says, "You obviously receive a lot of books. Do you keep them all, or do you give most of them away to avoid filling up your house with books? I struggle with all the books I acquire, even as new bookshelves I get fill up before I know it, and I feel it's a sacrilege to throw any of them away." No, for a long time I've been giving away a lot of books, both books that I personally purchased years ago and no longer have much use for, but also, yes, people send me books, both because I'm a college professor and because I have a podcast. People optimistically send me books hoping that I will talk about them on the podcast or whatever. For the most part, actually, that doesn't happen. Just so you know, usually if I have a guest on the podcast who has a book out, either they or their publisher or their publicist will contact me and ask me if I'm interested in the guest. And if I am, they will send me a copy of the book. And those are often advance copies, they're not even, like, the fully finished copies of the book. And so I don't just throw them in the garbage, I try to give them away. There's various ways to give them away.

2:43:43.8 SC: I live very close to a big university, so there's all sorts of, like, these little local boxes outside, you know, a coffee shop that say "Give a book, take a book" or something like that, so I can put them there. There's a whole shelf in the physics department where people give away their old physics books and things like that. So you shouldn't feel guilty about giving your books away. I think that the guilt comes from, there's always a part of one's life, don't get me wrong, we have a huge number of books in our house. It's kind of a silly number of books in our house, and there's still books that are in boxes that we haven't unpacked even though we built more bookshelves when we arrived. When you're young and you're discovering books and you're all into it, books are super-duper precious and it's hard to get them, and you spend time in the bookstore, in the bookstores, in the libraries, just looking at the books and, like, dreaming about being able to have them all. And then when you're old and you've accumulated a lot of books and you have a higher salary, now you have more books than you need and you feel guilty about it. Don't feel guilty, just give them away to other people who can take advantage of them.

2:44:48.4 SC: Robert Mattson says, "I'm hoping to get a bit more clarity on the term "quantum information" as it pertains to black holes. What exactly do we mean by information being destroyed, and why are we concerned that it could be?" I think quantum information is not really conceptually that different from classical information. In both cases... By the way, of course, the word "information" means many things to many different people, but in the specific context, what the word "information" means is just the data you need to specify the physical state of the system, okay? So if it's a classical system of several particles, that data is the position and momentum of every particle. If it's a quantum system, it's the vector in Hilbert space or the wave function or whatever. And Laplace, and actually he wasn't even the first one, Bošković or someone like that pointed out before, realized that in this sense, in classical mechanics, information is conserved over time. What that means is if you give me the state, the data is required to understand the state of a system at one moment of time, the laws of physics deterministically say what it is at every other moment. That's what it means to be conserved.

2:45:54.4 SC: In quantum mechanics, as long as the state is evolving according to the Schrödinger equation, what we call unitary evolution, the information needed to specify the quantum state is also conserved. The Schrödinger equation is completely reversible. Knowing what it is at one moment, moment in time, you know what the solution is at any other moment of time. Things like measurements in quantum mechanics destroy information because they are not reversible, they do not obey the Schrödinger equation, unless you think that there's some hidden determinism there because the Schrödinger equation is actually completely correct, like it is in Many-Worlds, or for that matter, in Bohmian pilot-wave kind of hidden-variable theories. In both cases, Schrödinger equation is 100% valid, and the apparent collapse of the wave function is in fact merely apparent. Maybe that's true, but for any one actual observer, making a measurement effectively destroys the quantum information. But in physics, what we like to think, you know, if you have both general relativity and quantum mechanics, forgetting about measurements and observers and things like that, both of those theories by themselves preserve information over time.

2:47:03.8 SC: So it's at least a natural guess that when you combine them together to get quantum gravity, information will also be conserved in that sense, putting aside measurements. That's not a guarantee of anything, of course, you have to actually do the work. But a lot of people think that unitary evolution conserving information has a lot of benefits. There's a lot of things that could go wrong if you violate that, so they want to see to see it preserved. And maybe it's not. You have to accept what nature gives you at the end of the day. But when you're investigating the possibilities for what nature could do, you absolutely have your preferences and you try to make them come true, and then you see whether or not that worked.

2:47:46.4 SC: Derek Corwin says, "Looking at Doyne Farmer's work on complexity economics, there's a fascinating tension between his bottom-up agent-based models and the absolute necessity of coarse-graining to handle non-equilibrium conditions. When we apply Gödel's incompleteness, computational irreducibility, and undecidability to these interconnected networks, perfectly simulating emergent behavior from the base layer up appears mathematically impossible. Given these hard logical ceilings on bottom-up approaches, what is the underlying physical or thermodynamic justification for why macroscopic coarse-graining remains tractable and phenomenally successful even in far-from-equilibrium systems?"

2:48:23.8 SC: So there's a lot going on here in this question. Let me try to address different things going on. For one thing, Gödel's incompleteness has, as far as I can tell, zero to do with any of this. The Gödel's incompleteness theorem, the spirit of it, let's not worry about the precise statement, the spirit of it is: In a formal logical system, there are propositions I can write down that if the system is consistent, those propositions are true but unprovable, okay? Notice I didn't use any words about physics, about evolution, about predictability, anything about that. It's about the difference between truth and provability in consistent formal systems, okay? So that's just a different thing; it's not going to be relevant here.

2:49:07.4 SC: But the other stuff is relevant. The basic idea is that if you have a large system made of many small parts, and the small parts are individually complex, then in principle it's computationally super-duper difficult to predict what the system as a whole is going to do. That's absolutely true, and that's a worry. But there is also an undeniable feature of the world that sometimes, under the right conditions, despite the fact that in principle the behavior of the system as a whole is unpredictable, in practice it's pretty predictable. In practice, you can find patterns that emerge at the higher levels. And oftentimes, maybe you wouldn't have guessed those patterns until you did a simulation of the many, many things acting together. The whole idea of agent-based modeling, what that really means is you don't just average over what you think the behavior is for your little constituents of your big system. In statistical mechanics, thermodynamics, kinetic theory, all these classic physics problems, the whole reason they work is because you can average over what the atoms are doing. You don't need to follow every individual atom to understand the behavior of a box of gas.

