Harvard Physicists Expose Fatal Flaws in Many Worlds

I remember feeling quite shocked and upset about that. At each moment, the universe splits. There is no such thing as personal identity over time. In this staggering and intense conversation, Professor Emily Adlam and Harvard's Jacob Barandes dismantle the many-worlds interpretation by arguing that its promised elegance has been broken. Even further, without a process that selects which quantum branch you experience, you could believe almost anything about which future you you'll become. The probabilistic math gives no guidance. As this conversation progresses, questions about what it means to be you in general are explored, such as whether the

you that wakes up tomorrow is really you at all. This conversation culminates with explorations of consciousness and their passionate defense of philosophy in physics. Should physicists avoid overly philosophical questions? Well, these questions led to quantum computing, Bell's theorem, and decoherence theory, as they reveal many major physics breakthroughs involved Philosophers and physicists working together. Today, we need that collaboration again. Enjoy this Theolocution with Emily Adlam and Jacob Barandes. Alright, what I'd like to know is what do you admire most about one another? It's super easy. So I'll keep it short for a change. I think Emily

is one of the most creative and rigorous thinkers in philosophy of physics today. I am profoundly grateful that we're colleagues and to be more specific I think that Emily's analysis of deep questions at the heart of our best physical theories is second to none. And Emily also has a remarkable ability to synthesize complicated questions and formulate very cogent answers remarkably succinctly, a skill that I do not possess and admire enormously. Emily? Yes. Jacob is a very precise thinker. He wants to understand every step of every argument. He does not accept fluffy thinking or intuition or

any of these things. He has a very precise mind in terms of working out where the steps go and where they should end up. He's also very creative. He has this very, very fascinating new approach to quantum mechanics that I think resonates with a Lot of things that I have thought about for a long time but not known how to sort of formalize. And Jacob is doing it. He's showing how to make it formal and how to develop it into a real framework that you can use to do calculations. So that is really exciting to

me. And I think that sort of seeing the way he's worked through that and has turned that into something that is really workable and viable is really exciting. What would be an example of sloppy thinking that you didn't realize was sloppy thinking but after speaking with Jacob you said, okay, there's a hole in my argument here or I was vague and I didn't mean to be. Oh, good. Yes, well, I think we've talked about the Everett interpretation quite a lot and discussed, I think both of us have some worries about sloppy thinking that goes into

some kinds of arguments in that context. And some of the things that Jacob has brought up have been things that I hadn't thought a lot about. I think particularly some of your ideas about looking at how the theory can be reformulated in ways that make the wave function representation look very different and much less compelling. It Has been really helpful to start thinking about the way in which so much is being taken from one particular mathematical representation of the theory. For example, thinking about what a sort of zero mod squared amplitude means in the representation.

I think I'd always just sort of accepted the thing the Everettians say, which is that zero mod squared amplitude, it's not there, I don't have to worry about it. And Jacob has this nice argument to sort of point out that that doesn't necessarily follow that you should perhaps be more careful with that. So that has really given me a new way of thinking about that. If it's all right, may I elaborate a little bit on that point? So there's a reformulation of, so just stepping back for a moment, the standard axiomatic formulation of quantum theory

that you find in the textbooks features some mathematical parts and some parts that are connected with physical reality in the sense of connecting to what we see, obtain, measure in experiments, the empirical output of the theory. The Hilbert space axioms or the mathematical axioms and the measurement axioms are the ones that connect this mathematical Framework with empirical reality. The Everett approach claims to be able to do away with the measurement axioms. The measurement axioms suffer from the measurement problem, the ambiguity over exactly which processes count as measurements, and therefore leading to an ambiguity about when

to apply the measurement axioms. The Everett approach attempts to remove the measurement axioms and somehow build a sensible, empirically adequate, meaning capable of accounting for the things we see in all of our experiments, to build an empirically adequate theory or interpretation out of the mathematical axioms. There are many things that we take for granted when we look at those axioms as they are usually described. The axioms talk about a quantum state. The quantum state captures what we can say or know about quantum systems. In the Everett approach, one is supposed to take this quantum state

and notice that as it evolves through time, it gradually develops these roughly well-defined, macroscopic, big-picture, real-world realities. We call them branches. And over time, the branches start to become independent of each other. They don't interact Or interfere with each other. And in that stage, we're supposed to think of them in the many-worlds framework as representing distinct ways the macroscopic world can be. One is... supposed to just regard as intuitive that when one of the branches has an overall magnitude that's zero, in the sense that the number sitting in front of it in the branch is

zero, well then intuitively it's not there anymore. The only branches that are there are the ones that have numbers in front of them that are not zero. Those numbers are called amplitudes. The measurement axioms say that you take those numbers and you square them and they give you probabilities of measurement outcomes, but we can't do that in the Everett approach. We've gotten rid of the measurement axioms. We don't want to rely on an ill-defined notion of what counts as a measurement. So one of the things that the advocate of the Everett approach needs to do

is justify why those numbers in front of the branches that show up in this mathematical quantum state should be understood in a probabilistic way. One of the implicit assumptions is, like I said, that when one of those numbers is zero, We can ignore it. It disappears. It doesn't exist. It doesn't have any physical meaning anymore in some sense. The problem is that if one is just relying on these mathematical Hilbert space axioms that just talk about this abstract quantum state evolving in time, it turns out that this mathematical picture is, the word we use in

mathematics is isomorphic. Isomorphic just means there is a mathematical connection between this way of formulating the picture and another seemingly different looking way of formulating the picture, but they are mathematically equivalent. There is, for everything you can say in one picture, there is something you can say uniquely in the other picture and vice versa. They're connected by this mathematical bridge and both give a completely comprehensive, adequate way to formulate the mathematics. This other way of formulating the Hilbert space axioms is due to a number of people. It seems to have been developed independently by a

bunch of people. Franco Strocci worked on it in the 1960s. Andre Heslott worked on it in the 1980s. It turns out you can take this abstract Hilbert space picture and write It in a mathematical form that looks just like a collection of classical springs or pendulums. We call them harmonic oscillators, just systems that sort of swing back and forth. And everything you can say in the Hilbert space picture, you can say in this other alternative picture. Now, technically, this is easiest to do when the Hilbert space involves finite dimensional systems. That's a technicality. People have

extended it beyond the finite dimensional case, but it's easy to see in the finite dimensional case. This other picture lets you tell the same mathematical story in terms of springs that oscillate. And then in this other picture, for one of the numbers in front of a branch to be zero is to say that the spring is not oscillating. Now, a spring that's not oscillating is still a spring. It's still there. We wouldn't look at a collection of springs and say that only the springs that are going back and forth exist and the ones that are

not going back and forth don't exist. So in this equivalent mathematical picture, it's no longer obvious that zero amplitude means does not exist. And because there's Nothing about the Hilbert space axioms that fixes one mathematical representation over another, and if in that other mathematical representation, zero amplitude doesn't correspond to lack of existence, it's no longer obvious that we should throw away branches at a zero amplitude. Throwing away branches at zero amplitude plays an important role in making sense of the Everettian world picture. And so this alternative representation challenges that view. Okay. So some background is

that you all were just at a conference about why you're not Everettians. And I imagine an argument like this was raised, maybe not, maybe so. If so, what was the argument then raised to object to your objection by some of the Everettians in the audience? We didn't actually have enough time to get to the response to this particular objection, but have you presented this to Everettians before? Do you have a sense of what their response would be? I have not. No. I think this is the first time I've presented in a talk. Okay. Yes. And

I lack Amelie's succinctness, so there wasn't adequate time for this to come up in the discussion. So we're yet to find out what the Everettians are Going to say about this. Okay. If there are any Everettians watching this discussion, they should let us know what you think. So what else do you all agree on that puts you at odds with many of your colleagues? So there's plenty that you all will agree on that is just non-substantive, like math is useful for physics. Yes. Okay. No one's hunting you down for that. What is it that you

agree on that's contentious? Something that comes to mind is our views on self-location and the relevance of that. There is this very common idea that there are certain kinds of self-locating scenarios in which there are natural ways and correct ways to think and to hold beliefs. So scenarios where we're going to make 10 copies of you, 10 clones. We'll put nine of them in rooms with blue doors and one of them in a room with a red door. And then we're going to ask you, what is the probability that you're in the blue door room?

So I think most people have this quite strong intuition that it should be nine-tenths or something. And my view is that there's no rational constraints on what your credence should be in that situation, that you're allowed to assign Probabilities any way you want. It has no real significance. It has no practical meaning. You can do what you like. And that is a view that a lot of people really strongly resist. I get a lot of pushback when I present on this, but Jacob, I think, immediately understood where I was coming from and had very similar

views. And so that is something where I think it's important that we all agree on that. That is something where I think we're perhaps against the flow of the current overall, but quite strongly aligned on that one. Yeah. So many years ago, I was thinking along similar lines, maybe a brief just digression to talk about probability, credence, chance. So Ian Hacking described probability as a Janus-faced entity, that Janus was the Greek god of beginnings and endings, gave his name to January, right, the month that is the beginning and ending of the calendar year. On the

one hand, we use probability to talk about chance, to talk about how likely we think that a die or a coin that's tossed will show a particular value. Chance probabilities are sometimes called objective probabilities, sometimes they're called aleatory Probabilities. On the other hand, we use the language of probability to talk about how much we believe something may come to pass or how much we believe it's true. This is known as credence, like creed belief, credence probability. Some people call it Bayesian probability, named after Bayes, who centuries ago worked on the structure of probability and developed

a framework that people who use probability in this credence sense use to try to formulate statements about what they think is going to happen in the world. So you have these two different notions of probability, and when you have a physical theory that describes processes by which phenomena take place, and the physical theory is not able to tell you definitively exactly what outcome will happen, often these physical theories will tell you that what you'll expect to see is a pattern of phenomena, and this pattern of phenomena will be random looking but maybe with some bias

toward one thing or another, and so we say that this is a theory, a physical theory that's describing chance probabilities in some sense. If you have a theory like Everettian quantum mechanics, and Everettian quantum Mechanics is not unique in this regard, in which the overall picture provided by the theory is that the overall process is deterministic, there's no fundamental probabilities, no chance probabilities really happening, but maybe the theory predicts there will be many observers, maybe each observer will appear many times as copies, as clones. Then one way to think about probability in that case is

maybe we can bring probability in at the level of credences, of beliefs, and we use this kind of reasoning all the time. Even outside these exotic circumstances, when I want to talk about how much I believe that my favorite sports team is going to win a particular championship game, there is maybe one fact of the matter about what's going to happen, but I don't know what it is, and maybe the best way to describe in a probabilistic way how strongly I believe my team is going to win is to use this language of credences. Often

when we use language, the language of probability, whether we're talking about objective chance or we're talking about credence type belief probability, we often are attaching those probabilities to statements, statements that may or may not Involve people like Jacob or like Emily or like Curt. And sometimes we'll find it convenient not to say Jacob or Emily or Curt, but to say words like I or you or us. Words like I, you, me, us, these are called indexical words. Like an index, like a thing identifying a particular thing out of a set of things. Indexical words have

a meaning that depends on who is saying it. When Jacob says I, I mean something different. From when Emily says the word I or when Curt says the word I. If you think about it, it's kind of hard to imagine how we acquire a facility with indexicals because it's very hard to explain the use of indexicals to a small child. That's true, yeah. I think we just sort of watch people use indexicals and we, as small children, and we develop a facility for them just sort of by observing how they're used in practice. Sometimes we

use indexicals in statements we attach probabilities to. So for example, we might say, and now I'm borrowing an example from Emily. I think I'm going to wake up at 9 a.m. maybe. How likely, what probability would I attach to the proposition, the statement that I'm going to Wake up at 9 a.m.? We could just as well, or I, me speaking, could just as well have said, what's the probability that Jacob Barandess will wake up at 9 a.m.? Perhaps I have removed the indexical and replaced it with a non-indexical, Jacob Barandess. I want to be more

specific, although there are no other Jacob Barandesses. I'm sui generis. But if there were more than one, I could say something like, the Jacob Barandess who was born in New York City, and I could add further details until I specify that Jacob Barandess uniquely. And then I wouldn't need a use of indexicals. Now, specifying that long list of details is a lot of effort. It's just easier to say I. But the use of I is not fundamental. It's just a shorthand, at least in this case. In much of the science that we do, our use

of indexical expressions like I, us, we, you, can be replaced with third-person objective descriptors, Jacob Barandess, Curt Jaimungal, Emily Adlam, maybe with further details as needed. So we have abundant experience with such uses of probability in science and beyond science. What bothered me about the use of probabilistic statements, generally of a credence kind, A belief, degree of belief kind, that's the kind that usually comes up in these circumstances, in scenarios where there are many copies of ourselves, maybe infinitely many copies of ourselves, maybe in a universe that's infinitely big in which there are infinitely many

