Theories of Everything with Curt Jaimungal - Sara Walker: Aliens, Assembly Theory, Information

Episode Date: March 15, 2024

This is the full episode, ahead of time, as a thank you for supporting TOE.

Where is all that information? Please tell me. We want to go out to space to look for life. What we should be doing is looking here on Earth in chemical space. Because where aliens exist is in the possibility space of things that our universe can construct, and we can do experiments on this planet. This is a presentation by the wonderful and impavid Sarah Walker, hot off the press, recorded just a few weeks ago at MindFest Florida Atlantic University 2024, spearheaded by Susan Schneider who's the director of the Center for the Future Mind. Sarah Walker is a professor at Arizona State University of Astrobiology and here she talks

about the hard problem of consciousness, the hard problem of life, the hard problem of matter, alien life, and of course assembly theory. Stuart Hameroff is coming on next and all of the talks from the AI and Consciousness Conference, also known as MindFest, are in the description as well as the link to the website to the Center for the Future Mind is in the description. I recommend you check it out. You should also know that right now there's a two- video on the mathematics of string theory which was just published. It's string theory like you've never heard it talked about before. The link to that is also in the description.

For those of you who are unfamiliar with this channel, welcome. My name is Kurt Jaimungal and this is Theories of Everything, where we delve into the topics of mathematics, consciousness, physics, and artificial intelligence with depth and rigor that's distinct to this channel due to us not as chewing technicality in favor of a wider market. If this fastidiousness is interesting to you, then you're in safe hands here at Theories of Everything. Alright, now enjoy this presentation by Professor Sarah Walker. Guess we can get started. For our second speaker at MindFest, we have Sarah Walker. Guess we can get started.

For our second speaker at MindFest, we have Sarah Walker. Sarah's a theoretical physicist and astrobiologist at ASU, Ayrton State University. She's the Deputy Director of the Beyond Center, an external faculty at Santa Fe Institute as well, and a couple other centers out there. You know, her main focus is original life research, constructor theory and you know looking for assembly theory, assembly theory. My mistake, my mistake. And looking at see if we can find signatures for life on other worlds, techno signatures which is really interesting. Thanks for coming Sarah and and turn it over to you.

Great. I'm thrilled to be here. I work primarily on origins of life, actually not origins of mind, but as we heard earlier this morning, those problems are probably deeply related, and I am interested in both. I just think the origin of life is so tractable

and solvable within the next few decades that I really wanna make it happen. My training is in theoretical physicists, as a theoretical physicist, and I'm really interested in fundamental laws. And so a lot of people that have approached the origin of life in the past have tried to do it from the perspective that known laws will be sufficient to explain the origins of life. That is not my take on the problem. My take on the problem is we're missing something fundamental and that there are

new theories to learn about how reality works when we're studying things like life and mind. So I went through like the whole formal education in physics, but when I actually really started working on this problem, I realized that there was not a particular shoe that fit this problem. And so I've been spending most of my career trying to think about what would be the structure of laws of physics that would explain life. So that's just to give you a little bit of framing for how I think about the problem.

And what I want to talk about today is this theory that I've been working on with my collaborator Lee Cronin and a bunch of grad students and post-docs and other collaborators at Santa Fe Institute called Assembly Theory, which is an attempt to explain life from a perspective of a more fundamental understanding. And it has some really interesting philosophical properties of the theory that I'm hoping to touch on today, which I think will frame some interesting questions on the nature of what minds are as products of evolution and the fact that things that are complex are only structured by an evolutionary process and an informational process.

And we need to take that as part of the fundamental physics. So when I approach the questions of minds, the thing that I think is most interesting as a physicist is not what a thing is but what does it do. And what I mean by that is does it have any causation in the universe? Are there measurable consequences for a thing existing? And so I've been thinking about the question of consciousness in minds for a while now, but the kind of question I really focus on is not the one of subjective experience, but if there are some things that exist only because minds exist and they would be a measurable consequence of a mind existing.

And how does that actually structure reframing the question and how we think about the nature of mind and how we measure it? So as a preface, I'm not even convinced that consciousness is measurable as a property of individual minds. I think it measurable as a property of individual minds. I think it might be a property of societies that you can measure over time, which is pretty radical, but it's consistent with some other things, I think. So we're used to thinking about the hard problem of consciousness, and it's been excellently framed in many ways as thinking about what it feels like to exist, at least for us. There's other hard problems that are related to this

and may be structured in similar ways, like the hard problem of matter, which is kind of a play on the hard problem of consciousness. So I think that the hard problem of consciousness is more familiar to people. But the hard problem of matter is the idea that we actually can't measure anything,

not just subjective experience, but anything about an object in the universe outside of interaction with another object. So even matter in some ways is equally mysterious to consciousness because you can't know its properties except when you measure them. And with my colleague Paul Davies, like, thinking about the hard problem of consciousness and the hard problem of life, we came up with this idea of the hard problem of matter. We came up with this idea of the hard problem of matter. We came up with this idea of the hard problem of life. And this really came from thinking

about the origin of life. What is unique about the origin of life? What happens when life arises that we don't see in any other kind of physical system? And the idea we were really centered on was this idea that information seems to play a causal role. So I might do an experiment. I won't do it.

