Making Sense with Sam Harris - #388 — What Is Life?

Episode Date: October 21, 2024

Sam Harris speaks with Sara Imari Walker about a scientific understanding of life. They discuss the contributions of physics to this topic, Erwin Schrödinger, the inadequacy of standard definitions of life, the prospect of "artificial" life, the role of information, constructor theory, assembly theory, the space of all possible structures, a "block universe," the existence of abstract objects like numbers, the Fermi paradox, the likelihood of life elsewhere in the universe, experiments that could decide how likely life is to emerge, the possibility of a Great Filter, the number of Earth-like worlds, and other topics. If the Making Sense podcast logo in your player is BLACK, you can SUBSCRIBE to gain access to all full-length episodes at samharris.org/subscribe. Learning how to train your mind is the single greatest investment you can make in life. That’s why Sam Harris created the Waking Up app. From rational mindfulness practice to lessons on some of life’s most important topics, join Sam as he demystifies the practice of meditation and explores the theory behind it.

Welcome to the Making Sense Podcast. This is Sam Harris. Just a note to say that if you're hearing this, you're not currently on our subscriber feed, and will only be hearing the first part of this conversation. In order to access full episodes of the Making Sense Podcast, you'll need to subscribe at samharris.org. There you'll also find our scholarship program, where we offer free accounts to anyone who can't afford one. We don't run ads on the

podcast, and therefore it's made possible entirely through the support of our subscribers. So if you enjoy what we're doing here, please consider becoming one. Well, we're about two weeks out from the presidential election in the U.S., and there's going to be a lot of politics on the podcast. I think I have three episodes that I will drop before the election, all more or less focused on politics, But not today, because today I'm speaking with Sarah Walker. Sarah is an astrobiologist and theoretical physicist who focuses on the origin of life and also the possibility of discovering alien life in other worlds. She is deputy director of the Beyond Center for Fundamental Concepts in Science

and a professor in the School of Earth and Space Exploration at Arizona State University. She is also a fellow of the Berggruen Institute and a member of the external faculty at the Santa Fe Institute. She is a recipient of the Stanley L. Miller Early Career Award for her research on the origin of life, and her research team at ASU is internationally regarded as being among the leading labs aiming to build a fundamental theory for understanding what life is. And she has also written a very interesting book titled Life as No

One Knows It, The Physics of Life's Emergence, which is the focus of our conversation. We discuss the contributions of physics to this topic, Erwin Schrodinger and his famous book, What is Life?, the inadequacy of standard definitions of life, the possibility of artificial life, the role of information, constructor theory, assembly theory, the space of all possible structures, the concept of a block universe, the existence of abstract objects like numbers, the Fermi paradox, i.e. where is everybody, and the likelihood of life elsewhere in the universe, experiments that could decide how likely life is to emerge, Robin Hansen's concept of a great filter, how common Earth-like worlds might be, and other topics.

In all candor, this is a pretty dense conversation, but I loved it. And some of you will, too. And now I bring you Sarah Imari Walker. I am here with Sarah Walker. Sarah, thanks for joining me. Happy to be here. My wife, Annika, is the person who puts you on my radar, and she says hello, by the way. Hi. Tell her I said hi. It's always nice to hear from her. Yeah, I know you've already spoken with her. She's got an audio documentary that is soon to be born, and I know she spoke to you for it. But let's talk about your book, which is really fascinating. Perhaps you have other

topics you want to touch here, but your book, Life As No One Knows It, is where you address this question of what is life and how could we recognize it and how could we come to understand it in the context of physics and chemistry and a concrete conception of the universe. Before we jump in, how do you summarize your background academically and intellectually? and intellectually? I am trained in physics and I try really hard to not let that training bias my thinking, but I really love the rigor of theoretical physics and thinking very deeply about the nature of reality. So I guess my training is very traditional in terms of a basic undergraduate physics education. And then going into grad school, I was still in a cosmology group at that time, but I have decided to work on problems that are probably very non-traditional from the perspective of physics. But I think they're very much problems for the way we think very deeply