2:50:25.3 SC: The whole thing behind agent-based modeling is maybe that doesn't work. Maybe you have to actually take seriously what the individual actors do. It doesn't mean there can't be some emergent patterns, but the thing is that the only way to find those emergent patterns is to simulate how the agents actually behave with each other. And if you find some emergent behavior when you have a million agents in your simulation, then maybe you can extrapolate that to a billion or ten billion agents in your simulation. That's the hope of something like agent-based economics. So there's a little bit of what a scientist would call phenomenology in here. Maybe you can't derive the emergent higher-level behavior from the lower-level behavior, but maybe you can still find it. And honestly, this is very much what Phil Anderson was talking about many years ago when he said, "More is different." Anderson, Phil Anderson, condensed matter physicist, you know, famously says more is different and the emergent behavior at the higher levels can be treated separately from the individual lower-level microscopic physics. His paper is often put forward as some kind of rejection of reductionism, which is hilarious because if you read the paper, Anderson says multiple times, "Of course reductionism is true. What do you think? We all know reductionism is true." What he's arguing against is what he calls constructivism, the idea that the right way to find the emergent higher-level behavior is to start with the small-level behavior and somehow do some coarse-graining, okay? In principle maybe you can, but even when the constituents are very simple, in practice that can be really hard. In practice, it just is often easier to go to the higher level directly. So the hope is, I don't know, I don't have any theorems or results about how realistic this hope is, but the hope is that agent-based modeling is a way to find the emergent higher-level patterns just by doing brute-force computations rather than guessing them or whatever, or for that matter, not just guessing them but analytically carefully deriving them, for that matter.

2:52:32.4 SC: Okay, Carolee Cantor says, "Do you ever experience a slight quantum disturbance when, of all people, fellow physicists pronounce the G in Schrödinger with a soft J, like 'Schrödin-jer,' I suppose? I was watching your conversation with Neil deGrasse Tyson, who consistently used 'Schrödin-jer,' and I found myself quietly hoping that the wave function might eventually collapse in favor of the hard G." Well, I've got to say, I can't in good conscience be bothered by that. And I would love it, it would be great if every person, not just physicists but every person on Earth, when they pronounced words that come from a foreign language, pronounced them in the right way as viewed by listeners of the other language. That's just not going to happen, that's just not realistic. I certainly don't do it. You know, in high school I took German as my foreign language, and therefore I'm pretty good at pronouncing German words like Schrödinger. I've heard even German speakers pronounce Schrödinger in different ways, sort of, there's a scale from "Schrödinger" to "Schrädinger," something like that. And of course, non-German speakers don't speak it very well. But that's okay because, number one, when it comes to, like, names in French or Chinese or whatever, I'm sure I do terribly bad. And number two, even with names in German, I cannot help but say "Max Planck," okay? Because I just look at the word "Planck" and I know that it would not be pronounced "Plank" by a real German speaker or Max Planck himself. But I can't be on a high horse about that because I say "Einstein"; I don't say "Ein-shtine." I've known physicists who said "Ein-shtine," and I can't blame them because they're pronouncing it like Einstein would. But that's just, like, such a common English word, everyone talks about Einstein now, that it would seem pretentious to keep saying "Ein-shtine." So Max Planck is a little bit less well-known, so I can get away with that one. But anyway, it's okay. As long as you know what people are talking about, I'm not annoyed by people pronouncing things in slightly different ways.

2:54:36.4 SC: Mike Pencil says, "I recently heard about a 2021 study that caused a tardigrade, the little tiny organism, to be placed in a quantum superposition. If that's true, does it mean you could in principle perform the double-slit experiment with a tardigrade gun? Would you expect an interference pattern on your detector made of tardigrade-shaped specks?" I think the phrase "in principle" is doing a lot of work there. So for one thing, if I'm remembering correctly, the study was not... Well, yes, it is true that in some sense a tardigrade was placed in a quantum superposition. What was actually meant by that, again, if I recall correctly, is that the magnetic state of the tardigrade was entangled with the magnetic state of something else. It was not put in a superposition of being in very different locations in space, okay? So there's not, like, two different places the tardigrade could have been that are separated by more than the size of a tardigrade and it was really in a superposition like that. That, as a practical matter, would be incredibly hard to do because of what we talked about before with decoherence and photons and whatever. You'd have to put the tardigrade in basically absolute zero. If you could do that, we'd be much further toward building a practical quantum computer than we are now. So just being in a superposition doesn't mean that you could be placed through the double-slit experiment. But in principle, sure, you could put tardigrades through the double-slit experiment, human beings, whatever you want. And you would in principle get an interference pattern. My former colleague David Pollitzer, Nobel Prize winner at Caltech, he used to... He was not into the foundations of quantum mechanics, but he did have an opinion that was really close to what I would call Many-Worlds. You know, he took quantum mechanics seriously; he thought that you and I have quantum wave functions, or at least density matrixes, matrices or whatever. And he used to talk about his colleague who said, "You don't really think that when you walk through a door you interfere, do you? You have a diffraction pattern." And he said, you... Sorry, you diffract, I guess is the right way of saying it, and he said, "Yeah, I think you do, but you do it by a little tiny amount, so it's invisible."

2:56:47.4 SC: All right, I'm going to group, let's see, two questions together. One of them is long, one of them is short. Remember, short questions. I love short questions, they're much easier to read out loud. If you want the AMA to be mostly me giving answers than reading questions, then keep your questions short. Anyway, Tim Giannitsos says, "In the July AMA, you remarked on how we can characterize a system of particles by a few numbers like pressure, temperature, density, which are easily surveyable and which can be used to predict the future of the system for the most part. It's noteworthy because we're throwing away a lot of information about the system, but our descriptions can stay in terms of those numbers without losing the ability to make predictions. But it seems that the quantities you mentioned are kinds of averages of the particles, and thus it is no more mysterious than the fact that the expected value of multiple summed random variables is easily computable in terms of the individual expected values, even when we have no clue what the original full distributions were. It's not perfectly analogous, but I wonder if averages are just generally easier to reason about. Am I misunderstanding the statement? Isn't it just a matter of averages?"

