AMA | March 2024

0:00:00.0 Sean Carroll: Hello everyone. Welcome to the March, 2024. Ask Me Anything edition of the Mindscape Podcast. I'm your host, Sean Carroll. Just a couple of quick housekeeping, bookkeeping kind of announcements before we get into the AMA. First, the good news is we have awarded this year's version of the Mindscape Big Picture Scholarship. For those of you who don't know every year, well, anyway, this is the second year in a row, Mindscape listeners and I have donated to a scholarship fund. This is run by bold.org, so you can go to bold.org/scholarships/mindscape. And the idea is we collect some money and we look for student applicants, high school students generally, or early college students who are studying science, philosophy, math, big ideas in some way or another. And we were able to offer them a $10,000 scholarship. We gave two scholarships last year. This year we're just gonna give one scholarship, but I think we're on track to give multiple scholarships in the future.

0:01:03.1 SC: It's not gonna completely pay for college, but it might help someone who wants to really dig into these big ideas that we'd love to talk about here on Mindscape, but which are not necessarily the most practical job oriented things. We're really looking for people who want to understand the world better. So this year's winner is Ryan, if I get his name right, Ryan Funakoshi. Ryan lives in California and his application mentions he does a lot of volunteer work and things like that, runs marathons, the usual high-achieving kind of guy, but also really, really interested in biology and in particular, molecular and cell biology and the relationship to physics, even in the sense of radioactive elements causing mutations that can affect organisms, that can affect evolution, etcetera, for a first year college student. That's, great stuff. So congratulations, Ryan.

0:02:01.9 SC: And also, I wanted to say thank you to the many people who have donated to the scholarship. You can go to that address that I just mentioned, or if you wanna donate, you can just go to the podcast webpage, preposterousuniverse.com/podcast. I think that more people should visit the podcast webpage. There's a lot of good stuff there. I know a lot of you listen to the podcast on mobile devices or whatever. So a website is not immediately accessible, but there's transcripts, there's show notes, there's links to books and things like that. And on the sidebar, you'll find a link to the Mindscape Big Picture Scholarship. So you can donate there. And I wanted to say that there's been just tremendous support for this scholarship effort. I started it, the bold.org people reached out to me, so I started the scholarship. I said, sure, why not?

0:02:49.9 SC: I was completely unsure whether anyone would donate money, but the results have been truly heartwarming. And some of you, some folks out there have given truly substantial amounts of donation. And I don't really have any easy way to thank you individually. So don't think that, I don't know, and I'm not appreciative. I think it's really, really wonderful what this Mindscape community has come together to do. And I think we're gonna keep doing it for a while now. So that was one little bit of housekeeping. The other one is just that these Ask Me Anything episodes, which you're about to listen to, are funded by Patreon supporters. So feel free to become a Patreon supporter of Mindscape. If that's how you roll, just go to patreon.com/seanmcarroll and you can join up, pay a dollar or even more if you like, for every episode of Mindscape in return, you get ad free versions and of course, you get the feeling of belonging to the community.

0:03:44.2 SC: And not to forget to mention, which I always do for Patreon listeners, I do after every regular episode. So like not after a solo episode and not after it's just me talking at an AMA, but when I'm interviewing someone, I will do a little video, just a few minutes reflecting on that episode, amplifying on something that happened, talked about it. I think it's interesting. I think that people like it. I don't know. I'm not exactly sure what the feelings are out there, but I enjoy doing it. So that's a Patreon exclusive, if you're into that. Again, never any reason to feel bad if you're not a Patreon supporter. Mindscape, as far as I'm concerned, will always be out there for free in one form or another. But if you'd like to support it, that is very much appreciated. Keeps me going, keep me doing these podcasts and of course, keeps me doing these AMAs. So let's go.

[music]

0:04:50.5 SC: Eric Chen asks, can you explain what your new paper on holographic phenomenology via overlapping degrees of freedom is about? Sure. I love it when the AMA gives me an opportunity to talk about my actual physics research or even my philosophy research for that matter. So this is a new paper, just came out. You can look it up on the Internet's out there, Holographic Phenomenology Via Overlapping Degrees of Freedom. So Oliver Friedrich, who's a postdoc in Germany, he was the lead author on this project. When you have multi-author papers, they don't always do exactly the same amount of work. Everyone does some work, otherwise they wouldn't be an author, at least on theory papers. So, but Oliver was definitely the guiding principle behind this particular paper. And it's actually a fun idea. I got to get a little technical here. Sorry to start off the AMA with the technical physics thing.

0:05:40.2 SC: But that's the given back and forth between technical stuff and just fun, silly stuff is part of the joy of the AMAs. So here's the idea. We have something called the holographic principle. And the holographic principle is slightly ill-defined. There's no once and for all complete, perfect formulation of it, but there are different versions that have implications in different regimes. The ADSCFD correspondence is a very nice implementation of holographic ideas. But the original holographic idea, which was sort of firmed up by Raphael Bousso, former Mindscape guest, says that the amount of entropy that can pass through a region of space is not arbitrarily large. Which you might think from quantum field theory and quantum field theory, every region of space has potentially an infinite number of degrees of freedom because there's waves and waves are smooth, continuous things, they can wave in any possible way, right?

0:06:41.2 SC: But the holographic principle says that there is a maximum entropy that can go through a region. And I'm skipping over some technical details here, but roughly speaking, that maximum entropy is proportional to the area of the boundary of the region. Famously, of course, in a black hole, Bekenstein and Hawking said that the entropy is equal to the area of the boundary of the black hole, the event horizon area that's measured in Planck units. The holographic principle says that's an upper limit to how much entropy you can fit inside a region. And that could be maybe a little surprising because ordinarily, when you're not dealing with gravity or quantum field theory, rather you're dealing with a box of gas, then the area is not relevant to the entropy. The entropy that you can fit in a box of gas depends on the volume of the box of gas.

0:07:31.8 SC: And in principle, the same thing would be true in quantum field theory in flat spacetime. But in gravity in particular, once you have gravity, something happens and it limits the entropy you can have in a region. Okay, so who cares about the entropy you can have in a region? Well, in quantum mechanics, the entropy in a region tells you something about how many degrees of freedom there are in that region, which is just to say how many different things can happen, how many physical ways can the system wiggle or change itself, right? That's what a degree of freedom is. So a degree of freedom might be a qubit, or it might be a particular way that a quantum field can oscillate, which again, is typically infinite in a regular quantum field theory. So the upper limit on the entropy in a region from the holographic principle seems to say that in every region of space, there's only a finite number of degrees of freedom, a finite number of things that can happen.

0:08:27.7 SC: A finite dimensional Hilbert space describing what's going on, if you want to get into the technical quantum mechanics of it all. So that means that the universe is not a quantum field theory in a very real sense, at least it seems not to be. It's always very hard in these games to draw very definitive conclusions, 'cause we don't know what the final complete answer is, right? But let's take that seriously. Let's take seriously that in any one region of space, in a theory of gravity, only a finite number of states can exist. Okay? That seems to be the straightforward reading of this holographic principle. And in fact, it tells you how many can exist proportional to the area of the boundary. So what does that mean for everyday physics is the question. And that's a very hard thing to answer because the answer is it's not clear.

0:09:16.5 SC: It doesn't immediately imply anything about any everyday physics. You have to ask why are there only a finite number of degrees of freedom? And what are those degrees of freedom, right? Those are questions that are not immediately answered by the holographic principles. So this kind of paper is taking a stab at modeling that, at figuring out, well, what would you mean by a finite number of degrees of freedom? What might they be? And in if that guess, if that model, if that hypothesis, if that conjecture is right, what would be the consequences for experiments? Okay, so how do you do that? How do you model this small number of degrees of freedom? Well, you could say, how many degrees of freedom would there be if quantum field theory were right? Right? Quantum field theory, there's an infinite number of degrees of freedom. That sounds bad.

0:10:07.0 SC: That's too many. So you start putting cutoffs, you say, all right, quantum field theory is a theory of fields. It's a theory of waves in some very real sense. Let's imagine that there's a minimum wavelength that we're going to consider. So instantly that helps you. That's called an ultraviolet cutoff, 'cause ultraviolet is small wavelength, right? There's a cutoff that says no wavelength is smaller than a certain amount. And if you have a region of space, then you can also say, no wavelength is larger than a certain amount. So good. Just by those two ideas, just those two constraints you're putting on your theory. You have a finite number of possible ways your quantum fields can wiggle. There's a largest wavelength and the smallest wavelength. Now, there's a technical distinction between bosons and fermions that we're gonna ignore here. Let's do fermions. Okay, let's think about electrons, neutrinos, things like that.

0:10:57.4 SC: So there's a finite number of wiggles, but if you do the counting, the number of ways that a quantum field theory can wiggle is still much larger than would be implied by the holographic principle. So somehow you have to get rid of some of those quantum field theory modes, as we call them, those ways that the quantum field theory can wiggle, because quantum field theory is telling you have more degrees of freedom than holography allows. So what we did, was use something that other, that mathematicians have noticed a long time ago, which is a fun mathematical fact. If you have a vector space in quantum mechanics, the space of states is always a vector space. So vector spaces have the following property. If you have an end dimensional vector space, let's say a three dimensional vector space, then you can only fit in three orthogonal vectors, three vectors that are completely perpendicular to each other, right?

0:11:54.9 SC: So there's only three-perpendicular directions in three-dimensional space. There's only two-perpendicular directions in two-dimensional space, etcetera. That's a very well known mathematical fact. Here is a less well known mathematical fact. You can try to put in vectors that are almost orthogonal to each other, and then you can fit in more than the dimensionality of space, right? If your vector space is two-dimensional, then I can fit in three vectors that are not 90 degrees apart, but they're 120 degrees apart. Well, that's not great. That's not exactly orthogonal to each other. But as you go from two to three to four to some huge number of dimensions of the vector space, it turns out there is a mathematical theorem that you can fit in a huge number more than the dimensionality of the space of almost orthogonal vectors, where they're really, really, really close to being orthogonal to each other.

0:12:50.3 SC: So that is what the overlapping degrees of freedom in the title of our paper refers to. It's the idea that we are suggesting that all of these huge number of modes of degrees of freedom that you would get in quantum field theory are actually not completely orthogonal to each other. They're not completely independent. They're not completely separate non-overlapping degrees of freedom. They overlap just a little bit. So the actual number of degrees of freedom in the region is much smaller than quantum field theory gives you. But you're tricked into thinking that quantum field theory is the right description 'cause there's a little tiny overlap between these states that would be completely separate, completely independent in quantum field theory. They're not quite in holography. That's our suggestion. Okay? And again, people have made suggestions like that before. What we did, what Oliver and the rest of us did, was to say, okay, let's get down and dirty with that.

0:13:53.5 SC: Let's write down equations of motion. Let's figure out how the actual propagation of a particle or a vibration in the quantum field across the universe might be affected by the fact that it has a tiny overlap with other kinds of quantum fields. And so we showed that if you have, let's say, a neutrino, and this is, this goes from these crazy high level ideas of quantum gravity down to observable things, which is the kind of thing I like to do. You have a neutrino traveling across the universe, and ordinarily a neutrino traveling across the universe can just be a neutrino. There's again, technicalities with neutrino oscillations and things like that. But just imagine you have a non oscillating neutrino. In our model, there's a tiny probability that that neutrino will sort of scatter itself without even bumming into anything. It will decay or scatter itself into other kinds of neutrinos moving in other directions, because there's a little bit of overlap, because they're not truly independent states.

0:15:00.5 SC: And so you can play the game of looking at whether or not that is compatible with current observations. And I'm not gonna go into all the details, I've already gone into more details than you wanna know, but you use some constraints from high energy physics, like the Large Hadron Collider, other constraints from neutrinos, from the sun, or from cosmic rays and things like that. And we find that it's very interesting. There is, in fact, a prediction that there should be a cutoff on the highest energy neutrinos from far, far away in the universe. And where that cutoff is predicted by us is exactly where the data become not good enough to actually see whether there are such neutrinos. So we're making a prediction that is right at the edge of observability. The neutrinos we're talking about are observed by the IceCube experiment in Antarctica.

0:15:51.9 SC: They look for neutrinos all over the sky, and then you have to do a lot of difficult analysis to figure out, are they from nearby or from far away and so forth. The highest energy neutrinos that have been detected by IceCube are just a little bit less energetic. Then what we predict should be the upper limit on the amount of energy from these neutrinos because of holographic effects. So, look, I'm gonna be very, very honest about this. This is a stab in the dark. This is the conjecture. This is a guess. We're saying, okay, there's this great idea called holography. Its implications for real particle physics are hard to pin down, but here's one way that it might show up and it is experimentally accessible. So we give some suggestions for where to look for it. I think it's a lot of fun.

0:16:36.4 SC: We'll see what other people think about it. That's what the paper's about. Sorry if it got too much into the weeds, but trust me, I did not get nearly as much into the weeds.

[laughter]

0:16:46.7 SC: As we could have gotten with a paper like that. Harry Zumwalt says, priority question, how does self-directed education via online resources outside of the university sphere fit into your picture of our future education landscape? So remember that all as, Patreon supporters get to ask once in their life a priority question. So usually I get more than twice as many AMA questions that I'm actually able to answer. So I can't answer all of them. But if there's something you really, really wanna know and you're a Patreon supporter, you can label it priority question. I will do my best to answer it in a sensible way. So Harry wants to know about self-directed education.

0:17:26.4 SC: Look, I'm a huge believer in all sorts of education. I'm always a huge believer in a multi-pronged approach to a difficult thing like education. You should go to college if you wanna do that, but you should also take online courses, you should read books, you should listen to lectures online, you should listen to podcasts. There's a million different ways that you can improve your education. Okay? So at the loosey-goosey level, I think that self-directed education is extremely useful. The fact that we can do it so much more easily now than ever before should be revolutionary in how people approach their post-college careers. Or even if you're in college and the college that you're at doesn't give you the course that you wanna get. Maybe it's online somewhere, right? There's very, very high level, very, very good courses from edX and Coursera and places like that, like MIT and Stanford and all these very, very good universities have offered a tremendous number of online content.

0:18:22.7 SC: So you should check it out. At the same time, it's not a replacement for regular college for a number of reasons. Number one, going to a college is a life shaping experience, right? Being in an environment where you can talk to people, where your peer group is learning kinds of the same kinds of things, where you're exposed to new ideas, where there are course requirements that force you to take certain courses that maybe you wouldn't have taken if it were up to you. But that turn out to be really interesting. These are all things that actually can have a huge impact on your life. I was forced to take philosophy courses as an undergraduate, and that had a huge impact on my life. Not to mention a course in, I don't know, solving differential equations that is kind of really necessary, but maybe not the sexiest thing that you want to do voluntarily.

0:19:14.6 SC: So I think it's all there. I think that formal education as well as informal education online as well as in-person education, all of these things are important, should be pursued by individuals, should be supported by institutions, but they're not replacements for each other. They're just different things that are important in different ways. Bob Toro says, particle accelerator detectors have to ignore data. Ignore data never leaves the detector and never gets recorded. How confident are we that there isn't any new physics in the ignored data, is fiddling with the triggers a thing? So this takes us back, but over 10 years ago, I wrote a book, right? The particle at the End of the Universe, which talks about the Higgs boson and the LHC. And in there, among other places, you can read about the need for triggers in modern high-energy physics experiments.

0:20:06.5 SC: The point is that there are way more events in a typical particle physics detector than can possibly be detected, can possibly be recorded in real-time, way more. So typically, if I'm remembering the numbers correctly, the number of events or the number of things that happen that you actually record to a hard disc in a particle physics detector is one in a million events. And that's actually not as bad as it sounds because most of the events are boring. Most of the events are just reminding you of good old fashioned physics that you already knew. They're not brand new in any way. So, but that does imply the existence of the need for a kind of art form, which is looking at the event really, really, really, really quickly in real-time and making a quick decision: Should we, is this a precious event that we should write down to tape or to disc or whatever?

0:21:05.1 SC: Or is it just a common thing that we can throw away? That's called the trigger: Should you trigger on this and record it or not? And yes, there's an enormous amount of work that goes into making the triggers be the right ones. As far as I know, fiddling with the triggers is not a thing. I mean, of course you kind of like update a little bit, a tiny bit, make sure you're on the right track and things like that. But I don't think that particle physicists generally, dramatically change the triggers from day to day or month to month thinking that they're gonna find something different. My impression as not an experimentalist, sorry about this, but my impression, someone who knows this better than I do can chime in, is that the strategy is to do the best you can, making the triggers good, and then let them run and then collect all the data you can because luminosity, the total amount of data that you can collect per second or whatever, is crucially important.

0:22:01.0 SC: So you don't wanna throw away potentially interesting data because you're just trying to be cautious and looking for collecting data you think is just the Standard Model doing its usual stuff over and over and over again. That does raise the possibility that we're wrong, right? That we're missing something very, very important. But that's why all of science is an interplay between theory and experiment, because the theorists have the job of figuring out, oh, you better look for this kind of thing. Here's the thing that you really should kind of look for. And the experimenters then have to go, okay, how do we look for that? How do we find this? What are the triggers? Or what are the kinds of detectors that we would build to do that? For very, very exotic kinds of physics, typically, you just need a different kind of experiment, for the kinds of new physics that is being looked at at the Large Hadron Collider, for example.

0:22:52.8 SC: We kind of know what we're looking for. And so I think that it's pretty reasonable to think that the triggers we've built are good ones and don't need to be fiddled with. But it always pays to be cautious in these things. Douglas Long says, my sister keeps insisting that we should listen to the MAGA crowd and find a common ground. I disagree in that. It seems we have gone separate ways. Do you think there's any future in engaging with them? I actually think this is a question that is super important in its most general consideration. It's easy to be overly simplistic about questions like this. It's easy to say yes, all of these people are completely past the bend, no point in talking to them. We just gotta fight the good fight for ourselves. It's also easy to say, you should always try to reach out and find common ground with other people.

0:23:43.7 SC: I think people are different. Even if you just say the MAGA crowd. For those of you who are not in the United States or 500 years in the future, that's the Make America Great Again crowd. The Donald Trump supporters that we have here in 2024, I think there's a spectrum of people. I think people are different. But more importantly, I do think that right now in the state of our democracy, we are finding it hard to do a very basic democratic thing, which is to persuade people to agree with us or to support the same policies that we do, right? You want to, in a democracy, have your favorite policies enacted. That means, have more people be in favor of them than not. So they will vote for representatives who will enact those policies. This is pretty basic stuff, but the actual discourse that happens online and elsewhere is almost never trying to reach across to people who disagree with you and persuading them.

0:24:47.0 SC: It's usually making fun of them, mocking them, being angry at them on both sides, right? I'm not saying this is one side or the other. There's just very little space right now for actually trying to find common ground. So the point is that I do think that finding common ground, or for that matter, just sticking to your ground, but persuading someone to join you on your ground, this is absolutely crucially important. At the same time, some people are not going to respond to it, and so you have to figure out who the ones are, who can respond to it, there are some people who just don't act in good faith, right? They're not actually, after figuring out and thinking about what the best policies are, they want to achieve a certain goal and are going to do whatever it takes, including taking advantage of your outreach to do that.

0:25:36.0 SC: So you have to figure out who's worth spending time on, but there must be people who are worth spending time on. Otherwise, your favorite policies cannot possibly win the day, at the end of the day in a democracy. Ron Gebbler says in the February 2024 AMA, you said that your view on locality in quantum mechanics is that it's surprising that physics looks local to us at all, since this is a consequence of the locality built into the Hamiltonian, which is well understood mathematically, your question seems to be, why does the Hamiltonian have locality built in? To me, this sounds like a rock bottom question to which I wouldn't expect any further answer. What kind of answer are you looking for? So, just to give a little bit of background, this is gonna be a technical question, I know, but for those of you who are not quantum mechanics experts, you might have heard that the fundamental dynamical equation in quantum mechanics is the Schrödinger equation.

0:26:27.0 SC: You have some quantum state defining a system. The Schrödinger equation tells you how that state evolves over time. It's the quantum equivalent of Isaac Newton's F=MA. So what the Schrödinger equation says is that the time change, the rate of change over time of the quantum state is driven by something called the Hamiltonian, which is basically a quantum mechanical object that asks how much energy is in different parts of the Wave Function. So essentially in quantum mechanics, the Hamiltonian simply is the laws of physics. Just like in Newtonian mechanics, if you tell me what all the forces are acting on an object, I can tell you how it's gonna move. In quantum mechanics, if you tell me the Hamiltonian of that system, I can tell you how it's going to evolve over time. Now, it is a feature as far as we can tell of the Hamiltonian of the world in which we live in that it is local, by which we mean the Hamiltonian, which is again, asking the question, how much energy is there in this quantum state, takes the form of at every point in space, asking what the energy is at that point, and then adding up all those points.

0:27:36.2 SC: And there are interactions, right? If you poke the quantum fields that make up the universe, the consequences of that poking vibrate out into the universe. But those consequences are local, which means that what is happening at one point interacts with what happens right at the next point, infinitesimally far away, but not immediately with some other point very, very far away. There is no true direct action at a distance in the Hamiltonian. Quantum measurements seem to feature spooky action at a distance, but that's an entirely separate question. We're not talking about measurements or the measurement problem here. So Rob is asking this question about why the Hamiltonian is local. This seems to be just a fact about the world, right? It's not a fact that you could search for an explanation for, it's just a fact that is at the bottom of the laws of physics as we understand them, what kind of answer could there possibly be?