and abstractly about the nature of physical reality. So I guess I've just gotten really excited about that stuff. But I didn't have any science or anything like that in my background before starting college. Do you call yourself an astrobiologist or a physicist? What's the one word descriptor of your specialization? I used one word, which is actually two, is theoretical physicist. But I work in problems in astrobiology. But I think at the core, I feel much more of a theoretical physicist than anything else. As I said, you are focused on the question of life. Obviously, physicists have done this before. You discuss Erwin Schrodinger's book, What is Life?, in your book. What was Schrodinger's contribution to this topic? Yeah, I think the way that he asked the question was really structured. And I think

it was the first time that really anyone had laid down with like the discoveries at that time, thinking in a very logical manner about how we might reason about the fundamental nature of life based on what we understood from physics. And so his major contribution in that book was actually to think about the nature of genetic heredity. And what he had talked about was the fact that, you know, in order to specify all the information in a cell, you really require a lot of information. And the most robust kind of information storage we know in physical materials are crystals. And so he, but crystals tend to be periodic. So this is where he came up with this idea of an aperiodic crystal, something that had a non-repeating structure so it could encode a lot of information.

And this is, you know, heralded as sort of like a prediction of DNA as the genetic material, which is, you know, in some sense, a quote unquote non-periodic crystal. So that was one thing that he said in that book that was really interesting and is the one that usually people cite as being kind of, you know, like a major sort of insight. But I think what really caught my eye about that book was how much he really went back and forth about what physics could and could not say about the nature of life. And one of my favorite quotes toward the end of it is about this idea that other laws of physics that we haven't established yet might actually be necessary to explain life. And that was really kind of a big motivator for me when I read it,

because I really deeply felt that. And I didn't really see a lot of people talking about it that way. So what is our best current definition of life? And then I guess, what are the edge cases where the definition seems to fail? There are a lot of definitions. So in my own work, I tend not to take a definitional approach, but other people have. So I wouldn't say that I agree with these kinds of definitions. But the one that you'll usually see in astrobiology is that life is a self-sustaining chemical system capable of Darwinian evolution. And at first pass, that seems pretty reasonable because

you're talking about life being chemical systems that can evolve. But as I talk about through the first chapter of the book, if you focus on each word in this definition, they fall apart at some level. And they're really simple things like evolution only happens to populations. So are individuals alive or not? Is life chemical or not? We can think about memes as evolving systems or is technology actually a part of life or not? Viruses are kind of a typical definition. They're not self-sustaining or challenging. They typically challenge definitions. They're not self-sustaining on their own. Are they alive when they're in the cell and not alive when they're out of the cell? And even self-reproduction is a bit problematic. So there's always the examples

of things like mules, but I like honeybees as a better example because it's very clear that a colony of bees is a living thing, but you have individual members of the colony that can't reproduce. And so all of these things seem to suggest that if we want to draw this kind of hard distinction around life, we have all of these edge cases, as you're pointing out. You just raised the question whether technology could be a form of life. And this poses an interesting paradox. Maybe it's only just a semantic one, but the phrase artificial life sounds like a contradiction, right? Because we distinguish between biology and manufactured objects and processes, right? And so we can use a phrase like artificial intelligence without any contradiction because clearly we've been able to

instantiate the function of intelligence in a substrate-independent way into our machines. But the idea of instantiating life into our machines or building machines that are in fact alive, but they're non-biological, that just sounds like a contradiction in terms. So, and obviously the way you're analyzing life begins to ignore that apparent semantic boundary. How would you think about that? But how would you parse the phrase artificial life? I view that phrase more as a challenge, like in the sense that could we understand life

enough to actually instantiate it in a machine or recognize that we had? Because I think one of the issues, you mentioned artificial intelligence, you know, one of the issues that we have and why people are massively debating about whether they're intelligent or whether, you know, our technologies could become animated in some way we might view as alive is because we actually don't have a fundamental understanding of what these things are. And so usually I see the distinction between like when we say things are artificial, usually we mean that they're human created. And that doesn't mean that they're not natural and that the things that we do aren't part of some underlying fundamental description of nature. And so I guess I don't really like the word artificial because it makes