2:57:53.4 SC: And Luca says, "Can we imagine theories at high energies which do not reproduce the Standard Model of particle physics or the Standard Model effective field theory at low energies because they only apply to higher energies? Basically a patchwork of laws instead of the Standard Model being fully contained within the deeper theory's domain of applicability. Did my PhD in effective field theories... I did my PhD in effective field theories, and I was wondering about that. In matching calculations, we assume both theories have the same infrared physics." So these sound like different questions, but they're actually closely related. Let's do Tim's first about: "Should we really be surprised by the effectiveness of this higher-level emergent phenomenon when, after all, it's just a matter of taking averages?"

2:58:35.4 SC: So there's two things that are going on. One is, you have to take an average of the right thing. That's the miracle, okay? So indeed, when you go from a large number of particles to macroscopic things like temperature, pressure, density, you're taking a tiny little region of space and you are averaging over them, either the number of particles or their velocity or their force that that they would exert on an imaginary boundary or something like that, to get things like temperature, pressure, density. That's not the only average I could imagine doing. I could make an average in momentum space and take every set of particles, no matter where they were in physical location, but particles with the same velocity, the same momentum, or something like that, and I could average over those. I would get some variables just like I get temperature, pressure, density, but those variables would be zero usefulness in making predictions because physics doesn't work that way. Or for that matter, I could imagine... Even in space, I could imagine taking an average, but if my distribution was... Imagine a bell-shaped distribution, okay? So a bell curve, a Gaussian distribution of particles, and I take the average and I get a point in the middle, right?

2:59:50.4 SC: That's fine. But now I imagine that I really have a superposition of two Gaussians, one of which has all the particles going to the left and the other one of which has all the particles going to the right. Now the average position just remains constant, just remains stuck at some point, okay? But once the two little wave packets move apart from each other, I'm no longer making good predictions about anything physical even though I've computed an average. So it is not just a matter of the fact that when you take averages, you get robust features that sort of average out a lot of idiosyncrasies. It is a rather miraculous fact that in the real world, there's a specific way you can take averages under the right conditions and be left with a bunch of data that is enough to make predictions about what goes forward. That's a highly non-trivial fact.

3:00:41.4 SC: And to Luca's question, the relationship here is that he's talking about quantum field theory, not about particles and thermodynamics. But it's the same thing, when you create an effective field theory from a high-energy quantum field theory in the ultraviolet, there's a very specific way that you are coarse-graining effectively. In quantum field theory, we would call it renormalizing or something like that, but it's really coarse-graining, and you're really coarse-graining out what is happening at short distances and high energies. And there's just a non-trivial fact about quantum field theory that there's a well-defined way of doing that that still gives you predictive physics at low energies. So for example, again, you're not going to be convinced until I tell you how it could go wrong, and I remember this example vividly because I got mistaken about it when I was a graduate student just learning these things. We often talk in physics, in quantum field theory, about integrating out some heavy particles. What that means is you have a particle like the W boson, which plays a role in the weak interactions in beta decay. When the neutron decays into a proton, electron, and antineutrino, we know now that what really is happening is one of the down quarks in the neutron emits a W boson and turns into an up quark, and then that W boson decays into the electron and the neutrino.

3:02:06.4 SC: They didn't know that. Fermi didn't know that when he invented the Fermi theory of beta decay, he had an effective theory which was just neutrons, protons, electrons, and antineutrinos. And so effectively, to go from the fundamental electroweak way of talking about that with the W boson to the Fermi theory, we are integrating out the W boson. The reason we can do that is because the W boson is heavy, and it therefore only influences things over short distances. And I was once talking to a quantum field theory professor, and I said, "I want to integrate out the photon in this particular physics system that we were talking about at the time." And he looked at me like I was insane. Like, you can't integrate out the photon; it's massless! If you did... I mean, you can do it, you can try, what happens is you can get wildly non-local physics. It is a feature of the real world that I can integrate over small regions of space and get an infrared effective theory that is still local. That's the equivalent of saying that when I integrate over a cubic micrometer of particles to get a fluid description, that fluid description is still local. It would not have been if I'd integrated over momentum space instead. So it is a feature of the real world that space matters, that there is a way of constructing emergent higher-level theories by averaging over what's going on in tiny regions of space. And that is kind of miraculous in a very real way.

3:03:39.4 SC: Leland Beaumont says, "I've heard that a qubit in superposition can be thought of like a flipped coin spinning in the air. The state is unknown until the coin lands as either heads or tails. Is this an accurate analogy?" No, this is a very, very bad analogy. This is exactly the analogy you don't want because it decreases your understanding of what's going on. In the flipped coin example, of course, the reason why you're tempted to think of it as an analogy is because, just like a qubit in superposition, you don't know what the answer is going to be until you measure it. The difference is that for the qubit, you really have an actual physical superposition. It's not that you don't know what the answer is going to be, that's true, but that's not the point. The point is there isn't an answer until you do the measurement. For a flipped coin, there is always a state of the coin. And in fact, in principle, classically, if you knew the state of the coin and you knew exactly all the air molecules, et cetera, you could predict what the answer is going to be. A good magician or sleight-of-hand expert can flip a coin in a way that they know exactly how it's going to land even though it spins when it's in the air. So there's a huge difference between ontology and epistemology, between saying that the true physical state of the qubit is a superposition, that's the whole point of the Schrödinger's cat thought experiment, versus saying that, "Oh, it is something, I just don't know what it's going to turn out to be when I look."

3:05:01.4 SC: Scott says, "Is there anything we could observe at a high-energy collider that no quantum field theory could account for in principle? If so, I further ask if there's any observation that no quantum theory could account for. I'm well aware of how successful these frameworks are, and I'm curious about how flexible they are in a high-energy regime." It's a little bit... This is a good question, but it's a little bit of a difficult one because, you know, when you say, "no theory can account for something," or, "no quantum theory," or something like that, there's a lot of different possibilities. So it's hard to say anything with absolute certainty. In the real world, we tend to compare better-defined theories than that, even though in principle we say, like, "Oh, I have some credence that the real answer is just a theory I haven't thought of yet." In practice, those are very, very hard to compare to experiments, right?