Earths that all look like our Earth, populated by people who look like the people here at this table, and with Jacob Barandesses. Or in the Everettian world picture, where the universe is splitting and there are many, many copies of all of us, is that now to make probabilistic assertions, to attach probabilities to statements, it seems like you have to use indexicals in a much more fundamental, irreducible way. Now, other people have worked on this notion of irreducible indexicals before me. And there's an interesting philosophy literature that I encourage people to read about. It's very, very

interesting literature. But I was just very uncomfortable with trying to engage in rigorous probabilistic reasoning about statements that feature indexicals in an irremovable, irreducible way. If there's a universe in which there are infinitely many copies of observers who Look just like me, who are named Jacob Barandess, who live on planets called Earth, who have all the same details we talked about, they were born in a city called New York City, everything is the same. If I want to ask, which Jacob Barandess am I? We now run into kind of a tricky question. I can't replace

I with Jacob Barandess anymore because then the statement reduces to a tautology. Which Jacob Barandess is Jacob Barandess? No, wait. Which Jacob Barandess who was born in New York City is Jacob Barandess who was born in New York City? You see that there's something qualitatively different about those kinds of statements. We can't remove the indexicals no matter how hard we try. And it wasn't clear to me that anything we've learned about the use of probability for statements that don't involve indexicals in an irremovable way, it didn't seem to me that anything we'd learned, any experiences

we had using probabilities in those kinds of situations could tell us anything reliable about how to use probability in situations that involved irremovable indexicals. They just seemed to me to be fundamentally different kinds of statements. And I didn't, I gave this some thought, but I never really formulated anything cogent. And then I saw a paper by Emily, and at one time I was a little upset that I'd been scooped. But on the other hand, I couldn't think of anyone I was happy to be scooped by. And Emily's paper on this, again, self-location paper, which I

highly recommend that people read because it is like all of Emily's papers, brilliant, but also very engaging and a wonderful thing to read. Emily paints this dichotomy much more clearly than I was able to and gives really compelling arguments for why these two kinds of probabilities. Here I don't mean objective chance versus credence, but probabilities about statements that involve indexicals in a fundamental way and those that don't, or as Emily puts it, statements that involve pure self-locating uncertainty. These are the kinds that involve indexicals you can't remove, at least the way that usually I would

talk about it, and what Emily calls superficial self-locating uncertainty. And your dichotomy between these and your explanation of the difference between them, And then your arguments about why pure self-locating uncertainty, there's much less structure that we can impose on it, was just a brilliant argument. And I want to stop talking because I think Emily will say much more clearly what was going on in that paper. One of the most overlooked variables in cognitive performance is hydration, especially the kind that actually supports your brain. That's why I use Element, spelled L-M-N-T. I drink it first thing

in the morning, far before any coffee. It's a zero-sugar electrolyte drink mix with science-backed levels of sodium, potassium, and magnesium. There's no fillers. There's no BS. Element was co-founded by Rob Wolf, a former research biochemist and a New York Times best-selling author. It's trusted by Navy SEALs, pro athletes, and Stanford researchers. But most importantly, it works. I use it while fasting. Many people don't know, but I fast prior to podcasts. So technically speaking, I use it to keep me sharp during these rigorous interviews. I also use it for the preparation, the deep research sessions. And

currently, in Toronto, you may not know, there's a heat wave, Thus this is an indispensable tool. Plenty of people in the TOE audience, whether you're into low-carb or whole-food diets, or just want some sharper focus, they've found that it's made a noticeable difference. Right now, Element is offering Theories of Everything listeners a free 8-count sample pack with any purchase. Try all the flavors. I gave my sample pack to my in-laws, and they love it. My sister-in-law in particular loved the flavor, as sometimes water alone doesn't cut it for hydration. Go to www.drinkelement.com slash theories. That's

www.drinkelement.com slash theories. It's totally risk-free. If you don't like it, they'll refund you. No questions asked. Visit my link at www.drinkelement.com slash theories. And I want to stop talking because I think, you know, Emily will say much more clearly what was going on in that paper. Yes, sure. I mean, one of the main things the paper is doing is making this distinction between the pure and superficial cases, which, I mean, Jacob puts it in terms of irreducible and dixicals. I think the way I described it in the paper was in terms of whether the possible

locations you're considering Belong to the same possible world or different possible worlds. So with the waking up in the morning case, in every possible world, there's exactly one time at which I actually wake up in the morning. And so that means that all of the different times I'm considering when I'm wondering what time is it? When did I wake up in the morning? Those are all different possible worlds, whereas with the cloning kinds of cases where you're just making 10 clones, and you could be any of these clones, those clones all exist in the same

possible world. So that case where they all exist in the same possible world is what I call the pure self-locating uncertainty. And I think that these types of self-locating uncertainty are often kind of treated as the same in discussions of this topic, but they are quite different because in the superficial case where it's different possible worlds, ultimately that's just a question about which possible world is actual. That's the kind of question we ask in science all the time, and we can address that by going to some scientific theory and getting probabilities from it. Whereas in

the cloning case, there's no sort Of process which picks which clone you are, you just are whoever you are. So there's never going to be, you can't sort of go to a scientific theory and look at a description of a process and try to get probabilities from that. So wherever those probabilities come from, or those credences that you're assigning, they have to come from somewhere quite different than how they would come from in the superficial case. So that's why I think it's important to see those as different types of cases, because the reasoning for those

cases is going to be quite different either way. And then the second part of the paper was about trying to understand for these pure cases, can there ever be any way of assigning credences which you are sort of rationally obliged to take, or which is at least rationally convincing or compelling? And I sort of went through some reasons you might try to say that there are rationally compelling ways of doing this, and argued that none of them succeed. So in particular, one thing we'll often do in these kinds of cases is try to make an

argument that if you imagine sort of making bets about where you're located in the world, Then certain strategies will be more successful. And so those are the ones that you should have. But the problem with that approach is that before you can say the strategy is more successful, you have to define what counts as success. And so in the self-location case, that's going to look like saying something like, I want to maximise the winnings over this whole set of individuals, or I just want this one to win and the rest, I don't care about them.

Some choice like that has to be made. And what you can see is that as soon as you've made that choice, that just immediately dictates what the credences are that you should have. If you want to maximise winnings over all of the observers, that means you should assign equal probability or credence to all of them. If you want to just make this person over here win, then you should assign probability one to this person and zero to the rest of them. So in this case, there's kind of no room between what your goal is and

what your credences should be. So for me, what that tells us is that these credences aren't really beliefs, they're not really probabilities. What they are is just a sort of measure of a way Of expressing what your practical goal is or how much you care about these different observers. And insofar as you sort of agree that what we care about for what our goals are isn't dictated by rationality, it looks like there's not going to be any sort of rationally compelling way that you have to proceed in those cases. And that I think sort of

is problematic for a variety of ways in which these things are applied. Self-locating credences are applied quite widely in physics and philosophy to cosmological multiverses, to the Everettian multiverse, to questions about Boltzmann brains, to questions about the simulation argument. All of these things, I think, depend quite sensitively on the assumption that there's some correct way of assigning your credences in these situations. So I think once you buy this kind of argument, you're going to say, well, maybe there's not, in fact, a rationally compelling way to do this. And therefore, we should be more cautious about

how we're using these things. This is great. The setup. So we're on a ping-pong table. And I like the fact that we're on the same side of the ping-pong Table. That seems somehow fitting. That's dispropriate. Yes. Okay. So two questions. One, you use the word possible world. And earlier, we were speaking about Everettian many worlds. Okay. So I assume that's a subset of all the possible worlds. So does your argument in that paper, does it work for what sorts of possible worlds? Great. Yes. So, I mean, the way I think about it, I'm inclined to

say that one Everettian multiverse is a single possible world. And therefore, the various different locations you could be within the branch in universe, that's a case of pure self-locating uncertainty. And I think that's right because the sort of differentiating factor between pure self-locating uncertainty and superficial self-locating uncertainty is that in the pure case, there's no process which decides what the outcome is. There's just all of the outcomes just exist, and there's nothing that picks which one is right. That's kind of what happens in the Everettian multiverse. In an Everettian measurement or something, all of the

outcomes occur. You will find yourself in one of those branches, but there's no specific physical Process which picks which one you're in. You're just in whatever one you're in. So that's why it makes sense to think that that should be a case of pure self-locating uncertainty and not superficial self-locating uncertainty. If I can add a couple of things. So this language of possible worlds may not be familiar to all your listeners. And I think also just for someone who hasn't spent a lot of time thinking about all of this, there's a lot of intuition I

think that one brings to these sorts of questions. This is exactly what Emily's been challenging. We have a lot of intuition from instinct, experience, evolutionary programming about how probabilities work in the case of, as Emily puts it, superficial self-locating uncertainty, the kind of probability we attach to statements that we can describe in objective third-person terms without the use of, you know, the need to use fundamental indexicals in my language. We have a lot of intuition about it, and so it just seems very natural, very natural to take that intuition we've developed and extend it to

these other kinds of probabilistic statements, the pure self-locating statements, Especially because we often use the same language for both kinds of statements. We often use indexicals in both kinds of statements. It's just that because we're not aware always that the indexicals we're using in one set of statements are not as important as they are in the case of pure self-locating uncertainty. So part of this is just Emily's challenging the intuition that people often appeal to. They'll give an example. They'll say something like, well, how do we fix the kinds of probabilities we want to use

in a multiverse scenario with lots of copies of ourselves? Well, consider this following non-multiverse scenario, you know, in which we have experience with probability, we know how it's supposed to work, and then extend that intuition to this other circumstance. But there really is a qualitative separation here. You know, there are a lot of problems in philosophy that you can call sororities problems or heap problems. How do you decide when you're, you know, when you switch from one thing to another? Often there are smooth gradations between them, right? How many strands of straw do you need?

To have a heap of straw. One Strand is not a heap. Two is not a heap. Three is not a heap. A billion, we would call that a heap, maybe a very big heap. But like, where's the line between them? A lot of problems in philosophy are like that, where there's no sharp line. This is actually one of the circumstances in which I think there is a bright line. And remarkably, despite the fact that there's a bright line, it's very easy not to see it. Wait, a bright line is the same as a sharp line?

Yes, a sharp, a sharp, right. Yeah, exactly. There's a sharp demarcation, you know. So for example, what is science? What is pseudoscience? This is, this is called demarcation problem in the philosophy of science. That's a hard problem. Where exactly is the line between what counts as science and what doesn't? People have been debating this for a very long time. So many of the problems one finds in philosophy are like this, where there's no sharp dividing line between things that demarcates what's on one side and what's on the other. This is, like I said, one of

the cases in which I think if you look really carefully, you'll see there is a very bright line. Some statements Are about where we are in one possible world, and the other is about which possible world we're in. And these are just different kinds of statements. This use of possible worlds, by the way, in logic, just to keep things simple, a possible world plays a formal role in formulating certain kinds of logical statements. If you want to ask whether a certain statement or proposition is true or false, well, it may depend on various circumstances. Maybe

a certain proposition is true, given some circumstances, and false, given other circumstances. If you want to assign all well-formed, reasonable propositions true or false values, then we say what you've done is you've picked out one possible world. A possible world is just the assignment of true or false to all sensible propositions. And if any of those propositions change from true to false or vice versa, we say that we're talking about a different possible world. Intuitively, we're supposed to imagine that possible worlds are like complete realities, complete ways that all of the world across all of

space and all of time could be. And once it's all been fixed, Then all sensible propositions have a well-defined value. And one of the jobs we're trying to do in science is kind of figure out which possible world is the world that we inhabit by learning which propositions we think are true and which are false. Even if you've established which possible world you think is the actual world, there's still this further question about where we are in it. The notion of a centered world captures this idea that in addition to saying which possible world is

the actual world, there's this further question about what time we're talking about, where we're talking about, which individual we're talking about. And if you're talking about a possible world that's some kind of multiverse filled with many copies of ourselves, then which copy I am is not a statement about which world is the true world. That's been established now, I suppose. It's a question about where in that world I am, which copy I am. And that's now a question of pure self-locating uncertainty. It's important to distinguish this notion of possible worlds from the worlds of many

worlds. In the many worlds interpretation of quantum theory, We imagine we have this universal wave function that is branching into these macroscopic realities that are not interfering with each other. When you give them enough time, they become sort of independent realities. But they're all connected to each other. They're not really separate worlds in a fundamental sense. There's just one world in this picture, one actual world. There's one universal wave function. And it's just convenient for humans to think of this in this sort of branched way to make sense of this reality. So which branch you're

on is not a question of which possible world you're in. It's, as Emily put it, a question of where you are in this single possible world. And so it becomes a question of pure self-locating uncertainty in Emily's language. It's just that we use these words like worlds in so many different ways, and it's very easy to get confused about their different meanings. The last thing I'll just say is, when people first approach a multiverse theory, whether it's a cosmic multiverse, an actually big, big space-time with lots of regions, maybe it's so big that Earth has

occurred many times with People on it like us, or maybe you're talking about a multiverse in the sense of a branching Everettian multiverse, I think we all have this intuition. Maybe we don't state it. As Emily puts it, we think we have a kind of Cartesian ego. There's a kind of essence that it is to be this particular Jacob Berendes, or that particular Emily Adlam, or that particular Curt Jaimungal. And that essence is like a karaoke ball, you know, on a particular word in a karaoke song that identifies this is the word we're talking about.