it seem kind of like it's an epiphenomena and not telling us something very deep about the nature of what we are and what we're doing. And that, you know, the kind of physics of the universe that created structures like us might operate even through us to create other things that are living or intelligent. And so this gets more into an evolutionary continuity between biology and technology where you see less of a distinction and you see more of a progression of what evolution does at a fundamental level. I want to make another pass on that terrain because I think our intuitions, or at least about what other people believe about the meanings of words, might be a little different here. I think the distinction most people assume is real

is between biological and non-biological systems. So for instance, if human beings created an artificial form of life or a synthetic form of life based on biology, right? If we, you know, through synthetic biology created organisms that had never existed before, but they were nonetheless wet and cellular. I don't think anyone would hesitate to believe that those organisms were alive. And I also don't think anyone is hesitating to believe at this moment that our artificial intelligences are in fact intelligent. I mean, we haven't achieved general intelligence yet, and people are wondering whether we're going to do that, and that we might have, that a proper definition of life really should be substrate agnostic?

Yeah, I think that's exactly what I think. I think it's very provincial. And I think even, I probably don't ascribe the level of intelligence to current algorithms that other people do. So I maybe don't agree that we actually even have narrow forms of intelligence. I think it's very easy for us to recognize things we think of as intelligence because we see ourselves mirrored in our technology. But the technology has been selected to mirror us. So it's very unclear to me how much the substrates themselves are embodying things that we might view as intelligence versus being kind of a false mimic to us. But I think that's a really interesting question. And I kind of take both sides of it depending on the day of the week. But this question you're asking

about the distinction between chemistry and silicon, for example. So if we could imagine if we ran an original life experiment and we evolved a life form, say we could actually solve the original life and it had completely different chemistry than any life on earth has a completely different lineage of evolution and information processing. It, you know, and it was, it was, it happened to be cellular. I think probably cellularity is, is probably deeply intrinsic to, to having open-ended evolution. So I'm okay with that as kind of a feature that we put on this thought experiment. Most people, I guess, might be okay calling that biological life and saying perhaps that was more similar to what cells evolved on earth than a computer program evolving in a computer is to cellular life. But it is an

interesting case that there's a direct lineage from the last universal common ancestor of all life on earth. So the first populations of cells that were evolving from the early geochemistry of our planet all the way up to modern computers. And if I wanted to write into a sequence of DNA right now, I could encode it with some information. I could take that sequence of DNA and I could try to amplify it by PCR and read it into a computer. And it might just happen to be the case that with that information content, once it gets into the computer, I actually create a computer virus and I infect somebody else's computer, right? So there is a possibility of having a direct, this is just sort of a thought experiment, that there's a direct line of

information between us and the technologies that we're creating. They're not independent lineages. And this is also the case with large language models. Why is it that they have the structure of human language? It's because we were the environment that selected them. And human language had to evolve first in the substrate of human minds and in our writing on paper and then into our computers. And so the kind of information that they have is actually a direct lineage from biological life evolved over billions of years. And so I actually, I think there's more of a continuity between our quote-unquote biological life and our technological life on this planet than there

is potentially between biological life on this planet and a different instantiation of alien life and a different chemistry, which would exist in a totally different space of possibilities and a different kind of chemical makeup. So you've invoked the concept of information here and lineage and the causal continuity between ourselves and our technology and somehow that being significant. How do you think about the role of information here? our technology and somehow that being significant. How do you think about the role of information here? And is the boundary between life and non-life more likely to be informational than

merely physical or chemical? Yes, with the caveat that I think information, when we understand what it really is foundationally, is actually a very physical feature of reality. It's just, you know, like the things we call material are often things that we can measure and things that get regularized into our theories of physics. So, you know, in order to talk about things like mass and charge, you know, we had to invent a lot of technology to be able to measure those properties and they became relevant because they were the ones that we could formalize into theories. So information to me is certainly related to the boundary between non-life and life.

And early in my career, when I first started thinking about the original life and really thinking that theoretical physics was the right approach for addressing the transition from non-life to life, my first sort of sets of conjectures about this were that life is the sort of boundary in the physical universe where information has to take on a causal role. You actually have to consider it as a physical feature of the systems under study. And it's really hard to do that when we talk about information as such an abstract property. So what we've tried to do,

you know, really with the formal work, and this is really a foundational feature of assembly theory that I'm working on with my colleagues, is figure out how it is that you would make information as a feature that constructs these kinds of evolutionary objects, these things that we see in living systems, necessarily require information and storage of information, processing information in order for them to be selected to exist? How do we turn that into something that we can actually understand as a physical property and then talk about this boundary where these kinds of objects can't exist unless there was information acquired over time in order to construct them. There was some kind of selection that happened.