0:28:32.8 SC: So that is entirely fair. But of course, we don't know the answer to this question. So the question now is, is the fact that the Hamiltonian of the universe has this locality property that it seems to be made up of individual contributions at all the different points in space and then adding them up? Is that something that we can explain or we just have to accept? We don't know the answer to that question. Maybe it's something we just have to accept, but maybe it's something we can explain. How could we possibly do that? Well, the minimal thing is that we can try to understand how easy it is or how hard it is or generic or rare or whatever for a Hamiltonian to be local. So is there always a way that you can sort of rearrange things, given any Hamiltonian so that it looks local from someone's perspective?

0:29:23.2 SC: I think the answer there is no, I think that that's actually a very special feature. Locality is a very special feature of the laws of physics as we know them. But that's a tricky question. I think you have to actually put some work into it. But then the other thing is, could you either within quantum mechanics or by generalizing quantum mechanics, find some reasons why the laws of physics would have evolved to this particular state? That's like one of these super ambitious crazy questions, right? It's very, very plausible that the answer is who knows? We don't know. Like, can you come up with a theory in which this happens? I don't know. You can try, right? So part of me wants to try, I wanna think about, I have some vague ideas. I'm not even gonna say what the ideas are here because they're too vague and I need to think about them myself.

0:30:12.9 SC: But it is a question that I'm interested in, given that the locality laws of physics as we perceive them seems to be non-generic, seems to be special, then it raises the question, is there some deeper thing going on that could help explain that? Whether it's anthropic or dynamical or statistical, whether it's within quantum mechanics or you need a generalized quantum mechanics. I don't know. Like I said, I have some vague ideas. Stay tuned.

[chuckle]

0:30:41.5 SC: I'm very slow. So stay tuned for years, okay? Don't come back two months from now and say, what's the answer? I'll let you know if I have any good ideas about this particular question. Okay. I'm gonna group two questions together. One is from Garth Brantley, and it says, is it possible to prove that quantum computing like speedup is impossible with a classical system, or is it proven?

0:31:07.1 SC: And then AJ says, can you describe the kinds of AMA questions that are better directed at ChatGPT or another large language model versus you? So the reason why I am grouping these two questions together, even though they're completely separate questions, is because when I read Garth's question, is it possible to prove that quantum computing like speedup is impossible with a classical system? One of my initial responses or my initial thoughts was I bet you can just ask ChatGPT that, right? Or one of these large language models, why not try that? So, of course, I tried it. It's an empirical question, right? And you all know that on the one hand, I'm presuming you all know that at the current state of the art here in March 2024, large language models are actually quite good at answering questions that some human being knows the answer to and is out there in the corpus of texts that all these LLMs have read.

0:31:58.5 SC: So at face value, even though it's a slightly technical question, this kind of question, is it possible to prove that quantum computers definitely can speedup questions over classical computers is right in the wheelhouse of what AI, large language model things can ask, can answer. So I turned to Claude, which is the new offering in the large language model gained from Anthropic AI. Anthropic is a highly regarded competitor to OpenAI, which is in charge of ChatGPT, etcetera. And the people who I know and who actually follow these things very carefully tell me that Claude is right now even better than OpenAI's GPT. So I asked Claude, I said, I basically asked this question, is it possible to prove, or I reworded it, but it was basically this question, are there provable speedups from quantum algorithms over classical algorithms? And I know the answer to this. So this is, it was not...

[laughter]

0:32:53.0 SC: The reason why I was a good test, it's because I do know the answer. The answer, by the way, Garth, for your question, is yes, it is possible to do that, but you have to be super careful about what exactly it is you've proven. So Claude comes back to me and says, yes, it is possible to prove that quantum computers speedup over classical computers. And then it gave me a list of examples, and the very first example was Shor's algorithm, sho Peter Shor, spelled S-H-O-R. There's no C or E anywhere in there. S-H-O-R, Peter Shor made waves back in the nineties by writing down an algorithm that can factor large numbers into their prime factors in a polynomial number of steps. So the big of all the things that you care about in quantum computers, the big target is taking something that takes an exponential number of steps.

0:33:44.4 SC: So if you have an N digit number, it would take E to the N or 2 to the N, something like that number of steps in order to solve this question, what is it? What are its prime factors? And what you want to do is reduce that to a polynomial. So N to some power rather than some number to the power N. When N is very, very large, polynomials are going to be much smaller than exponentials, even if it's N to the power 10 compared to 2 to the N, that's gonna be a very small number when N is a million or 10 to the 10 or something like that itself, right? So Shor's algorithm says, here you go, I'm giving you a polynomial algorithm for factoring large numbers. And as Claude correctly said, the fastest known classical algorithm is actually exponential, even that's not exactly right because it's sort of slightly sub-exponential, but more than polynomial for subtle mathematical reasons. But okay, but the answer was wrong.

[laughter]

0:34:42.0 SC: Because that is actually not provable. The question was, is it provable? And I put into my prompt, is it provable in the mathematical sense? And as often happens in these worlds of algorithms, we have not proven that the fastest classical algorithm is exponential or is even bigger than polynomial. Okay? So there is seemingly a large speedup offered by Shor's algorithm, but we haven't proven that it is a large speedup. So Claude was just wrong. Then it gave me a couple of other examples. The next one was Grover's algorithm. And Grover's algorithm is a search algorithm. So if you search a list of N entries, classically, guess what, you're gonna take of order N on average and over two. But there's no better way to search through a list than to just look, is it the first one?

0:35:38.2 SC: Is it the second one? Is it the third one? Whatever, you're looking for some particular answer. Whereas Grover's algorithm can do it in the square root of N, which is smaller than N when N gets very, very large. And so on the one hand, we have proven that there is real speedup for Grover's algorithm at the quantum level versus the best classical algorithm. And the reason why we can do that is because the classical algorithm is so simple, the classical problem has no structure in it. You have a list of numbers, it is unordered, search through it until you find a certain number, okay? Or a token or word or whatever. So it's actually possible to prove what the fastest classical algorithm is, unlike prime numbers and factoring where there's structure in there. And number theory is hard and it's hard to prove things.

0:36:27.7 SC: So for Shor's algorithm, for factoring large numbers, the speedup is large, but you haven't proven that is actually better than classical. For Grover's algorithm for search, the speedup is small because the original problem is not exponential. It's just n going down to the square root of N. But it is provable, okay? So after Claude gave me that answer, I said, are you sure about that? I'm really looking for things that are provably faster. And if you have played around with these large language models, you know that very, very often they'll give you a wrong answer. And then you say, are you sure? And it'll say, oh no, you're right. Here's the right answer. So Claude says, oh no, you're right. Actually, we have not proven any quantum algorithms are faster than classical algorithms. Thank you for the correction.

[laughter]

0:37:10.8 SC: That was also wrong because we have proven in the case of Grover's algorithm that it is faster. It's not exponentially faster, but it's faster. So I said it again, but no, I thought we had proven Grover's algorithm, and again, it's like, oh yes, you're right. I was wrong. So to answer AJ's questions, what are the kinds of AMA questions that are better directed at ChatGPT or another LLM? Right now, none of them.

[laughter]

0:37:35.3 SC: I still I've said it before, the current status of these large language models is kind of like a super good Wikipedia. It's super good in the sense it's actually less accurate than Wikipedia, but it is easier to find things like that. That's the real thing. Like in Wikipedia, maybe there's some fact and you just don't know where it is. Searching through might make it difficult or Googling, just in general, it might be difficult to find what you want. The nice thing about the large language models, you can just ask a directed question, it will give you an answer, but don't take the answer as true.

0:38:06.0 SC: You can use the tool very effectively by using the language the LLMs answer to then go search. So if Claude did correctly bring up Shor's algorithm and Grover's algorithm as important examples here, and then you could Google those or go to look at Wikipedia about those and learn for yourself whether they really are doing what the large language model tells you. So I think AJ, many questions that one gets asked during an AMA could be answered with five minutes of online work. I tend not to pick those in general, like if you're frustrated 'cause you're asking questions and I'm not picking them. If I think you could Google around, or ask around on the internet to get the answers, then I'm probably not gonna pick that one. Deepthi Amarasuriya says, would you please explain what happens to the Hubble Constant under the current accelerating expansion model of the universe?

0:38:58.4 SC: Yes, I think this is a frequently asked question. I'm sure I've answered this before, but I can say it once, more just to let everyone know. The constant, the Hubble Constant asymptotes to a constant value in the current accelerating expansion model of the universe. So according to Einstein's equation, you have the Friedmann equation that governs the expansion of the universe. And the Friedmann equation relates what the Hubble parameter does. The Hubble parameter's not really constant, it's the derivative of the scale factor divided by the scale factor itself. So it is potentially going to be fast or slow depending on what the rest of the universe is doing. The Friedman equation relates that Hubble parameter to the energy density of the universe in the early universe. In particular, it says that the Hubble parameter goes as the square root of the energy density.

0:39:43.7 SC: So in the early universe, the energy density is very high. Later in the universe, it's slow. The Hubble parameter is in fact decreasing over time. It's called the Hubble constant, because today its value is not changing very rapidly, right? So it's over a human lifetime. The Hubble parameter is nearly constant. But it's also true that if the universe is dominated by a cosmological constant, then the cosmological constants energy density is constant. That's the name. And the Hubble parameter goes as the square root of the energy density. So the Hubble parameter goes to a constant, and that counts as accelerating precisely because the Hubble parameter is not just the slope of the scale factor, it's not dAdT where A is the scale factor, and T is time, and that's the derivative. It's dAdT divided by A. So if you take a dot over A and set that to a constant, the solution is A, goes as E to the T exponential expansion.

0:40:39.6 SC: Richard Burgess says, the question I've thought about a lot but have never seen discussed is what the relativity of simultaneity would imply about how a materialist theory of consciousness would work? Now, a brain is an extended object, meaning different observers will have different conceptions of what the configuration of your brain is at give at a given point in time. To me, that would imply your consciousness would be the locus of causally connected timelike separated events making up the brain converging at a single point as opposed to say, some non-local field thing. You know, I don't think, this is actually a very hard question. I get it, it's sort of a clever question because the idea being that before relativity came along, you knew what it meant to say an extended object like the brain at one moment of time, because everyone agreed on what it meant to say at one moment of time.

0:41:27.8 SC: After relativity comes along, different observers or different coordinate systems will parse the phrase at one moment of time differently. So to someone moving near the speed of light, the collection of atoms and neurons and so forth in your brain at a single moment of time is a slightly different collection than the ones to someone who is moving stationary compared to you. But the fact is that at a practical level, this is completely irrelevant. It is only relevant... Let's put it this way. Think about the speed of light and how long it takes light to traverse the distance that is your brain. It's a very tiny amount of time.

[laughter]

0:41:39.7 SC: It's a very tiny amount of time. So what that means is that the difference between being in one reference frame or another to how you describe what is the brain at that moment of time is completely negligible. The timescales of a brain are actually quite slow, right? Milliseconds at the fastest, and a millisecond is much, much longer in time than the time it takes light to travel across your brain. So the differences in what you call the brain from one observer to another as far as relativity is concerned, are not that different. Now, I do think that there's something interesting and important going on in what you, how you think about a materialist theory of consciousness and personal identity. Because I do think that if you're a human like myself, if you're sort of bare bones, all the world is, is a bunch of things happening. And then at an emergent higher level, there are things like people and consciousnesses and brains and so forth. You have to be a little bit careful because there isn't any essence. There isn't any, here's the true consciousness locus sitting in your brain somewhere, okay?

0:43:14.7 SC: It's an approximate description of a higher level thing. And so those approximate descriptions are always going to be fuzzy when you try to push them a little bit too hard. This particular relativistic fuzziness is actually one of the easiest to account for. Okay? I'm gonna group two questions together. Abanish Nala says, what is the physical mechanism for the second law of thermodynamics? I understand that appropriate coarse-grain dynamics imply the second law, but why are we allowed to perform such coarse grainings? And then GS says, one part of the concept of entropy that has always confused me is that there seems to be some human interpretation/subjectivity involved in determining whether a particular state of a system is high or low entropy, is the concept of maximum entropy of a system of fundamental property, of a system, but the current entropy of a system, a matter of human interpretation? Well, it's in between.

0:44:03.2 SC: It's in between being just a human interpretation and being a fundamental property. So you have to, in this game, accept the existence of things that are not completely set by the fundamental laws of physics, but also not arbitrary. That idea that that space in between is absolutely central if you want to allow yourself to use higher level emerging concepts. And there's nothing special about entropy here. I'm talking to you from a little table in my office, and here's a table, and it's made of atoms, right? And we all agree that it's a table and it's made of atoms, but there's a famous paradox. I always forget how to pronounce it, sorites or something like that. What if I remove one atom from the table? Is it still the table? Well, yes, but if I move enough atoms, it stops being the table.

0:44:53.9 SC: And there's no bright line there, okay? There's no point at which I removed enough atoms to say, now it's not a table anymore. And yet the concept of a table is perfectly useful to us in our everyday lives. We know what you mean when you look at it. And entropy is like that. It's a matter of usefulness. So to define entropy, you need to coarse-grain somehow, or alternatively, you need to have some probability distribution over micro states or different ways of doing it. But the point is, in either case, the human need for something convenient comes in either because I don't know the exact state, I only know what probability or I coarse-grain and consider a lot of different possible configurations of atoms and molecules to be macroscopically indistinguishable from each other. And if you didn't do that, if you didn't need to do that, if you said, I'm just gonna follow the micro state, I'm gonna follow what every single atom and molecule and particle and quantum field does, that's great.

0:45:52.4 SC: Then you are Laplace's demon and you can predict what's gonna happen in the future, good for you. But you're not Laplace's demon, nobody is. You don't have access to that information. So you coarse-grain, you say, well, I don't know. When I look at this cup of water, I don't know what exactly the position and momentum of every molecule of water is. What do I know? What can I see? What can I measure? Well, I can see where the water is, I can see the shape it takes. I can see I can measure its temperature. I could measure its density and things like that. There are various things that I have access to in the real world, and that's not just a matter of my flaws, right? That's a matter of what I am as a physical system and what physical manipulations I have access to. Okay? So it's ultimately down to the laws of physics that determine what I, as a macroscopic being, can reasonably measure about this system. And then the very obvious natural and commonly used way of course-graining microscopic states is to say two microscopic states are in the same coarse-grained macro state if they macroscopically look the same, if they have the same temperature and volume and things like that.

0:47:07.0 SC: Okay? So the point is that the coarse graining is something that is done by human beings, but it is not done arbitrarily. It is done for good reasons because we coarse-grain with respect to the actual things that we have access to, and it is with respect to those coarse graining that we have access to that the entropy of the early universe was very, very small and entropy has been going up ever since. Patofix says, what are your views on the prescription of pure or impure placebos? I recently read an article about placebo which changed my mind about it. The article concluded that it is not moral for a doctor of medicine to prescribe a placebo as one, it can generate distrust in medical professionals, two, it hinders the ability of the patient to make an educated decision, three, it neglects the consent of patients, and four, it increases the paternalistic behavior from medical professionals. Well, I don't know this for sure, but I don't think that doctors do prescribe placebos. I think when you think about placebos and tests on the efficaciousness of placebos, those aren't done by random doctors giving you prescriptions. Those are studies being done in laboratories and medical schools where the subjects of the study consent to being in the study.

0:48:27.8 SC: And part of that consent is they know they may or may not be given a placebo or some equivalent piece of knowledge. So I would certainly be very, very much against doctors telling you that they're giving you a useful medicine for your illness and instead giving you a placebo, but I don't actually think that that happens. Diane Russell says, what forms of social science, if any, do you read or follow? What do you find worthwhile in social science? Well, many things, I would say look at the Mindscape podcast. I talked to sociologist John Skrentny who was just recently on, was a sociologist. I talked to economists. I talked to psychologists. I talked to all sorts of social scientists. Look, I think that that's a very simple and glib answer, but the reason why I wanted to answer this question is because I think almost any academic field has things that are extremely worthwhile in it. Many academic fields, maybe all of them, also have some things that are less worthwhile, but it's just a huge mistake to dismiss entire fields of intellectual activity because you don't like parts of them if that's actually true.

0:49:35.5 SC: I think that it is the reader's job, the consumer's job, let's put it that way, to judge whether certain aspects of a different kind of academic discipline are interesting and useful or not. The social sciences are much harder than the natural sciences because people are complicated, because you can't just give them placebos, you have to actually do controlled trials, which is very difficult and there's a lot of noise in the system, so completely unsurprising that the actual results one gets in the social sciences are less definite, less counterintuitive, less reproducible than those in the natural sciences. But I'm a person, social sciences are the sciences of people, I care about people, I care a lot about their psychology and their sociology and their economics and so forth, so I think that doing the social sciences is extremely worthwhile for just about all of them. Dave Grundgeiger says, I've heard it said that space-time isn't fundamental but emerges from the Wave Function. I've also heard it said that the Wave Function evolves over time implying that time is fundamental. Does that mean that space and space-time are emergent but the time is fundamental? Well, I want to be very careful about the phrase I've heard it said.

0:50:49.5 SC: You know, I've said things like this but I've said things like this in the context of being speculative about the fundamental nature of space and time. These are all things we don't know about. So again, the reader, the consumer, has to beware here when a scientist or philosopher or thinker of any sort is talking about speculative ideas at the edge of knowledge. You can't just translate, well, maybe this might be true, to I've heard it said that this is true. Okay, you got to understand that there are hypotheses, there are conjectures, that's how science goes forward. So we don't know any of these things, we don't even know the wave functions are fundamental. But we have an open mind. So the Schrödinger equation of quantum mechanics has time as a fundamental parameter in it. That's what it does. It tells you how the Wave Function evolves over time. It does not have space as fundamental in it. Therefore there is certainly a point of view that says that time is fundamental, but space is not. So space emerges from the Wave Function in that kind of picture, and then it's perfectly fair in that kind of picture to say that space-time emerges from the Wave Function, 'cause there wasn't any space-time. That is fundamental. Time was, but space-time wasn't. Or maybe not, right?

0:52:10.2 SC: It is absolutely possible that both space and time are emergent, or that neither are. I personally think that space being emergent is almost true, like it's almost guaranteed. Not quite, because again, we don't know, and we're hypothesizing, we don't know the final answer to these questions, but the evidence that space is not fundamental seems pretty convincing to me. Others don't think that at all. Tim Maudlin, who I recently had on the podcast, is an example of someone who would laugh at the idea that space is not fundamental. But all the options are on the table, okay? It's possible space is fundamental and time is not, time is fundamental and space isn't, neither is, both are, we just don't know. Mark Kummeri says, priority question, it seems that most theoretical physicists believe in eternalism, the block universe, as it is a natural consequence of special relativity. An analogy that is often given is to think of space-time as a DVD where all the events are fixed even though the appearance of time flowing as the DVD is played. This always troubled me as it seems to imply that my birth, my death, and my decision to make this a priority question couldn't have been any different as the past, present, and future are all fixed in the block universe.

0:53:19.6 SC: But might the many worlds interpretation of quantum mechanics allow for a way out of this severe limitation on free will? What if the right analogy is not a DVD but rather a choose your adventure story whereby all the branches of the many worlds reside in the block universe and our consciousness allows us to choose among these branches? So no, I don't think that that is actually a very good way of thinking about it. The block universe, by the way, it is true that some physicists or some people think of eternalism or the block universe as because of special relativity, because when special relativity comes along there now is no natural way, no objective way to divide space-time into space and time. There just is space-time. I would personally say that even before special relativity, the block universe, the eternalism viewpoint is the most sensible one just because the laws of physics don't pick out anything special from moment to moment. Laplace's demon in classical physics says that once you know all the information about the universe at any one time, you know about it at all the other times also.

0:54:26.0 SC: So that's just classical physics for you. And I don't think, I've said this many, many times, I think that you shouldn't worry about free will because of that. As Mark says in the question, it seems to imply that my birth, etcetera, couldn't have been any different. What does that mean? You have to be very, very clear about what you mean by couldn't have been any different. Of course, they could have been different. If the initial conditions were different, then they would have been different. And of course, they could have been different relative to what you actually know. Again, you are not Laplace's demon. You don't know what exactly the universe is going to do. So relative to your knowledge, they absolutely could have been different. So for all intents and purposes, your life, your decisions, absolutely could have been different. And the block universe has nothing to do with that. Now about quantum mechanics, the tricky thing here is that it does look like if the universe is branching and there's different possibilities, that now there's some freedom there that wasn't there before. But that's completely an illusion. There's no more freedom there because it is not true that you're choosing your own adventure.

0:55:33.4 SC: It is not true that you are in any sense deciding what branch of the Wave Function you will end up with. In the many worlds interpretation of quantum mechanics, the fundamental dynamics are just as deterministic as they are in Newtonian mechanics. Even though there will be several branches in the future, we can determine precisely what the branches are going to be and they will be there with 100% probability. There will be a future version of you that's all the spin up and a different version of you that's all the spin down. There's no freedom there, there's no choice, there's no extra volition that comes in over and above ordinary classical mechanics.

0:56:14.3 SC: Adam Rotmil says, how much more complex do you think the universe can get before equilibrium? Well, that's a loaded question. We don't have a simple once and for all metric or measure or way of characterizing the complexity of the universe. Of course, we can sort of approximate it in different ways or we can have proxies, I should say, for it that may or may not be useful in ways. But I think the crucial thing that makes this a difficult question is, do you judge the complexity of the universe by something like the average complexity over many parts of the universe, or do you judge it by the maximum complexity? So let's say that we human beings here on Earth are the pinnacle of complexity of the universe. And it's possible that there is no other life, no other advanced intelligent life anywhere in our observable universe. It's also possible that we will destroy ourselves here on Earth, right? So maybe the peak maximum complexity is right now, right? Me and you and the other listeners of Mindscape and the other people out there in the world who are not listening to Mindscape, we collectively make up the maximum complexity the universe will ever see. That is possible, okay? It's possible. I mean, it's even plausible that in that sense, the world is more complex now than it was a billion years ago.