This is the specific Jacob, or Curt, or Emily that we're talking about. And in an Everettian picture in which the universe is splitting, I think a lot of us have this intuition that this karaoke ball, this Cartesian ego, is hopping. At each moment, the universe splits. There's now many, many, maybe uncountably many copies of us, and this Cartesian ego, stochastically or probabilistically, jumps from the past pre-branching self to one of the post-branching selves in kind of a random way. And this hopping is happening as the universe is branching, and this is where the probabilities come

from. But notice this is a process now. It's a physical, metaphysical process of something actually hopping according to some rules. And if that were the case, if we believe that were the case, then we would have a process to which we could begin to assign something like probabilities in a more well-defined sense. We could take indexicals and say, no indexicals, take the Cartesian ego, the unique Cartesian ego, and ask which Jacob Barandess it inhabits. Now this becomes a statement that doesn't involve, you know, just, it isn't just a statement that involves this irremovable, irreducible indexicality.

But... Most advocates of Everettian quantum theory don't believe in a Cartesian ego. And it's important that people who are bringing that Cartesian ego intuition with them into the Everett approach, that they should be aware that this is not how the Everett approach is formulated. Now, I think it's an interesting question, could you formulate the Everett approach with something like a hopping Cartesian ego that is really choosing between observers in some probabilistic way, which we can talk about probabilities. That's an interesting question. But without it, you can't appeal to the kind of intuition that we take

from a hopping Cartesian ego, and you're stuck in the situation Emily was talking about. There's no process, there's no choosing going on. And so we can't appeal to the intuition we have in that case in order to make sense of probability in these circumstances. I think, does this capture... Yeah, right, exactly. It's the lack of any selection process, which makes it difficult to see how you could have a sort of physically grounded notion of what your credence should be. And thus, in the absence of such a thing, it looks like anything should be rationally permissible.

Now, earlier when you were speaking about the possible worlds, and you were saying it's difficult to rationally say what is the best strategy, you're able to pick out different strategies, say strategy A, B, and C, and say that it's either inconsistent or what have you, so it doesn't work. Are you able to show that any strategy would produce some inconsistency, or can you only say that strategy doesn't work, that one doesn't work, that one doesn't work? Yeah, so what you can do in a self-locating case is you're Imagining a scenario in which you have, say,

your 10 observers in rooms with different coloured doors, and you're asked, they're each supposed to place a bet on the colour of their door, and the question is, what should you bet, given that you don't know which one you are? And I think many people think, well, you should sort of bet with nine-tenths odds for blue and one-tenths for red. And that is the right thing to do if what you want to do is maximise the sum of winnings over everybody, if you want to have the highest total winnings across all of the observers weighted

equally. But you don't have to do that. There's no sort of rational obligation to try to maximise the sum of winnings over all observers. You could say, actually, I prefer for the guy in the blue room to win. And in that case, the right thing to do in that case is to bet everyone should bet blue. And so the point I'm sort of trying to make here is that before you can decide what is a sensible bet to place in that situation, you have to make a choice about what am I trying to achieve, who

would I like to win the bet. And as soon as you've done that, that will immediately fix what The credences are that you're supposed to have. That will immediately determine, if you wanted to maximise winnings over everybody, that means the credences should be equal across everybody. If you want to maximise winnings for the blue person, assign probability one to that person. And this is different to the non-self-location case, right? For example, if we do a sort of series of coin flips and I'm placing bets on that, I can say, well, my goal is to maximise

my winnings over all of the bets. And then there's still a further question after that, what will in fact maximise my winnings? What odds should I bet? And we can only answer that by empirically testing the coin and finding out about its properties and understanding that process. There's no step like that in the self-locating case. As soon as you've specified what you're trying to achieve, that immediately fixes what you should bet. There's no sort of empirical considerations or theoretical considerations that could possibly be relevant. Your goal just immediately dictates what your bet is. So there's

kind of no room for sort of rationality to play any role here. All that's Really happening is that you're making a decision based on your personal values and goals about what you would like to achieve, and then that will just immediately fix all the bets. I think we have this intuition and one finds this intuition expressed frequently in the research literature on these kinds of problems, that obviously we want to maximise the success or utility of the greatest number of our future copies. That this is somehow obvious or intuitive or self-evident. And it just isn't.

And I think one of the things that Emily has done very nicely is highlight the difficult to justify assumptions that go into that reasoning. And I think that reasoning we want to maximise for as many people, I think it comes from this intuition we have that we have this Cartesian ego hopping around randomly. And if it's hopping around randomly and picking one of these people at random, we would want something like uniform utility across all of them. But again, without something like this hopping ego, sorry Descartes, this ego hopping around, it's actually very difficult to

justify that. So are there Everettians who believe in the Cartesian Ego? There is the many minds interpretation, which I think is a little bit like that, except instead of one hopping ego, there's a whole bunch of hopping egos and you could be one of the hopping egos and you're going to go into different branches. And I think, in fairness, I think that does solve some of the problems that I have with the Everett interpretation. I think you can sort of straightforwardly understand what probability is and how we get predictions out of that. So that addresses

some problems. It introduces a lot of other problems though, because one of the main selling points of the Everett interpretation is supposed to be that you just have the standard formalism. You don't have to add any extra structure. You don't have to pick out a preferred basis. You don't have to have any additions. You just have to have this underlying simple quantum mechanical structure. So as soon as you add in either one hopping Cartesian ego or a bunch of hopping Cartesian egos, you can't make that claim. You're no longer just using the simple formalism. And

perhaps the most problematic thing at that point is that Once you've added in some number of hopping egos or something like that, it becomes a bit unclear why we have a many world scenario at all. If you've added the structure, which sort of picks out well-defined pathways through the worlds, but let's just get rid of all the worlds and have just one pathway through the world. And that's what actually happens. So I think it's quite important for the sort of Everettian dialectic to resist the idea that you should add something to sort of single out

probabilities in this way, because as soon as you do that, it sort of seems like the argument for having multiple worlds at all just goes away. Yeah. And if you have just one sort of funny looking trajectory, perhaps an indivisibly stochastic trajectory. Right. Yes. You might end up with something like that. Perhaps you have an alternative approach to quantum theory. The indivisible stochastic approach, which is one of the things that we've talked about. The many minds interpretation goes back to David Albert and Barry Lower. And if people are interested, you should read about it. It's

very interesting. I just spoke to Barry, But I don't know if he's a proponent any longer of the many minds interpretation or version of many worlds. Is he? I'm not going to speak for Barry. That's the thing I've learned long enough to do. Barry is wonderful. But he should definitely speak for himself on this matter. Yeah. So I'd like to tease out some of the differences between you all. And we've spoken about some of the agreements. But since we're on the same physical side of this ping pong table, why don't you, Jacob, instead of stating

a disagreement, somehow phrase it in the form of a question for Emily? Do we have any disagreements at the level of philosophy of physics? It's the personal identity teleportation. But that's more of like a philosophy of mind metaphysical problems. Yes. So we differ on the teleportation problem. Emily would step on the teleporter. I would not. Emily, why would you step in the teleporter? Okay. So my thinking about this is in some ways quite similar to thinking about self-location. I think that a lot of people have an idea of personal identity as a sort of well-defined

thing that you can track through time. And there are very well-defined Facts about I'm going to, I'm this person now, and I'm going to turn into this other person in the future and so on and so forth. And so in particular, you can have scenarios where you're imagining sort of various kinds of splitting and branching cases. And I think many people have an intuition. There's got to be some fact of the matter about which one of these teacher people is me. There's got to be some sort of extra fact which singles out who I am.

And for much of the same reasons as in the self-location case, I don't think there are any such facts. I think that there are the physical facts about there's this body which is causally related to this other body in the future. And usually, this body is going to be causally related to only one body at each future time. Maybe in some particular strange scenarios, it could be causally related to two bodies. But I think once you've sort of laid out all of those physical facts about where the bodies are and what they're doing, there's nothing

further to be said. There's no further question of personal identity or whether this one is really me. So for that reason, I think that my relationship to my Future self in the ordinary course of events is not really significantly different to my relation to my future self if I'm to go through this teleporter. If I go through the teleporter, the idea is that I'm going to be dissolved and reconstituted somewhere else. That reconstituted version is still going to be causally related to this body in pretty similar ways. It's going to have all the same memories.

It's going to have all the same kind of psychological continuity. So it seems to me that that reconstituted version of me in the future is me in just the same sense as I'm going to be me tomorrow in the ordinary course of events. I don't think there's any deep and principled distinction between those two cases. And so if you're not afraid of going to sleep and waking up tomorrow, then you also shouldn't be afraid of going into the teleporter because your relation between yourselves at those times seems to me to be pretty much the same.

There are many variations on this teleporter problem. Yeah. Like let's say you thought it was going to teleport you, but it cloned you instead. Okay. Or you thought it was going to teleport you, But then it killed you or created a clone and then killed you. Yes. So no matter what, you'd go through the teleporter and you would still feel like it's you at the other end of the teleporter? Well, I guess it's this idea that there's a sort of fact of the matter about whether it's me or not is what I kind of want

to deny. I think that once you have described what is going to happen in this situation, that's all that you can do. So you have set out your story, you have said this will kill you and will create a similar copy at the other end. There's no further fact about whether that is me or is not. But in much the same way, there's also no deep fact about Emily tomorrow being the same as Emily today. We're just two bodies which happen to be causally related in some way. So although it sounds radical and scary to

say, oh, I'm going to kill you and create a copy of you, I actually don't think there's any very significant difference between that case and the case where I just go to sleep and wake up tomorrow. There's no fact over and above the physical facts about whether that is me or not. Who am I speaking to right now? Who are you speaking to right Now? Speaking to a person with a certain causal history, who's often referred to as Emily Edlund by others, who grew up in New Zealand and moved to the UK and then moved

over here. That's all there is to be said. Yeah, so my question to you is, suppose you lived in a country that had developed the ability to scan a person, replicate the person, basically create a teleporter. But the protocol was that they would first create the copy. And they would run all kinds of checks on the copy to make sure the copy was a faithful copy. They even have the copy say hello to the original. But the law was they can only be one of you. So as soon as the copy is made and confirmed,

they have to kill you. Suppose we go through this process. They make the copy. They check the copy appears to be pristine. It's the quality seal of approval. They open a video link. You say hello to the copy. The copy is like, wow, I'm here in this new place. Then they look at you and they say, okay, now we're going to kill you. You would not have any fear. I mean, I cannot claim that I would not have an emotional reaction. But I think as we've been, I think fear, emotions Are a kind of intuition.