So you mentioned assembly theory, which is your theory. Let's talk about that and its relationship, if there is one, with constructor theory, which is David Deutsch's approach to, I guess, adjacent matters. I know you're a fan of Deutsch's, as I am. He's been on the podcast several times. Is there a relationship between assembly theory and constructor theory? And perhaps you can give a potted description of each. Sure. I think there's definitely a relationship. And Lee Cronin, who's my collaborator that developed assembly theory and I'm working closely with, and I have both talked with David and I've had a, you know, and I also have worked with Chiara Marletto. So I'm, you know,

over many conversations over many years, you know, we're still trying to dig into the relationship between their view of reality and ours. But I think there are a lot of interesting parallels, and I think they're scratching at the same kinds of fundamental understanding. So the sort of short description of constructor theory is this idea that the laws of physics should not be cast in terms of initial conditions and laws of motion, which many of us think because that's sort of an inadequate description for life for many reasons, which might be a tangential conversation. But David's very adamant about this not being a

final description of nature. And instead, what he advocates, and Chiara wrote this nice book about also called The Science of Can and Can't, is this idea that we should be talking about things that are possible to do and things that are impossible to do and why they're possible or impossible. And so this reframing is really intended to focus on this idea of constructors as causes for things to happen. But the theory of constructor theory actually doesn't deal with constructors directly. What it does, it talks about tasks, which are things that can be caused to happen. So a classic example is a chemical reaction that might not happen unless you had a chemical catalyst, and the catalyst would be a constructor for the reaction. So the reaction, impossible, unless you have a constructor. And so actually actually it's a possible reaction to happen. So

it's not actually physically impossible, which is very different than something like a perpetual motion machine, which we have laws of physics that say that literally cannot exist. So they want to classify things that cannot exist as entirely separate from things that could be caused to exist. And this idea of can be caused to exist, I think is fundamentally important because I think when you get into biology, if you get into evolution and you deal with things that have knowledge, which was my understanding is David's main interest in developing constructor theory was in part to account for knowledge as a constructor that allows things to be possible that wouldn't be possible unless you had entities like us that understood really basic features of how reality works.

So there's this whole space of things that could be caused to exist, but they require a constructor to exist for them to happen. And so that's the whole premise of constructor theory. They abstract away constructors and then they just end up talking about possible and impossible tasks, but they have a lot of ways of using that to be able to describe really interesting features of physics that are not possible to describe in standard approaches to physics. So that's constructor theory. Assembly theory is really specifically developed to tackle the problem of the original life. That's our main interest. But I think throughout my career, I felt that that problem was very conceptually deep,

and so deep that whatever would explain that would reveal fundamentally new physics. And I don't know why I had such conviction on that, but I just remember being a PhD student, studying theoretical physics, cosmology, particle physics, quantum field theory, and thinking that, and at the same time starting to dig in the original life literature and just fundamentally thinking that there was something about these explanations of nature that didn't fit this problem. And so what assembly theory says is that anything that requires a lot of complexity, like any object that's very complex, has many independent parts, doesn't happen in the universe for free. There's no such thing

as spontaneously fluctuating into existence. An object like a cell phone or DNA. These things are things that need to be constructed over time. They need information to exist to construct them, or more specifically, they need other objects that set the constraints to enable their existence. So if we think about this idea of constructors and this reaction can't happen without this catalyst existing, if you build a structure of physical systems that relies on that property, you see that there are certain things that can and cannot happen in the history of an

evolutionary process. And so assembly theory is really trying to get at this idea of how would we say that this particular object, if we just looked at it in the universe, was necessarily a product of evolution. Evolution being the process of objects constructing other objects over time and basically getting a hierarchical stack of all of these kind of constructors that can only exist because the other things lower in the stack exist. And so it's kind of a way of physically embodying information by actually asking, could this thing exist without a very specific history for its existence? There's one counterintuitive claim you make in your book about at this point, which is