3:05:52.4 SC: So anyway, for quantum field theory, QFT, the super important difference between quantum theories generically and quantum quantum field theories in particular is that quantum field theories are local, okay? I'm tempted to say they're Lorentz invariant, but that's not true. Relativistic quantum theories are Lorentz invariant, but there's plenty of quantum field theories, such as, you know, in condensed matter physics, that are not Lorentz invariant. So the really important thing about QFT is that it's local, by which we mean if we poke the field at one particular location in space, those influences of that poking are going to spread out slower than the speed of light, okay? They're only going to affect the exact nearest neighbors right away, and they're not going to affect things instantaneously very far away. So I haven't done any serious thought about it, but my guess is that if you really wanted something which was a dramatic violation of quantum field theory, the most straightforward way to get it would be by finding something non-local. Like, I have a switch that is a light year away from whatever is happening at CERN at the LHC, and by doing something with that switch, I instantly affect what is going on at the LHC. That might be something that you could try to do. That's very, very hard to do. This is why people won the Nobel Prize for testing the Bell's inequalities, because how do you know there wasn't some common influence in the past that affected both what you did a light year away and what is going on at the LHC? So it would be work, but maybe you could do it.

3:07:26.4 SC: In terms of just quantum theory itself, I am not sure that how to test... How to do an experiment that no quantum theory could account for. I mean, it's in practice it's very easy, right? If you could just construct a spin in an eigenstate of X-spin and then observe it, and then it always came out spin-up in the Z-direction, that's not what quantum theory predicts, and that's in contradiction. But then what you're asking is, "Could I invent some weird different quantum theory that might explain that?" That's harder to say. I'm not really sure.

3:08:04.4 SC: Linus Melberg says, "I know that the Lagrangian and Hamiltonian formalisms of classical mechanism mechanics are equivalent, but couldn't the past hypothesis be treated differently between them? I'm thinking there could be a Lagrangian of the universe that in general forces a low-entropy bottleneck, a Big Bang, between two random high-entropy boundaries. I'm guessing I'm just hiding the past hypothesis somewhere, but I don't know where." The short answer is no, I don't think that that's different. So even if you could do what you... So there's two issues here. Number one is: Could you do what you, you suggest, to have a Lagrangian of the universe that gives you a sort of low-entropy Big Bang? And is that different than the Hamiltonian formalism? So I think even if you could do that, it would not be different than the Hamiltonian formalism. There's literally a mathematical demonstration of how to go back and forth between Lagrangians and Hamiltonians. It is not... There's always exceptions, loopholes, things like that, usually sets of measure zero, but basically these two things are just mathematically demonstrably equivalent.

3:09:07.4 SC: As to whether or not you could make a Lagrangian that forces a low-entropy Big Bang, I think that that's very hard to do. I mean, unless what you mean is something like what Jennifer Chen and I proposed 20 years ago, with a universe that has a U-shaped entropy curve, generically there will always be a minimum, and the entropy will grow without bound both to the past and the future. But that's... I mean, that's only in a very loose sense a feature of the Lagrangian. It's really a feature of the state space, the idea that the entropy is finite at any one point but has no bound, can grow forever. If you have those two conditions, then it's generically true, unless you really try hard, that entropy will increase without bound to both asymptotic directions of time.

3:09:54.4 SC: John Schoning says, "My understanding is that we often treat potentials as just bookkeeping, since the only gauge-invariant quantities are observable. But effects like the Aharonov-Bohm effect suggest potentials can have physically real consequences. How should we think about the ontological status of gauge potentials? Are they merely descriptive redundancy, or do they reflect something genuinely real in the structure of the theory?"

3:10:20.8 SC: So what John's referring to is, if you talk about, for example, the gravitational field, okay? You say Isaac Newton says there's an inverse-square law for gravity, that the gravitational force that the sun exerts on the Earth is instantaneous and it's given by GM1M2/R^2, where M1 is the mass of the sun, M2 is the mass of the Earth, and R is the distance between them, and G is Newton's constant. And then Pierre-Simon Laplace, one of our heroes around here, comes along and says, "Well, I can do exactly the same thing. I can reinvent Newtonian gravity, but instead of saying there's a force that goes off as an inverse-square law, I can say there's something called the gravitational potential field that exists at every point in space. And I can write down an equation that it satisfies."

3:11:12.4 SC: And now the potential field has the property that the force is the derivative of the potential. And in fact, the same thing goes true for a ball rolling down a hill, right? You can actually treat the height above ground of the hill as a potential, and the force that the ball feels is like the slope of that potential, okay? And this is sort of an amusing mathematical thing in traditional classical mechanics. But then when it comes to electromagnetism and gauge theories, now you have an electromagnetic potential, and different derivatives of the electromagnetic potential give rise to the electric field and the magnetic field, which are physically observable. But there is some gauge invariance. You can take the electromagnetic potential and shift it by an amount without changing the electric field or the magnetic field. It is the equivalent of saying, if I have a hill and that the slope of the hill is the force, if I take the hill as a whole and move it up by 100 feet, the force on the ball rolling down the hill doesn't change because that only depends on the slope of the hill, not on the height of the hill. That's the sort of Newtonian particle-in-a-potential equivalent of gauge invariance in electromagnetism.

3:12:21.4 SC: But then people, very smart people like Aharonov and Bohm, realized that they could find situations where particles move through the vacuum, that is to say, no electric field equals zero, magnetic field equals zero, and yet you can't correctly describe what they do without using the gauge potential. And the reason for that is because Aharonov and Bohm imagined a solenoid with a magnetic field in the middle, and an electron moves around both sides of the solenoid. So even though the electron doesn't travel through a region of non-zero electric or magnetic field, the loop that goes around the entire electron path on either side of the solenoid includes a magnetic field inside. So I think that in some sense the gauge... So the usual lesson people take from the Aharonov-Bohm effect is the gauge fields are real. There's something real captured in the existence of these potentials, not just in their derivatives that give us the electric and magnetic field. But it's something global, right? It's something that is not true at every point in space. At a point in space, it doesn't matter what the gauge field is doing, it only matters what its derivatives are doing.

3:13:37.4 SC: But globally, if you go around a circle in space, then it matters what happens at other places along the way. That's not just something weird about electromagnetism, gravity is exactly the same way. The metric tensor in general relativity is very much a potential for the curvature. You know, it's only derivatives of the metric; if you just added a constant to the metric, it wouldn't change the curvature at all. But there are circumstances under which you can show that what the metric is doing actually does matter over and above the curvature. So I think that it's just, you know, a matter of, it's not that the case gauge potentials are real in the most obvious way, but there are features of the gauge potentials which typically show up as a global phenomenon rather than a local one, where there's something real about them that is not completely captured in their local derivatives. Something like that.