0:57:38.5 SC: It's also possible that that's the right way to think about it, but we will not destroy ourselves, that we will generate a lot more complexity in the future. Then if you want to say, you know, how much more complex could human beings, how much more complexity could human beings create or give rise to? I think the short answer is a lot more, even if you only thought about the existing human beings. We're individually complex, but the structures that we fit into are not that much more complex than our individual selves, if at all. So, you know, if we became one hive mind with eight billion individuals making one collective consciousness, I bet that under most metrics that would be much more complex than any individual human being right now, right? I think would have to be. So, we're nowhere near that, and that's only with just eight billion human beings here on Earth. That's nothing to say if we colonize the galaxy or something like that. So, there's plenty more room in the future for much, much, much more complexity than we have.

0:58:42.3 SC: On the other hand, if you would prefer to think of complexity in terms of something more like the average complexity of the universe rather than just looking at the special points where things are very complicated, then the picture is a little bit more grim. You know, the one objective fact that I can bring forward is that star formation is mostly done. I mentioned this in the solo podcast, The Holiday Message About Immortality. Most of the stars that will ever be created in the universe have already been created. The rate of star formation has been decreasing for a while now. So, if that's a proxy for complexity, which I could see you wanting to make that effort, then we're already fading, right? You know, most of the stars have already been made. We're making a few more, but also stars are dying out, so we're in a little bit of a race. Are stars dying faster than they're being made? They're probably right now being made faster than they're dying, but that won't last forever. So that's a different point of view that gets you a different answer to this question. I think you need to be more specific in asking a question like this before you can a very reliable answer.

0:59:49.5 SC: Malta Ubel says, I was surprised that the solo podcasts on AI Thinks Different had the word emergence in it zero times. Surprised not only because of the frequent occurrences on the podcast in general, but also because given the simple design of transformers, essentially all behaviors we are seeing are emergent. Do you think there's a path of further emergent behavior such as having a model of the world that can appear within current AI architecture? I am not the person to ask what can and cannot appear within current AI architecture. As I said in the solo podcast, the thing that AI, that large language models are very, very good at is sounding human. And I think that the evidence is that they are sounding human not by mimicking the way that human beings think, but by other tricks that they use. Okay. I thought, as I said in the podcast, I think it's possible to imagine that we just sort of tune and tweak the current large language model paradigm so well that the only way for it to sound human is to think like a human being. Okay. I think that is conceivable that that could happen. I don't see any evidence that it is happening right now. And I don't think, honestly, that we're very close to it.

1:01:05.0 SC: So I do think that it is, you know, I'm a physicalist at the end of the day, a materialist about consciousness. I think that there's no obstacle to creating computers that think like human beings and act like human beings. I just don't think that the architecture of large language models and predictive text processing is the right way to get there. So I suspect that if and when that happens, it will be through a different kind of technique. Justin Wolcott says, is it our best guess that A, there are some universal laws of physics applicable to all universes in the multiverse? B, any universe in the multiverse can have any laws of physics, even stuff that's impossible in ours] C, we don't have enough reason to believe one way or another, or D, other? Well, of course we don't know. That's the short answer. We don't know much about the multiverse, if it exists, what kind of multiverse it is. The simplest thing is certainly to imagine that the underlying laws of physics are the same in all the universes, if there is a multiverse out there. They might show up differently. But we think of that as sort of different phases, just like solid ice and liquid water and water vapor are different phases of water.

1:02:10.8 SC: There can be different phases of space-time itself that give rise to different local laws of physics. The cosmological multiverse often in scenarios takes advantage of this. That's why you can have different constants of nature and things like that, but it's the same underlying laws. In the simplest versions of the many worlds interpretation of quantum mechanics, the laws are exactly the same in all the universes. It might be a little bit more subtle in more advanced versions, but again, the true answer is we don't know. Michael Wall says, in your conversation with Christoph Adami, I was struck by his example of a human submerged in water. Humans and other animals have many adaptations to being submerged in water, and I would argue that the information content of a human in water system is not much different from a human in air system, at least for a little while. Does an organism with more adaptations intrinsically contain more information, or is it only relative to the moment-to-moment situation? I think that Chris Adami's idea is that the total human genome is actually very highly attuned to the environmental niche in which human beings find themselves. We are attuned to eat certain kinds of food, breathe certain kinds of air, flourish in certain kinds of temperatures, avoid certain kinds of predators, find certain kinds of resources and food and things like that.

1:03:32.5 SC: All of that is information contained in our genome. We can go underwater for a little while, but not for too long before we're gonna die, because our genome does not contain the information that would build an organism, a human body, that would survive underwater for very long. So that's the kind of thing that he means, that the information contained in our genome doesn't just give us enough information, give us enough specificity in our bodies to last moment to moment, but also to anticipate future problems and things like that, because when our ancestors didn't anticipate those problems, they died. So all the complications of a human being in this way of thinking are there in the genome and it's chosen to be that kind of information because of what our environment is and because of our desire to adapt to it. Even such things as what we're allergic to and what we can eat and so forth. So the one thing that Chris emphasized very strongly in his view of information is that it is only ever relative to some other situation. It's information about something. The exact same string of symbols can have a lot of information if you have the capacity to interpret those symbols and learn something about the outside world. If there is no way to do that interpretation then for all intents and purposes there is no information in those symbols.

1:05:03.4 SC: Tomer Hakohen says, what happens when you put a black hole in a thermal bath of photons? My intuition is that if the temperature of the bath is higher than the Hawking temperature then the energy going in will cause the black hole to increase in size thus decreasing its Hawking temperature and increasing the flux in ad infinitum. This seems kind of weird to me. Well, okay, it seems weird to you but it is exactly what happens. The weird thing is not that you can increase your size by accreting energy and matter from the universe. Not only black holes do that but lots of things could do that in principle. A planet or a star could accrete matter from the outside world. The difference is of course, the planet or star could also lose matter to the outside world whereas, at least classically, the black hole is a one-way street. The black hole does quantum mechanically have a Hawking temperature and it's giving off radiation and the interesting thing about black holes is this fact that when the black hole gets bigger it gets colder. So technically we say that that means the black hole has a negative heat capacity. If you have a chunk of iron and it just remains the same size but you shine photons on it to increase its thermal energy, it will get hotter. Right?

1:06:16.0 SC: Most things when you shine energy on them get hotter. Black holes go the other way, negative heat capacity. You shine light on them and they get colder. They get bigger, more massive but that actually gets, implies a lower temperature but there's nothing super weird about that. It's a little bit surprising at first but gravity has all sorts of weirdnesses about it. It's kind of magical that gravity works as well as it does. This is just one of those examples. Anonymous says there's a meme going around social media where people are asked to consider which they would find more surprising, hearing a knock at the door and finding a fairy or hearing a knock at the door and finding a walrus? Some people argue that fairies being real changes everything they know about the world which makes the fairy more surprising. Others explain that so many improbable things are implied by walrus showing up at their house that the fairy is less surprising even with implications that magic is real. Which one would you be more surprised by and why? I'm a little surprised that this is controversial.

1:07:16.3 SC: I'd be more surprised by the fairy showing up because my credence that fairies exist is very, very, very small. But I think it's kind of a cool question because it is trying to balance two small numbers, two low probabilities, two low credences, which is a notoriously difficult thing to do. And in some sense, it's an intuition pump for certain ways we have of thinking about probabilities. In the case of a walrus knocking on your front door, you know that could happen. You know exactly the kinds of steps that would be involved in it happening. A walrus somehow getting to your front door, decide knocking on it, all these various things, and you know that they're all individually quite unlikely. So despite the fact that you know that it's possible for this to happen, you can even see how it might happen. Just because all of those individual possibilities are small, you end up giving it a very small credence. For the fairy, you have no idea how that would happen, really. In the real world, let's face it, you don't know that fairies exist. So if they did exist, if they existed in the good old magical fairy tale sense of being a fairy, your intuition is out the window for how likely that is or it isn't.

1:08:34.8 SC: Your overall credence that the whole story is right of their existing fairies, etcetera, that might be very small, but it doesn't get operationalized as the product of a lot of small probabilities. So I think you can sort of trick yourself into thinking, well, I only need to buy one weird thing in the fairy story, whereas I need to buy several weird unlikely things in the walrus story. I do believe at the end of the day that the many weird things you have to believe in the walrus story are still much, much more likely overall than the one big weird thing you have to believe in the fairy story. But I get that it's a difficult thing to estimate. Rad Antonov says, is there anything in the laws of physics that precludes the existence of black holes with masses comparable to those of asteroids, about 10 to the power 20 grams? If not forbidden, can you think of mechanisms for their formation? No, there's nothing absolutely that forbids black holes that big. The Planck scale, the Planck mass is roughly speaking what you would think is the smallest possible black hole. And the mass of the universe is the largest possible black hole. So 10 to the 20 grams is definitely in between there. If I'm remembering correctly, the Planck mass is about 10 to the minus five grams. So 10 to the 20 is much bigger. Everywhere in there, you can imagine black holes.

1:09:52.5 SC: But the interesting thing is black holes are kind of hard to make. And that's because matter doesn't want to be squeezed in to such a small region of space. Black holes are inevitable if you squeeze an enormous amount of matter into a small region of space. That was the result of the singularity theorems proven by Penrose, I should say former Mindscape guest Roger Penrose, and Stephen Hawking and Bob Gorosh and other people. But actually getting that much matter into that much space turns out to be hard in our actual universe. This is one of the reasons why I and others are very skeptical about Lee Smolin's idea, Lee Smolin, yet another former Mindscape guest. He has an idea of cosmological natural selection where black holes lead to baby universes with slightly different laws of physics and therefore the universe becomes tuned to make as many black holes as it can. To most of us we look around and the universe is just obviously not tuned to make as many black holes as it can.

1:10:53.9 SC: It's kind of hard to make a black hole. You need a star exploding or many, many, many stars tumbling into each other or something otherwise dramatic. So for a asteroid sized black hole, the actual size, the physical size of an asteroid sized black hole, asteroid mass black hole would be incredibly tiny, right? I mean, I don't know exactly what it would be but less than a millimeter across. Getting the mass of an asteroid into that much size is a very difficult thing to do. So the short answer is there's no obviously easy way to make black holes of that size. There's a big lacuna in this reasoning which is that in the early universe maybe a whole bunch of things happened that made a lot of black holes. There are scenarios for making primordial black holes that people think about and they're basically unconstrained.

1:11:44.0 SC: So other than, don't make so many black holes, you would have seen them already, you can imagine black holes of any size. So yeah, you could imagine it's very very speculative but you can imagine in the early universe making a whole bunch of black holes about the masses of asteroids. Craig Vandervest says, in a recent podcast you mentioned music. What do you think people enjoy, sorry, why do you think people enjoy and respond emotionally to music? What evolutionary advantage could this possibly have provided given there were no pianos, violins or electric guitars in the wild? You know, I actually have done podcast episodes on this. Indre Viskontas back in episode 54, I'm looking at the webpage right now and David Rosen and Scott Miles back in episode 104, they all talked about neuroscience and the brain and its relationship to music.

1:12:36.1 SC: Indre in particular was talking about what music does to our brain. David and Scott were talking more about creativity, but actually which reminds me, like I've been toying with the idea maybe I should bring into the podcast feed some classic episodes. Like what if in addition not replacing, but in addition to the regular Monday ones like on Thursdays I released a favorite old episode or something like that, re-released it. They're already there. You can get them right now. It's not hard, but maybe it's hard to find them so I could do something like that. I don't know. It's just an idea. I would have to dig out the files and probably I would have to try to clean up some of the audio to maintain my current standards which are higher than they were when I started out, but that could be a fun thing to do. Anyway, you know the point is that electric guitars are spandrels. Spandrel is the term that was borrowed from architecture into evolutionary biology by I think Lewontin and Gould back in the day when they pointed out that if you think that the point of natural selection is to increase the fitness of organisms, but natural selection works in certain ways, right? It takes what it already has and mutates it in slight ways, small ways, and that's how it ends up where it wants to be. It doesn't see ahead of time what would work.

1:13:56.9 SC: It just starts from a starting point and looks at different possible small variations. And as a result, in order to get where it goes, accidents sometimes happen or you get tuned to one thing, you think of some certain signal or some state of affairs as good and you continue to think of it as good even if the context is completely different. So, rhythm, melody, harmony, things like this, I can imagine that in the wild those were considered good for all sorts of various reasons even though musical instruments and formally written down songs did not exist back then. After all, we human beings are not the only animals to do music, right? The birds outside my window are definitely very interested in musical sounds, not to mention packs of wolves howling in unison and things like that. So whether it's communication or just plain having a good time, I can easily imagine our former ancestral selves enjoying music and that growing into the music that we know and love today. Nikola Ivanov says, in your solo episode discussing the emergence of space from the Wave Function of the universe, you emphasize the importance of the concept of locality. Discussing the example of the Schrödinger cat, if I understood you correctly, you indicated that the superposition of an asleep and awake cat is a superposition of a spatially coherent configuration of states.

1:15:27.7 SC: My question is about the expression, spatially coherent. Do you have a hypothesis of why entangled quantum states can decohere only in spatially coherent configurations? Yes, I have more than a hypothesis. It's actually this is something that is pretty well understood on the basis of work by people like Wojciech Zurek and others through the 1980s and beyond and it's a pretty simple thing to understand. But the critical point is to understand that in quantum mechanics some interactions between things lead to entanglement, others do not. So entanglement is not just interaction. It's a certain kind of interaction. It's the kind of interaction where different parts of the quantum Wave Function of one system interact differently with parts of the quantum Wave Function of another system. So in the case of the cat, if you imagine the cat is in a superposition of awake and asleep and by that we mean the physical cat is doing two different things. The awake cat is standing up and walking around, the asleep cat is lying on the ground. Okay? Then a photon in the room, in the box or wherever the cat is, can interact differently with those two parts of the cat. It could be absorbed by the cat that's awake, for example, and just pass right on by. The cat's asleep because they are physically in different locations. That's where the locality comes in.

1:16:52.8 SC: The photon interacts with the cat when the photon and the cat are in the same part of the universe, the same location in space. And so what happens is, once you confine your attention to a branch of the Wave Function where the cat is spatially coherent, that is to say it is not in a superposition of being in different locations. It has one obvious consistent spatial location, then it can interact with photons, but it doesn't become entangled with those photons because all the photons interact with the cat in the same way once the cat is spatially coherent. So the point is that the laws of physics have this property of locality, as we talked about earlier in the podcast, and that picks out certain kinds of states to be what Zurek calls the pointer states. The states that kind of look classical, the states that the Wave Function apparently collapses onto, the states that you and I actually run across in our everyday lives. Cats in definite positions, not superpositions of different locations in space. Fran Pla says according, in re-social media following your advice I switched from Twitter to Bluesky. But on the other hand your fans also appreciate and enjoy very much the Instagram accounts Sean M. Carroll and Ariel and Caliban where we see personal things like the food you cook, the kittens and wonderful photos of Baltimore and fun things like Lord Ravens scaring Halloween kids, etcetera.

1:18:29.0 SC: So my question is, what is the Papermoon Diner right next to Hopkins? Do you recommend that if we ever visit Baltimore? So what Fran Pla is pointing out here is that I do have Instagram accounts, in fact, I have one for myself and there is another one Jennifer and I run jointly behind the scenes under the name of Ariel and Caliban. Our cats, of course and To be super honest We are not that active, like if I post to Instagram once a month, it's a miracle. It's less than that. But we use it for different purposes. We don't use Instagram for professional purposes, I'm not giving science tips or talking about quantum mechanics on Instagram. I'm just posting pictures of my life or whatever it is in sort of ways that are kind of touristy and average and nothing very special about it. Lots of cat pictures, of course, so we had to make once we got Ariel & Caliban We quickly had to make a separate Instagram account, otherwise 98% of our own individual Instagram accounts would have been pictures of the cats. So if you want more pictures of Ariel and Caliban, then you get as Patreon supporters. That's the other benefit of being a Patreon supporter, of course, you get extra pictures of the kitties for every Ask Me Anything episode.

1:19:40.0 SC: But if you want even more you can get them on Instagram under Ariel and Caliban. So the question involves the Papermoon Diner, which is a diner very close to Johns Hopkins, which is I put a picture of it, a couple pictures on Instagram. It's this crazy over-the-top kitschy Baltimore diner. I absolutely recommend it if you like, you know colorful settings and fun. It's actually hard to get into like they don't, I don't think they take reservations. Maybe it depends on the size. But anyway, if you go there early enough for off hours, it's easy enough to get into but you know, the food is diner food.

1:20:15.0 SC: It's fine. It's perfectly good. You get a perfectly reasonable lunch there, but the it's decorated in all sorts of ways, in crazy colors. I don't even want to try to describe it. I wouldn't do a good job. If you're that interested and if you google Papermoon Diner in Baltimore, and you'll see what we're talking about. Jared Sage says in an interview with Fields medalist Maryna Viazovska mentioned a promising young teacher mathematician she knew who was killed in a missile strike on Kharkiv during the Ukraine invasion. She said when someone like her dies, it's like the future dies a line I think about a lot as a never ready and I often wonder about possible futures and with current events I've lately found myself thinking about all the futures we've lost, particularly at the hands of injustice. There is a genuine grief I experience when I imagine what we as scientists lost when Alan Turing's castration led to his suicide. When Agnes Pockels was denied a formal physics education as a woman, when we failed children ad nauseam.

1:21:16.5 SC: I do not expect this grief to wane anytime soon, but my question to you is this, do you think there's any utility in mourning the loss of possible futures especially in contexts of injustice like these? I do. I do think that there is utility in doing that. I don't really think this has anything to do with quantum mechanics or many worlds or whatever, the possible futures that are relevant here are hypothetical possible futures. Things that could have happened and they're just as hypothetically imaginable in a single classical universe as they are in a quantum many worlds context, but this is precisely where ideas of loss and grief come to be real and important because our imaginations are, we can imagine situations that are different and better than the reality that we face. In particular we can imagine a future that could have arisen out of a certain past But didn't for various tragic reasons. And I think that you know, it makes us sad to imagine what we've lost in these circumstances and you might say to yourself, well, it just makes us sad. It's not actually useful to do that, let's buck up and you know pick ourselves up and keep going. I don't think it's quite that simple, honestly, I think that there is a purpose in regretting these kinds of losses for the simple reason that it makes us more motivated to prevent them in the future, you know.

1:22:41.9 SC: I think that rather than brushing off these tragic incidents, we should take them seriously not let them drive us to inaction or paralysis but use them to motivate us to prevent those that kinds of things from happening in similar circumstances in the future. You know, it's like a pain reflex when you put your hand on a hot stove. That is bad. It's unpleasant, but it's also educational. It teaches you not to do that anymore, hopefully, and I think that in a slightly more deep sense the same thing is true about loss and regret and grief that they are Teaching us something that hopefully, we can use to make the future better. Mark Foskey says whenever people give examples of events that would cause the world to bifurcate it's always a textbook measurement like measuring the spin of an electron.

1:23:31.8 SC: But isn't every time two air molecules collide a kind of measurement on that scale? Well, yes and no. Not, there are degrees here so for a couple of reasons, number one, the reason why we always talk about the spin of an electron is because the number of possible measurement outcomes equals two whereas if you're talking about the position of an atom or an electron, then the number of possible measurement outcomes equals infinity, and just conceptually it's harder to wrap your mind around how to talk about infinity, especially if you want to know like how many worlds are there, and the right answer to that question is there's no right answer to that question in situations where Hilbert space is truly infinite dimensional and there really are an infinite number of possible measurement outcomes.

1:24:17.8 SC: It's exactly the same kind of question as how many numbers are there between zero and one? The answer is an infinite number, same thing for the number of worlds. But that doesn't stop you from relating the relative number of worlds with certain features versus other features. Just like in a given measure, it's perfectly sensible to say that there are one tenth as much space between zero and one as there is between zero and 10. Even though there's an infinite number of numbers in both cases, likewise in many worlds you can say there is a certain number of worlds where the electron is here versus the electron being there to within some error bars. So that's the big complication. For the specific example of two air molecules colliding, mostly those air molecules are behaving classically. Okay, unless you really, really kind of set up the experiment carefully so that they hit right on and have a spherical wave or something like that. If you have two air molecules hitting each other at an angle with a certain momentum you can kind of use classical mechanics to do a pretty good prediction.

1:25:25.8 SC: It's not a measurement in the direct quantum mechanical sense. Remember, a measurement is when a quantum mechanical system that is in a superposition becomes entangled with its environment. That's what a measurement is. When you have just two air molecules, the environment isn't there. It's just not being relevant here. You have to think a little bit more carefully. I don't mean to gloss the question under the table, thinking more carefully is actually hard for these kinds of things, but it's not a deep kind of challenge to our understanding of many worlds or to quantum mechanics. James Swift says, I recently heard Frank Wilczek discussing time on The Joy of Why podcast when he mentioned various arrows other than time, in particular an arrow of radiation.

1:26:08.9 SC: I never heard this term before and I would love it if you could give a short explanation of what that even is. So I didn't hear the episode, but Frank Wilczek, former Mindscape guest was on The Joy of Why podcast, The Joy of Why is a podcast hosted by Quantum Magazine and it has, it used to have one host, Steven Strogatz, former Mindscape guest. Now it has two hosts. There's a co-host Janna Levin, who's a former Mindscape guest. You see there's really not that many human beings in the world, and many of them have already been on Mindscape. So you see the same names appearing over and over again. The point is, I suspect that these arrows are all arrows of time, but they're different arrows of time. So it's not that there's the arrow of time and there are other arrows, is that there's the thermodynamic arrow of time, that's the direction in which entropy is increasing, but there are also other arrows of time.