that you can't talk about a single object in isolation here. So like a single screwdriver, you need more than one to get this theory rolling. Is that the case? It's not that you need one to get the theory rolling. I think it's an observational fact that single screwdrivers don't exist. And what we're trying to get at- Okay, but what about like the ceiling of the Sistine Chapel? So this is interesting because it's not, there are some objects that are very refined that like might exist in only one structure, but all of the parts of the Sistine Chapel exist in other objects in our biosphere. So the way they come together as a Sistine Chapel is probably a very unique structure, or even if you think,

like I would use Gaudi's churches, like that one's totally crazy. I don't even like know how you would possibly reproduce that. But if I can take a step back for a second to explain a little bit more of the structure of the theory, it might become a little bit more evident what this copy number feature comes in, which you're talking about, because this is actually the hardest and most conceptually deep feature of the theory that I think is really hard to wrap our heads around. Because it seems like when we talk about complexity, that seems obvious to us. When we start getting into this issue of copy number, reproducibility of structure as being necessary to understanding that physics, that gets a little bit more difficult. So the sort of key conjecture

of assembly theory is if you imagine, again, this sort of key conjecture of assembly theories if you imagine again this sort of idea of all possible things that could be caused to be constructed or could could possibly exist assembly theories conjecture is that objects can only be made from objects that already exist and the way they so if you wanted to try to build into a space of new objects you have to do it recursively. You have to use structures that are already selected to exist. So we're talking atoms, molecules. Sure.

Yeah. So if we think about in chemistry, if you want to think about, yeah, you want to think about making a molecule, you have to use, you know, like you have to use atoms and the atoms have to come together to make bonds. And the conjecture is if you're just thinking about fragments of molecules, if you want to make the next structure, you have to have pieces already in place to make the next structure. So it's sort of a very abstract view of reality. It's a little easier to think about Lego. You put two pieces of Lego together, and then you can use that to build the next structure and in a hierarchical way to get to a more complex object. Now, this is important because if you imagine you're randomly searching this space, every time you combine two objects, the space is super exponentially growing because you have a lot of diversity of components that you could stick together. And it ends up being such a large

space, even for very small molecules, that the universe couldn't exhaustively search the entire space to instantiate even one copy of every object. So you're talking about the space of all possible objects that could be built from those parts. Yep. And so I think this is a very underappreciated fact of our reality that there are far more structures that could exist that don't than will ever exist. And we can imagine a lot of that space, which I also think is an underappreciated feature of our reality and is telling something that's quite deep about how much information is

encoded in us as physical structures. But just to go back to this idea of this possibility space of, you know, like thinking about all possible molecules, for example, what assembly theory says is that there's actually a hard boundary that the universe cannot cross, a threshold complexity, what we call, we call it assembly index is the measure of this. If you take an object and you take it apart to basic building blocks, and you try to do this process of building it from those atoms by making bonds, or if you're sticking your Lego together, there's a minimal number of steps that you have to take to get to that object. And we call that the assembly index. It's the size of the minimum space necessary to construct that object from elementary parts. And our conjecture is that if

the assembly index is too high, the universe can't make that object spontaneously. Because if you imagine every step you could have an error and the space is exponentially growing, it's becoming exponentially less likely that you would hit that specific object even once. And if you want to make it twice, that's a double exponentially less likely by a random process with no selection. And if you want to make it three times, it's even less likely. And so yet, on our planet at least, we see complex objects, things that take many steps for their construction, their assembly index. The minimum path, the shortest path for producing them is quite high. And we see them in high abundance. And so our conjecture is that there's actually a boundary in the space of all possible objects, all possible molecules if you want to use that as an example.

Above which you need to have physical systems that can constrain the space of possibilities to construct specific objects. You need to have constructors themselves that persist in time. The things that can cause this thing to exist themselves need to be persisting in time. So you end up getting these structures that their coexistence necessitates that they need to be able to produce each other. And actually, their existence at all, they need something else to exist in order to exist. They need to be caused to exist by another physical object. So that boundary, we think, is the origin of life.