I do not generally trust intuitions in this kind of situation. So I have no doubt that in that situation, I would be upset and I would feel fear. But at the same time, I hope that my philosophical part of my mind would be comforted by the knowledge that in fact, I still exist and will go on existing in just the same way as I always would. Is there a rational reason to care about your persistence? Well, I think sort of in order to say what is rational, you know, rationality doesn't tell us what we should

care about, in my opinion. Rationality tells us how to get the things that we care about. So for evolutionary and emotional reasons, I do care about my future self. And having established that, I can then go and ask, you know, what are steps I should take now that are going to help my future self flourish and rationality will help me decide what those steps are. But I don't think it's the job of rationality to tell me whether or not I should care about my future self. You have to decide the goal first and then you

can go and ask what does rationality say I should do to achieve the goal. So I'm going to follow up on That question before. Yes. You said you would for emotional reasons feel fear. Yes. But you would be skeptical of those feelings. Yes. But you don't have the same fear when you go to sleep. Yes. Or in other circumstances in which you're moved from one place to another. Yes. Why in this case would you feel fear but not those cases? I think because this is being presented as death and I think we all have very

strong emotional associations with the idea of death. So what if they just said it's not death? We're going to end your life functions, this copy, but there's another copy. I mean, would that make you no longer afraid? I mean, I think it would be difficult to present this in a way that it wouldn't sound like death. And I think, you know, I think as soon as you say to someone that you're going to put them through something, which is qualitatively quite similar to death, that gets all the emotional reactions and resonances we have around death

going. And so it would be difficult, I think, to avoid having those kinds of emotional reactions. Even if I do believe that in sort of some less emotional sense, this is no different from going to sleep. The associations we have With death and sleep are very different from each other. And so those are going to feel different. So in particular, you also see no meaningful distinction between this sort of transport scenario and someone building a wormhole that you can traverse in a spatio-temporally connected way, like walking through a portal that's actually connecting two space-time regions

where there's no duplication, there's no elimination of any copy. You see that as similar in all meaningful ways to the case of being actually copied and then the original destroyed. Yes. I mean, I think obviously the types of causal relationships are different here. It's not the same kind of causation. And so, I mean, the question I guess you're asking is, are some kinds of causation more relevant than others for establishing personal identity over time? And I think not, because I think there is no such thing as personal identity over time. There's no sort of fact

of the matter about that. All that can be said in this situation is that you can describe what the causal relations are. You can tell me what the physical facts are. I can make a decision about whether or not I'm happy to accept that other person as a future version of me, but ultimately that's a choice that I'm making. There's no sort of fact over and above the physical facts about whether that really is me or not. What is your view on the hard problem? We're going to do a hard problem. I think the hard

problem is very hard. The hard problem is difficult for me because... Everyone knows the hard problem is the hard problem of consciousness. It's not merely the so-called easy problem, which is not actually easy, but one can imagine in a way connected with modern science how we would solve it. The easy problem of consciousness is how do we model a very complicated system like the human brain well enough that we can predict the kinds of things brains will do in detail when it's conscious, when it says it's conscious, right? Of course, that problem is well beyond

our current ability right now, but one could imagine science reaching that point. The hard problem is how do we get from that objective third-person picture of a system simply acting and behaving to having experiences, the subjective sensation of color and emotion and feeling and sound, the Inner experience of being a conscious being? How do we cross the so-called explanatory gap between what we can describe descriptively and get across the actual experience of these sorts of feelings? And is that phenomenal conscious experience, that subjective first-person experience, something that we can reach or solve using the tools

of science as we know it? That's called the hard problem. And I guess there is a dispute in the community of people who work in philosophy of mind and beyond about whether the hard problem is really a problem, whether it's distinct from the easy problem. And so, yeah. The thing I find very difficult about the hard problem is there are many difficult problems in philosophy and in physics. And for most of those problems, I have a sense of what the answer might look like. I don't know the answer, but I have a sense of what

form the answer might take, what kind of answer might satisfy me. And then I think about the hard problem and I can't really even just form a concept of what kind of answer could possibly be satisfying or what form that answer might take. So, it's not a matter of looking through the possible options and trying To figure out which one is right or anything like that. It's really just a case of I can't see how any possible answer could ever resolve this question, which I guess in some ways does make me tempted to sympathize with

those who say it's not really a question because if we can't envision what the answer could possibly be, then perhaps it just isn't a meaningful question. But at the same time, I guess the options are really either it's not a question at all or it's a question that's so beyond our current cognitive capacity that we just can't even envision what a good answer to that question would look like. But yeah, I think for me, it differs quite qualitatively from other sorts of questions I worry about in my work because of all of those other kinds

of questions, I have some idea of what I'm looking for and then I think about the hard problem and I just have no idea what that would look like. I don't like when people tell me I can't or shouldn't ask questions just because they're hard problems. So the mere fact that the hard problem is such a hard problem and I, like you, don't have any sense of how to approach it, to me that's not a reason to deny That the hard problem exists. It's a reason to take it seriously and to exercise some humility that

there's some things we might not be able to answer. Consciousness feels so close to us. There's a sense in which it is the closest thing to us. When you become aware of your own conscious experience, it feels not just close, but also enveloping. It's the most familiar feeling. It is the root of what it is to feel. And to think that this thing that is so close to us and so familiar to us is so mysterious is deeply strange. And I think you're right. One view is to just deny that it's a problem because you

can't address it in some operational way, some positivist way, something like that. And I take it very seriously like you do. Can I ask, the first time we spoke in an interview, you asked me what got me interested in philosophy and I talked about coming to grips with my own conscious experience as a child. I remember. Got me interested in philosophical questions and although I didn't work in philosophy of mind, in large part, as I said in that interview, because I didn't feel like I had the tools to do it, although I'm very interested in

It, I didn't think it was where I could make the most meaningful progress. It is still something I think about. It's very important to me. And I'm just curious, when did you first encounter in an explicit way an awareness of your own existence, your own conscious experience? Did you remember having this experience the way that I did? And did this play an important role in your development as someone who thinks about philosophical questions? Gosh, that's a great question. Yes, I mean, I definitely remember having moments of being surprised and sort of, I guess, upset about

the fact that I will only ever experience this one conscious experience, and that I'm never going to have any direct access to any other kind of conscious experience. I am limited to what's going on in my mind, and in some sense, what's going to happen in the future of this, what's called causally continuous with me. But the sense of being confined to this one very specific spatio-temporal region, and that there being this big world out there that I can only ever learn about inferentially, I remember feeling quite sort of shocked and upset about that. And

that sort of coming as a big realization that there's something very specific about the way in which we are sort of confined to this region, that this is a boundary that we can't get past no matter what we might do. We do have this very specific kind of limitation in terms of our spatio-temporal existence. I wonder if the comments are going to have lots of suggestions of psychedelics. I think it's so interesting that you've described the experience of coming to grips with your conscious experience as one of boundedness and constraint and limitation. That's very interesting.

The guys were always very curious about other people's minds and how other people think and how other people experience the world. I ask people about these things and I read books and I try to understand, but there's limitations to that. You're never going to know in the purest detail what it is like to be another person. You're only ever going to know what it's like to be yourself. Why is that so interesting to you? I just think that there's so many ways to engage with the hard problem or philosophy of mind or questions about conscious

experience. I Didn't experience my conscious experience as a kind of limitation or boundary between me and other people. It's just very interesting to talk with someone who encountered these kinds of philosophical questions, but from a very different kind of perspective. It's almost like we don't all have the same conscious experience. I think it's the spatio-temporal thing that really struck me. The very specific limitedness of my consciousness supervenes are like this one region in my head. There's this whole world, all of the space and time out there. For some reason, I'm attached to this one little

bit of space. It's very strange. This is interesting. This now resonates with me. I have this image, this metaphor of a giant recording needle that just landed on one particular body, one particular massive object called Jacob Arendis. This needle just said, okay, this body's experiences will be your experiences. While I don't know that this is a probabilistic kind of a needle drop, I think it better not be. There's still a sense in which this is the single most important thing that ever happened to me. I don't understand it at all. I like the fact that

Although you began talking about your engagement with your own consciousness in a way that felt very different from mine, now we found common ground. I don't understand that recording needle, and I don't understand the hard problem. That is why I would not go to the teleporter. I don't feel like I have enough of a handle on how this is all supposed to work to trust that the teleporter is going to do the job. I mean, I just don't think there's anything to trust. I think there's no question. Once you know what the teleporter is doing,

that's it. There's nothing further to say. What questions do you have for Jacob? Questions I have for Jacob. For one, I imagine you all disagree on causation. Do we disagree on causation? I don't think we do. Okay, I just imagined it. That macroscopic causation is an emergent thing. And we've been talking a lot about how there may be a notion of microphysical causation that may be distinct in some ways, but may be connected in other ways. I think actually we're pretty much in agreement about that. I think perhaps you maybe have more sympathy for the

idea that it could be a directed notion Of microphysical causation. Is that correct to say? Oh, I see. That at the microphysical level, causation is directed. Yes. And your view is that there's a causal-like connection at the microphysical level, but it may not be directed. I think now we're getting to really hair-splitting differences. I mean, if we have to say, well, we agree on all of these things, but at the microphysical level, I think causation can be directed and you think directed. I mean, that seems pretty... In some ways, quite a significant difference, though. Okay,

fair. Because I think for many people, the directedness is at the heart of causation. I have a quote, I've forgotten who it's from, but from someone that I use a lot, which says, the concept of causation just is about asymmetry. That's the whole point of it. So I think that holding the view that there's something like causation, but that it's not directed or asymmetric at a microphysical level, forces you to reject a lot of the standard ideas about what causation is. And also, forces you to ask a lot of questions about how do we end

up with the sort of well-ordered structure despite The absence of any directedness or asymmetry. So although it seems like a small difference, I do think that in the devil's in the details in this particular case, it's quite an important difference. Fair enough. I would say that my view was a little more eliminativist about microphysical causation in the sense that I didn't think there was a well-defined notion of causation at the microphysical level for a long time. My views changed because of this project I've been working on, this indivisible stochastic approach, which provides a probabilistic dynamics

for the microphysical degrees of freedom, microphysical configurations of nature. And these conditional probabilities that form the laws in this theory naturally have a directedness. It's important to note that this directedness does not single out an arrow of time. It threads a needle here. One view is that in order to have a notion of causation, you have to have an asymmetry between causes and their effects. Effects are supposed to happen after the causes. Causes are supposed to precede their effects. And it's a little difficult to understand how you can ground this notion of Causation in our

best physical theories, our best microphysical theories, like the standard model of particle physics, in which there doesn't appear to be at a logical level any distinction between the past and the future. There's some time asymmetry in some esoteric processes in the standard model, but not of the kind that would underwrite the kind of temporal time asymmetry that seems to be necessary to talk about causes preceding their effects. In this indivisible stochastic approach, we get an asymmetry between causes and their effects. The conditional probabilities point in a direction from causes to their effects, but without preferring

a direction in time. The idea, the fact that a direction of time seems to occur in our world appears to be a contingent fact related to the initial conditions of the universe. And an arrow of time emerges in indivisible theory in the same way. It's not there fundamentally in the laws any more than it would be otherwise. I think if I had not encountered microphysical laws that were able to ground a notion of directedness, I wouldn't believe that there could be or should be a directed notion of microphysical Causation. So I would say that my

willingness to entertain this idea has changed because of this project. Otherwise, I probably wouldn't. I'd probably be more in Emily's camp on this. Yeah, I think one thing we do agree on perhaps is that I also have always been inclined to be quite eliminativist about causation and think it's a macroscopic thing. But what I do think is important is to say that clearly the microscopic world has some kind of structure. And although I wouldn't necessarily call it causal, it is similar to causation in the sense that it seems like it is related to counterfactual statements

that you might make. It seems like it's modal in the word philosophers would use, but it seems like it's not just sort of bare correlations. It has some kind of law-like force. It has to be that way. And I think if you're going to eliminate causation from the microscopic world, I'm happy to do that. That's the thing that makes sense to me. But I think you want to appreciate that there's other kinds of structure present there. When you get rid of causation, that doesn't mean there's no structure. There's various other kinds of robust Law-like structure

you could be seeing down there. What question do you have for Jacob that would persuade him to have your view of mind? My view of mind? I don't know if I have a very concrete view of mind because I don't know how to solve the hard problem of consciousness. Unfortunately, I'm the same way. Okay, here's a question about mind that I have thought a lot about, which is about how mind is related to the experience of time. And in particular, why do our minds experience time at a certain time rather than sort of spread out