1:27:01.7 SC: At least many people talk that way. Personally, I don't think that talk is very useful. I think it's really all the same. There are different versions of the thermodynamic arrow of time. So the arrow of radiation is simply that if I take an electron and I shake it, there's a solution to Maxwell's equation that says electromagnetic waves propagate out into the future. But you can guess, if you really understand that the laws of physics are time reversal and variant, there is also a solution to Maxwell's equation where electromagnetic waves were coming in, focusing in on that shaking electron. They were coming in from the past and they were precisely designed so that they would be absorbed by the electron as it moved. And no radiation is emitted to the future. That is a perfectly valid solution to Maxwell's equations in electromagnetism, and we never see it in the real world.

1:28:01.0 SC: Why? Well, because, just like I said, you would have to imagine that incoming radiation was precisely designed to cancel out the outgoing radiation that we would ordinarily get. We don't think that that is very likely. Why? Well, because there's an arrow of time, because we think that we know something about the past and we call that thing we know the past hypothesis, the idea there was a low entropy past, and that doesn't include very, very finely oriented radiation that cancels out the radiation you would ordinarily get from shaking electrons. We don't put any analogous boundary condition in the future, so there's no problem with the radiation moving to the future. So I think that that electromagnetic arrow of time is exactly the same as the thermodynamic arrow of time, but for some reason people like to make a big deal out of it. I think that if you really understood the arrow, you would describe all of these different consequences of it in a unified framework.

1:29:02.2 SC: Edward A. Morris says in your discussion with Christoph Adami about information in biology, I guess he must have had something like Kolmogorov complexity metric in mind because he said that a genome with redundant copies of the same gene doesn't really contain any more information than if it only had one. But is there any place for such a complexity metric in the context of physics and statistical mechanics where we'd like to say there's a negative correlation between the amount of information in the system and the amount of entropy? Or are we always only talking about the amount of raw information in those contexts regardless of how complex versus compressible that raw information might be? So I think there's a lot of things that need to be unpacked in this question. Edward, you say, where we'd like to say there's a negative correlation between the amount of information in a system and the amount of entropy, well, would we like to say that really, that is assuming a very specific notion of the word information, and the word information, like the word entropy, is actually one that has different definitions in different contexts.

1:30:05.5 SC: And that's fine. You can't just say, my definition is the right one. You have to be careful about what context you are in and what meaning you are supposed to be attaching to these words. You can't just argue over what the right one is. When we have a Boltzmann idea of entropy, that is to say, we think about entropy as coarse-graining and say there's a lot of micro states that go into a macro state, and the entropy is the logarithm of the number of micro states in the macro state we are in. Then there is a very real sense in which low entropy means high information that senses that if the state is low entropy, there's not that many micro states that look that way. That gives us a lot of information about what kind of state the system is in. Whereas if it's a high entropy state, there are many, many micro states that look that way.

1:30:56.9 SC: So in that sense, we have very little information about what specific micro state we're in, but that notion of information is just not at all what Christoph Adami had in mind. As you said, he has something more in mind like the incompressible amount of information because he is thinking biologically, he wants to have information that is useful for some specific biological purpose. That's why he doesn't care that much about mutual information, mutual information being the amount of information in one system that tells you about another system, because for him, that's the only kind of information. It's sort of ironic that he doesn't care about it that much because to him, what he means by information is the mutual information between two systems. That's why he says that the context is always very, very important. And so as thinking biologically, if the information you care about is information that helps you react to the world or predict what you're going to see or survive in a hostile environment, then taking exactly the same information and just duplicating it doesn't help you at all.

1:32:05.5 SC: That doesn't add to your store of useful techniques for surviving in the world. So yeah, he's using information in a different sense that's perfectly valid. You just have to ask people who are being ambiguous, what definition of information they have in mind. Siddhartha says, could you outline what it would mean for time to be emergent instead of being fundamental perhaps by contrasting universes where time is one or the other? Isn't time somehow fundamentally necessary for any sort of change to happen? Yeah, time is, well, you stuck the word fundamentally in there. Time is necessary for any sort of change to happen. Yes, but that doesn't tell us whether that time is fundamental or emergent. Fundamental in this sense means an idea is part of the most basic, most comprehensive description of nature at its deepest levels. Emergent means that this whatever concept we're talking about appears in some approximate course-grained higher level description.

1:33:07.1 SC: So it might be true. Many people think it is true that the deepest layer of reality does not need time as a fundamental concept, but maybe time is emergent. So maybe in, so to me, and different people think about this differently, so don't necessarily attribute my thoughts on this to anyone else. To me, quantum mechanics makes this kind of idea at least interesting, as opposed to just a waste of time. The reason why is because in quantum mechanics, you can take two different situations, two different states, and you can make a superposition of both of them. In classical mechanics, you can't do that. The, particle is at some location is not anywhere else. There's no such thing as a superposition of those locations. So in quantum mechanics, I can take, I can say, well, I can imagine a clock embedded in a bigger system, and there's a correlation, when the clock reads something, then the system is doing something, okay, and I can just let time flow and the clock reads different readouts, and the system is doing different things. But then I can just take time slices. I can say, here's what that system did at 10:00 o'clock and 10:01 and 10:02, and I can take those specific states, which don't have any time dependence in them. Okay?

1:34:24.0 SC: Okay, so I'm just taking a state at a moment of time and then I make a superposition of all those states. So I'm making a state that is not itself evolving in time. I'm inventing new laws of physics. The new laws of physics are, time doesn't pass, but I'm constructing a quantum state in which time doesn't pass. But that quantum state is a superposition of different clock readings and different things going on in the universe. In that quantum state, you can say there's a correlation, there's entanglement between the state of the clock and the state of the rest of the universe such that it looks like the rest of the universe is evolving in time where time is what the clock says. So that is a way that time could emerge out of a quantum mechanical description. And this is something that has been explored.

1:35:12.7 SC: I didn't just make that up. This is an existing idea out there. I don't think it's completely understood, well developed, anything like that. I think it's a speculative idea that people like to think about is worth taking seriously. But we don't know. This is one of those things we have to be careful about and explore before declaring victory. Soonest Mended says, I really enjoyed your wine episode from way back. Do you have any reasonably priced $20 go-to reds that you just keep on hand for a regular Friday night? We don't actually have, this is of course the classic question. When we had Matthew Luczy on the podcast to talk about wine way back when we did ask him precisely this question and he said, yeah, that's what everyone wants to know. I mean, maybe today it'd be a $30 bottle. I think that when I was your age, $20 bottles were the target.

1:36:03.8 SC: Things have gotten more expensive since then, but I don't have one... I'm not the kind of person who just wants to drink the same kind of bottle over and over again. Honestly, that's one of the pleasures of wine is it's not like drinking soft drinks where you're drinking Coke or Pepsi or whatever. For wines, you could have a different kind of wine everyday for your whole life, really. So what I tend to do is pick a good wine store. That's the crucial thing. If you have a good wine store, then you can figure out that, oh, they carry certain kinds of wine that you really like. The other thing to keep in mind is, of course, different people like different wines. So I tend to like reds. Some people like whites or roses. I tend to like French wines.

1:36:41.5 SC: Bordeauxs are my favorite. I also sometimes like certain kinds of Italians and Californians and Riojas from Spain and things like that. But my go-to is a good French Bordeaux. And one of the reasons why I like them the best is because they keep their age better. An older Bordeaux is more likely to be good than an older California cabernet or something like that. And I like that aged taste. Of course, the older they are for a good Bordeaux, it's harder to get them in the $20 or $30 price range, this is where the good wine store comes in. If you have a good wine store, you can rely that they're going to find good wines in that price range, in that category. So, and it doesn't need to be like some exquisitely charming mom and pop store. We have one right near where we live, the Remington Bottle for those of you who live in Baltimore, I can recommend that.

1:37:33.9 SC: But there's also Total Wine, which is a national chain that we have here in Towson that I always used to go to in LA 'cause there was one in Pasadena also. Now there's one in Baltimore and well, there was one in Baltimore. Now we're in Baltimore and it's great. Total Wine is great 'cause it has a huge selection and they're very, very good about describing the wine. So it's not just like okay, here's a wine from Côte-Rôtie or whatever. They will give you the flavors, okay? They will try to describe it. They will tell you if it won any prizes, things like that. Is it dry? Is it more sweet? That's what you gotta do. You gotta find a good wine seller, learn to trust them, and then figure out what are the $20 bottles that keep you happy. Chris Murray says, in the latest mindscape, Matt Strassler mentioned multiple times that laser light is made of photons.

1:38:27.7 SC: But I've seen an experiment showing interference pattern with laser light between two paths of vastly different lengths, even when the light is attenuated to what should be just one photon at a time. What's the right way to think about these laser photons? So this interference is possible. Yeah, this is called quantum mechanics. So that's the whole point of the double-slit experiment is if you could imagine doing this double-slit experiment with electrons, which by the way is very hard to do because electrons are charged particles and they tend to interact with things, but you can do it. And even though electrons are individual particles, they're not lasers, right? They don't pile on top of each other like photons do to make a classical electromagnetic wave. You still see interference patterns 'cause those electrons have a Wave Function and you're seeing the interference pattern in that Wave Function.

1:39:16.6 SC: Exactly the same thing is true for individual photons. They're no longer describable as classical electromagnetic waves, but they are describable as individual quantum particles with a Wave Function. That is what is leading to the interference pattern that you can very readily see. Anonymous says, I've heard that gravity could end up being possibly derived from entropy and would not be fundamental. Could you explain how that would happen? Well, yes and no. I mean, I can explain some aspects of it. I've written papers about it, Grant Remmen and I wrote a paper called What is the Entropy in Entropic Gravity. I can't do a very good job of explaining the whole thing, but I have one go-to analogy that I'd use to describe the idea of an Entropic force. So an Entropic force is an idea that predates any specific application to gravity, but it's not a very common idea, but it's a contrast to a mechanical force.

1:40:13.4 SC: So you imagine an oscillator, let's say you imagine a spring that is attached to a wall and there is a weight on the other end of the spring that is moving on a frictionless surface, okay? Typical physics setup. Now in that kind of setup, there will be a place where the weight can just sit still, right? The equilibrium resting configuration of the spring. And if you push the spring together, so you compress it, there'll be a force pushing in the other direction. Also, if you pull the spring away from its equilibrium there'll be a force pulling it back, and if you let it go, it will oscillate back and forth. So that's a mechanical force because it's truly the mechanical operation of the spring that is doing the job. For an entropic force, imagine that instead of a spring, you have a very, very lightweight chain.

1:41:04.6 SC: So you have little links of chain that are connecting the weight to the wall, such that ordinarily at zero temperature, the chain would just be lying on the floor and there's no force acting on the box at all, acting on the weight. But now in your mind, you imagine heating it up so you really heat it up, you make it very, very hot. Nothing dissolves or anything like that, but the chain is so lightweight that now in the hot environment it's bouncing around, right? It is jiggling forward and backward depending on what the temperature is. So it's no longer just lying on the floor. What you notice in that case is that if you try to push the weight towards the floor, there are more ways for the chain to push back, right? The chain wants to jiggle and it doesn't want to be squeezed into a little ball, so it will push back and exert a force back on the weight.

1:42:00.5 SC: Likewise, if you pull the weight so that the chain is straight, the chain will want to pull back because it wants to be in a certain configuration where there's at least a little room where it can jiggle. So there is once again an equilibrium location for the weight that depends on the entropy. The equilibrium location is where the entropy of the chain is maximal. So this is not a mechanical force, this is not a spring pushing or pulling. This is the chain wants to have the largest number of degrees of freedom, the largest number of places it can wiggle, room to wiggle where it's neither confined to near the wall, nor stretched into a straight line. And that is called an entropic force acting on the weight. So the idea is that gravity is an entropic force. This is an idea that was popularized by Eric Verlinde.

1:42:52.9 SC: It's very similar to previous ideas by Ted Jacobson and other people. And it's basically suggesting that there is some set of degrees of freedom underlying gravity with the property that they're pushing and pulling things in order to try to go to their maximum entropy configuration. Now, what those degrees of freedom are, utterly mysterious. We don't know what that is, but one of the nice things about thermodynamics is that you don't need to know what the atoms are to know the basic ideas of temperature and energy and heat and things like that. So maybe we can figure out what the basic ideas of gravity are without knowing what the underlying degrees of freedom and their entropy really comes from.

1:43:34.7 SC: Michel Pickle says, do you view entropy as correlated with the arrow of time or as a cause? I view it as a cause. I think, I forget whether this is a future question or whether I didn't answer it or not, but sometimes people say, is entropy time? And that's not right. Like time exists just like space exists. Time has an arrow. The arrow of time is a property of time in our world, and space exists without any arrow of space. So time is not the arrow of time, the arrow is a feature that time has, that it happens to have in our world. It doesn't necessarily happen to have it in the space of all possible worlds. So the origin of the arrow of time is the fact that the matter of the universe is evolving in a certain way, in irreversible way from the macroscopic point of view and an irreversible way such that entropy is increasing. That increase of entropy.

1:44:36.7 SC: It gives time its arrow. So I would say it is the cause of what we call the arrow of time. Keith says, a central premise of panpsychism arguments against physicalism is along the lines that physics is only the business of telling us what an electron does, not what an electron is. My question is, what do you see as the strongest arguments against this panpsychic premise that physics is gated by telling us only what the stuff does or more generally, what do you think of this is/does distinction? Well, I think that it's a bad distinction, honestly. So the way that I would respond to this argument, physics is only in the business of telling us what electron does not what an electron is would be to just be careful about what the presumptions are, okay? So when you make that argument, you make that statement that physics is not in the job of telling us what an electron is.

1:45:33.1 SC: You are presuming that there is something called what the electron is that is not simply a restatement of what it does. That might be true, but that might not be true also. So it completely begs the question of whether or not there is anything over and above what the electron does to simply say, you're only telling me what the electron does, not what it is. It's absolutely a valid perspective to say that's because that's all there is to say about the electron. Once you've said what the physical universe does, you're finished describing the physical universe. Now maybe there's more you would like to say, and maybe that's perfectly legitimate. Maybe you would like to describe it in certain ways that are helpful to you, that are algorithmically useful, that describe some emergent higher level properties. That's all fine, but that doesn't mean that you have to use any of those extra terms to describe it.

1:46:30.7 SC: So this is entirely compatible with my perspective as a human that what the universe is is a bunch of things that happen. In other words, it's nothing more or less than what the universe does. Rick Dewitt asks a priority question. He says, I am fractal faculty at The Physics Monastery in Logan, Utah, residing near the Salish Sea. It is a moral imperative that we talk. My priority question is this, when should I fly to Baltimore to appear on Mindscape for the scientific purpose of improving natural philosophy? Since 2013, I have quietly developed a new model of time and gravity that emerges consistent with both general relativity and quantum mechanics. So the answer is never. You should not fly to Baltimore to appear on Mindscape. I am glad that you have developed a new model of time and gravity that emerges consistent with both general relativity and quantum mechanics.

1:47:27.0 SC: And if you're at all serious, write a paper and publish it. That's what you should do, not worry about going on podcasts. So there, I've saved you the plane ticket and increased the likelihood that your theory will gain scientific respectability, submit it to a journal and see what happens. Steve Welton says, non-expert questions. Some of your recent podcast guests have touched on the possibility we might not have evidence of prior civilization sufficiently back in time, e.g. Due to physical processes. This got me wondering about a civilization forming in the far far future where the galaxies are incredibly far apart. Would this hypothetical advanced civilization, say at our technological level, have any evidence that the universe itself is anything more than the surrounding galaxy cluster? Well, if you go far enough in the future, I think the answer would be no. In fact, the galaxy cluster that we're in, the set of gravitationally bound galaxies will merge into one kind of big galaxy before too long.

1:48:27.7 SC: And so they will think that there is just this one galaxy and the rest of the universe is empty. In principle, there's still light being given off that they could detect from other galaxies that gets more and more redshifted. When we say that galaxies or other things leave our observable universe, it's very much like falling into a black hole. They cross a horizon, but in fact, the light from them, just something going into a black hole or something crossing the cosmological horizon, from our perspective, it just looks like they slow down and become more redshift. However, any actual technologically advanced civilization will have a limit on what is the longest wavelength of radiation that they can possibly detect. And ultimately, there'll be a point where there is no more radiation from those galaxies left to detect. Now they might get lucky because there's more than one way to keep information consistent through time.

1:49:24.6 SC: You don't need to have it be in the form of radiation. It could be in a book. So their ancestors might have written a book saying, oh look, there's all these galaxies in the sky, or taken photographs or whatever, and passed those down in much more mundane ways. But I think what you're getting at is yes, if you wait long enough, a new civilization that didn't have ancestors that told them anything about the universe could find itself in a situation where it didn't know what happened back in our fun part of the history of the cosmos. Jay says, I'm really enjoying The Biggest Ideas in the Universe. At the end of chapter three, you introduce least action and the Lagrangian, which is kinetic energy minus potential energy. This is a very interesting alternative to Newton, but it's not as intuitive. Could you give us some insight, intuition on why it's kinetic energy minus potential energy?

1:50:15.6 SC: Yeah, so for those of you who have read or will soon read Volume 1 of The Biggest Ideas in the Universe, we talk about something called the principle of least action. The action is the integral. So the sum over different moments of time of something called the Lagrangian, which is made out of the energies of the system, but it's the kinetic energy minus the potential energy. So it's not conserved because kinetic energy can turn back into potential energy and vice versa. Kinetic energy plus potential energy is conserved, but the Lagrangian is something different. And there's this remarkable property that if you take the Lagrangian integrate it over time, then if you have boundary conditions where the system starts at some location at some time and ends at some location at some time, then the path that it will legitimately take between those two starting points and ending points minimizes the action over the space of all possible paths that it could have taken.

1:51:15.2 SC: It sounds a bit like precognition, right? Like how did it know what is the minimum action path? Of course, you can show mathematically it's completely equivalent to ordinary Newtonian physics. So of course, it didn't know where it was supposed to go. The whole point of the least action principles, is there's a future boundary condition that it's aiming to. So it's kind of a global view of the dynamics of a system rather than a local view. The specific question here is why is the Lagrangian and the kinetic energy minus the potential energy? Honestly, I'm not gonna do a very good job of explaining why that's what it is. It works. That's the thing that ultimately matters. In the Biggest Ideas, I try my best to give you an intuition for why it works. So I don't remember exactly what I said, but one nice thing to think about is just think of a ball rolling on a hill.

1:52:05.0 SC: So there's some potential energy, the height of the hill, there's no friction or anything like that. If the so-called hill is just flat, if it's a flat plateau, the ball can just sit there. The energy is perfectly well conserved, but we know that if the hill is tilted, if there's a slope, then the ball could not just sit there. Now that's interesting because if the ball did just sit there, energy would be conserved, right? But we know that if it's on a tilted hill, it will start rolling down or it was rolling up. Those are the two possibilities, or it... There's a third possibility that it could just be at the peak of its motion and then start rolling down again. But let's imagine the following boundary conditions that at some moment of time the ball is at one location. Okay? We don't, in this way of thinking, give the momentum.

1:52:54.3 SC: We just tell you the location of the ball is at a certain point, and we say that at some other point later, it is at the same location. Okay? Those are the boundary conditions for the beginning and the end, and the whole potential is just a straight line with some tilt, with some slope. And you might say, well, I'm confused because the ball should start rolling down the hill and it will never come back up if it's just tilted downward. But you're forgetting that I didn't tell you the momentum. So I'm not saying that you start the ball stationary at that point, maybe the ball was rolling up the hill. So indeed there is a solution to the equations where the ball rolls up the hill. It always was, it was starting with some velocity. You just didn't say so in the specification of the initial conditions.

1:53:39.7 SC: So it rolls up the hill, reaches a maximum and rolls back down so that it's at the same point, same location and space at the future time that you've mentioned. So think about this in terms of kinetic energy minus potential energy. If you're trying to minimize kinetic energy minus potential energy, well, first you wanna minimize the kinetic energy that sounds like stay still, okay, don't move too fast. So it would seem like even though you're on a tilted hill, maybe just staying at the same point is good. That minimizes the kinetic energy, but it doesn't minimize minus the potential energy, minus the potential energy being minimized means you maximize the potential energy. So in some sense, this is saying that the ball wants to be up there at a higher potential energy. So rather than just stay at one physical location, you could minimize minus the potential energy by making the ball zip up to a much, much higher potential energy, very, very quickly, stay up there and then zip back down.

1:54:42.2 SC: But of course, that would not minimize the kinetic energy. So to minimize the kinetic energy and to maximize the potential energy at the same time, or at least to compromise between those two things, what the ball actually does is roll up the hill gently, reach a maximum and then roll back down. That's the best I can do. I don't think it's very satisfying myself, but this is a law of nature. Who's to say that our intuition should line up with it? Elif and Lucas ask a question together. If we understand correctly, gravity is emergent from some kind of entangled something. So our question is this, if gravity is emergent from some kind of entangled something, would you expect the other forces to also be emergent in the same way? So the short answer is yes. Again, going back to an earlier question that we addressed, we don't know whether gravity is emergent from some entangled something, that is an idea that is being hypothetically pursued.

1:55:37.9 SC: Okay? But maybe it is, maybe it's not. But in that picture, especially in my version of it, where it's sort of maximally emergent, that is to say the fundamental thing from which gravity is emerging doesn't look like space-time or quantum fields at all. It's just some Wave Function of a quantum state. Then everything has to be emergent. Certainly, the other forces, the other fields, the other particles, all of that is emergent. I put less effort personally into thinking about the ways in which it could be immersion, but that is absolutely the idea. Yes. Jonathan asks, what is your favorite sauce for chicken wings? Hmm. That's a very good question, but I think the immediate answer is the classic buffalo sauce is my favorite. If I order chicken wings, which I do, which I've been known to do, yeah, Buffalo sauce is the best.