And it's in a very abstract space. It's in a very large causal space, which we call the assembly space. And it gives us a way of formalizing this boundary. And in fact, when we look at this property for molecules, you can measure the assembly index for molecules using mass spectrometry, NMR, and infrared. And Lee's lab did an experiment where they went in and they found the assembly index for a whole bunch of molecules from abiotic and biological samples. And they were able to show that there is indeed a threshold above which they only find molecules in life of a given assembly index. So that's sort of the basic structure of it. But I think this helps build a little bit of the intuition about copy number, because what happens in this kind of physics is

in order to even get to high assembly index structures, you have to have reuse of parts and you have to have objects that persist in time long enough to make the same structures again and again. And this feature means that in order to even get to that space, you have to have a reliability of all of those construction processes all the way down the causal chain, which means that most of these objects will never come into existence once. They'll come into existence multiple times because the constructors that are necessary to make them must themselves persist in time. So of course, we'll get some novelty on the edges. And one way I think about it is we have

these sort of diagrams we draw of the assembly space where you're just adding parts by taking parts of the history and stacking them on top of each other to build more and more complex objects. And so most of the things that structure creates, when you actually look at the causal depth in time, they have the same exact history in their construction. It's only the things at the tips that start to have a lot of novelty and variation. And so the copy number feature is actually all the way down that causal history. And so the copy number feature is actually all the way down that causal history. And when we get things at the tips, you know, they'll be genuinely novel objects, but they won't persist in time into the future unless they can actually be entirely reproduced,

all the information in them. So the things that we care about in assembly theory are also very similar to the ones in constructor theory in the sense that you want the transformations that are reliable. You want the causation that the universe has selected to continue to exist, to actually be the core of the theory, because that's the thing that actually ends up being the content of evolutionary lineages over billions of years. Right, yes, that final sentence is worth reiterating, because if you're going to imagine that Darwinian principles are at the bottom of this process, you're by definition talking about many copies of things, right?

You're not talking about single novel objects. Yes, and I think this is also where this physics really demonstrates that having spontaneous design or spontaneous fluctuation of objects is not explanatory. And I think the copy number is actually at the crux of that. The fact that when we see complex objects, we see them in abundance and we don't just see spontaneous formation of things that are very deep in time that have a very large assembly index just forming instantaneously, but they actually require evolutionary lineages. They require physical time. Time instantiated is a physical material in the objects to construct

them. It means that there's not really a possibility for the information content for those objects to exist at every point in time. The information is embodied in the constructors that also must be physical that construct those objects. What would an infinite universe or an infinite many worlds multiverse do to this picture? I mean, given how big infinity is, doesn't that give us the possibility of high assembly index objects like cell phones spontaneously appearing out of nothing? I think that interpretation of reality is not correct. So I think this is one place where maybe the deep foundations of assembly theory and constructor theory diverge, because I know that David really firmly believes in many worlds as the most explanatory interpretation

of quantum mechanics. And in assembly theory, I think we intuitively feel that there is one universe and it is constructed over time, but the possibility space that's folded up in physical structures that's actually embedded in time as a physical dimension is so large that when we interact with physical objects, we're only interacting with the tips of their actual, that structure. The tips of their lineage that embodies all the information required to create them. Yeah. Yeah. So I, you know, like if I, if I think about, you know, what I am as an evolutionary structure, you know, like it's easy to think about me as a, a three-dimensional object that's, you know, about five, three in height. And, you know, like I have a, you know, a spatial extent, but we don't really think about the fact that I'm, you know, partially or like parts of me are literally 3.8 billion years old because they've been constructed on this planet that long. And I think that's actually a real physical feature of what I am.

And so some of the things that we see in quantum mechanics that I think lead to the many worlds interpretation, I think are looking at the structure of physical reality from the wrong end. So when we do quantum experiments, we look at objects that are very small. They're instantaneous in time and have no memory. And so like elementary particles, and the universe has, you know, it doesn't require an evolutionary history for the universe to create elementary particles. But when we think about things like quantum uncertainty or entanglement, it seems to be the case that when you start to build causation and contingency in the system, you have these elementary particles that are kind of existing in this non-deterministic

underlying reality that doesn't have a lot of structure to it. When they start interacting with each other, determinism emerges out of that. And the causal structures actually are self-constraining. And so my view of it is there's a non-deterministic underlying reality, and determinism is an emergent property of causal constraints that emerge in the observable universe, what we interact with. And the parts that we actually can interact and have structure with are the things that we actually have those deterministic properties.