3:14:31.4 SC: Chris Chautard says, "There's something I don't understand with the usual 20th-century physics narrative. I've read time and again that Heisenberg invented a matrix-based quantum mechanics theory, and that as he was initially ignorant of matrix calculus, it was Max Born who in fact oriented him to matrices. But general relativity was discovered years earlier, and it heavily relies on tensor calculus, which encompasses matrix calculus. Does this mean that Heisenberg didn't have the math toolkit to understand GR?"

3:14:58.9 SC: This is a great history of physics question. I'm only going to be able to guess and conjecture. I didn't look up the actual answer to this, I'm sure someone has known it. But there's two things. Number one: It's totally possible that Heisenberg did not understand general relativity. Remember, we're talking about Heisenberg doing his work in 1925, and general relativity was put on the scene in 1915. So it's only ten years, right? It's the equivalent of talking about something that happened in 2016. Not every working physicist here in 2026 knows all of the cool stuff that happens happened in 2016. There was too much cool stuff happening, especially when general relativity is a totally different set of ideas than quantum mechanics, right? You don't need to understand GR to invent quantum mechanics. So maybe he didn't. But maybe he did. Plenty of people did. Wolfgang Pauli famously wrote a whole book about general relativity when he was, like, I don't know, 20 years old or something like that at around that, this time in history.

3:15:54.0 SC: The other thing is, Chris says in the question, "general relativity relies on tensor calculus, which encompasses matrix calculus." In some sense it does, but that doesn't mean that the sense is obvious to you. When you learn general relativity, you have tensors, and you usually describe tensors using all of these indices, right? So you have g_mu_nu, where mu and nu are indices that go from zero to four, the four dimensions of space-time. You can write g_mu_nu as a matrix, and you can imagine contracting one tensor with another tensor and analogizing it to multiplying matrices. But doing that in your head isn't obvious. Like, maybe you didn't. Like, maybe you just knew there were tensors, but you always treated them with indices and you had summation rules and things like that, and you never stopped to think, "Oh, this is like multiplying matrices." Again, I have zero idea what was actually going through Heisenberg's thought process, but it is tricky to think about what was known and obvious to physicists 100 years ago or much longer ago.

3:17:00.4 SC: Andy Kearney says, "Bayes's theorem and Bayesian reasoning comes up in many of your episodes, and when explained, it always seems to make sense; it's even natural. But I hadn't really thought about there being any really formal alternative, so I was surprised recently when I came across frequentist reasoning for the first time. Would you be able to explain how they differ and discuss if there are any circumstances where frequentist reasoning would be preferred method to use?"

3:17:23.5 SC: So I think that there's two things you have to keep in mind. There's sort of reasoning, and there is... I don't know, what do you want to call it? An approach to the ontology of probabilities or something like that. Okay? So Bayes's theorem is a theorem. Everyone uses it. It doesn't matter whether you're more Bayesian or more frequentist in your philosophy of probability. Bayes's theorem says when you have some set of credences and new information comes in, here's how you update them. You don't even need to call them credences. Bayes's theorem is really just a relationship between different marginal and conditional probabilities in a big giant probability distribution, and it's true for everyone no matter what your philosophy is.

3:18:09.4 SC: The difference in philosophy is really about answering the philosophical question: What is a probability? What do you mean by a probability? Okay? And the Bayesian... The person who does sort of Bayesian philosophy, gives a different answer to that question than the person who is a frequentist. The frequentist says the only real meaning of probability is imagining that we could do something an infinite number of times and showing that a certain thing happens a certain fraction of the time and a certain other thing happens another fraction. This goes very hand-in-hand with the origin of probability theory, which had a lot to do with gambling, with rolling dice or flipping coins or playing cards or something like that, where you have a well-defined probability that something happens that can easily be understood in the infinite-case limit. If you flip a coin and say it's a 50/50 chance of being heads, maybe what you mean is if you imagined flipping it an infinite number of times, half of the time it would be heads, okay?

3:19:13.4 SC: And now the Bayesian says two things. They say, number one: Clearly that can't be right. You can't do this thing an infinite number of times. If you say it's going to happen 50/50 an infinite number of times, that's because you already have an opinion about what probability means even before you define that, right? And no one... People talk about probabilities all the time in cases where you don't do infinite numbers of trials. And the second point is, in some cases, you can't do an infinite number of trials, but nevertheless we talk about probability all the time. The probability of someone winning a presidential election, the probability of a certain country winning the World Cup in soccer, okay? You're not imagining doing that hundreds of times or an infinite number of times, you are expressing your personal degree of belief. That's the best you can do in those cases. It's not frequentist, but the Bayesian says it's still probability because those credences you have obey the axioms of probability, right? If there's 24 teams playing for the World Cup and you have a credence of each of them winning, those credences better add up to one, they better all be non-negative numbers between zero and one that add up to one, and you better update your credences when new information comes in using Bayes's rule. So let's call them probabilities. And the fundamental philosophical difference is exactly that, whether or not you're able to call your credences probabilities or not. Bayesian says yes, frequentist says no. I'm definitely on the Bayesian side of these things, but there's a long conversation to be had about that.

3:20:51.4 SC: Josh Dobbin says, "I have a question of the 'do I understand this correctly' form. The speed of light is the speed limit of stuff in the universe, but expansion of the universe itself, empty space, can expand faster than light, right? Like the space between distant galaxies is moving faster than light, effectively not so much pushing them away faster than light, but kind of increasing the distance where no stuff is. Isn't that a de facto moving the stuff away at a speed faster than light?"

3:21:16.6 SC: So no, you don't quite understand this correctly. You're on the right track, you're getting there, you will understand it very soon. But the point is: The speed of light is a speed limit of stuff in the universe; that part is completely right. Empty space can expand faster than light; that part is not right. That is completely wrong. Don't ever say that. And the reason why I can be so definitive about saying don't ever say space can expand faster than light is because the expansion of space is not measured in units of velocity, right? The velocity of what? The whole point of the expansion of space is the Hubble law. The Hubble's law says the velocity is the Hubble constant times the distance. So the velocity that I observe a distant galaxy to have depends on its distance. A nearby galaxy isn't moving away very fast, a distant galaxy is moving away faster. So what in the world should I assign to the expansion of the universe that would have a velocity? The velocity of what? Different galaxies have different velocities. The units in which the expansion of space are denoted is one over time, okay? That's different units than velocity, which is distance over time. So they're just different things.