1:56:26.0 SC: It's spicy. I like the spicy food. It's not too spicy, like a real... Someone who came from a culture which truly valued real spices would not be impressed with buffalo wing sauce. It's more vinegarette than chili based, right? It's not based on chili peppers that much, but that's okay. I like it. I think it works well, especially with the blue cheese dressing. The classic, what I believe is the original version of buffalo wings, all in favor of being creative with different kinds of sauces, as long as they don't go in the typical American direction of making them all goopy and sweet and corn syrup-y, like why would you do that? If it's something based in spices or even herbs or garlic or whatever, then that's great, but don't, yeah, put some goopy sauce on your chicken wings. That's just a little bit too American for my tastes.

1:57:20.0 SC: Igor Kapelov says, do you think artificial general intelligence has a coherent definition in the context of human intelligence? You've mentioned a few times that we shouldn't think of smartness as a general thing that people have more or less of. Is there a similar mistake in how people talk about AGI? Yeah, I absolutely think that that is a mistake. In fact, it's a much worse mistake because it's taking something that wasn't even true about human beings to say there's something called general intelligence that is a useful concept. And then to extending it to machines that are completely different than us, as if like there's something called intelligence out there in the world that we all share of, and we're teaching the machines how to partake in it in some way, that is entirely different from what is actually happening. And I think that it is part of the tendency to anthropomorphize these machines more than we should, and the corresponding tendency to think of them as more human than they really are.

1:58:15.4 SC: They work differently, they're gonna behave differently, and we need to take that seriously. Blake Brasher says, do you think there is a moral responsibility for individuals to abstain from using social media if they think the social media company is causing harm? I'm not one who thinks there are many moral responsibilities along those lines, and in particular, I don't think it is usually helpful to assign moral responsibility to individuals in their participation in broader structures. Well, is it morally irresponsible to drive a car that uses gas or to fly in an airplane that burns fuel or things like that? I don't think that's the right way of thinking about it. Like maybe you can make an argument one way or another, but if you want fewer fossil fuels to be burned, change the system, change the incentives so that the companies that are making and flying and driving these vehicles have some incentive to give you options that are less harmful to the environment.

1:59:20.0 SC: Likewise, I'm not going to blame any individual for using a social media app just because they think the social media app is causing harm. You individually might feel better about yourself if you think that the company is causing harm by abstaining from participating in whatever that company does. This has nothing specifically to do with social media companies. However, this could be true for any company, and I think that you would find that it's going to be very, very difficult to be consistent about this. If you don't want to get yourself involved with any company that causes harm, that's gonna be a very difficult way to live. Not to say that you shouldn't, but it is difficult. Just to warn against picking and choosing where we're going to place our outrage. I personally, like I said, I've not stopped using Twitter entirely, which I suspect this question has something to do with, but I basically only use it for promotional purposes.

2:00:24.7 SC: I mentioned when there's a blog or I'm giving a talk or I have a book coming out or whatever, I have my conversations on Bluesky, which is just a lot more fun for me to do. And also, yes, I feel good about the fact that I am not contributing to a social media company that I do think is causing harm, but I'm not judging other people for making other decisions about that. Kyle Kabisarez, sorry, Kyle Kabisarez says, I think I've heard on occasion your NBA team is the 76ers. So who was your favorite player growing up? My money would be on Dr. J, but curious to know your answer for sure. Oh, yeah, that's a very easy guess. And you guessed correctly that it was Dr. J. So when I was 10 years old was when Dr. J came over from the Nets to the 76ers, so perfectly timed for his peak years as a basketball player in Philadelphia to be my sort of formative childhood years in terms of watching sports and things like that. And Dr. J was both immensely entertaining to watch and super successful. I think that people tend to forget how many games they won. They only won one championship, so he's sort of not given quite as much credit as he deserves. But in the middle years, in the early years when he first got there, they had a lot of star power that just didn't play well together; in the middle years, late in the '70s, early '80s, they didn't have that many other good players. Dr. J took them to the NBA finals in 1980 with no other all stars on the team.

2:02:01.1 SC: Finally, they got some other good players, or they had young players like Maurice Cheeks who really grew into their roles and Andrew Tony and so forth. And then they won the championship when Moses Malone came over in 1983. But Dr. J was a good role model. He was wonderful to watch on the court. The team was always competitive. That was an easy choice. I was not like clever or iconoclastic enough to pick Bobby Jones or Mo Cheeks or whoever to be my favorite player. I just went with the obvious choice. David Maxwell says, DeepMind co-founder, Mustafa Suleyman's, recent popular book, The Coming Wave places synthetic biology as the other part of The Coming Wave of transformational technological development along with AI. What do you reckon? I'm not gonna go into detail here because I vaguely have plans to do a whole podcast about vague predictions about future technological developments, but I certainly think that synthetic biology is going to be part of it.

2:02:54.3 SC: The idea that we can design organisms to do what we want. Yeah, that's gonna be absolutely massive. Overall biological transformations I think will be at least as important as computer science ones. Michael says, I have heard that for some of Einstein's discoveries, if Einstein had not made the discovery someone else would have, eventually. I'm wondering if, for example, Edward Witten hadn't come along, do you think someone else would've made Witten's mathematical discoveries? Yeah, absolutely. I think this is true for any scientist. Like that's the thing about science. You can't... It's absolutely sensible to give people credit when they are the ones who actually did make the discoveries, but it's just wrong to think that no one else would have made them because they're all studying the same underlying nature. Some people might understand it sooner or slower than other people, but eventually we would do it and I think that's also true for mathematics as well as for physics and other sciences.

2:03:53.0 SC: Larry Rossi says, one possible explanation for Fermi's paradox is that we are truly alone in the galaxy. If that's the case, is it morally right for us to attempt to colonize the galaxy as broadly and as fast as possible since intelligent life is rare and maybe even unique? Even if we colonize via robots and not biologically as, that would certainly be faster. I don't think that there's any moral obligation. I think you mean moral obligation more than morally right, or moral direction, I guess, to go fast, I'm not a utilitarian. I do not think that there is some number called utility that we are here to maximize. And I don't think that that number, even if it existed, would be like the sum of all experiences that were experienced by conscious creatures or anything like that.

2:04:45.6 SC: I think that's just not a good extrapolation away from true human experience. And furthermore, in general, in principles of broad scale futurism kinds of questions, I think it's almost always much more sensible to think short term than to think long term. Short term here might be a few hundred years [laughter] rather than a few hundred thousand years, because what do we know about what life will be like by the time that we have the technological ability to colonize the galaxy? Like this is fine for science fiction, speculation kinds of things, but I certainly am not actually motivated here in the real world to drop everything and colonize the galaxy. I think that there's far too many unanswered questions between now and then to talk about that in any reliable way. Natalie Lyons says, is it possible to have an arrow of time in a universe without a beginning?

2:05:38.4 SC: Is it even possible to have a universe that has been around forever that hasn't reached equilibrium? So yes and yes. In fact, this is basically the idea at the heart of the scenario that I put forward with Jennifer Chen back in 2004. It is precisely a universe that doesn't have equilibrium. There is no equilibrium state to be in, as a result of that generically if you imagine such a universe, so the analogy we used is similar to what we just said with a principle of least action. Imagine a ball on a hill. The hill is a straight line with some slope, and there is no bottom. Okay, so now we're not talking about least action anymore. We're just talking about the space of all possible trajectories for a ball on a hill where the hill is a straight line with some slope and no bottom.

2:06:28.0 SC: They all look the same. Every trajectory looks the same, which is at T equals minus infinity. The ball was very, very far away, and it was moving up the hill. At T equals plus infinity, the ball is very, very far away, and it's moving down the hill. And at some point in between, the ball reaches a turning point with zero velocity and goes from moving up the hill to moving down the hill. So it goes forward, stops, and then goes back. And that's the only thing that it can do for all of eternity. Okay? And the idea there is that in that space of trajectories, even though that that space of trajectories is not what we call normalizable in the physics literature, you can't actually put a well-defined probability distribution on it. But the idea is who cares? Because they all look the same.

2:07:15.0 SC: So you don't need to care about which specific trajectory it is. They all have this feature of a turning point, one specific moment with zero velocity, and it was moving upward in the past and moving downward in the future. So what if entropy is like that? What if there is no maximum entropy state, but you just let the universe roll forever, then you will have the same kind of thing In the past, entropy was decreasing from our point of view. Of course, from the point of view of someone who lived in the far past, it was increasing, but they define time in the opposite direction from what we do. But the point is that the curve of entropy versus time is very high, goes down and then goes back up again, just like the curve of X versus time in the ball rolling on the hill.

2:08:03.0 SC: So that scenario purports to explain why we observe an arrow of time, because almost all points in history have a very strong arrow of time, just like almost all points in the history of the ball rolling on the hill have a non-zero velocity. Many, many details remain to be worked out about this kind of scenario, but yes, I do think it is possible. Chris K. Says, I heard a quote of yours lately about life in the universe, and it was something like if intelligent life existed, we would expect it to be most likely to be either in zero places or everywhere, all over the universe. It seemed like your implication was that this means there likely isn't intelligent life out there, but doesn't the existence of humanity mean that the option is not zero? The answer already is more than zero. If we are observing from an unbiased lens instead of a human focused lens. Well, I'm all in favor of observing from an unbiased lens, but we are biased by the fact that we are human beings.

2:08:58.6 SC: We are selected by the fact that we would not be having this discussion if we didn't exist, okay? Therefore, I would argue that our existence is essentially zero information other than the fact that the laws of physics allow for us to exist. It tells us almost nothing about the probability of us existing. In particular, imagine that the universe is much, much, much, much, much larger than the observable part of the universe. That's very plausible. In fact, it's almost likely in some set of possibilities. And imagine that the probability of intelligent life, such as ourselves existing in any one billion light year sized patch of the universe is very small. Imagine that it's like 10 to the minus a hundred, but there's so many patches that it's bound to happen at some point, okay? And we just happen to be in that point. Well, then we'd be having this discussion and we'd be talking about, oh, look, we are here.

2:10:00.8 SC: Therefore, it's evidence that life is not that unlikely. But we'd be wrong about that because we're biased because we're already here. That's old evidence. You cannot use our existence to update your credences on how likely it is that life can exist. So sure, the probability of life existing is not zero, that I will buy, but the probability of life coming into existence on any one planet could easily be 10 to the minus a hundred or 10 to the minus a thousand or something like that. We don't really have any evidence one way or the other about that. Jonathan Peretz says, can you help me understand the black hole information laws paradox better? It seems to me that for observers outside the black hole, the book that falls in never gets past the horizon. And observers in falling with the book, or those that are behind the horizon are free to read the book until they run out of time.

2:10:49.6 SC: So why do we say information is lost? Well, the story you're just telling there is the story of a classical black hole, not one that is radiating and evaporating, and there is no information loss paradox in a classical black hole. The information loss paradox comes because once you believe in Hawking radiation, black holes radiate out to the universe, they shrink and they disappear. So there's some moment in time in the future where there is no more black hole. It is only the radiation that it turned into. And the question is, how can the information from the book get into that radiation? I hope I said that correctly. How does the information in the book get into the radiation that comes outside? And once you go into the details of the size of the black hole and where the book is, and what it's doing, etcetera, things like the quantum no-cloning theorem, it is very hard. It is essentially impossible without some kind of non-locality to get the information from the book into the radiation. That's why it's not a paradox, but it is a puzzle.

2:11:53.2 SC: John Stout says, can you give us a detailed update on your recent work on complexity and emergence? Well, not very detailed, no. I don't like to talk about research that is completely in progress because it can change and we don't know what the answers are until we're done yet. I can tell you generally what I'm working on. I have a paper in progress with an undergraduate Adhrit Perola... [2:12:24.3] ____ Adhrit Perola, sorry, Adhrit, where we're trying to understand emergence better. And it's a very simple, like it's just a short, philosophically oriented paper. We're not trying to do something. What Anil Seth did, a former Mindscape guest where he and his collaborator Lionel Barrett try to define emergence in some mathematical way, give a characterization of when emergence happens.

2:12:47.0 SC: We're just trying to make sense of the definitions from a philosophical perspective by removing any sense of judgment or human agency from the definitions. So a lot of times in definitions of emergence, you will hear things like a property is emergent if it exists at the macroscopic level, but its existence would be surprising if all you knew was about the microscopic level. What is that supposed to mean? I hate those definitions. How do you know what's surprising and what's not surprising? Sometimes it makes it sound a little bit more rigorous by saying, if something is derivable from the microscopic level. But again, that sounds a little sciencey, but it's very hard to make rigorous. How do you know what's derivable or not? You know What you've derived, but you don't know that complete set of things that are or are not derivable.

2:13:36.6 SC: So that's actually more fuzzy than it sounds. So we're just trying to come to understand, try to invent conceptions of both weak and strong emergence that do not rely on words like that, which do not rely on human judgments. And on the complexity side of things, I'm working with [2:13:56.5] ____ Golche Cardez who's a graduate student at the University of Colorado, actually, on following up on my work with Scott Aronson and others on Complexogenesis, how complexity comes into being over the course of time. Scott and I had this little cellular automaton kind of thing where it was coffee mixing into cream. So Golche and I are trying to be much more realistic. We're trying to look at more realistic models of physical interactions with things like photons and things. We're trying to ask, what is the minimal set of ingredients you need to develop these complex structures in a universe that is slightly more realistic than just an automaton on a lattice?

2:14:38.2 SC: And in particular, we really want their... One thing that turns out to be super important are photons, [laughter] because photons can be... Photons are massless, which means they can have as little energy as you want, which means that non photons, atoms, the things that are gonna come together to form planets and DNA and things like that can fall into lower energy configurations by emitting photons. Right? And this is something that obviously is just taken for granted in ordinary physics. It's obviously there. What we're emphasizing is it's crucially important that the physical laws allow things like that to happen, which you wouldn't if you didn't have photons, if you only had massive particles. There'd be far fewer ways for dissipation to occur and therefore for configurations to find these lower energy, but more complex meta stable equilibrium. So we're working on developing how this occurs over cosmological time and how information is stored in the structures that form and things like that.

2:15:43.5 SC: Andrew Goldstein says, in a previous response, you indicated that the definition and understanding of complexity needs greater consensus, including perhaps when it begins and why it increases, complex systems seem to eventually result in equilibrium thermodynamics. Is it reasonable to suggest that if complexity has a purpose, could it be the acceleration of entropy by the destruction of energy gradients, or could there be other explanations? I think that these kinds of suggestions are tempting, but they usually don't work. Which is not to say they never work. These kinds of explanations, by which I mean talking about the emergence of complexity and so forth, as if it has a purpose, as if it is there to do something. The laws of nature do not have purposes. They're not trying to do things. They just obey the laws of nature.

2:16:36.3 SC: Now, there are various times in nature where we human beings can understand the behavior of a physical system in terms of either maximizing some quantity or minimizing some quantity, like the principle of least action, like we were talking about before. And in thermodynamics and statistical mechanics, these principles can be very useful. Certainly, if you just have a box of gas and let it equilibrate for a long time, it will go to its maximum entropy configuration. And so there are all sorts of proposed cases in statistical mechanics where under certain circumstances, either entropy is maximized or entropy is minimized, or the rate of entropy creation is maximized or minimized, etcetera. But there is no overall single theory of that. It's a very specific case by case basis. And that makes perfect sense because in nature, sometimes entropy goes up quickly and sometimes it goes up slowly and sometimes it remains constant.

2:17:36.2 SC: So to sort of say after the fact that this particular configuration is there because it's accelerating the increase of entropy or something like that, hmm, I don't think that that's the best way to think about things. Maybe there will be some specific circumstances in which that works, but I am doubtful that's the general way of thinking about it. I'm happy to be wrong if someone comes up with a great principle that I don't know about yet. And the other thing is you say the destruction of energy gradients, that's not what happens in the universe. Energy gradients come into existence in the evolution of the universe because gravity and gravity is different. Remember we were talking about gravity having a negative heat capacity. If you let a box of, not gas, but stars evolve over time, the energy gradients increase rather than decrease.

2:18:31.1 SC: So entropy is still increasing in either way. You just gotta be very careful once again about what exactly that means. Sandra Stuckey says, why is quantum locality of the Bell's theorem kind not a problem for relativistic quantum physics, EG and QED? I'm happy to get an answer from the Everettian perspective, bonus, if you have an answer that a Copenhagener would accept. Well, some people would say it is, Tim Maudlin would absolutely say that it is. So what's going on here for the non-experts is that there's two kinds of evolution in quantum mechanics as John von Neumann famously clarified for the world. There's the evolution that happens when you're not looking at it. When you have a quantum system that is just obeying the Schrödinger equation, there's absolute locality in that, in what we call the unitary dynamics. The field theory, whether it's QED or the Standard Model, whatever, they have local unitary dynamics, then there's a whole separate thing that happens when you measure the system and the wave functions collapse.

2:19:32.5 SC: This is what EPR and John Bell and his theorem showed as necessarily non-local. There are non-local correlations between the outcomes of measurements. So the point is that those really are two different things. When particle physicists or quantum field theorists construct models like QED or the Standard Model, they just don't worry about quantum measurements. They just think about the evolution of the system in its unmeasured state. And that evolution is entirely local. So you write down a Langragian and like we were talking about before, and you get local evolution from it, and there's no problem there. It is still quantum mechanics. So the evolution you get at the end of the day describes systems, which when they are measured can give rise to non-local correlations. But that's true for non-relativistic quantum mechanics, just as much as it's true for relativistic quantum mechanics. So the Everettian has no problem with any of this.

2:20:28.2 SC: Of course, the Everettian just says you do a measurement. The Wave Function branches, there's a little bit of a subtlety because you can choose to define your branches simultaneously with respect to some reference frame, or as David Wallace likes to do, to define it within a light cone or something like that. But it ultimately doesn't matter, you get the same set of experiences for observers. Either way you do it. For Copenhagen Of course, the problem with Copenhagen is it's just not a well-defined theory. It doesn't say when wave functions collapse or how they collapse or anything like that. So I don't know what those people would say. Mike Briggs says, John Skrentny in Mindscape 265 decries a tech executive's churn and burn attitude towards tech employees. I agree that that is a rather harsh point of view, but it's sort of a two-way street. No? College grads plan on lots of job hopping. My question is, which attitude came first, churn and burn, or job hopping? Well, I'm not the right one to ask here, but these are highly asymmetric situations, okay? A person who has a job versus a corporation who has employees, it's just not the same kind of situation, to that person, that job is enormously important for their life. Trying to find the best job for you, the job that keeps you happy.

2:21:56.7 SC: The job you can make the most money, just makes absolutely perfect sense for a corporation. If a person, an employee leaves, they just find another one, right? And they probably have many, many employees. The status of any one employee is enormously less significant for a corporation than the status of a person's job is for an individual person.

2:22:13.0 SC: Now, you can say that the corporation is just trying to maximize its profits, it's trying to maximize its shareholder value, that's just what John Skrentny complained about, that if your goal in life is to maximize shareholder value as a corporation, you're going to be absolutely terrible to your employees, and one can argue there are things in life that are worth doing other than maximizing shareholder value, certainly the maximum happiness that human beings can achieve in the world is not achieved by just trying to maximize shareholder value. Arne says, I enjoy listening to the podcast, but then the guilt overwhelms me and I realize that something has to be done by thinking people to try to prevent Donald Trump from becoming the dictator that he wants to be in just a few months. Do you think we all should be dropping what we're doing and focusing on this one existential question to make sure that Donald Trump does not become president again? Well, no, I don't think that. Or I would be doing it, right? So I guess that's a sort of a... You knew what the answer was going to be. And I guess the real answer is, Can we justify that or is it just that we're fooling ourselves and ignoring the real problems of the world by trying to get on with our lives? Look, I've never thought that the right thing for people to do is to find the one single biggest problem in the world and tackle that at the expense of doing anything else.

2:23:38.7 SC: I think that both individuals have the capacity to do more than one things with their lives and societies have the capacity to let different individuals do different things, so I think it's okay to do theoretical physics while other people are housing the homeless and other people are trying to stop wars and other people are defending democracy, I think that all these are okay things to do, we need to balance the rate or the amount of effort we put into each of them. I do think that the democracy that we have here in the United States is incredibly fragile right now, and it's completely plausible that will end in a short period of time. I think that would be disastrous. I don't personally know what to do about it, maybe giving little podcast rants about how important democracy is, is something I can do about it, I don't fool myself into thinking that's a very important or influential thing I can do about it. I don't know what else I can do about it. I donate to various causes and people, I think will be better. I try to make the point in conversations and talks and podcasts that this important thing is happening, I'm even writing a book on The Physics of Democracy, it will be too late for the 2024 election, but part of it will be valuing the importance of democracy, and so hopefully, reminding people of that value, but maybe the answer would be different if you had some way.

2:25:10.8 SC: Like if there were some specific things you could do that if you put a certain number of hours a day into doing it, you would guarantee to save democracy, then I might feel differently. I would certainly feel differently about myself, but I don't know exactly what that would be, so I think that it's important to live our lives while recognizing the important challenges to the world and putting some effort into trying to combat them. Anonymous says, suppose you and a team of scientists find yourselves in the library of an ancient civilization located in our own galaxy, you find a few records that seem really important, working with limited time, the team is only able to decipher rough translations for some of the records, which are the following: Black holes, exactly what is inside them, what came before the Big Bang, ship design for traveling faster than light, dark matter and dark energy, a full explanation, an atlas of wormholes and a compendium of intergalactic civilizations. For reasons of brevity, this is still the question, avoiding over-complication and missing the point of this fun hypothetical, you and the team find you must leave immediately, never to return, you are only able to take one of these records on your way back to earth, which one would you choose and why? Okay, I'm probably gonna surprise people, remember the options are, these are various sets of records I can save from the oncoming disaster.