So it's very counterintuitive, but in some sense, what we think in assembly theory is living structures are the most deterministic things in the universe. They have the most causation in the universe because of this feature of the fact that they wouldn't exist without all of this causation built into them. I'm wondering whether the very notion of possibility is just that, a notion that doesn't actually map on to reality. What if there is no possibility space? What if the only thing that is possible is what is in fact actual and there is what happens and everything else is our idea about counterfactuals that aren't just that, not factual? or factuals that aren't just that, not factual. What if the future is not only determined, but just as real and as viewed from above as the present or the past, and there are no such things as events or processes, there's just a single object? Is there something in modern physics that's discredited that idea? I think it's an interpretation. Again, it's sort of like many worlds. It's one way of building a philosophy out of current theories

of physics. So I don't know anything that's refuted that idea, but at the same time, nothing's refuted many worlds. And the reason is because these are philosophical interpretations of theories that give sort of a broad explanatory framework that's consistent with the data and the structure of the theories that we built that correspond to those sets of data. So I think the black universe is maybe not as popular as it was, but I think the idea still has some favor in the foundations of physics. And I have colleagues that certainly view reality that way. Does it play well with quantum mechanics or is it just, does it somehow ignore quantum mechanics? I think it lives in a separate space,

at least for me conceptually. There may have been people that have thought about the two together, but I haven't really meditated on thinking about the block universe and its correspondence to quantum events. What I tend to think about the block universe is that it's easy to view reality that way if you can take a God's eye view and think you exist outside of the universe and you can write down laws that also exist outside of the universe and initial states. So again, it goes kind of to this idea that David was rolling against about what he calls a prevailing conception, that initial conditions and laws of motion, you know, are really not the right framing of the physical universe. And if you take that seriously, most of modern physics eventually needs to be

thrown out. But a direct consequence of that entire chain of the evolution of theoretical physics from Newton up to Einstein is the creation of the black universe idea. But I think, you know, the reason I never felt very comfortable with that conception is that living inside reality and not thinking that our laws of physics can exist outside of it. I tend to think that, like, my conception of laws of physics is very different than most physicists, but I think about the laws of physics as information that our biosphere has constructed by intelligent beings like us, happens to be us in this case, that have regularized a large set of observations we've seen in the physical world to the point that we can say we feel like these are objective features of the

physical world, but the laws themselves actually are also constructive processes in our biosphere. So an example I give in the book is the fact that we understand things like Newton's laws of gravitation or Einstein's theories allows us to do things like build satellites and other technologies that wouldn't be possible without that kind of knowledge. And to me, that's much more interesting. The fact that we have a description like the block universe allows us to do other things that we wouldn't be able to do without that kind of description. So I tend not to think that any of our theories of physics are platonic ideals that really describe the way reality works.

And I think this even with assembly theory, assembly theory is a very constructive theory of physics. I literally have a conception of the universe that it's constructing itself. And even the theories I build are a part of that process. So I would expect that theory to play a role for a certain time in our understanding of reality, but eventually it will be replaced by something else that's better. But I think it's to me the best explanation for the nature of life right now that I can find. And I really do take seriously the fact that we can imagine counterfactuals as being causal to the reality that we live in. We see that

every day as humans, it's like literally a part of our human experience. And that's the part of physical reality that I want to understand is the part that's us. Well, so imagining a counterfactual is obviously that act of ideation is, in our case, physically instantiated in our brains, right? So it's an operation we're performing in spatio-temporal terms. And I'll grant you it seems to have causal consequence, right? Because obviously we can talk about it as we're doing now. We're talking a lot about objects and information and causation. This is, of necessity, a conversation that's pushing us into metaphysics. What does it mean for something