or at different times at once? Why is temporal experience always of, located at a time in this way? So I don't know if you have views on that. This is something I worry about quite a lot. If you step back and look at the lifetime of a human individual from a four-dimensional space-time point of view, or 10 or 11 dimensional, depending on how many dimensions of space you think there are. But from a space-time point of view, you can point to the event that corresponds to the beginning of the life of this person, maybe in

some coarse-grained sense, and then a point in Space-time that corresponds to the death of this person. And then arranged in something like a line, some people call this the space-time worm, there's all the moments that make up your life as some kind of linear shape, a worm-like shape extending from the past to the future. The past is down here, the future is up here, time is pointing up in this depiction. And if you could imagine stepping outside of the universe, stepping outside of time, and somehow probing this individual at different points along this, we call

it a world line in special relativity, or general relativity, in relativity we call it a world line. Imagine probing this person at different moments. You're just looking at the contents of this person's brain, the patterns, the structures in this person's brain. We'll assume that this entity outside of the universe, having the power to be outside of the universe also has the discerning and resolving power to read the patterns in this person's brain. From what I understand about contemporary neuroscience, you would see that the patterns in this person's brain are different at these different points along

that world line. Those patterns in some way encode or represent memories, experiences that this individual has had up to that point. And if you could somehow ask the person at that event what they remember, what they're experiencing right now, they would describe feeling like they were in the moment, having had memories only in a past-directed sense. And so at least operationally, again, if you can imagine this entity outside of the universe probing this person, it seems operationally that there's a meaningful way to think about this person having that experience. But whether that operational standpoint is

enough, whether it makes sense to appeal to this sort of external entity, these are all difficult to justify, and I'm not going to claim to justify them. The best I can say is, especially as I've gotten older, it kind of feels like I'm on a ride, like in an amusement park, right? And this world line is like the track. It just is. There's no notion of it becoming or changing. I know I'm using the present tense in is, and there's a whole argument among some philosophers who work in this area that merely to use present

tense is In conflict with talking about this worm or this world line just existing. I acknowledge that our language is limited here, and it's possible the limitedness of our language is actually telling, and it's more than just a question of our language being limited. I acknowledge all of this. But to the extent that, if we can suspend our disbelief and just allow me to talk about this world line just being, sometimes it feels like my existence is like a ride at an amusement park, and I'm just traveling along this groove. And my sensation that I

am making choices is just one of the illusions, like all the others that I'm having along this track. Along this track, I have memories forming, I have feelings that I experience, at least I think I do, depending on one's view of the hard problem. I think I'm having those experiences. And I have the experience or at least the sensation that I'm making choices, even though perhaps in some sense I'm not. And this is all part of the ride. And maybe at the end of the ride, I'll find out it was all a ride or something

else. I don't know. See, I think I draw quite a different conclusion here because, you know, I think this Sort of model of consciousness is, you know, just passively looking in at the state of the brain at various times. That's quite intuitive. I think many people do think that way. But I think if that's how you think of consciousness, there's kind of no reason why you should only see the state at one time. Why couldn't the sort of the being outside of the universe look in at the states at several different times and hold them

all in awareness at the same time? And I think to understand why that's not the case, I think the way to understand why that's not the case is to think of consciousness as being more active than I think that picture suggests. Because, you know, there's no reason why I shouldn't be passively conscious of what's going on at several different times. But if I'm going to be acting at those times, there is a good reason why I shouldn't be acting at and conscious of several different times at the same time, because that's going to lead to

all sorts of very strange causal paradoxes and causal loops and things. So I think if you think of consciousness as being more active and more agential than I think perhaps This looking in picture suggests, that gives you perhaps to me a clearer sense of why we have this, why our experience has this particular temporal form, why we experience being located at times, why we feel as if we're moving through times. So for me that motivates perhaps thinking about consciousness a little bit differently, less as a looking in and more as an active involvement in the

world, if that makes sense. It's interesting how you talked about that external entity being able to probe different points along the world line at the same time. Yes. You used that word at the same time, this entity is outside of space time. So what does time even mean for this entity? People have talked about whether there are other directions of time and an entity that were outside the universe could experience a notion of time that's somehow orthogonal or distinct from the one that we're having. These are obviously all very speculative. I think what we're in

agreement is that given the structure of this world line and the structure of the brain in this world line, I think we're in agreement that it makes sense that at no point along the world line would this Individual be aware of their memories that they're going to have in the future and in the past. There's just no point along the world line where their brain contains those structures. Right, but I think if you look at the way consciousness works, it seems like there's some consciousness of a whole spatial region that's not a single space time

point. So you might think, okay, if consciousness in some sense supervenes on this whole spatial region, why can't it also supervene on several times at once? Our experience of time seems quite sharp in a way that our experience of space is not. We sort of seem to be very much located at one time and to not have this sort of ability to hold in our awareness several different times simultaneously. And so I think understanding the temporal structure of consciousness to me does require telling a story where consciousness is not purely a passive looking but is

very much involved in agency and in making decisions. And you can sort of see how that's going to lead to a certain kind of temporal structure of consciousness. I think some viewers in the comments may point to how these sorts of questions have been handled In fiction. Yes, exactly. In The Watchmen with Dr. Manhattan who experiences time in this more holistic way. Yes. In Arrival, the film which was based on the Ted Chiang short story. Yeah. And I, of course, recommend all of these people. They're all very interesting. Of course, in those cases, at any

given moment along the world line of those characters, their brains seem to contain information about the future and the past. And I think we're in agreement that at one moment, the brain structure of the person in question doesn't contain information about what is to come. So this is where I have trouble understanding what it would even mean to feel like you could experience multiple times given that they don't appear to be the brain structures at any point in the person's world line. There aren't the brain structures that would seem to encode information about what has

yet to happen. Right. But I think, I mean, you're sort of assuming that the brain has to supervene or the conscious experience must supervene on some slice. And I agree. If it's just supervening on a slice, then you will only have consciousness of the time at that slice. But I guess my question Is why does it have to be supervened that way? Why does it not supervene on several different slices or on some kind of blurring across the slices? And I think the answer is, for me, closely linked to the agential nature of consciousness. And

I think that the fictional examples are helpful because if you sort of think about, you try to imagine a fictionalized scenario in which you are sort of conscious of your whole life in some sense. I think what you can sort of see imagining that is that the more knowledge you have, the more you become conscious of your whole life, the less you are an agent. The less you can really be deliberating and acting and making choices. So I think for me there has to be some kind of quite tight link between our nature as agents

and the fact that we experience time in this way. I feel like you're channeling Curt Vonnegut here in Slaughterhouse-Five. Yes, exactly. They see all of time the way that we would think of seeing a landscape. Yeah. They talk about how to be human. It must be like someone strap you to a railway car. Right. And put a helmet on you with a very narrow tube. Yes. And you can only see one Tiny part of the mountainscape and then someone kicks the cart and you always see one little tiny part of the mountainscape. And only humans

with their very limited view. Yes. They're the only intelligent beings in the universe, according to the aliens, the only intelligent beings are humans who talk about free will. Yes. Who talk about this agential notion of consciousness. So it's very interesting that you're connecting our limited awareness of time to our sense of agency. Because you're right, if we could somehow be conscious of our whole world line, it would be very difficult to sustain a notion of being a free willed agent in the world. I mean, you want to know my view on the hard problem of

consciousness? This is not a very well formed view. But I think all of these things are connected to understand consciousness and why it is the way it is that needs to be linked to time and our experience of time and our nature as agents. I think those things are closely tied together and you're not going to fully understand what consciousness is until you understand the way it's linked to our nature as agents and the Way that shapes our temporal experience. It's very clever of you, Curt, how you've gotten us to talk so much about a

topic that both of us have openly acknowledged is not our area of expertise. That's the DMT in the air can and the water. So I want you to speak more about consciousness and this agential element. What do you speculate the link to be? Because it sounds as if there's a free will element here, but I don't imagine you believe in free will. You're right. I don't really believe in free will. Emily, what do you mean by free will? I don't believe in free will in the strong sense. I don't believe that there is a self

outside the physical world that is somehow able to reach in and choose in a way that's unconstrained by physical considerations. I'm sort of happy to accept that there's some kind of compatibilist form of free will such that if your decisions are mediated through appropriate causal pathways in your brain, we can call that free will. The reason I ask is that in describing your definition of free will, you use terms that are equivalent to free will. It's very difficult to provide a non-circular definition of free will. So one of The first questions you have in a

philosophical problem is, can we adequately define the terms in question in a way in which we're in agreement? And one of the rules is you're not supposed to define something in a circular way because then it's not really a definition. Do you think you have a way of talking about free will or presenting some kind of definition of free will that doesn't, in the definition somewhere, talk about choosing or decisions or something that means the same thing as free choice or free will? So I think at a compatibilist level, for sure, I can do that.

If I try to say what the sort of strong concept of free will is supposed to be, no, I can't do that. And that's one reason why I think the whole idea is a bit incoherent. I think it's very hard to state what free will is actually supposed to be in this strong sense. And for sure, if you do succeed in stating what it's supposed to be, it sort of seems like it's not what we wanted it to be in the first place because presumably free will is supposed to be something to do with making

decisions based on your personality and your experiences and your memories. So if it were really sort of separate from the physical world and all the things that encode my personality and memories, then it wouldn't really be the thing that we actually want. So yeah, I think I'm in agreement that the sort of strong concept is just not very coherent. Of course, you can say the same thing about consciousness, too. Consciousness is also extremely difficult to define, and some people have used exactly that problem. Oh, I can't define what consciousness is, therefore it's incoherent. I put

it in a similar bucket. Free will and consciousness for me are examples of things that are very difficult to define, and these aren't unique. I mean, probability is another example of something that's extremely difficult to define in a way that doesn't refer to something that's equivalent to probability. But I think probability is out there, not in the sense, I mean, you can ask what I mean when I think it's out there. I think that it's meaningful to talk about probability in the world. I think that the philosophy of probability is a very interesting question. It

plays clearly a very important role in our physical theories, And there's some meaningful sense in which when we talk about probability, we're talking about something that's worth our discussion. And I think it makes sense to ask, do you believe that things are probabilistic or not is a question you could talk about. Even though we don't have maybe a rigorous, non-circular definition of what we mean by probability, I feel the same way about consciousness, and I also feel the same way about free will. Just because something is difficult to define in my view doesn't mean that

it's not possibly there or worthy of our attention. So with free will, I think there's a specific problem. It's not just that it's difficult to define, it's that if you start to try to pin down what it is supposed to be, you arrive at something which is not what you wanted it to be. Because when I say I'd like to have free will, what I mean is that I would like to make decisions based on my personality, my memories, my history, all of these things. I would like to make decisions for good reasons. And all

of those reasons, the personality, the history, all of that is going to be encoded in my mind, it's going to be encoded In my physical brain, it's going to be constrained by the laws of nature. And when you look at this idea of free will as something that's supposed to be somehow unconstrained by the laws of nature, you end up with this very weird picture where it's this, somehow it's supposed to be the ability to make choices, but not for any reasons. Not based on who I am as a person, not based on my memories.

And once you sort of describe it in that way, it's kind of unclear that that's what we wanted it to be. So I think when you really try to pin down the sort of ideal concept of free will that people have aspired for there to exist, it's not just that it's hard to define, it just turns out to be the wrong thing and not what we were looking for. Similar to my question about rationality earlier, where I said that you can say strategy A doesn't work, strategy B doesn't work, and so on. Are you employing

a similar argument with regard to free will, saying that anytime you instantiate the concept of free will, it's incoherent, it doesn't work, it's not what you intended it to be, same with here, same with here. And you're able to make that argument for any conceivable definition Of free will? Or are you just saying that... the definitions that have been offered to me so far, I haven't found satisfying, and I also don't think the person who's offered it to me finds it satisfying? Well, I think there's a sort of a fundamental conflict inherent in the notion

of free will, because what people want free will to be is somehow unconstrained by the physical world. They want to make choices which were not dictated for them by physical reality and the laws of nature and all these kinds of things. But at the same time, people want to make their choices for reasons. People don't want their choices to just kind of randomly happen out of the blue. You want to think that I, Emily, because of who I am, made my choices. And I think those two things are just fundamentally in conflict. You can't reasonably

say that you want your choices both to be unconstrained by anything else and also made for reasons. So you can perhaps imagine a choice that is some mixture of being made for reasons and some kind of arbitrary, non-physical thing that's happening that's not based on any reasons. But all that seems to be happening there is that you're Making the choice more random or more arbitrary. You're not really adding in freedom in any sense. So I think it's not just that individual definitions don't work, but more that there's this sort of fundamental conflict at the heart

of what people mean when they talk about wanting to have free will. And Jacob, it's my understanding that you're more open to it, that you see that you can't define it in a satisfactory manner. And you don't even know what a good definition would look like, similar to the hard problem. You don't know, at least, well, Emily, you said you don't know what a good explanation would look like, such that if someone handed it to you, you could say, check, this is the hard problem solved now. I don't know if you also have that same

view of the hard problem, but regardless, do you also have the same view of the definition of free will? The fact that I can't come up with a satisfactory definition, a non-circular definition of free will, I guess only makes the question more fascinating to me. I'm especially fascinated by things that I can't pin down. There have been a lot of arguments over the years about free will. Some people who Are watching this may be familiar with Peter VanWagen's argument, the consequence argument for why there cannot exist free will in a universe based on deterministic laws.