2:26:33.3 SC: One about black holes, one about the big bang, one about traveling faster than light, one about dark matter, dark energy, one about wormholes, and one is a compendium of intergalactic civilizations. The answer for me would be compendium of intergalactic civilizations. Now, why would I say that? Because the other ones are all kind of sciencey questions, the compendium of intergalactic civilizations is somewhere between history and politics, I guess, right. It's a specific set of facts about the universe that I might not otherwise be able to learn, if I read in the library of an advanced civilization, a title of a book called ship design for traveling faster than light, then I... If I believe it, if I don't think they're trying to fool me, now I know it is possible to build ships that can travel faster than light, by the way, it's not. So I don't think this is a hypothetical, very plausible, but okay, what if it were? Then scientists could get to work on it, we could try to figure out how that would do it, and as I said before, eventually you would figure it out. We don't need that book, we can figure it out for ourselves, and likewise for dark matter, for black holes or whatever, we can figure those things out, but we can't just by doing science, figure out what intergalactic civilizations exist or maybe what did exist, which ones did exist in the past, so to me, that compendium would be absolutely priceless, irreplaceable knowledge. That's what I would save.

2:28:01.4 SC: James Allen says, in your discussion with Philip Goff at Marist, you mentioned there was a first date question you needed to ask if you were going out with a panpsychist. So by the way, the first day question was, Does your version of panpsychism entail we must modify the Standard Model of particle physics? So, no one wants to answer that question, and Philip, did not wanna answer it during the debate, but anyway, James continues, most conversations either in person or online are in courts of law, the other person isn't under cross examination, you're not required to answer questions, what do you think is the appropriate way to proceed when someone is hand-waving, changing the subject or otherwise failing to answer the question without acknowledging they're not answering the question? Can the conversation continue? Or are you better at shrugging and walking away? Yeah, I think it entirely depends on the attitude of the other person, very often you will be better shrugging and walking away or just shrugging and changing the subject, you don't have to dig into and find agreement or disagreement about every single thing that you might disagree on. Sometimes certain topics of conversation are just not worth pursuing, but you need to know, you need a judge for yourself, whether the other person is interested in coming to some kind of well-reasoned, clear consensus about this.

2:29:18.5 SC: Maybe they just don't wanna talk to you about it, there's all sorts of possible reasons why this might be a frustrating situation, I don't think there's a simple cut and dried strategy here that fits for all of them. RPD asks a priority question. Is the common understanding of backwards time travel wrong? And is this the more accurate version? By the way, this is now Sean speaking, this is not a good kind of priority question, trying to use priority questions or other questions to say, Please agree with my idea. That's not how it works. I really encourage anyone asking AMA questions, to ask questions that are actually trying to get information from me, not trying to convince me of something. Anyway, RPD continues, moment travel occurs when an individual travels into a replicated moment in time while maintaining their current physical state, time is still on its continuous linear path for the individual, no traveling through time occurs in Back to The Future Marty moment travels. True time travel occurs when an individual returns to a previous physical state and they lose all memories they had between the years of time travel, for them, a segment of times linear path is cut out and they truly do travel through time. In Back to The Future everyone other than Marty would have time traveled. So no, I don't think that's a more accurate version at all.

2:30:43.1 SC: Of course look, you're welcome to come up with any definitions you want, time travel into the past as far as laws of physics know, is impossible, it doesn't happen in any way, so you're welcome to come up with all sorts of imaginative scenarios, you're not gonna prove or disprove them scientifically. Usually, the closest that we can imagine to the standard science fiction trope of time travel is travel along a closed time-like curve, which is ordinary travel through space, but where space-time is sufficiently warped that you end up repeating an existence in what you thought was the past, even though it is now part of your personal future, neither one of those descriptions that you've given here are that sort of respectable kind of time travel, the idea of things like that to the future where you hop in a car, disappear, poof, and then you're in the past, that is not very respectable in any sense, it's not very clear what that would even mean, so if you try to push it too hard to make sense of it, you're going to come into trouble. This other kind of time travel that you're trying to define where an individual returns to a previous physical state, losing all memories. I'm not even sure what that would mean. In what sense have you returned if you are the same collection of atoms you were before with the same memories that you had before?

2:32:02.3 SC: I'm not someone who believes in some ineffable essence of consciousness that travels around the physical universe, disconnected from the atoms. I think that there's just atoms doing their various things. So I would not really qualify that as time travel at all. Bart Shipper says, I was recently listening to an episode with Nick... The episode with Nick Bostrom again, where you discuss among other things, the doomsday argument. I always thought the argument doesn't really make sense, what bugs me about it is I can easily imagine doing the same thought experiment three orders of magnitude smaller, a very smart caveman could in theory have concluded that 100 billion people have been alive, and so the chances of him existing and also humanity reaching 100 billion in total were vanishingly small, and yet here we are. Do you have any thoughts on this? Is it not simply the same case when thinking about 100 trillion people? So the doomsday argument for those of you who don't know says, look, if a certain number of people who have already existed in history, it's about 100 billion, and if I think that there is a finite number of human beings who will ever exist, then chances are very, very small. Well, chances are 1% that I would find myself in the first 1% of those people. Chances are also 1%, I would find myself in the last 1% of those people.

2:33:21.9 SC: And therefore, I can actually predict how many human beings will exist in the whole history of humanity to a 99% confidence, using this kind... If I'm a typical human being within that ensemble. Now, my own view is that this is not correct, and the reason is very simple, you are not a typical human being within that ensemble, there's no justification for thinking that you are, so the whole line of reasoning fails, but for Bart's worry that is actually not a worry. I see why, I know why you would think that. You would think, Well, if I was one of those people very, very early on in human history, I would use this logic and I would be wrong, right? 'Cause I would predict that there wouldn't be more than 100 billion people in human history. The response on the part of the doomsday people would say, Yeah, you would have been wrong. But there are far, far fewer such people who were there around in the very, very early times of human history, than there are since then, the doomsday argument and similar typicality arguments don't claim to give you 100% reliability, they claim to be the kind of argument that will turn out to be correct for most people. So if you think that there is a finite number of human beings who will ever exist and all of those human beings think of themselves as typical, then it will be true that most of them will be between the first 1% of humans and the last 1% of humans.

2:34:52.7 SC: So 99% of them will get it right. Not all of them will get it right, but that's still more than good enough for this kind of reasoning, so I don't think the reasoning actually works, but the reason you suggest is not the one that I would choose. Mike Johnson says, If you could snap your fingers and undo one event in human history, what would it be? I'm gonna weasel out of this one, I honestly don't know because it's a little ill-defined and it's a little impossible, honestly to say, on the one hand, is the Holocaust an event or is that many events? Does that count? I'm not quick sure what I'm allowed to do by snapping my fingers, but also it's very, very hard to know about unintended consequences. If I snap my fingers and Saint Paul never thought that he ran into Jesus on the road to Damascus and got converted and then became a proselytizer and spread Christianity to the Roman Empire, history would be very, very different, right? Would it be better? Would it be worse? I'm not someone who actually thinks that they know the answer to these questions, I think there's unintended consequences in both directions, both good and bad, there is in science fiction stories and in time travel stories, there's sometimes the kind of conservatism that over-emphasizes the bad unintended consequences.

2:36:13.7 SC: You try to make things better and you end up making things worse, I'm not someone who believes that, I think that very possible for it to imagine ways in which we could make the universe better by changing history in different ways, but I'm not sure that you would make it better, and sometimes you would think you're making it worse and you end up making it better or vice versa. All of these things are possible because history is very complicated. So I don't know is the short answer. It would need to be much better defined and then I would need to sit down and think about what are those turning points in history that really could have turned out differently, because sometimes maybe if Christianity had not spread to the Roman Empire, some other monotheistic messianic religion would have. What do I know? And maybe there is some sort of attraction, basis of attraction that means that that was going to happen one way or the other, that monotheism would just... Was destined to supplant polytheism. I don't know. I really don't. So it would require a lot more thought than I've been able to give such questions. David de Cloet says, when do you record the episode introductions relative to the conversation with the guest? Well, that depends on when I have the time, I try to do both the episode introductions and the little reflection videos, the Patreon supporters get to see, I try to do them right after the conversation, but sometimes that's not possible.

2:37:34.9 SC: The reflection videos, I try very, very hard because I want it to be an immediate off-the-cuff reaction, the intro video sometimes are weeks later, if I have just been too busy trying to do things and I have to put it off. So, I try to do it soon, it's never before. I don't think it's ever been before, I don't remember having done an introduction before actually having the conversation, that would seem to be asking for trouble as far as I'm concerned. John Keller says, you are mentioned a few times in Robert Sapolsky new book on the lack of free will, called Determined. Have you read the book and has it influenced your perspective on free will? So I have not read the book as it actually is, Robert Sapolsky sent me a draft of some of the chapters so I could comment on them and honestly, it's super frustrating. So I made comments, I don't know whether those comments made any impact on the final book that actually got published, but the thrust of the comments was very simple, and it's the kind of thrust I could give to just anyone on the no free will side of things. There seems to be a difficulty in getting across what compatibilism says, and I don't know why it's a difficulty because it's not that hard that you can disagree with it or agree with it, but it's not that hard to say it.

2:38:58.5 SC: There seems to be this feeling in certain circles that if you just really, really say how the laws of physics determine what will happen, then compatibilism will go away or somehow be refuted. That makes no sense. Because compatibilism is literally the compatibility of deterministic laws of physics and free will, that's what it means. If you want to refute compatibilism, by all means, but you're not gonna do it by saying that things are determined, and that's what Sapolsky was trying to do in his book, and that's what many people who are on the anti-free will side of things attempt to do, not all of them, but it's a very common thing. So no, it has not influenced my perspective on free will. The other thing that is very frustrating is that, as I said this before, the anti-free will people somehow have convinced themselves that there can't be compatibilism between determinism and free will, and they don't want to believe in free will for whatever reason, or they don't believe in it, therefore, they want to defend the idea that the laws of physics are deterministic, but they're not. There is such a thing as quantum mechanics. Even if you're someone like an Everettian or a Bohemian who thinks that there is some hidden underneath determinism, it's hidden.

2:40:19.8 SC: We can't see it, for our perspective, the laws of physics are indeterministic, so they force themselves to have to jump through hoops to pretend that the laws of physics really are deterministic, even though there's a little bit of quantum indeterminacy. It's just wrong, it's just flatly incompatible with laws of physics as we know them, but they need to work very hard to do that, and I think that that's a waste of time because it has nothing to do with free will. If the laws of physics are stochastic versus deterministic, they're still the laws of physics. If you are determined to say that the laws of physics being deterministic are incompatible with free will, you can just as easily say that stochastic laws are incompatible with free will, there's nothing that is gained by the ability of Laplace's demon to predict the future. The laws of physics are what matter, not whether they are deterministic or stochastic. So it's all very frustrating to me, I gotta say. Rob Patro says, I was wondering what your thoughts are on the population collapse crisis/hypothesis. Should we be concerned or cautious about this, or is a declining birth rate, just something that happens when you educate a population, raise the standard of living and empower women to have a meaningful say in their own reproductive choices?

2:41:35.0 SC: Well, I am personally not that worried about the population collapse, I was young enough to remember when the population explosion was the big problem, none of these people, either the people who were worried about population explosion or the people who currently worry about population collapse have told me what the ideal population is that we should be shooting for. Is there some number out there that if we go above or below, things get bad, they're just somehow they work themselves into a fret about the rate of change. If people don't wanna have kids, then I think they shouldn't have kids. Honestly, that's fine. That's an individual choice. Again, I'm not into these moral obligations of the individual fitting into the larger system, I think that's a misplaced way to put your moral obligations on the world. But if the world only had 1 billion people would that be so bad? It has 8 billion now, I'm not worried that it's gonna dip down to one billion people any time soon. I think that there are two things going on that worry people about the declining population, and both of them, I think are a little sketchy. One is frankly, racism. They say, well, they don't say, but in the backs of some people's minds, the real problem is not that the world's population is diminishing, 'cause it's not, but that the wrong people are breeding and that the right people are not breeding fast enough and I think that's just kind of racist and also wrong for all sorts of reasons.

2:43:03.5 SC: It's a wrong, not factually wrong, but morally wrong thing to think, I think that you put yourself into a bad place when you're thinking like that. The other thing that people think is, well, you know, there's a lot of good things like art and music and scientific discovery and innovation, and we need people to do that, and if we have fewer people, we would do less of it. Again, what is the right amount of scientific innovation that you need, I think if you had fewer people, they would be innovating also, and eventually they would get the same place that more people would get, maybe it would take them longer to do it. I don't really see what the problem with that is. The same number of people would experience the innovations after all, so I don't ever, I have not yet seen any actual logical persuasive thing to worry about there with... By the way, the population is not declining, the rate of growth of population is declining, maybe that will eventually lead to population decline, but that's not something that I'm very concerned with within my lifetime, let's say. Paul Hesse says, why does the universe seem to come into focus at certain specific scales of description, for example, we can describe things in terms of quantum scale, atoms, molecules, chemistry, human scale objects, planets, solar systems, and beyond. It seems like tuning a radio where you hit a specific channel, everything comes into focus and in between there're vast stretches of noise.

2:44:24.2 SC: Yes, this is absolutely true. And either you've been listening to Mindscape a lot or you had this idea separately, but this is a feature of the universe, it presents itself to us in layers, I talk about this in the Big Picture and elsewhere. Why does it do that, is the question? That I don't know, I think that it would be nice. We were talking about emergence before and attempts by people like Anil Seth to write down an equation which tells you what has emerged, so I can imagine that the perfect version of that equation would sort of find peaks of emergence in some landscape, of course granting you, imagine all the different ways to coarse-grain the world, and some of them give you immersion descriptions and some don't, and so there's some peaks in that landscape where you get a lot of good emergent description with this particular kind of coarse-graining, what exactly determines where those peaks are, and why does the world have so many of them? That I don't know. These are good questions. That's exactly the kind of thing I would like to understand better. Kyle Hicks says, I'm interested in the intersection of ontology, specifically monism, dualism, pluralism, and particle physics as outlined by the Standard Model, in your view, does the Standard Model lend support to any of these ontological perspectives? Furthermore, considering the distinct nature of fundamental particles, would you argue the Standard Model suggests a pluralistic ontology?

2:45:48.2 SC: So I'll confess. I've heard people talk about monism versus dualism, versus pluralism, etcetera. I don't get it at a very fundamental level. If I have an apple and an orange, I have two objects. There's an apple and an orange. So there's two, there's a duality, right? But also I have a collection of two fruits, an apple and orange, I have one collection, one set. So is that monist, or is it dualist? I know that's a silly example, and you can say, well, what I mean by monist is blah, blah, blah, blah, but I would need to know what do you mean? I can imagine different meanings to those terms, so for the Standard Model of particle physics, the particles, the fields that make up the Standard Model are all vectors, are all elements of some vector space and you could just combine them into one big vector space if you wanted to, with obvious ways to subdivide them, but it's like the apple and orange example, but much harder to be sure that there's only one way to subdivide them. Indeed the whole mechanism of spontaneous symmetry breaking in the Higgs mechanism can be thought of as taking one field that you would have thought of as part of the Higgs boson and it's eaten by the W and Z bosons as you will read about in Volume 2 of the Biggest Ideas in the Universe when it comes out. So it's not even obvious how to divide up those fields, so you can do it, but it might not be only one unique way of doing it.

2:47:23.8 SC: I personally think of the universe as a vector in Hilbert space, which certainly sounds like a monistic ontology, but then again, the first step after you say that is to ask, Okay, but how can I divide it up into subsystems, right? And how can I recover the real world that way? So honestly, my answer is, I don't know whether these categories are relevant to thinking about the fundamental aspects of nature as we currently understand them, Ahmed Hindawi says, You've mentioned a number of times in the past few years, something about wanting to write an undergraduate textbook on quantum mechanics, did you make any progress on this project, is it something we should expect in the next year or two, perhaps? I made a little bit of progress. I hope my publisher is not listening, I've not made nearly as much progress as I had hoped, there's other books I gotta get written first, for better or for worse, the textbook is the easiest one to delay, so I would've loved to have finished it by now, I'm nowhere near finished, do not expect it in the next year or two, certainly expect it the next five years, unless something terrible happens to me, it will be out in the next five years. That's all I'm willing to tell you, right now. Letme101 says, I'm new so forgive, if this has been asked, I am endlessly confused by the horizon problem, everyone I've heard explain it seems so certain inflation is needed to explain the homogeneity of the universe.

2:48:45.5 SC: But what I don't understand is, if the laws of physics are the same at both sides of the observable universe, then it seems feasible to me, that parts of the universe outside of causal contact with each other could evolve in the same way by following those laws with the same homogeneous end result, regardless of exchange of information. What makes cosmologists so sure that it can't be explained in any other way? So there's a bunch of things to say about this question, one is, please, do not ascribe certainty to cosmologists or surety, sometimes they might talk that way, but not always, and honestly, they shouldn't. We don't know these questions about the super early universe, we need to maintain a little humility here, the horizon problem is an example of an attempt to convince you that the initial conditions that we need to make sense of the Big Bang model are more surprising than you might have thought, that's it, that's all it's trying to do. Inflation might be the right solution, that makes it seem less surprising or it might not, we really don't know, you have to be careful and have credences and balance them and be willing to upgrade them when the moment happens. So for those of you who don't know, the horizon problem is simply the statement that without Inflationary Cosmology and standard Big Bang cosmology, if I look at different parts of the cosmic microwave background.

2:50:03.5 SC: So features that were there in the universe a few hundred thousand years after the Big Bang, but far away from each other, then those spots in the microwave background on different sides of the sky share no common past. That is, if I take one point in the CMB and I look at its past, so all the light cones that go toward the past and hit the big bang, they do not overlap. The light cones that describe the past to points elsewhere in the CMB, if they're more than about one degree away from each other on the sky, they do not have any events in common in the past, and yet they are the same temperature, so the size of the region that is their past is called the horizon, that's why it's called the horizon problem. So how did these different regions know to be at the same temperature and you might, people have like, Letme101 is saying, say, Look, it's just the laws of physics, of course they're gonna be the same temperature. What else could they be? That's not right, because of course, it is the same laws of physics, or let's say that it's the same laws of physics, but the point here is that it's dependent on time, right? The thing that you're observing when you see the cosmic microwave background is not just the temperature of the radiation that is given off at the moment of recombination, because that's the same everywhere, that just depends on the number of photons and the number of baryons in the universe.

2:51:26.7 SC: And you can predict using the laws of physics what the temperature of the light was when it was emitted. But guess what, you're not seeing that temperature, it's been red-shifted because the universe has expanded along the path that the light took from the CMB, cosmic microwave background, to you today. The question is, why is that amount of red shift the same in different directions on the sky, in other words, not only were the conditions nearly the same at these different points of the universe, but the universe started expanding and cooling at the same moment, in some very well-defined way of slicing the universe in two moments of time, the universe looks smooth, it looks homogeneous, and it sort of all started at once. That's the mystery, not that the fundamental laws of physics would have been different, how did it know to start at the same time, it's really at the end of the day, just rephrasing the fact that the universe is very smooth, it's smooth on scales that are so large that without inflation, these scales never would have been able to coordinate what they're doing at the same time, that's supposed to be the mystery. Murray Don says, can you recommend a good treatise on the moral arguments for and against abortion and use of laws to attempt to prevent abortions, and if you please, if you have time, expound on your views.

2:52:47.6 SC: Well, I don't actually know a good treatise on the moral arguments, I think... As far as I know, I might be super out of date on this, but the classic book that sort of looks at different arguments is by Laurence Tribe, the law professor at Harvard. Ah, I forget what it's called, but it's something about the fact that the people on different sides are not gonna agree with each other because they have fundamentally different starting points, and that's a problem in a democracy where you're supposed to try to agree with each other. And I think it is hard because there is a distinction between the arguments people give and what they actually think, so you can't necessarily take people's views at face value. People who are pro-choice, the people who think that you should be able to get abortions will often accuse the pro-life people, the people who think you shouldn't be able to get abortions of actually not caring about fetuses or embryos or children, but caring about controlling women's bodies. And I think that there's an extremely good case to be made that they are right, much of the time. Not all of the time. I went to Villanova as an undergraduate, a Catholic school, the biggest charity on campus was Villanovans for Life, the anti-abortion group, and so I absolutely know people who just very, very sincerely believe that life begins at conception, that the human life...

2:54:10.0 SC: Well, look, they're not materialist, they're religious, they think that the soul enters the body at the moment when the sperm fertilizes the ovum, okay. They literally believe that. And they say that before doing that, before that happens, it's just a cell after that, it's a human being and to end its life is to kill it, but there's plenty of other people who don't believe that or don't care about that, what they really wanna do is write down restrictions on what people are allowed to do, especially what women are allowed to do, and the evidence for that belief is that if you really believed that protecting young unborn children was the point, that would naturally go along with a whole bunch of other policies about care for children who grow up in poverty and things like that, and those do not actually correlate with the pro-life position very, very well. So that's why it's complicated to know what people actually think. I do think... You can lookup a blog post that I wrote with a weird title about How Metaphysics Matters or something like that, but the point was to say, on the other side, forget about for the moment, the accusation that many pro-life people are actually just trying to control women's bodies, because some pro-life people are perfectly sincere. And I think that the pro-choice people give essentially no time to talking to the people who are sincere on the pro-life side.