to exist? What is an object? In what sense can a law of physics exist and be causally effective, right? And be something other than merely its many instantiations, right? Like, is the law of gravity something beyond the fact that every object that has been dropped has fallen already? Or like, how does it impose its will on the next object that I let go from my hand if it has to be something in addition to the instance of that object falling, or so it would seem? I guess I'll ask you, having vomited all that philosophy on you, let me just ask you a simple question. How do you view the existence of abstract things like numbers? How do mathematical objects fit into your ontology here? Yeah, I've thought a lot about the physicality of math

almost my entire career. And I have a very intuitive understanding of what I think math is that I don't really see reflected in the way that I see most people talking about math. And just to take a step back from that, one of the things that I think about a lot, and it comes, I have a formal way of talking about it now with the structure of assembly theory, but it's something that I've thought about a lot longer than that, assembly theory, but it's something that I've thought about a lot longer than that, is this idea that what we call information is deeply tied, like what we think of as abstract things in our environment, human language, mathematics, I don't know, the information content of genomes, like things that seem substrate independent because they can be

copied between different kinds of physical materials. You know, that feature seems very perplexing. This is one of the reasons that information has been really hard to understand as a physical property. And I think what we're really talking about when we talk about this property of information is we're talking about objects that have a physical size and time, because what you're looking at when you see something that's informational is you're seeing something that was the product of a history to generate it. And so I have this sort of philosophical interpretation of what we're doing in assembly theory, which I think, which comes from the measurements and also just the ontology of how the theory is set up, that evolutionary objects are deep in time and the size and time the assembly index is actually a physical feature of the object.

And so if you have things that have a, you know, like a smaller depth in time than you do that you can isolate as physical structures and look at them, you know, they look very physical to us. So, you know, we're built out of elementary particles and cells, and it's easy for us to see those as physical objects because, you know, we're larger than them. We have enough capabilities within the bounded physical structure we are to acquire most of the information about those objects. So they seem physical to us. But things like human societies or human language or mathematics, I think, are much larger physical

structures than we are. They're much larger in time. And so as smaller bounded objects in time, we can't possibly observe all of them at once, and they look very abstract to us. And so the sort of idea of a platonic world, to me, is just saying that we actually are embedded in this massive causal structure. I'm going to give you an abstract object that is going to stay abstract, and I think it's nonetheless identifiable, at least it's going to seem so when I utter the sentence. So, you know, we have a sense that, a very deep sense mathematically,

that there are an infinite number of prime numbers. And if you just assume that we are finite and our interaction with these numbers is going to be finite in time, however many millions or billions of years we do math, there's going to be a final prime number that we interact with or instantiate on any of our hard drives or otherwise use in our cryptography, etc., etc. So there's a largest prime number that any member of our species, and let's just for the moment assume we're alone in the universe and no one else is going to do math. There's a largest prime number that we're going to talk about,

write about, use in some way. But of course, there's a next prime beyond that one, right? So I'm talking about that next prime beyond that one. Does that exist? It can exist in the future. But I'm talking about that. We know that however long we live, we're never going to, there's a last one. It exists as an idea,

which I think is still a physical object, but the actual embodiment of that idea doesn't exist yet. And I think sort of, it's hard to think about numbers as physical things. So you gave me a very abstract example. I'll give you another very abstract example. And then after that, it might be good to talk about time getting bigger as part of this process, but I'm just going to put a pin in that because I can get to that later if we come back to it. But hold on, but just before we go to other examples, when you say it exists as an idea,

it's a little bit more than that because one, you know, we know if we know anything about numbers at this point, we know that it exists in that it is potentially discoverable, right? And we know that it doesn't end in two. Right. Yeah. No, this is a good point. I think I actually like that you're pushing on this quite a lot. This is very fun. So I think if I was going to be more precise in my language, I would say that the knowledge to generate that structure exists. So a constructor that can build that next prime exists,

but it hasn't actually mediated that transformation. If you'd like to continue listening to this conversation, you'll need to subscribe at samharris.org. Once you do, you'll get access to all full-length episodes of the Making Sense podcast. The podcast is available to everyone through our scholarship program. So if you can't afford a subscription,

please request a free account on the website. The Making Sense podcast is ad-free and relies entirely on listener support. And you can subscribe now at SamHarris.org.