I think that argument is particularly elucidating about some of the questions I think about when I think about free will. Imagine that we live in a universe in which the laws, the microphysical, the most fine-grained laws of nature are well-defined and are logically reversible. I say logically reversible because, well, for one thing, they don't appear to go backward in actuality, and it may also be that the laws maybe look a little different going forward and backward. In the standard model, there is a very small amount of time asymmetry and things don't look exactly the same

going forward and backward. Logically reversible just means that one could retroject just as well as one could predict with the laws. You could imagine running them in reverse and getting a unique trajectory just as you could run them forward and get a unique trajectory. Although I don't know whether VanWagen was thinking about logically reversible deterministic laws, Certainly logically reversible deterministic laws are the kinds of laws that we're familiar with from our best physical theories up until we got to the 20th century. And I think what troubles people is the idea that the deep past, together

with the, in this case, assumed to be deterministic laws of physics, already determined everything that we would ever do, separate from our personality, separate from who we are. That idea that it's not us making choices, but that all of our behavior has been determined by something that took place well before we ever lived is very troubling to people. And the VanWagen argument formalizes this as a set of premises followed by a conclusion. One premise is that the laws of nature, that one premise is that the laws of nature, I'm going to phrase it in my

own way, but this is essentially the argument. The laws of nature are deterministic. The laws of nature together with the past uniquely determine the future. We don't have control over the laws of nature. We don't have control over the past. And we also don't have control over the fact that the laws of nature together with the past fix the future uniquely. And given those premises, or some combination or some division up of those premises, one arrives at the conclusion that we don't have control over the future, which is the thing that we don't like. We

don't, people react negatively to this. This is an argument against compatibilism, an argument against the idea that free will is sustainable in a universe of this kind. And I actually think this argument is really nice because I think it highlights, like a good argument should, it highlights what are the principles that have been put, so to speak, on the table so we can analyze them and dissect them and decide if they make sense. Well, all of them, however you phrase them, deal with this notion of control. Yeah. Now, in the way that Van der Waal

originally wrote it, I don't think he used the actual word control, but it was something equivalent to control. We can't control the past. We can't control the laws of nature. We can't control that the laws of nature together with the past uniquely dictate the future. But what do you mean by control? Right. Right, control itself wraps in the idea of free will to begin With. Yeah. If you could define what control is, I would have a better sense of what free will is. Yeah. Can you phrase those premises without something like control? And I'm not

sure that you can. But I would actually even go farther. You see, the laws of nature aren't all that you have. The laws of nature need to be combined with initial conditions. And I think there's not enough attention paid to the role of initial conditions and how we think about our best physical theories. Without initial conditions, the laws don't tell you anything. David Albert has this really lovely example in his book, After Physics. He says there is absolutely nothing in the laws of Newtonian mechanics. Good old Newtonian mechanics. Nothing in the laws of Newtonian mechanics

that would prevent a collection of rocks from spontaneously assembling into a complete set of statuettes of the British royal family. Yes. You might go, well, of course, Newtonian mechanics doesn't allow that. Really? Well, if you started with a bunch of statuettes and dropped them, they would shatter into lots of little rocks. Newtonian mechanics is time reversible. So there Exists a well-defined process in which a bunch of rocks assemble themselves into the statuettes. We think that's unlikely, but that's not a statement of the laws, it's a statement about the initial conditions. We think the initial conditions

of the universe were very generic in some way. Very random or boring, uninteresting, typical in some way, that would make it very unlikely they'd be precisely fine-tuned to give you this particular outcome. But in a world in which, I sound like a movie director, in a world in which, in a world in which the only laws are, the microphysical laws are logically reversible. The idea that the past determines us and the future is no more sensible than to say that we determine the past and the future, or that the future determines us and the past.

Yes. You could just as well say that when we make choices, we're determining the initial conditions of the universe. That would be just as sensible as saying the initial conditions of the universe are determining us through the laws. And so I think you can actually recover a notion of compatibilism in this sort of a universe by saying that there's no sense in which the past Is determining us. It could just as well say that I make choices, and those choices are fixing the initial conditions through the laws. Yeah. Even to say our fixing involves the

present tense, which isn't even a well-defined thing here. But if you try to pare back these arguments against compatibilism, they either involve circularities themselves or they involve reasoning that doesn't necessarily hold up to scrutiny. And I'm not saying that any of this is original to me. I mean, many people have tried to parse these arguments about free will. Yes. I just think that they typify how thorny this question is, and how open it is, and how mysterious it is. And I really like these kinds of mysteries. These mysteries convince me that we're just never gonna

get to the bottom of everything. If we ever did, I think that would make things very boring. We'd lose our jobs. There'd be nothing for us to do as well. Yeah, I do think a lot of these arguments about free will, as you say, there's a sort of implicit assumption about time that's going into them. There's this model of the universe that's starting at the beginning and rolling forward to produce the Course of history. And I think if that's the picture you have in mind, then I guess, yeah, this determinism does sound quite scary, that

it was all fixed long ago at the beginning. Whereas I think if you have a view of time where you don't think the initial state is that special, the kind of suggestion you make seems much more compelling, that there's no particular reason to be worried about what the initial state is, because it's no more special than what happens now. I guess I do wonder if the kind of view you have set out, I wonder sort of how much freedom there is, because you might think that there's this space in some sense for some set of

choices to be free. But can we think that they're all free? Or is there going to be too much constraining that? Do we have to have some kind of, do we choose a subset of choices are free? Or do we just adopt some kind of view on which they're all sort of perspectively free in some sense? This is sort of interesting questions about if you're not going to put the freedom in at the beginning of time, where do you put it instead? And how do you sort of model the sense in which there's some kind

Of arbitrary and a sort of freedom in what the course of events? Freedom is just another word for nothing left to lose. I don't have answers to those questions. I'm not claiming that these are well formed views. I think I've been pretty clear that I don't work on free will. I don't think it's an area of particular expertise for me. But like anyone who thinks about philosophical questions, whether you're a professional philosopher or not a professional philosopher, I'm very interested in these questions. The mystery is very appealing to me. And at the very least, I

think I know enough to be able to identify weak points, I think, in arguments that have been put forward. Again, not to make any claim that I'm saying anything original about those problems, but yeah, so I don't dispute what you've said. Like I said, I don't have a sufficiently well-defined positive view about how to think about free will that I think I can provide any kind of defense. I do very much enjoy the discussion, though. Yeah. What other thorny, open problem in philosophy do you think about? Thorny, open problem in philosophy do I think about?

Most recently, I've been thinking quite a lot about emergence. I think we have this sort of aspiration in physics that we're going to write down a microphysical theory, and then we're just going to see how the macroscopic world arises out of that, and that's all going to work very nicely. I think the more I think about that, the more worried I get, because the microphysical theories that we have have been arrived at via measurements we have made using our macroscopic instruments. They certainly incorporate a lot of the things that we've learned about microphysics, and they

certainly incorporate some features of our macroscopic perspective in ways that are quite hard to tease apart from the microphysical theories. So when I think about how emergence works, I start to get worried that there's something circular happening here where we're trying to write down a microphysical theory and see how the macroscopic stuff arises from it. But the microphysical theory we're using is building in all the stuff that we have we're sort of already helping ourselves to the existence of the macroscopic world by using a theory that's formulated in this way. So looking at emergence And

thinking about the story we want to tell as physicists, I start to get worried about how can we fully separate out our macroscopic perspective from this description so as to really be able to tell this story, genuinely tell the story in which we start from a purely microscopic description and the macroscopic world arises out of it without being presupposed at the start. So this question about emergence is very important and has been highlighted by a lot of people in particular in the context of physical theories. So it will come as no shock to anyone who's

seen any of our previous discussions that I'm not an advocate of the Everettian approach to quantum theory. Oh, are you? If you read Hugh Everett's unpublished draft of his dissertation, you can find this online, maybe Curt you could put a link to it. It is beautifully written. It has a very early but fully formed version of the Wigner's Friend Thought Experiment, which we've talked about before. In that dissertation and in some of his correspondence, a particular letter that Hugh Everett wrote to Bryce DeWitt in 1957, Bryce DeWitt was a theoretical physicist who became Aware of

Everett's early work, had a number of questions, was initially very skeptical, eventually became a very strong advocate of Everett's approach. At the time in 1957, he was very skeptical and had written to Everett, Everett wrote back to him. And there, I think, in particular in that letter, Everett makes clear his problem with the then existing dominant paradigm for how to think about quantum theory, the Copenhagen interpretation. And Everett says very clearly that one of his problems is the Copenhagen interpretation. So one of them is this hybrid nature of sometimes you have an impersonal deterministic theory

of the wave function evolving, and then other times you have this very personal, observer-oriented, probabilistic version of the theory when measurements are happening. And he disputed the idea that it was good to formulate physical theories in this way. Interestingly, at the time he wrote about how he would have been happy to have a stochastic theory or a deterministic theory of the kind he was forming. He just said it should be one or the other, not some Frankenstein's mixture of both. But one of the other shortcomings he saw in the contemporary paradigm, the Copenhagen picture, was

that it provided no way to understand, even in schematic broad outlines, how to get an inter-theoretic reduction of classical physics to the microphysical quantum world. Inter-theoretic reduction just means an accounting of how we're supposed to understand classical physics as being derivable in some limit or some regime from the more correct, supposedly fundamental theory of quantum mechanics. Because the Copenhagen interpretation presupposed classical reality as one of its axiomatic ingredients and just foreclosed on principle the idea that you could somehow explain or derive the classical world from the microscopic quantum world. He found this untenable. I'm not

saying that providing a specific impersonal microphysical theory of the kind that, as you know, I've been working on, is enough to finish the project of understanding the emergence of classical reality. But I think at least it gives us hope that we might be able to do it. So I'm curious, as you talk about you think these sort of problems with emergence, Do you see these problems as fundamental obstacles that will preclude our ability to get emergence to work forever? Or do you see them as maybe transient challenges that we could, in principle, overcome with more

work? I definitely think they are overcomable, but I think they are perhaps more difficult to overcome than we immediately think. Because, you know, I think we write down what we take to be a definite impersonal microscopic theory. But I think it can be quite hard to see sometimes the ways in which what we think is an impersonal microscopic theory, in fact, depends on specific features of our perspective as observers. For example, sometimes people write down microphysical theories that involve something like time evolution or something temporally directed. That is something that I am intended to say

is a feature of our macroscopic experience that's not there in the microscopic world. I mean, that's just one example, but I sort of worry that, you know, it's hard to know in advance what's going to happen in future. And it's very hard to know, sort of separate out the bits of our theories that are sort of really describing The microscopic world versus, in fact, sort of artefacts of our perspective. And so that makes me worried about any attempt at sort of describing emergence. Like how do we know that we have really removed all the macroscopic

stuff and ended up with just a microscopic theory from which we can sensibly describe the emergence of the macroscopic world? I mean, there's a sort of nice quote from Jaynes about quantum mechanics involves the inferential stuff and the causal or real stuff all being scrambled up in a big omelette that no one can unscramble. But I think it's not just quantum mechanics. I think this is a problem throughout science that it's very hard in practice to distinguish things that are sort of independent of us versus things that are artefacts of our perspective. And if you're

not sure that you've completely unscrambled the omelette, can you really be confident that the emergent story you're telling is capturing all those elements correctly? Right. I mean, there was a time when people would speak of Newtonian mechanics as a theory of cause and effect. Right. Oh, Newtonian mechanics is a theory of cause and effect and quantum mechanics Is not a theory of cause and effect. You see this in popular depictions of quantum theory even today in some cases. But this is also some of the language that was being used back in the 20s and 30s.