3:22:27.4 SC: Now, of course, as I alluded to earlier in the AMA, we informally talk about the velocity between galaxies all the time. And that's just because we're being informal, that's all. That's really the only reason. To make it a little bit more legitimate, we can see the light coming from a distant galaxy, we can calculate its redshift, and we can say, "What would the velocity have to have been in order to get that redshift?" And then we can talk about the fact that... We can talk about it as if the galaxy is moving at that velocity. But really, like I said earlier, unless two objects are at the same location in space, the rule in general relativity is there's no such thing as their relative velocity. So certainly space expanding faster than light is not the right way to think about it.

3:23:15.4 SC: Paul Hess says, "Do you think the Electoral College is a useful construct or a pointless anachronism? Many people decry it as an undemocratic distortion each election cycle, but it seems to me it serves a useful purpose by preventing a candidate from winning by simply catering to a small number of populous areas to max out votes there without having to appeal more widely in many areas. That would make extremism in politics even more of a winning strategy. What are your thoughts on this? Maybe it needs to be modernized but not jettisoned."

3:23:42.9 SC: My thoughts are it needs to be jettisoned. I think the Electoral College is a terrible idea. And I'll tell you a very... There's a very simple reason why. Like, so the worry that you have expressed and other people have expressed, if it weren't for the Electoral College, which for the non-Americans or for the people 500 years from now who just read about American politics in history books, the way we elect a president is that each state elects Electoral College members. In the original conception, this was supposed to be literally people who would be given the responsibility of then going and talking to each other and picking a president. That was soon abandoned, and the idea now effectively is that there's just a number, which is the number of electoral votes every state has. And who is actually the electors is almost completely irrelevant. It's just a number. And most states, not all, but most states, have the policy that whoever gets the most votes for president in that state gets all the electoral votes in that state, okay?

3:24:46.4 SC: So California, Texas, New York, Florida, states that have huge numbers of people in them get a lot of electoral votes. States like Wyoming or Idaho or whatever, who have relatively few people in them, get relatively few electoral votes. But it's actually not proportional. So Wyoming has a lot more electoral votes than it should have given its number of people. So the worry with a system like that, you might think, "Well, who cares? It's just like a step of coarse-graining, but still, the more votes you get, the more likely you are to win." Empirically, in the last few elections, we've had examples where people have lost the total number of votes in the United States but won in the Electoral College. The justification for this is that by giving representation to smaller states, presidential candidates have to spend time campaigning and caring about the values and interests of people in those smaller states. They can't just go to the big cities and campaign there. Now, I have very little patience for that argument because, number one, it's clearly not true. Small states like Wyoming has very few electoral votes. How much time do presidential candidates spend campaigning in Wyoming? Zero.

3:26:08.4 SC: And the reason is because in many states, there's many more Republicans than Democrats or vice versa, so the states are not competitive. So even though in California there's an enormous number of Republicans, Republican voters in California, but there's even more Democrats in California, enough that very safely these days, the Democratic candidate is always going to win California. And likewise the opposite in a state like Louisiana, okay? There are Democrats in Louisiana, they're never going to vote overall as a state for the Democratic candidate. So those states are just completely ignored. There's no presidential campaigning going on in California, Alaska, New York, or whatever because we know how those states are going to vote. What matters is the small number of swing states where there's approximately equal number of Democrats and Republicans. And that is wildly unfair to the rest of the country. I think that the Republicans in California should get a voice, I think the Democrats in Idaho should get a voice, et cetera. So the Electoral College just doesn't do the job that it was described to do.

3:27:16.4 SC: And the other thing is, of course, the world is a very different place than it was 250 years ago. Much campaigning is national. If you didn't have the Electoral College, you would be able to, you would care about every voter, right? Because they weren't coarse-grained away in the Electoral College process. Every vote would count equally. The vote of a Republican, the vote of a swing voter in California, which is not a swing state, would count exactly as much as the vote of a swing voter in Ohio or Pennsylvania, which are swing states, okay? So that would be a much more fair system, I think. There's really no more justification for having the Electoral College anymore.

3:27:58.4 SC: David Kudaverdian says, "I am not a native English speaker, and while reading your books and listening to Mindscape, I sometimes notice how certain words and phrases previously unknown to me seem to enter and then leave your vocabulary. Do you ever notice these kinds of changes, or do they pass unnoticed? For example, when Mindscape started, you frequently used the word 'oomph,' and then you stopped using it. Or in your book about the Higgs boson, you used the expression 'nothing to write home about' several times, but then you didn't use it in your other books."

3:28:26.9 SC: This is fascinating to me. I love this observation; I'm sure it's true. I don't notice it myself, but I'm completely unsurprised by it being true. And I think that this all that happens, it's not conscious. It's just that certain words or phrases are in your brain at certain different times of your writing or speaking. It reminds me of... There are claims, and I don't know how legitimate they are, there are claims that we can figure out which plays Shakespeare acted in of the plays that Shakespeare wrote. You know, Shakespeare was a playwright, but he was also an actor, so he played roles in his own plays. And people have said that since we know the order in which the plays were written, what happens is if... And Shakespeare was not the lead actor, right? So he was not playing the biggest parts; he played some smaller parts. He was busy writing the plays. And he of course invented all these crazy words, right? Like, he was endlessly creative that way. So the claim is that you can look at the words spoken by certain parts, by certain roles in one play, and if Shakespeare was the actor playing those roles, those words appear with an ominously high frequency in the next play that he's going to write because they're in his brain, okay? Again, I have no idea whether that's true, but it's cute. I hope no one does that with my podcasts or books.