2:55:36.9 SC: Because they don't actually try to talk you out of thinking that life begins at conception or anything like that, they just don't take that seriously. And so there's almost no dialogue back and forth about exactly that. When I was an undergrad at Villanova, I helped organize a little symposium on exactly this question. We had priests there. We had biologists there. I remember one of the biologists, said like, I don't know why I'm even here, there's no interesting biology questions here we know what the biology is, and then he heard some of the non-biologists give their little panel discussion talks and he goes. Oh, actually, there's a lot of things that a biologist needs to explain to some of these people. It's a very, very messy thing. The biological reality of conception and childbirth and things like that. So I don't think that there's a lot of good faith talking back and forth in this. For example, let me just give you one simple example, the pro-choice person might say, I get to choose, it's my body, I am a woman, I can or cannot get pregnant, I should be able to choose whether or not to give birth to the child, if you actually believed that life begins at conception. And every living being deserves the same rights, then that argument has no weight for you because you have the right to choose what happens to you, but the unborn child has the right to choose what happens to them.

2:57:00.7 SC: They don't have the ability to do it, but they have the moral status to do it, and therefore there's an obligation to protect them until they are born, and then other things can happen. And I think you can sensibly argue that this is not a very sustainable position, are you really saying that you can be held hostage to support the life of another person in all circumstances, or is this something special to being pregnant that we're talking about here? But again, my point is just that people don't talk to each other about these things, so my only point is that it actually matters what you think about what it means to be a human being, and when personhood begins, this is why I write books like the Big Picture, where I try to convince people of naturalism and physicalism and say, There is no soul that enters the body at the moment of conception. And once you understand that, then when you actually look at this little tiny group of cells, you're not even tempted to assign it urgency or personhood or moral status, it's just a couple of cells, right? And we all know from the biology that cells are... That ova are fertilized and not implanted, and therefore do not grow up into children all the time, and this is not considered to be some great holocaust of death or anything like that, this is just part of the natural human condition.

2:58:27.0 SC: So I do think that getting the physics and metaphysics and philosophy correct is very helpful in this, and I think that the physics and metaphysics of philosophy that's been handed down by the Catholic Church over thousands of years is probably not your best guide. Stewart Hane says, I think paraphrasing that you had a low credence for intelligent life elsewhere as, given billions of years, there is time for self-replicating machines to pervade the universe, if so, can you do your best to refute your own argument? So I wouldn't even say this is my argument, this is an argument that goes down to, in this sense, von Neumann, that self-replicating machines could easily have filled the galaxy and they don't, therefore, I do think that is evidence against there being technological civilizations elsewhere in our galaxy, but it's not definitive evidence for the very obvious reasons that maybe there's something that prevents intelligent civilizations from doing that. The reason why that's not an easy thing to adjudicate is because you don't need... Even if it's only 1% of the intelligent civilizations that do that, the universe, the galaxy would still be filled with these machines. And it apparently is not. But as I said before, these are making predictions about the behavior of civilizations that are way more advanced than us, so that's a very tricky thing to do.

2:59:47.9 SC: So I don't think it's a matter of refuting the argument, but just realizing that it's not an air tight argument, it's an argument, and I absolutely believe that it is true that... Let's put it this way, it is a feature of Bayesian reasoning, that if there are two possible pieces of data you could observe X and not-X, and you think that had you observed X, it would increase your credence in a certain proposition, then it must be the case that observing not-X decreases your credence in the proposition, they can't both increase or they can even increase for one and stay the same for the other. Okay, so if you agree that actually observing self-replicating machines that were clearly the product of an intelligent civilization would have increased your credence in the existence of those intelligence civilizations, then the absence of any such evidence must decrease your credence. There you go, that's a good argument. Pete Faulkner says, Your recent one dream course on many worlds has really helped to clarify the idea for me, however, it's clear that one of the core challenges within the many worlds interpretation lies in its explanation of probability. The branching multiverse suggested by MWI creates the problem of the 100% certainty that each specific outcome is allowed by the Schrödinger evolution of the Wave Function actually happens.

3:01:13.5 SC: The theory addresses this challenge by introducing weight to differentiate the likelihood of different outcomes, however, this concept of weight appears to rely on pre-existing knowledge of the Born rule, the probabilistic framework, it's supposed to explain, doesn't this dependants create a circular argument within many worlds? Good. The answer is no, it does not, because number one, the concept of weight is not introduced to differentiate the likelihoods, and number two, it does not rely on pre-existing knowledge of the born rule, it's just linear algebra, it's just Pythagoras theorem in some sense, it's just the statement that the components of a vector with length one are... What to say? The components squared added up gives you the length of the vector squared, that's where the square comes from, in Born's rule, it's just from Pythagoras theorem, and the weight that we're attaching to these different branches is just the amplitude squared. It's just geometry, it's just trigonometry, it's not invented to make probabilities come out, it's there without anything to do with probabilities. The idea of probability in many worlds is fundamentally subjective, it has to be for exactly this reason that all of the branches actually happen, but they happen with different weights, again, there's nothing to do with probability, they do happen with different weights, according to the Schrödinger equation.

3:02:39.5 SC: The appearance of probability is something that human beings add to help make sense of the theory, and the argument you need to make is that the probabilities should track the weights, the weights are there anyway. And guess what, there's many good arguments, that the probabilities should track the weights, for one thing, the weights are a set of numbers between zero and one that add up to one, they satisfy the axioms of probability, the total probability, the total weight is conserved over time. There's many other things, so many other arguments you could put forward, more sophisticated than that, but you always get the same answer, the right probabilities to assign are given by the amplitude squares of the branches of the Wave Function. Yohanne Fox says, Concerning large language models having an internal model of the world, a paper called Moving the Eiffel Tower to Rome describes how researchers investigating GPT2 and they locate the network nodes representing the Eiffel Tower, Paris and Rome, they then weaken the connection between the Eiffel Tower in Paris while strengthening the one to Rome, the resulting language model then seems to believe that the Eiffel Tower is in Rome, for example, describing the Eiffel Tower as a symbol of Rome and being located across St. Peter's Basilica. Do you consider this evidence that training large language models make them build up an internal model of the linguistic world?

3:04:00.4 SC: No, I don't. And remember, there are two alternatives being compared here, and you have to actually compare the alternatives. One, in either case, as we've said before, the large language models are able to make sentences that sound meaningful and human sounding. Okay, the question is how they do that, the way that they're explicitly trained to do that is next token prediction, they're, input some words and they're predicting what words might be associated with them that are most likely to appear next. A different way of getting those answers that sound human-like would be to have a model of the world, which means that you have some concepts and the different concepts have different properties and the properties kind of fit together, think about Legos or something like that, or Tinkertoys, if you were my generation, and then you would fit together the different pieces and they would be a logical structure there that would enable you to do things that go beyond the kinds of sentences you'd ever heard before, you can extrapolate in ways when you have a model of the world that you can't, if you're really strictly limited to just predicting kinds of sentences that you've often heard before. So in this case, I don't know the paper, so I don't know how reliable the paper is, etcetera, but let's take at face value that they have strengthened the connection of the Eiffel Tower and Rome, weakened the one between the Eiffel Tower in Paris.

3:05:27.7 SC: To me, all this is saying is that in the texts that the language model was trained on, it was very frequent to hear the Eiffel Tower mentioned in the context of Paris and not Rome, and you tricked it into thinking otherwise, so that it now thinks that the Eiffel Tower often appears in the context of Rome and not in Paris, and it is the least surprising thing in the world, that it describes the Eiffel Tower as a symbol of Rome and being located across St. Peter's Basilica, which is another thing that often is mentioned in the context of Rome, what you want to do, if you wanna figure out the difference between just next token prediction and an actual model of the world is think about the kinds of mistakes that the model makes, like you say, and again, I haven't read the paper, I don't know, you say The model says that The Eiffel Tower is located across from St. Peter's Basilica, that's a very specific location. Is it also located next to the forum in Rome, because according to the language model, because if it is, then it can't be located across St. Peter's Basilica 'cause those are in two different places. You might think that it... You might end up saying that it is, if all you're doing is pushing around next token prediction. But if you actually had a model of the physical geometry of Rome, you would not be tempted to say things like that.

3:06:47.1 SC: The point I tried to make in my podcast was not that it was not about the probability of getting mistakes from large language models because that is non-zero, but it can easily become smaller with time, I believe that, but it's the kind of mistakes that it makes, the kinds of mistakes that the large language models make, I would argue, and maybe I'm wrong, as I said before, and you can disagree with me and have evidence, the kinds of mistakes that they make are the ones that would be easy to avoid if you just had a model of the world, but are very natural to make, if all you're doing is seeing what kinds of words and sentences follow each other logically, and I think this particular piece of evidence doesn't change my feeling about that much at all. Colin asks a priority question. Aaron Bushnell self-emulated at the Israeli Embassy in protest against the Palestinian slaughter. I want to ask about self-harming acts of extreme protest, whether they can be anything more than simply tragic, reactions have run the gamut from much dumb to dismissal, to ridicule. So I don't know a lot about this case. I know that it happened, but I haven't been spending a lot of time digging into the details.

3:07:58.6 SC: It is a natural human tendency to when someone lights themself on fire to understand what was going through their mind, are they in a good mental state, is it truly a principled objection to some geopolitical events, or is it a reflection of their own struggles inwardly, etcetera, etcetera? And I truly don't know in this case, what is going on, so I can speak only in very, very general terms. I do think that these kinds of wildly dramatic protests can have an effect. I think that they're visceral, they have an effect in the sense of leaving an impression on people, and the effect might be big enough to make it worthwhile or not, that I don't know, I wouldn't ever ridicule them because if the choices are either, they were doing a principled protest against a terrible event in the world, or they were suffering from mental illness, neither one of those seems to be appropriate for ridicule, so I think it can be tragic one way or the other that something like that happened, but that doesn't mean that I think that they were necessarily wrong to do it, from their own perspective. I think that it's absolutely possible that something like that is an internally coherent moral stance, I don't know whether this one specifically actually was or not.

3:09:18.7 SC: Daniel Bagli says, Richard Carrier uses Bayes' theorem to offer odds on whether Christ really existed. What do you think of using Bayes' theorem to offer probabilities on historical accounts, when we normally think of them as simply being true or false in general, and on Christ's existence in particular? Well, this is another one where there's a lot of things going on. You say, when we normally think of them as simply being true or false in general, I think historical events are precisely the kinds of things you should not think of as simply being true or false, they were simply true or false, but we don't know. So we need to put credences on them, just as we do for things in the future, who's gonna win the next presidential election, when is the first date we will land on Mars or whatever, there is some fact of the matter about those things, but we don't know it.

3:10:07.0 SC: So we put credences on it. Same thing for historical events. We have to be a little humble there. We should never say, a 100% true or 100% false. Now, there is an obvious problem that anyone could point out with "using Bayes' theorem" to offer probabilities for anything. Because that's not what Bayes' theorem does. Bayes' theorem doesn't tell you a probability. It tells you how to update your existing probabilities when new evidence comes in. So the not precisely true but pretty memorable and pretty close to true motto is everyone is entitled to their own priors. No one is entitled to their own likelihoods. So in Bayes' theorem you have a prior probability that you start with for a certain set of propositions and then you have a likelihood function which says under those propositions what is the likelihood that certain new information would be obtained. If your propositions and your theories of the world are carefully spelled out you can rigorously and objectively predict the likelihood functions, but you can't rigorously predict or collect priors. Different people will have different priors.

3:11:20.0 SC: So the only sensible thing to say in a Bayesian sense is, this information should change your probabilities, change your credences in a certain way. It can't ultimately tell you what those credences should be. Steve Trettle says, as a mathematician who's a longtime listener to the podcast, let me first say thank you for helping me learn so many beautiful ideas from physics. Here is a mathy question. Do electrons around atoms really live in orbitals? Precisely, even though wave functions decompose into superpositions of eigenstates of energy, since the Schrödinger equation preserves such a superposition, it seems an electron is not only, sorry, it seems that an electron not already in an eigenstate will not evolve into one. So Steve offers some scenarios, but I'll tell you the right one. The right one is the electron is not alone in the universe. So what Steve is pointing out is that if you have a Wave Function that is in a superposition of different states, and those states are energy eigenstates, so those states are states with definite energy, which the orbitals in an atom for an electron are energy eigenfunctions. They are states of definite energy. So there's the lowest energy state, the next highest energy state, a set of different energy states corresponding to different orbitals that an electron could be in.

3:12:38.6 SC: And if you express your Wave Function as a combination, a superposition of different energy eigenstates, then according to the Schrödinger equation, those energy eigenstates just stay the same. Their amplitudes, their weights, if you want to put it that way, do not change over time. And that will be true in a world where the only thing in the world was an electron in an atom. But there are other things in the world, in most particularly the electromagnetic field. Remember when I told you, Bob, when we were talking about complexity that photons are super important, so it's true, everyone knows photons are super important, but here's an example. An electron in a higher energy state will tend to emit a photon and fall to a lower energy state. An electron that is in the lower energy state will possibly absorb an ambient photon and be bounced into a higher energy state. That's what makes electrons change their wave functions. And as a result, because entropy tends to increase, if you have a bunch of electrons that are in not the minimum energy state that they could be in, it's a higher entropy configuration for all those electrons to fall down to their lowest energy allowed states and in the process give off a bunch of photons 'cause all those photons add to the entropy of the universe.

3:13:55.0 SC: So there is sort of an attractor mechanism here where things like to settle down into their lowest energy states. That's why you think that, you know, a ball rolling down a hill goes to the bottom and stays there. Electrons in atoms are the same way. Laurent Delamere says, are you optimistic or pessimistic regarding the climate crisis, I.e., do you believe that by 2050 it will dwarf all other world crises or even the current climate deniers, and even the current climate deniers will be forced to admit we're in big trouble? Or that we will have found a path to cap the global temperature increase? I'm pretty pessimistic about this, so I don't know whether it will dwarf all other world crises. Guess what? Because there could be all sorts of other world crises that we don't know about. You know, I'm super worried about biological either warfare/terrorism or just experimentations gone awry. I know that people worry that COVID-19 was an example of that. I don't think that's especially likely, but there are plenty of other possibilities that are very worth worrying about. So not to mention, you know, rogue states or actors getting good old-fashioned nuclear weaponry. So all sorts of bad things could happen other than the climate crisis, but the climate crisis, climate change has this different character, right? Because it is happening, it's happening right in front of us.

3:15:14.6 SC: It is gradual in some sense, and it is optional. We could take action to stop it, and we don't quite have the collective willpower as a civilization to do that. So we're letting it get worse. You know, that's more or less pretty clear right now. We're doing a little bit. There's some room for optimism. The Hannah Ritchie podcast talked about that a little bit, but not nearly as much as we could or should be doing. So I think that bad things will happen because the climate will continue to warm. I think that more action will be taken once those bad things become more and more evident. I think that it will be far too late to stop really bad things from happening, but I don't think it's gonna be an existential crisis. I think it's just gonna cause a lot of misery and poverty around the world as these things tend to do. Sid Huff says, I've noticed that just about every guest you've had on Mindscape is able to speak very well, to express themselves clearly, stay on topic, not mumble or become confused in their expression, etc. Many seem also be good at infusing some humor in their talk. Do you do anything to vet in advance the general speaking ability of your guests?

3:16:28.6 SC: Yes, I do, you know, because I care about you, the listeners. I do want, look, it's okay to admit that not everyone is equally good at expressing themselves, at communicating, at talking. In particular, the distribution of intellectual ability, research ability, scientific ability, creative ability, etcetera, is not correlated in any obvious way with the ability to be a good speaker and communicator and things like that. So, when I look for podcast guests, there are multiple criteria going on in my mind. Of course, the most important one is that they have something interesting to say. But it also matters that they can say it in interesting ways. And, you know, again, some people are gonna be better at that than others, no doubt. But in this day and age, most of the people I have on the podcast are either people who I know personally or have met, maybe not close personal friends, but I've heard them give a talk or have talked with them, or people I can Google and find them giving talks on YouTube and I can see, oh, yes, they're good at this kind of thing. So, yeah, I do care about that.

3:17:42.9 SC: I want people who will be good podcast guests. That, at the end of the day, is the most important criterion. Ken Wolf says, your discussion with Benjamin Breen on the career of Margaret Mead was interesting in its own right, but a tangential thought that really struck me was just how profoundly her talking to a particular person at a particular time impacted the development of so many ideas and streams in science. It got me to thinking about how much in our current world is so profoundly path dependent. So to look at this from the other side, is there anything in the modern world that strikes you as being the opposite? That is to say, not necessarily inevitable, but very unlikely to have turned out much differently? So yeah, that's kind of, I think, Ken, your question, I had read it, of course, before starting the whole AMA, so that was in my mind when I was just talking about the what would I change in history question. I should have grouped these two questions together if I had been thinking about it. I think that both are possible. I absolutely think that there are moments, and this is part of the physics and democracy kind of idea, there are bifurcations, there are tipping points, there are symmetry-breaking moments, there are moments when things could have gone either way and a little tiny influence from a person or from a group or whatever can truly have an important change in the nature of society.

3:19:02.5 SC: There are other moments or other kinds of things which were gonna happen one way or the other. I think that Ken gives the example of capitalism, that's very a plausible one, but I think most scientific and technological developments are even better examples 'cause like I say, we would have discovered them, we would have discovered the steam engine, if not precisely when we did at some other time and that's going to, you know, there's not any plausible version of history, I think, where we have cell phones and the internet but we're still in horses and buggies, right? There are certain things that sort of naturally go hand in hand with each other. But having said that, in both cases, in the cases where a tiny influence has a big difference and in cases where there is some attraction mechanism that gets you to the same place whether or not, whatever your starting point might have been, it's really hard to actually identify those in practice, right? It is very, very tempting to look at history and to say, well, that's how it had to have happened. The ways that things turned out might seem to be more inevitable in retrospect than they really were.

3:20:11.4 SC: So, I think that it's something that we don't understand precisely because the social sciences are very hard, because human beings are complicated. We can talk about the possibility of these things, but the specific examples that we like to think about, I'd be a little bit, I'd be humble about them. I would have low credence on any particular example. Peter Spiker says, I imagine one of the hard parts of writing books like The Biggest Ideas in the Universe is finding the right balance between assumed knowledge of your readers and trusting them to follow into the more complex parts. When you are writing, how do you know you're getting that balance right? Well, I don't. I have no idea whether I'm getting that balance right. I mean, of course, one develops intuition over the years or experience in trying to do this. You realize that certain kinds of things work, certain kinds of things don't work. Another very good strategy is to literally give parts of the book to people to read, right? Having people who are not experts who are willing to read your stuff, crucially, crucially important for things like that, as well as having a good editor. Stephen Morrow has been my editor for a long time and he is invaluable.

3:21:18.5 SC: And I guess finally, there is the very well-known technique of conjuring up an imaginary reader, right? Having a very specific person in mind who you're talking to, because you might, if you're just thinking vaguely, you might say, well, you know, this concept I'm talking about, it's a little bit tricky, but I bet most people can get it. But then you say, no, I want to talk to Bob about this. And like, yeah, Bob doesn't, he doesn't get these concepts right away. I'm gonna have to work harder to explain it. Okay? Like that's a perfectly legitimate way of doing it. But at the end of the day, it's gonna be the readers who decide. It's not your place to figure that one out. John Tedesco says, my question is about entropy and the end state of the universe. After the last black hole is evaporated, after the last proton is decayed away, what is left in the universe that would account for entropy? If there's no matter left, isn't this a uniform state? Well, the simple answer is quantum mechanics. The harder answer is we don't know. The more accurate answer is we don't know. So the, let's take a black hole for example, okay? We think a black hole has entropy. Jacob Bekenstein and Stephen Hawking told us what it is. It is the area of the event horizon divided by four measured in Planck units.

3:22:32.6 SC: So, presumably that entropy has something to do with the number of degrees of freedom inside the black hole. What are those degrees of freedom? We don't know. Andy Strominger, former Mindscape guest, he and Camrun Vafa did a very, very nice and influential analysis of certain very special kinds of black holes in string theory where they could use dualities to relate them to brains etc. And they could calculate the number of degrees of freedom and they got the right answer, they got the right entropy for the black holes. But in the general case, we don't know. All we can say is space-time itself has degrees of freedom that are leading to this entropy. Same thing for the universe. That was the black hole discussion, but even in empty space we think that something like that is going to be true, that even a universe with nothing in it is still going to have some entropy 'cause there's still quantum mechanical degrees of freedom that account for the existence of space-time itself. We don't exactly know what those are, that's the best we can tell you. Helen Edwards says about recent progress in AI, what do you tell your students to master because a machine can't? I got nothing. There's nothing that I tell my students to master because a machine can't. I'm not very good at predicting what machines are gonna be able to master and what they aren't.

3:23:55.5 SC: And also, look, we're a little bit too early in the whatever revolutionary changes will be wrought by AI to really predict what machines are not going to be able to do. I think that the, you know, again, just like exploring the galaxy or staving off extinction and things like that, it's okay to think a little short term because we have much greater handle on what's possible in the short term than in the long term. I would tell my students to become comfortable with AI. It's not going away. Get used to it, you know, play with it, see what it can do, see if it can help you, but still the things that you have to learn as a student are more or less the same as they always were. You know, the fact that you're able to calculate components of the Riemann tensor using Mathematica, which is true now, or which was not true when I was a graduate student, doesn't change the fact that you should understand what it means to be a component of the Riemann tensor. That's still gonna be true. Leiland Beaumont says, "Why study math?" And then goes on to say, what is the best answer we can give our bright nine-year-old when she complains about learning the multiplication tables?