Wait, how did they justify that? It wasn't, I think, a matter of justification. I think people just collapsed the notion of determinism and cause and effect to mean this sort of similar kind of a thing. But around the time people were already beginning to question whether causation was a sensible notion at the microphysical level, right? There's this famous quotation of Bruce Russell, right? Causation persists, you know, only out of the mistake, like the royal family, only out of the mistaken impression that it does no harm. His words, quote that one a lot. Yes, yes. We

like the quotation. And now I think a lot of, for the most part, practitioners, both in physics and in philosophy of science, tend not to speak of Newtonian mechanics in terms of cause and effect, because those are difficult to define. We tend to talk about it in terms of deterministic microphysical laws. So that's one example, I think, where we've taken something That we take very, you know, every day we see cause and effect all over the place. I mean, cause and effect is at the center of how we do medical testing. We think about our,

you know, judicial system, culpability for, you know, moral acts, cause and effect. But I think we've gotten better at maybe not extending those ideas all the way down to the microphysical level. I think, so I very much agree with this. I think there are a lot of things that people take as intuitive that they take from our macroscopic experience and push all the way down to the microphysical level. I mean, Curt, before we've talked about Reichenbach's principle of common causes, a thing that we see all the time in the world around us. And I think

there's a tendency to push that idea all the way down to the microphysical level, where maybe it doesn't actually make a lot of sense, at least not the way that it's often formulated. There may be versions of it that kind of make sense at the microphysical level. So I think this is actually a really great point. When you say the microphysical level, there's two definitions of micro. So micro is like bacteria level, But then micro, what does microphysical mean? Does it mean fundamental? Does it mean UV complete? What does it mean? I think one can

use the word micro in a relative sense. Micro, relative to us, means the things that are the constituents out of which we emerge in some sense. So one can mean it in a relative sense. I actually mean it in a more technical sense. When I use certain words like probability or microphysical or dynamical law, I mean those, you should take all those to include implicitly the additional words on a specific physical theory. I don't know why philosophers say on theories instead of according to theories. It's an interesting preposition. We say at worlds. That's true, yes.

And on theories. I'm not sure about that. Not in worlds, not according to theories. It's just an interesting shibboleth, I think. Yes, you just pick up by being in the field. You pick up osmotically by being in the field. But you should add on that theory. So what does it mean for me to propose a physical theory? To propose a physical theory means that you propose some kind of stuff, physical stuff. You could call these the ontology, the matter, The degrees of freedom, the things that have configurations, whatever the moving parts are. I call them

like moving parts, the moving parts of the theory. And those moving parts, according to theory, are its fundamental micro-physical constituents. By definition, they're the things I call the micro-physical things. The things that are stipulated axiomatically by the theory as the basic moving parts. The basic elementary moving parts of the theory, I call those the micro-physical features of the theory. And then we have dynamical laws. The rules according to which those moving parts are supposed to change in some way. Either deterministically or probabilistically or something else I haven't thought of. And then we call those the

micro-physical laws. The dynamical micro-physical laws. Again, deterministic or probabilistic. That either govern or dictate or summarize depending on one's view about how laws are supposed to work. That relate those moving parts to the way that they behave. And then one can ask questions of emergence. What kinds of derivable macro-level Phenomena can we see emerge from this? Can we see tables and chairs show up? Things that were not put in in the micro-scale picture of the theory that come out contingently in some sense. Approximately contingently in some emergent way. A theory of atoms giving rise to

tables and chairs, for example, would be an example of this. Is there some more profound sense of micro-physical that is theory independent? I'm not quite sure. What I can say is that what's micro-physical on one theory might not be micro-physical on some other theory. For example, if you believe that your first theory is not a fundamental theory. But is itself obtainable through some inter-theoretic reduction to some deeper, more fundamental theory in some way. In a sense that you can phrase in terms of, say, supervenience. The only way that things in the less fundamental theory can

change is if there is some change in this more plain fundamental theory. That's a supervenience relation. Then it may be that the things that you were calling micro-physical according to your first description of that first theory are not micro-physical according to this other theory. On this other theory, sorry. And it may be turtles all the way down. That there's just a never-ending sequence of ever more fundamental theories. Each of which comes with its own notion of what the micro-physical constituents are. Or it may be this taps out at some most fundamental theory that we have

not yet discovered. And then its micro-physical constituents are the micro-physical constituents on which all other things and phenomena supervene. Yeah. I don't know the answer to that. But that's precisely what I mean by micro-physical on some particular theory you have of the world. Do you believe that there's an infinite regress of theories like this? And I mean out there. So for us, maybe. Maybe it's just we're not intelligent enough and we keep approximating something closer and closer. But there's something there in that model. Do you believe that there's something there? Or do you believe that

somehow nature does have this infinite regress of laws? I look at this and just see a mystery. And where I see a mystery, I want to go. I don't have any other presuppositions as to what I'll find. Emily? I guess I'm tempted to say there Must be a bottom level just because it seems just wildly inefficient to have an infinite series of layers going all the way down. And I mean just on the sort of principle that we shouldn't postulate more than we need to explain what's going on. It seems to me hard to believe

that it could really be necessary to postulate this infinite layer of things to explain what's going on. So I guess that's in some ways just an aesthetic preference rather than a real argument. But it sort of seems sort of like a bit excessive to postulate all of this stuff if it could be done with less. I mean there's so many different ways this could go. There's one of the opening couch gags on The Simpsons in which the camera zooms out from The Simpsons family as they sit on the couch. And it zooms out and you

see the earth and you see solar system and you see the galaxy and you see all the galaxies. And you keep zooming and zooming out until you see the whole observable universe. And as you keep zooming out you see atoms. And then molecules and eventually you pull out and you see, you know, and Homer, you know, he says, you know, wow, there's a similar thing that happens in an Episode of Adventure Time. There's a plug for Adventure Time when, you know, Finn puts on these, these, these glasses, the spectacles of Nerdicon, I think it's what

they're called. And you see the same kind of, you know, you pull all the way out and you return, you know, so it could be something bizarre like that where it isn't infinite. Yes. In a sense, maybe there's some kind of cyclicity to it, or maybe something we haven't even thought of. I mean, like I said, to me, exploring those mysteries is the most exciting intellectual journey I think that one can proceed along. That's why we do this, right? That's why we're in this field. Right. It's very exciting. Very exciting, yeah. It's fascinating. So let's

say there was no heat death of the universe and time could keep ticking forward infinitely. Okay. And at the same time, we have the premise that there is no difference between the past and the future. Yeah. So why do you care that there's an infinite regress in the past, but not an infinite progression to the future? To me, if there's no difference between the arrows, why is there some psychological problem with an infinite Regress in the past? Well, I don't particularly have a problem with an infinite regress into the past or the future. I think

we have reasonably compelling cosmological evidence that the past is not infinite. And I believe the cosmologists when they tell me that that's what the evidence suggests. But I don't think I would have a philosophical objection to time going on infinitely into the past. Indeed, I think if you are the kind of person who has a model of the universe where you put in an initial state and it evolves forward in time, then you are, I think, in some sense committed to there being a beginning of time where the initial state can be put in to

be evolved forwards. I think because I don't have that kind of view of time, I'm perhaps more relaxed about that possibility than perhaps others might be. The lack of an initial event does make it difficult to talk about initial conditions. That's true. And there are all kinds of conjectural theories, cyclic cosmology theories, eternally inflating multiverse theories, stuff that's even more bizarre in which our Big Bang is not a unique event. And maybe the universe will Go through some other kind of phase transition far in the future that we can't imagine. I think it's important though,

and I think it's always important when having discussions like this. There's a view among some people that philosophers like to engage in wild speculation. And as someone who has spent a lot of time in physics, in theoretical physics, and also in philosophy, I find that there tends to be less wild speculation in the areas of philosophy, at least the ones that I work in and Emily works in, than sometimes I find to be the case in high energy theoretical physics. One of the reasons why I found working in this particular area of philosophy, analytic philosophy,

philosophy of science, philosophy of physics, one of the reasons I found it to be so hospitable to the way I like to think about things is I actually tend not to like wild speculation. I tend to get lost very quickly once I feel like I'm in the midst of a lot of wild speculation, people stacking speculative metaphysical hypotheses on top of each other one after another. At some point I just sort of lose the thread and I have difficulty following. I like To be very careful in my reasoning, especially in circumstances in which we lack

data. And cosmology is a great example of this. Now we're living in a remarkable era. We're living in the era of precision cosmology. The amount of data coming in is profound. And this is the result of incredibly hard work by human beings painstakingly developing unbelievably brilliant devices. I mean, for anyone who doesn't know what's going on, you know, in cosmology, people who are working in observational cosmology, you're in for a treat if you just go look at some of the work they're doing. It's incredible. But for some of these very profound questions like where the

universe truly came from, whether there was anything before the universe began, where it's headed in sort of the long term, now we're in an area where we have to speculate and we have very little data at our disposal. And so this is exactly the kind of circumstance in which I think we have to be extra careful. And avoid wild speculation and proceed as methodically and rigorously as we can scrutinizing all of our steps as we go. That kind of rigorous scrutiny is one of the things I like the most about this Field of philosophy in

which we work. I think it's optimized for a certain kind of scholarly pursuits. Yes. And maybe not for all kinds of scholarly pursuits. But it appeals to me. It's one of the reasons I like doing the work that we do. I think there's a place for the wild speculation. Creative ideas come out of that. And I think it is important to have people doing that kind of thing. I think, as you would agree, it's important to keep track of the fact that that's what you're doing and recognize that you're in the process of creative idea

generation and perhaps not making a rigorous argument. And also recognize when is the right moment to transition from doing the wild speculations to trying to make it more rigorous and not go too far in that direction. I think one thing philosophy can do perhaps is provide a different methodological approach to these things and perhaps act as a check on what's going on and allow us to see those things from a different point of view. I think it's important, just to follow in your point, that we are careful to demarcate when we're engaging in one kind

of pursuit over another. But I think it's also very Important when we engage with the public to be very clear about when we're in one mode and when we're in another. I sometimes worry that people speaking to the public, which is a very important thing to do, we should all be doing it much more, are not always careful enough about being clear about what is rigorous and reliable and grounded in strong data, strong evidence, and what is more speculative. And I think it does everyone a disservice when we're not very clear about those lines. Your

point about the importance of people who work in our area of philosophy generally, but especially in our area of philosophy, in our engagement with science, I think is well taken and needs to be emphasized. We're all adapted for different kinds of, as I said, scholarly pursuits, different ways of engaging with scholarship, and to make progress on thorny problems, especially cases in science in which we may be lacking in data. I think more engagement between scientists and certainly people in our field would be very helpful. I know that there's this attitude that some people in any

discipline have, philosophy, mathematics, that we shouldn't be too focused on Whether our field is useful to other fields. You know, that our field doesn't exist merely to serve other fields, but has its own ends and its own meaning. And I agree with that. I think the kinds of questions that we often talk about, you, Emily, and me specifically talk about, but more broadly, people talk about in this field are interesting and worthwhile to pursue for their own sake, on their own merits. But the fact that thinking philosophically about physics, physical theories, physical problems can be

extremely useful in progress of science is not a demerit. It's something to be highlighted. And I don't think that we always do a good enough job highlighting it. Well, I think it's, you know, both you and I trained originally as physicists because I guess we were both interested in contributing to the progress of science and transitioned into philosophy. And I certainly didn't do that because I was no longer interested in contributing to the progress of science, but because I felt that that was the best way that I personally with the kind of person I am

and my skills could do that. And I would imagine for you, similarly, you're not, It's not that you have withdrawn from contributing to science, but that you see this as the best way to do this. Yeah, that's right. I talk about how I have two main modes, physical philosophy, where I'm interested in the areas of philosophy that are inspired by or constrained by or connected in some way to questions in physics, in science generally, but questions that are more traditional questions in philosophy and metaphysics. My other mode is I call philosophical physics, which is doing

physics, but using the methodologies that come from analytic philosophy. Rigorous scrutiny of definitions, of arguments, of assumptions, being very careful to avoid too much speculation, finding gaps in our theories, uncovering, lifting rocks, looking under them, scrutinizing, looking through things. And there's a prejudice of this leading to major progress in science. And when I have a conversation with, I'm interested to get your take on this as well. I'd love to hear from you what you think are the biggest contributions that philosophical thinking about physics, philosophy of physics, has contributed to the development of modern science. The

examples I would point to just in the limited case of quantum mechanics are the arguments over EPR and entanglement, which were done fully in philosophy mode. You read these papers and they were discussions about the nature of what was really out there. And although they were trying to address at some level practical questions, at some level philosophical questions, those philosophical arguments inspired by and generated important insights into thinking about how entanglement worked. But I would add to that collection, David Bohm's work on decoherence. You know, he decided in his textbook, his 1951 textbook, to probe

the structure of measurement from a foundational philosophical point of view, not to take for granted just what was handed to us from the Dirac phenomenon axioms. And in this chapter, chapter 22, he goes through this whole process. And in 22.8, he talks about, he introduces this idea of decoherence in a rigorous way, the first rigorous description of how decoherence works. He knew better than anybody how decoherence worked. And he was not convinced that that was enough to resolve the measurement problem. That's Why he introduced his hidden variables approach, partly out of conversations he had with