3:29:45.4 SC: Okay, I'm going to group two questions together. Paul B says, "Do you have a recommended way for latecomers to Mindscape to explore your amazing back catalog of regular episodes? I started listening a few years ago and kept up to date with all the new episodes while simultaneously archaeologically excavating my way through all the previous regular and solo episodes. It was interesting to travel two opposite timelines simultaneously, noting how your style of interviewing and presentation evolved. Also, pretty much all the other episodes seemed to be relevant, prescient, and illuminating, even though some were heading toward eight years old." And Paul Hess says, "I've always thought it would be really cool to make some sort of cluster or connection diagram of the concepts discussed in all the different podcast episodes. Maybe now with the new generation of AI tools, it might be a fun project to someday try to do that."

3:30:32.4 SC: So to Paul B's question, I don't have a specific ordering that you should go in or anything like that. I mean, I think, like you say... Basically what you're saying is that I have succeeded in the goal of making most of my podcast episodes somewhat timeless, right? As everyone knows, Mindscape is not about just the latest new wrinkle. Even if there's a new news story in science or a new idea in philosophy or even a new current event happening, the thing that I personally am going to try to emphasize in Mindscape are aspects of this new thing that will last, that are of long-term interest and not just ephemeral. So if the old episodes are still relevant and interesting, that's what I've been aiming for all along. So that's good.

3:31:22.4 SC: I love the idea of using AI or something, I don't know, to sort of do connections between ideas from different podcast episodes. I mean, sometimes the connections are obvious, if I talk to, I don't know, Raphael Bousso and Netta Engelhardt about black hole information, there will be similarities there. But I would love to see that, you know, there's connections between Raphael Bousso and Alison Gopnik or something like that, or Joe Walston, talking about completely different things. I have no idea what that would be. I suspect there would be... If it were done correctly, I bet there would be things that would come up in connections and with frequencies that you might not have expected that would be pretty interesting. Because after all, there is a common feature here, these are all people who I chose to talk to, which is not a completely random selection of people.

3:32:13.4 SC: Hail Zeus says, "Which combination of philosophy and science coursework would you recommend for an undergraduate who hopes to pursue a PhD in the philosophy of science? Closely related, which major or minor to pursue, such as a philosophy major with a science minor, science major with a philosophy minor, or even a double major, or a different suggestion entirely?"

3:32:33.7 SC: You know, my suggestions are always, if you want to do X and you know that field Y is closely connected to it, mostly do X. You know, the physics version of this, some physicists who want to be theorists realize that math is really, really important, so they try to learn all the math they can, even to the point where they're not learning as much physics as they could. And I suspect that it's easier to just learn the physics and embed that in your brain and then pick up the math along the way than it is to actually sacrifice learning physics for learning the math. Because when you take a math course from a math teacher, it'll be very, very good, but the purposes, the motivations, the interests of the mathematicians are different from those of the physicists, and vice versa. I think it's exactly the same for philosophy of science. If you want to become literally a philosopher of science, you should major in philosophy. You should absolutely do a lot of coursework in science, and maybe in one kind of science or maybe in multiple kinds of sciences. But guess what? The goals of the physics professor are not to teach you the philosophical aspects of physics, but teach you the physics aspects of physics. So you need to train yourself in philosophy as well as you possibly can, in addition to picking up up extra physics and things like that. A double major is a great way to do this if you can pull it off, then you get both sides and you can do the work of synthesizing them together. It depends on what university you're at whether or not that's a feasible thing to do in the real world.

3:34:08.4 SC: David Whitaker says, "Should scientists advocate for their point of view or beliefs or conviction? I ask in the context of the loss of confidence in the institutions, including science and academia, on the part of the public. If scientists are seen to be partisan on an issue or to favor a particular interpretation of the evidence, even if the issue isn't one that is in the political sphere, such as vaccine efficacy, their message may nevertheless be regarded as biased and might not carry the weight that an impartial scientific opinion would or indeed should." I mean, what can I say? Of course scientists should advocate for their point of view. What other option do you want to have? If there's some question in the public sphere that is relevant to scientific knowledge, then scientists should be very clear about what their best opinion about that question is. They shouldn't worry about the political ramifications of it. I mean, some... I shouldn't say downfall, the political ramifications of it. You know, sometimes we talk about scientists saying things, but what we really mean are, like, policymakers saying things. So if you're, you know, Anthony Fauci, your job in the Center for Disease Control or whatever is not to be a scientist, it's to be a policymaker and things like that. So he might have been a scientist also, but that was not the role he was in at the time. I do think that in some mild sense, scientists and policymakers during the COVID pandemic went too far in trying to pretend to be more certain than they were about certain facts about the disease and how it was spreading and things like that. I take this to be a very minor flaw. They were trying to do something good, which is to sort of remove the opportunity to be uncertain about these really, really important steps that individual people should be taking. But that's a relatively minor criticism.

3:35:57.7 SC: When it comes to the loss of confidence, confidence in scientists and institutions, I put 99% of the blame for that not on the scientists and institutions themselves, but on bad-faith actors that want to deny the scientific reality and therefore try to undermine confidence in these institutions. It's not the scientists' fault that people don't believe scientists. Again, at the 99% level, there's maybe 1% of their bad behavior that gets blamed for that, but most of it is because people are out there trying to undermine people's belief in what the scientists are saying. And some people are falling for it, and I think that's too bad.

3:36:36.4 SC: Steve H. Says, "Given your work on the multiverse and cosmological horizons, is there any physical observable that could distinguish between our universe being complete versus being a subsystem of a larger reality, or is that distinction permanently beyond empirical reach?" Well, it depends. That's a very vague setup. So, you know, like I just said, it's important as Bayesians that we assign credences to vague scenarios, but it's very hard to know exactly how to evaluate them. There's different ways to be a subsystem of a bigger system, okay? The real question is: Is there some causal influence that the bigger system has on our universe? But there's a secondary question is: What is the best theory of our universe all by itself? I do think that people sometimes have this overly simplistic picture of how science is supposed to progress, where we have ideas... And it's Karl Popper's fault in some sense, we have ideas, we do an experiment, we falsify or confirm the idea, right? But that's just not how science works. It's part of how science works, but the reality is more, much more nuanced and interesting than that. We want to explain the universe that we do observe the best we can.

3:37:54.4 SC: Sometimes the best theory that we can come up with to explain the universe we see implies that there is stuff that we don't see somewhere. Maybe there's dark matter, maybe it's going through my body right now. Maybe there's a multiverse out there somewhere else. These are all possibilities. And the question is not, "Well, if I can't see it, it's not real." The question is: What is the best possible explanation of what I do see? And then until you have a better explanation, you have every right to accept the existence of the theoretical implications that best account for the universe that you do see.