3:25:08.1 SC: Well, look, the multiplication tables are super boring. I wouldn't try to pretend that the multiplication tables are intrinsically fascinating, but they lead to fascinating things. I would complain about learning the multiplication tables. There's nothing wrong with that, but the world runs on math. The world is mathematical. Understanding the principles by which the world works, whether it's the laws of physics or economics or whatever, is enormously helped if you understand math. So if you want to understand the world, you need to learn math. That's the very, very short answer. I'm not up enough on the psychology of nine-year-olds to tell you whether or not that answer would work, but I think that's the accurate answer. Michael Wickman says, do you think the 76ers should rest Joel Embiid even if his knee injury is technically healed by the playoffs and what are their chances without him playing? Yeah, so for those of you who do not follow the Philadelphia 76ers as closely as I do, they have on their team the reigning most valuable player of the NBA, Joel Embiid, and he was actually playing even better this year.

3:26:16.6 SC: He was just destroying the league and the Sixers had a very, very good record, but then he got hurt. So he's hurt now, he had a torn meniscus and he had to get operated on, and now without him they are terrible. It's just more evidence about how good he is. They are truly, truly bad without him and they were truly, truly good with him. So the question is, the playoffs start in April, maybe, I'm not sure. But the timetable for his recovery is just about, you know, it's just annoying. Like if they thought he would come back and by now everything would be safe and sound. If he wasn't gonna come back for a year you could just shut him down, but the timetable for recovery from this injury traditionally puts him at about the time when the season is ending and the playoffs are beginning. So the question is, do you just let him recover, bring him back for the playoffs, hope that something wonderful happens, or do you say you're in danger of rushing him back and therefore risking both his health and the outcomes that the team could have? So I am not a medical expert here. I think this is a job to actually be handed over to doctors, not to podcasters, not to coaches, not to general managers.

3:27:32.2 SC: It actually needs to be a medical decision and of course, the player's decision, Joel Embiid should have a say in this. I do think, you know, it is strange and it's okay if you don't care about the answer to this question. I know most people listening here don't care about this, but the 76ers when Embiid was playing, when the team was overall healthy, were really doing well. Okay, they had a huge point differential, they were scoring a lot more points in their opposition, etcetera. And since then they've cratered without Embiid and they also have a whole bunch of other injuries that have really made it very difficult. Now if everyone heals up and they come back healthy, there's no reason to think they shouldn't be playing at the level of the best teams in the NBA by the time the playoffs come around. But because that hasn't been what you're actually seeing on the court or from the day-to-day perspective, people seem to have forgotten this. The Sixers are just being written off as contenders to do well in the playoffs. And I get it, because they're not playing well, so the simplest thing to do is to say they're not playing well, they're not gonna be a championship contender, but we all know why they're not playing well, and that fact is no longer gonna be the case, probably, in the playoffs.

3:28:44.9 SC: So I don't know, I think it'll be interesting to see what happens. I would say this, you can't predict how people will bounce back from injuries, and again, it's not just Embiid who's been injured, Tyrese Maxey has a concussion as I'm recording this right now, and it could be that they stumble into the playoffs with just as many injuries and flame out very, very quickly. But it's also completely possible that they're playing badly now 'cause they're very injured, they will heal up, all be ready to go in the playoffs and romp through the playoffs and win the championship, right? These things are hard to predict. That's why sports is kind of fun, or at least interesting, because you don't know what's gonna happen. Bran Muffin asks a priority question. How likely is it that an advanced civilization could master the strong nuclear force? Balankin proposed bulk nuclear matter similar to nuclear pasta found within neutron stars but stable at normal pressure and gravity. This form of matter would be able to withstand temperatures of billions of degrees and be a hundred million times stronger than steel per unit mass and a trillion times denser. Is this consistent with the laws of physics or not?

3:29:51.6 SC: Well, probably not, I want to say. It's hard to say, for sure, because the strong force as, as I like to say, you will all learn in book two of The Biggest Ideas in the Universe, is strong. And strong doesn't just mean the force is considerable, it's a technical term. In a weak kind of force, not just the weak interactions but other weak forces such as electromagnetism, you can do what's called perturbation theory. You can solve the equations for what the force does by first starting the force is just turned off, it's not there at all, and then gradually turning it on and you can see that there's a gradual buildup of what's happening. For the strong nuclear force, it is non-perturbative. You can't get a decent answer by starting with a situation where the force doesn't exist and gradually turning it on. It makes a dramatic, dramatic difference. That's why it's much harder to predict what happens when you have strongly interacting particles than when you don't. So is it possible, is it conceivable that there are these exotic states of matter that have very, very different properties than what we're used to?

3:31:03.2 SC: Sure, it's possible. Certainly, in my level of knowledge it is possible, but it doesn't seem at all likely to me. Like for one thing, why hasn't this stuff been created in a supernova explosion or something like that? Very often it's possible to imagine a state of matter, but it's just dramatically unstable and it all decays away in 10 to the minus 20 seconds. I think that most likely we're not going to get anything dramatically different in terms of physical materials out of the strong nuclear force than we already know about. For one thing, just very simple argument to keep in mind, protons are positively charged. That means they repel each other. It's hard to squeeze protons together. That's why there's only a small number of stable nuclei in the periodic table. You can make anti-protons, but they annihilate when they hit protons. So that's not a good way to get a bunch of protons together. You can try to make matter out of just neutrons, but neutrons are not stable. They decay in a few minutes. So there's an enormous number of obstacles to creating nuclear matter in any way different than ordinary atoms.

3:32:16.6 SC: Ari Moody says, the moon and life here on Earth seem to have formed around the same time. Does this mean life on Earth wouldn't have happened if the moon didn't exist? No, it does not mean that because they both happened early and we don't know exactly when they happened. So it's completely possible that the creation of the moon and the coming into existence of life have nothing to do with each other. I mean, maybe in this Bayesian sense it gives you a little more credence that they are related to each other, but to say that life on Earth wouldn't have happened if the moon didn't exist is just going way too far. We don't know enough about the conditions necessary for life to come into existence to make any claim like that. DMI says, can information be confined within a local region if quantum fields aren't? Sure, I've written books. Every book that I've written, I can pick up a copy of it and the information in that book is confined within a local region of space. The reason of course being that it is true that the books are excitations of quantum fields at the end of the day, the electrons, protons, neutrons, etcetera.

3:33:28.0 SC: But even though the quantum fields themselves spread out all over the place, the excitations within them can be confined to a local region and those can contain information. Brendan K says, is there any currently theorized path to quantum computers replacing current desktop computers and being better at running normal code? Not really, no. That's not the way you should think about quantum computers. Of course, anything is possible, technology can do amazing things, but in the current way of thinking about quantum computers there's an enormous amount of effort that has to be put into keeping entanglement coherent between different sets of qubits and there are various technologies for doing that, but typically they involve some kind of giant dilution refrigerator. You need to keep things at ultra low temperatures or something like that, okay? So even having a relatively small number of qubits requires an enormously big machine, not just because the qubits themselves need to be big like they would have been if you were thinking of transistors back in the '50s for ordinary computers, but because you need a big superstructure to keep them at a low temperature. Is it possible that there's some room temperature version?

3:34:46.3 SC: Yeah, sure, anything is possible as far as my knowledge is concerned, but that's just not what people are trying to do right now. The race to develop working quantum computers is not a race to putting them on your desktop. That's not what we're really aiming for here. Carlos Nunez says, Chris and Matt from the podcast Decoding the Gurus covered your solo episode on AI in their latest episode. You ranked very low on guruness, which is a good thing. It's a good thing that I ranked low, not a good thing to be a guru. Have you listened to the episode and to their podcast in general, and if so, what's your opinion on their content? Yeah, Decoding the Gurus is a podcast. I know that it exists. Look, I've said this before, and I'm not revealing any secrets. I'm not a big podcast listener. I spend my podcast bandwidth making my podcast, and now that I'm living in Baltimore and can walk to work every day, I have even less time to listen to podcasts than I could before. So there's a lot of good podcasts out there that I'm happy to recommend, but I don't listen to them on an episode-to-episode basis. You should all listen to Mindscape. That's all I can say about these other podcasts.

3:36:00.8 SC: But yeah, I think that they're doing a good thing. I take it, my understanding is that they talk about people who profess to be or aim to be, if they don't quite use those words, kind of secular gurus, people who can give wisdom by seeing through the conspiracies of the world and things like that. And they talk about the techniques these people use to develop an aura of wisdom and guruness. And so they used me as an example of someone who is not really like that, who is not actually trying to be more profound than he is. They seem to think that I admit it when I don't know things, which does not make you a good guru. A guru should never admit that they don't know anything. So, I mean, maybe I'm missing something. I'm sorry, I've not listened to the whole episode. I listened to a few minutes of it. Sounded good to me. So, I can't even imagine listening to two hours of people talk about me. That sounds like torture, even if they're saying nice things about me. I really just, not my idea of a good time. So thanks to Chris and Matt for doing that. Thanks to them for saying nice things about me. I can't judge what they say about other people. I'm guessing from the tidbits that I've heard here and there that I'll be largely sympathetic to what they say, but I can't speak to specifics.

3:37:27.2 SC: I'm gonna group two questions together. One is from Michael Honey, who says, looking around the world, we see many things going wrong, climate change, biodiversity loss, human conflict, but what if things go right? Imagine we could pull ourselves out of our current trajectory and make genuine sustainable progress on our major problems. What's your positive vision for the world, say a few hundred years time? And Paul Kohnhorst says, in pop culture it is much more common to see depictions of dystopias rather than the kind of scientific technical utopia you discussed with Benjamin Breen.

3:38:01.3 SC: This is unfortunate since the last thing we need is more cynicism and fatalism about our future. If you were asked to sketch out an idea for a series or movie which showed humanity using its intelligence to solve problems and attract a future, what would it be like? Well, let's take Paul's question first about the depictions of utopian optimistic societies in fiction. You know, they do happen. Star Trek is kind of an example. Iain Banks' Culture series is maybe an even better example. I'm sure there's other examples out there that I don't know about. There's a very basic issue, which is that it's hard to tell compelling stories about utopias compared to dystopias. Famously, when they brought back Star Trek to do the Next Generation, Gene Roddenberry, who was the creator of the original series, played a heavy role in that. And he said, you know, look, now we're doing Star Trek, but even further in the future, it should be even more of a utopia.

3:39:00.1 SC: There should never be disagreements between crew members on the ship. And the other writers in the studio are we need disagreements. That's what makes for stories. That's what makes for drama that people want to watch. So I think it's a very obvious reason why you see those kinds of depictions in fiction. In Iain Banks' stories, the way he does it is the culture is more or less a utopia, but the galaxy and the culture is a wide-ranging society with many species and many star systems involved, but it's not the whole galaxy. So there's many other species and individuals that are outside the culture that interact with it, and that's who he actually tells the stories about. So, what it would be like, yeah, I mean, let's answer Michael's question first, and then that will feed into it. What would the positive vision for the world be like? Look, I think that the sobering thing is that our productivity in terms of wealth and knowledge, vis-a-vis medical knowledge or food technology and agriculture and things like that, has just increased enormously over, let's say, the past 500 years, right?

3:40:18.0 SC: We've developed, we've created an enormous amount of wealth, we've created the possibility that as a race, as a species, human beings could make ourselves relatively comfortable. All of us, or almost all of us, like we have enough resources, that nobody should live in poverty, that nobody should not have access to education and health care and things like that. And we do, our, sorry, we do have those things, that is to say we have not organized society in such a way so as to eliminate those things. Why is that? And this is not, I don't want to hear any simple answers to this question because it's not so simple. If you think you have a simple answer, you're probably wrong. It's complicated; it's a lot going on. One could argue there's an inevitability. We were just talking about inevitabilities of social developments before. Maybe you could argue that society will always organize itself so as to be dramatically unequal and the worst off being very, very badly off, no matter how well-intentioned we are. I don't think that that's probably true. I think that's sort of a comforting thing we tell ourselves. But maybe it's natural that there are forces that push us in that direction because some people are greedy.

3:41:32.7 SC: If everyone had exactly the same amount of stuff, it would naturally dis-equilibrate very quickly, I think, because some people want more, other peoples are either less interested in having more or less good at keeping it, so inequality would develop. I'm not at all interested in perfect equality. I'm not arguing for spreading the wealth of the world completely equally throughout the world, but I do think there's a very good argument to be made that it would be a good thing to make the worse off members of the world much better off than they are now. So that's where I would put my effort into. I would not worry so much about the people who are well off and how to organize them, as I am very, very concerned with the people who are not very well off. And to go back to Paul's question then, that's my sci-fi future. I don't think it makes for good drama, necessarily, but there is always drama because people disagree with each other no matter what Gene Roddenberry might think. You could tell the story that's told in Pride and Prejudice, or Crime and Punishment, or Moby Dick, no matter how technologically advanced society was, human beings are just gonna be like that in various ways. So, I don't know what the specific answer to Paul's question is.

3:42:53.1 SC: What would a series or movie that showed humanity using its intelligence to solve problems and track the future would be like, but there's plenty of room for it in ways that are still going to be dramatic and interesting, set against a backdrop of true human accomplishment. I mean, Robert Heinlein's novels in an earlier era were exactly that, strong, competent people doing things. And Columbo was like that in some sense, but of course, Columbo was very competent, but he was showing all these terrible, terrible people who were wealthy doing terrible, terrible things. So again, that's where the drama comes from. That's okay. Tim Giannitsos says, gravity can be interpreted as an effect produced by the warping of space-time as opposed to being viewed as a force. Can the other three fundamental be viewed as a warping of some medium? Yes, yes indeed they can. Whether it is the most useful thing to do is less clear. The nice thing about gravity is that it's universal, right? The nice thing about gravity is everything feels it, everything causes it. So when we say that gravity is the geometry of space-time, not a force propagating on top of space-time, we really are referring to the uniqueness of gravity in precisely that way. There's not positively charged and negatively charged gravitational particles.

3:44:13.0 SC: Everything is charged in the same way. Other forces are not like that. But once we started understanding in let's say the '50s that you could think about forces of nature as gauge theories, not gonna explain exactly what that means, but guess what? You can read about it in Volume 2 of The Biggest Ideas in the Universe. A gauge theory has a very natural geometric interpretation. Let's put it this way, in general relativity, in our theory of gravity, there's a fundamental role played by what is called the connection. That is to say, if I have a vector at one point in space and I want to move it, keeping it parallel to itself, there's something called the connection that literally connects one point to another so I can move vectors back and forth. Same thing is true for the other forces of nature. The electron can be moved around using a connection in the electromagnetic field. The electromagnetic field in some sense is a connection and I don't know, yeah, it's gonna depend on what your background here is. There's something called the vector potential which gives rise to what you and I recognize as the electric field and the magnetic field and there's a perfect analogy, not, perfect is an exaggeration, there's a very good analogy with gravity and the connection.

3:45:32.3 SC: The vector potential is kind of like a connection between two different points. The electric and magnetic field are kind of like the curvature of that connection. So this is probably a very unhelpful, sorry it's late in the podcast, I'm not at my most crystal clear here, but things like the electric field and the magnetic field in gauge theories of fundamental particles do indeed have a very direct geometric interpretation as the warping of something, as the warping of the underlying connection field that is part of the definition of these gauge theories. For more you got to buy the book, sorry about that. Paul Conti says, many astronomers and astrobiologists believe that life arose relatively soon after the formation and cooling of the earth, although life on earth only represents a sample size of one, life remained as very primitive, bacteria, microbes, etcetera. For at least a couple of billion years before the sudden appearance of more complex multicellular organisms. This leads some to suggest that primitive microbial life might be quite widespread and even common through our galaxy, but complex life would remain relatively rare.

3:46:39.7 SC: I'm uncertain if this is what is known as the rare earth hypothesis, but what are your opinions on this idea regarding the presence of other life forms in our Milky Way galaxy? Well, I think that the second part of the hypothesis that it might be very, very difficult for primitive microbial single-celled prokaryotic life to develop into more complex forms, whether eukaryotic life, so you have a nucleus and a cell around it, or multicellular life, even more complex than that. These things came late, right? And so I would say that it is at least extremely plausible that those things are rare in the universe. Again, we are the aftermath of those things happening, so the fact that we're here, the fact that they happened in our past, is almost no information whatsoever, other than to say that they're possible. It doesn't speak that much to their probability, except that they're not zero, the probability. So whether or not it was easy to make the microbial life in the first place, I'm not sure that the fact that it happened early is that much evidence. It's a little bit of evidence, sure, but you know, we've looked at some other planets here in the solar system and we haven't seen any evidence yet for this microbial life, so it's not perfectly obvious that it appears everywhere very commonly. So again, I'm just gonna want to be humble about this.

3:48:07.6 SC: I don't think we know. I think it's completely plausible either that life is almost nowhere in our galaxy or that simple microbial life is all over the place but not more complex life or that somehow more complex life is also there. All of these are on the table. I don't think any one of them are ruled out by things we know here on Earth. We need more data. We need to know more examples of the development of life and we need a better theoretical understanding of how it actually happened. That would be enormously useful. Eamon McGee says, in the Big Bounce Theory, is there a continuation of anything from the previous universe? Well, the short answer is no, but the longer answer is there is no single agreed-upon thing called the Big Bounce Theory. There's a bunch of different scenarios that involve some kind of bouncing in the history of the universe. None of them is well established. None of them is especially likely, honestly, at this point. But typically, within those models, the actual bouncing point is extremely simple. And when a configuration is extremely simple, there's no way to convey a lot of information from them. There's a conveyance of, you know, matter and energy and maybe the laws of physics. Those things go through the bounce in some simple way, but no detailed information about what life was like in the previous universe.

3:49:31.0 SC: Again, in the simplest versions of these theories. Schleyer says, do you think there is a relationship between complexity and morality? There's a feeling of wrongness around the destruction of very complex systems, not just sentient systems like people and animals, but also things like ecosystems and culture, but I'm not sure why. I should have put this question earlier. This is a deep and good question, and I don't know is the short answer to it. I get why there is a feeling that there should be a relationship between complexity and morality because we think of us living creatures as value bearing, right? If you take a rock and you break it into many little pieces, you might feel different feelings about that, but you're not like upset because the rock's feelings are hurt or you destroyed some value in the configuration of the rock. I'm thinking of a non-precious rock. No one cares about this rock, okay? Whereas if you destroy some living being, you feel differently about it and that has to do, I think it's very natural, that that has to do with the complexity of the living being because that complexity is necessary for the living being to have thoughts and feelings and memories and goals and desires to whatever extent it does have those things.

3:50:51.7 SC: So I don't think it's surprising that there's naturally, intuitively, informally in our minds a relationship between complexity and value and therefore complexity and morality. But the reason why it's a really good question is, I think that there's a deep issue in how to extrapolate questions of morality from our very local, very tangible, very real world experiences to very different sets of circumstances. So in our experience, the other living beings that we know and love and care about are themselves complex. So maybe it's tempting to say, well, I'm gonna extend my care, my interest from the complex creatures that I call my friends and loved ones to all complex creatures everywhere or even more complex systems like you say, ecosystems and cultures. That's shaky ground. You know, that might be right. I'm not saying it's wrong by any means, but I'm saying that it's always dangerous to go from these particular instances that we're very comfortable with to extend them very far when we don't have a complete theory of everything, morally speaking. Okay. So I think these are interesting, open questions.

3:52:12.7 SC: I get where the feeling comes from, but I'm not gonna claim to have a full theory of how to think about these issues. Raj says, my understanding is that the arrow of time is emerging from entropy. Is the arrow of time different from time used in physics calculations? The corollary to this question is, when you say Schrödinger's equations evolving with time, do you mean evolving with the evolution of entropy at a deeper level? So good. I should have, again, I should have grouped this with the earlier question 'cause I partially answered it before. The arrow of time is completely different from time. It is the difference between a house and the color that you have painted your house. The arrow is a property that the passage of time has due to the configuration of matter in our universe. It is not a deep down fundamental feature of time itself. In particular, in the Schrödinger equation, just like in Newton's laws or Hamilton's equations or whatever, there is no arrow of time. There's no directionality to time built into any of these equations. So when we say in the Schrödinger equation it describes the evolution of the Wave Function with time, we mean that. We don't mean anything about entropy.

3:53:29.4 SC: Entropy doesn't even exist or is not a useful concept when you have a single atom or a single electron where the Schrödinger equation is extremely useful in talking about what happens. So keep, separate the arrow of time from the fundamental nature of time itself. And then the last question for today's AMA, for this month's AMA, comes from Stevie CPW who says, could you please provide a cocktail party level explanation of why the many worlds interpretation is more plausible than the belief in God? Yes I can. The answer is that the many worlds interpretation simply consists of taking the Schrödinger equation for which we have ample evidence of its empirical validity and believing it, taking all of its predictions seriously, saying the Schrödinger equation is all there is, believing what it says, and what it says is there are gonna be many worlds. The many worlds interpretation is just a matter of having a belief, a credence, that the equation that works in conditions where we do see it work continues to work in other conditions as well.

3:54:39.3 SC: God is not based on any equations at all. He is quite the opposite. He is based on very different ideas. There is literally no useful comparison between these two ideas. And with that we've reached the end of today's Ask Me Anything. Many, many thanks as always to all the Patreon supporters who help keep Mindscape going. Especially extra thanks to everyone who donated this year to the scholarship fund. Congratulations to Ryan Funakoshi for being the winner of the scholarship this year. Hope to see great things from you in the future, Ryan. Take care everyone. I'll see you next week with another podcast, next month with another AMA. Bye-bye.

[music]