Albert Einstein. But decoherence is now ubiquitous in papers throughout quantum physics, from high energy physics down to atomic molecular optical physics, quantum information, quantum computing. You can't pick up a paper without reading about decoherence timescales, how to protect systems from decoherence, the importance of decoherence and understanding the emergence of the classical world, but also protecting systems from decoherence so you can get the sort of delicate entangled superpositions you need to get successful quantum computation to happen. But this crucial idea that plays such an important role in some of these papers goes back to philosophical thinking

about these problems, problems that were considered too philosophical for serious physicists to think about, and ultimately had negative career repercussions on David Bohm and on Dieter Zee, who worked on decoherence in the 70s as well. Bell was inspired by the EPR argument to develop Bell's theorem. Bell's theorem was the sort of thing originally That Bell worked on kind of in secret. He published this work in these underground journals. He warned off people from working on this stuff because he said it would be damaging their careers. John Bell himself, his official job was that he was

a theoretical particle physicist at CERN. This was sort of on the side. This work was on the side. And now Bell's theorem shows up in contemporary physics all the time. We use Bell's theorem to certify the randomness of quantum random generators. We use Bell's theorem in quantum cryptographic protocols. We use Bell's theorem all over the place. And, you know, I would add to this list the no-cloning theorem, which was independently discovered by a bunch of different people, including some philosophers. Dennis Deeks independently discovered the no-cloning theorem. But I would add further things. No-signaling theorem, which

played a really important role in understanding the structure of how we make quantum mechanics compatible with special relativity. No-signaling theorem comes out of philosophical thinking about quantum mechanics, all the way up to things like the Elitzur-Vaidman bomb tester, Which was inspired by, you know, Everettian many-worlds thinking. When David Deutsch wrote his pioneering article in 1985 that initiated the idea of looking for quantum advantage and developing quantum algorithms to do certain tasks more efficiently than could be done on a classical computer, he says multiple times in that paper, which you should link to, that his motivation

was Everettian quantum theory. His goal was to find some smoking gun evidence that Everett was giving us the correct world picture. He says in the paper he believes that a genuine, working quantum algorithm that's more efficient than we can achieve classically would produce an untenable strain in any other interpretation than the Everett interpretation. How much is all this worth to the progress of physics? I would stack that list of contributions up against many contemporary research areas in physics. It was done in large part by people who had other official jobs or who suffered significant career

ramifications for working on ideas that were considered too philosophical for mainstream physics. It was done for almost No money. So there's a sense in which if you divide the output by the input, you practically get a division by zero error. So if anyone is thinking about, like, how to make the biggest bang for your buck in terms of making contributions to physics, I would argue that we need to invest more in philosophy of physics. I think every physics department should have a resident philosopher of physics there to hold accountable. Right. Just one. I mean, the

whole department, you have a department of five physicists, 10 physicists, the big ones have 30 or 40. At least the big ones should all have one philosopher of physics who shows up at seminars and like Statler and Waldorf in The Muppets, you know, calls attention to when people are saying things that that run beyond what the data allow or arguments that don't really hold up or engage in too much wild speculation. I think that would be an incredible service. Alone, separate from the fact that I think probing these fundamental questions is itself a creative enterprise,

right? The purpose of philosophy of physics is not merely to criticize or hold back physics that's that's not done as well as it Should. But like I've pointed all these examples, thinking carefully about our best physical theories has generated a lot of really important ideas. I would argue that the incredible ferment that we saw at the end of the 19th century into the beginning of the 20th century that saw the development of statistical mechanics, that saw the development of relativity, special and general relativity and quantum mechanics, that it's no accident this happened at a time

when philosophers and physicists were trained in each other's areas, were closely connected, that all the top physicists talked about philosophy and philosophers who were close to physicists, right, talked about physics. This softened the soil and generated a tremendous amount of raw creative input that has that has then, you know, extended all the way into the 20th century. I would argue that maybe we need another time like that, that this is a ripe moment for there to be more cross pollination between these fields to generate the ideas that will take us further into the 21st century.

Great. Would you add anything to that list? I think what I would say is that even Things that are taking place in physics, I think you often see, particularly at the inception of new research programs, a lot of thinking that I would classify as philosophical. If you read, I think, the early papers on holography, when some of those ideas are coming out, there's a lot of very philosophical thinking that is going into that kind of thing, that's thinking about the meaning of information or the meaning of a surface and sort of doing thought experiments, all

these kinds of things. I think if you look at the way these things are actually developed, philosophy in some sense is playing a major role in that. Perhaps that could be a fruitful way in which philosophers could talk to physicists more in that sort of period and sort of get more input into that kind of philosophical process that is happening at the inception of a research program, at the inception of new ideas. I think the idea that philosophy and physics are completely separate and that the physicists just calculate things and the philosophers just say words,

it has never been true. The physicists are doing very philosophical things, philosophers of physics Are doing calculations, are doing very technical things. I don't think we need to see these communities as being two communities that are opposed to each other, but we are both engaged in similar things using slightly different tools and methodologies and we can all benefit from speaking to each other more and from having more sort of cross-pollination between those things. Yeah, when you read Emily's papers, you don't notice any lack of mathematical sophistication. Yeah, and I think you're an exemplar of the

kind of person working in philosophy whom physicists should spend more time getting to know. I think it's not that physicists don't engage in philosophical reasoning sometimes. I do sometimes think that in any field, we can begin to think that we know well enough how to do everything and we don't need the help of anybody else. One sentiment that I've experienced personally spending a lot of time among physicists, certainly by no means among all physicists, but among some, is that physicists, at least some, feel like they can do philosophy better than philosophers and they don't need

philosophers, That physicists are so quantitatively trained, they're so smart, they can do all of this, they don't need any help from philosophers. And if you point them sometimes to philosophers, sometimes you'll hear them say, well, I've occasionally looked at philosophy papers and I find most of them not very good, so I've given up. There's this principle, Sturgeon's Law, named after Sturgeon was a science fiction author who felt bad because people in the literary establishment looked down on science fiction as not serious, not serious literature. And when he asked them why, you know, he would sometimes

hear them say, well, 90% of science fiction is crap. And his response was, he finally realized, wait a second, but actually 90% of everything is crap, right? 90% of everything that happens in academia, not academia, in the business world, corporate world, government, everything, 90% of it is crap. And you need to know a field well enough to be able to distinguish the quality 10% from the crap. It's inevitable. That's just, in any creative enterprise, you're going to generate a lot of ideas, many of them won't work, Some things won't be done correctly, but there'll be

a core of things that are good. My worry is that as philosophy and physics have diverged over the 20th century, the fields have gotten less and less capable. Each field has gotten less and less capable. The members, the practitioners of each have gotten less and less capable of identifying the quality 10% in the other discipline. And they just see more and more only the 90% that's not good, right? As these fields have moved apart, you begin to see kind of a blurry, coarse-grained picture of the field in which all you can see is the 90%

that's not so good. It seems more like 90%, it seems more like 99% or 100%. And as the fields get farther apart, and they begin to see each other as less good, they want to move farther apart. They tell younger people in their fields, yeah, this field is not useful to us, and the fields move farther apart, and you get this vicious cycle. I think the only antidote is for there to be individuals who are agential, who choose to take an active effort now at this particular moment in time. Taking advantage of the fact That

we are at one moment in time. Yeah, location right now, yes. To be agents right now, at this particular moment, to work to bring these fields together. I mean, you bring the fields together by having these kinds of conversations. And I think it's hard to imagine anyone alive today, frankly, who's doing more of this work than you are, Curt. I mean, your whole podcast is a celebration of exactly this thing. I mean, so you have obviously my immense admiration for the work that you're doing. But this is incredibly important work to bring these disciplines closer

together. For sure, yeah. People who are currently working in the disciplines, and also people not in the disciplines. Either people work in other areas, or young people who haven't yet made up their minds of what they want to do. They haven't made up their minds about what these fields are like, to get across the message that we need to actually bring them together, because it will be mutually beneficial for both of us. And as the fields come closer together, and we begin to see more and more of the quality 10% in the fields, then hopefully

there'll be a virtuous cycle. We'll Want the fields to come even closer together, as we see that they're more and more useful to each other. And I feel, in some ways, like this is a right moment for that. Emily and I have talked about how both of us are seeing among students lately, and also I would argue young faculty, a greater willingness to engage on these questions, a more openness to these sorts of interactions. And that's something I feel like makes this a ripe time for that kind of engagement. And we all need to participate

in it. None of us can say I'm too busy. It's part of what I think we need to do to push these fields forward, is to increase that level of engagement. Well, it wouldn't be anything without the indexical of you, and you as well. Thank you, Emily. Thank you, Jacob. This has been a complete delight, as always. Great fun, yeah, thank you so much. Any time I'm in town or you're in town, we have to always do this. My pleasure. I've received several messages, emails, and comments from professors saying that they recommend Theories of Everything

to their students, and that's fantastic. If you're a professor or a lecturer, and there's A particular standout episode that your students can benefit from, please do share. And as always, feel free to contact me. New update, started a Substack. Writings on there are currently about language and ill-defined concepts, as well as some other mathematical details. Much more being written there. This is content that isn't anywhere else. It's not on Theories of Everything. It's not on Patreon. Also, full transcripts will be placed there at some point in the future. Several people ask me, hey, Curt, you've

spoken to so many people in the fields of theoretical physics, philosophy, and consciousness. What are your thoughts? While I remain impartial in interviews, this Substack is a way to peer into my present deliberations on these topics. Also, thank you to our partner, The Economist. Firstly, thank you for watching. Thank you for listening. If you haven't subscribed or clicked that like button, now is the time to do so. Why? Because each subscribe, each like, helps YouTube push this content to more people like yourself. Plus, it helps out Curt directly, aka me. I also found out last

year that external links count plenty toward the algorithm, Which means that whenever you share on Twitter, say on Facebook, or even on Reddit, etc., it shows YouTube, hey, people are talking about this content outside of YouTube, which in turn greatly aids the distribution on YouTube. Thirdly, you should know this podcast is on iTunes. It's on Spotify. It's on all of the audio platforms. All you have to do is type in Theories of Everything, and you'll find it. Personally, I gain from re-watching lectures and podcasts. I also read in the comments that, hey, Toll listeners also

gain from replaying. So how about instead you re-listen on those platforms like iTunes, Spotify, Google Podcasts, whichever podcast catcher you use. And finally, if you'd like to support more conversations like this, more content like this, then do consider visiting patreon.com slash CURTJAIMUNGAL and donating with whatever you like. There's also PayPal. There's also crypto. There's also just joining on YouTube. Again, keep in mind, it's support from the sponsors and you that allow me to work on Toll full-time. You also get early access to ad-free episodes, whether it's audio or video. It's audio in The case of

Patreon, video in the case of YouTube. For instance, this episode that you're listening to right now was released a few days earlier. Every dollar helps far more than you think. Either way, your viewership is generosity enough. Thank you so much.