3:38:31.4 SC: Okay, Marie Roscue says, "What is your opinion on the Wheeler-DeWitt equation?" The Wheeler-DeWitt equation is an equation that you get if you try to quantize general relativity, particularly in a closed universe. It's just simpler mathematically to imagine finite-sized universes than infinite ones. And the weird thing about the Wheeler-DeWitt equation, I encourage people to listen to the solo podcast I did about the emergence of time, because the weird thing about the Wheeler-DeWitt equation is it doesn't say that the Hamiltonian acting on the wave function gives you the time derivative of the wave function; it says the Hamiltonian acting on the quantum state gives you zero. So there's no time evolution. So this is called the problem of time in quantum gravity. And I don't want to rehearse everything I said in that solo episode, but let me just say, you know, the Wheeler-DeWitt equation, I think, has a mixed status in some sense.

3:39:22.4 SC: On the one hand, it's a brilliant result from very good physicists that is foundational in quantum gravity. On the other hand, it is what you get by starting with a certain classical theory and quantizing it, and maybe the universe doesn't work that way. And I think this is what we're running up against in the episode, the conversation we just had with Daniel Harlow. He's struggling with how to interpret the fact that his theory seems to predict that the Hilbert space of the universe is one-dimensional, and that in some sense can't be right. So you can either say it's just not right and the reasoning that led you to get there is flawed somehow, or you can say it's right but with certain twists and turns that we hadn't previously anticipated. And that's the route down which Daniel's trying to go, he's trying to imagine that there's a big Hilbert space that is somehow connected to the little Hilbert space, and there's a whole superstructure there. And I think that it's not really the way that I want to go. So maybe you have to give up on the Wheeler-DeWitt equation, I'm not sure.

3:40:24.4 SC: Okay, one last question from Martin Marcin Chady: "You like to address people who listen to your podcast 500 years in the future. What things that we consider normal today do you think that they will consider barbarian?" There's a great question to which I don't have a pet answer ready for you because there's too many examples. I mean, it depends on lots of things, okay? This is why you can't give a perfect answer to this. One thing is, the idea that people 500 years from now will consider something that we do barbarian seems to optimistically rely on the fact that we'll continue to have progress, that we'll get better. Like, I think that if you go 500 years ago, there's things that we now consider barbarian, like how we treat prisoners or something like that, or minority groups or women. And we have... In my view, we have absolutely progressed in certain ways in the last 500 years. It's by far not clear that in the next 500 years we will continue to progress socially or even technologically, for that matter. All sorts of bad things could happen.

3:41:30.7 SC: So I guess the right way to construe this question is: What are the things that we do right now that sort of obviously could be done better? I think there's a bunch, but they're not all in my head right now. So let me just say two things that immediately pop to mind, which are both sort of political in cast. One is what we were just talking about, the way we elect people in the United States and in a lot of other countries. You know, forget about the Electoral College, the idea of what is called "winner-take-all" or "first-past-the-post," where you have a political contest and whoever gets the most votes wins. Even if the number of votes is 30%, if there were just a lot of candidates and the other candidates got less than 30%, then whoever gets 30% wins. That's completely ridiculous. And there's an even worse version that the Californians have managed to come up with, where they have a runoff election, they have a bunch of candidates, and then the top two vote-getters in the primary election go into a runoff. And I think... I don't actually don't know the details here, I think this was set up by the Democrats because the Democrats kind of dominate California, and they wanted to allow for the possibility that the top two would be both Democrats and then have them fight it out, and that would be good. But in fact, here in 2026, we're in a situation where there's only two Republicans running for the governor of California, there's a bunch of Democrats who have split the vote, and as of right now, the two leading candidates for governor are both Republicans, even though the total number of Democratic voters is much larger than the number of Republican voters. This is just even worse than the first-past-the-post thing. So something like ranked-choice voting, or I think there's a lot of even more imaginative ways that you can sort of split votes and represent people in representative democracy that we have not really looked into. And that what we're doing right now is just completely ridiculous.

3:43:29.4 SC: The other thing that I would point to is income inequality. The fact that we have just such a huge dynamic range of amounts of wealth between people in this country and the world, I think is going to be looked at as pretty barbarian if we continue to make progress socially and politically. Not because there's anything morally wrong about having a lot of money, I've said that many times, I don't think that's true, but because it gives undue influence politically, socially, culturally, et cetera, to people who have too much money. And we're seeing this right now, where individual super-ultra-wealthy people can just, like, buy a TV network or buy a newspaper and then completely change the news and information-giving aspects of these institutions, which should be beyond the capacity for any one person to completely remake just because they have a whim to do that.

3:44:25.4 SC: So I'm in favor of some version of sort of modified welfare-state capitalism that includes putting big taxes on wealthy people, both on their incomes and their wealth and everything like that, and using that to solve poverty throughout the world and to, like, give people basic goods to make themselves better. The irony is, if you gave a lot of money to everyone in the world, not a lot of money per person, obviously, but if you gave some money to everyone in the world, everyone would benefit, right? The rich people would benefit as well because the whole country, the whole world would have a much stronger economy with even people who are now in poverty buying things and stuff like that. The only way to become rich is because you have a product that a lot of people want to buy. And if people don't have enough money to buy it, you can't become rich. [chuckle] Which is just a way of saying that, you know, the more income and wealth is distributed, the easier it is to have a healthy, functioning economy.

3:45:22.4 SC: But the fact that we don't, we all know why, it's because the people who have a lot of influence want to keep their money and they want to keep their imbalanced wealth distribution, and so we're not doing a very good job of dealing with that. I'm sure there are other things that I could say. I think that, you know, maybe it'll be considered barbarian that people didn't accept the Many-Worlds interpretation of quantum mechanics. I don't know. That would be a very interesting result that we'll have to see if that comes about.

3:45:47.4 SC: So thanks as always, everyone who both supports Mindscape on Patreon but also who asks these questions and also who listens to the answers to these questions. Does me no end of marvel that people want to hear hear me talk about all sorts of crazy things for three hours. So thanks for your support for Mindscape. I'll talk to you next time.

[music]