Well, actually, let's do an experiment. How about can everyone raise their hand right now? Thank you. Okay. So what caused that to happen? You know, this is not like a, it's not a totally mysterious property, right? Somebody might give you an atomistic description, but it seems that a better description would be there was some informational property that emerged, say, you know, in minds that developed

language over time. We all speak the same language, you understood what I said, that's embedded in a historical contingency that we share. And that informational property allowed me to actually cause things that you guys just did. And that sort of action at a distance that information seems to have is actually, I think, action across time, not at a distance, but I'll get to that. But anyway, this seems to be the mysterious property, right?

Information looks like it's causal. That doesn't sound quite right, because it doesn't fit with standard physics, and maybe we could explain that by some mechanistic, atomistic narrative. But so far, we haven't been able to reconcile these things. So I think when we're looking at these hard problems, whether we're talking about consciousness,

matter or life, they have some common properties. So a hard problem of consciousness, existing feels like something. Hard problem of matter is that we can't observe anything to exist outside of interactions with other objects. And the hard problem of life, the abstractions in information matter in determining what can exist or what happens. I think the last one is certainly true in the biosphere, right? Everything that has emerged in the biosphere over its four billion years of evolution is

the product of some information that was selected over time. So all of these hard problems in some way deal with the nature of existence, and this to me is actually the most fundamental and interesting unifying thing about them, and also one of the ways that it kind of confronts the standard way we talk about things in physics and why there might be problems. So the nature of existence, what it is to exist is consciousness. How do we determine what things exist is matter, and how some things can be caused to exist and others cannot, which is life. So usually when we talk about the problem of what exists, people like to think about all possible mathematical objects.

So at least this is, I don't know how many of you are trained in, how many people are actually trained in physics in here, so I know. Yeah, yeah, we're the problem children, okay? Excellent. Yeah, so if you go into sort of a mathematical physics description of reality, you start asking these questions about like, why do the laws of physics have the structure

that they have? And we actually don't know, and most of, why do the laws of physics have the structure that they have? And we actually don't know, and most of the explanation for the laws of physics just gets pushed outside of the universe. So people assume their laws, they assume they have the mathematical structure they have, and then we have to assume an initial condition. And so these are boundary conditions put on the outside of the universe that no one has an explanation for. And when we get into the deep philosophical discussion of explaining why these laws, you

know, one of the easiest explanations is to assume that all mathematical structures exist somewhere. So it's a different version of the multiverse hypothesis. Everything exists somewhere, right? This is also the many worlds interpretation of quantum mechanics. So usually our kind of scapegoat for explaining why some things exist and not others is to assume everything exists somewhere. And I don't find that particularly explanatory because it doesn't explain much about why things exist here the way they do.

And usually like it's easy to talk about all possible mathematical structures existing and think that that's a real thing. But if I asked you the question again, and I said all possible molecules, can all possible molecules exist? Does anyone think they can? Why can't they? Does anyone know how big that space is? Go ahead. So this one's a topological constraint. Do you mean you can't make them all because of restrictions on reaction chemistry? I mean, you could potentially have them in an extremely non-nucleic in space, but not enough sufficiently flat space time. Okay.

I'm not sure I understand the argument, but we could get into that later. It'd be fun. Yeah, so I think there's kind of, yeah, go ahead. The universe isn't big enough. Yeah, the universe isn't big enough. I know, I set you out there. No one knew I had a setup in the audience. No, I'm just kidding.

So we don't have enough time or resources in the observable universe to make every possible molecule. And actually, we don't have enough to make every possible small molecule, like a few amino acids size. So I think what people underappreciate is how big the space of possible objects is, even if you look at chemistry. So chem informaticians have no idea, even though they do this all the time, how many

possible molecules there are. Not every molecule will exist, let alone every human being. And I'm bringing this up because I think when people look at pictures like this of the Hubble Deep Field, they think space is big. And we're all taught space is really big, right? The universe is huge. It's billions of light years across. There's billions and billions of galaxies, and each one has billions of stars, and we

are infinitesimal in comparison to this huge, vast space, physical space. And what we don't appreciate is that combinatorial space is really large, right? So I was talking about the space of all possible molecules. This molecule that I'm showing here is Taxol, which was mentioned in the last talk. It's an anti-cancer drug. It is a very heavy, well, not very heavy, but it's a pretty big molecule.

But if you wanted to iterate all possible molecules with the same molecular formula, it exists in this vast space of other possible configurations. And you have to ask the question, well, why does taxol exist on earth and not another molecule out of that space? And it's not like there's, you know, a couple other kinds of options. It's like there are millions of them. It's an exponentially growing space every time you add an atom to a molecule. So this space is really large. And as it turns out, the only place in the universe

that we know that most of the space is even accessible is actually on planet Earth, because we have chemists that can do chemical synthesis. And now we have digitization of chemistry, which is using AI to automate the process and allow exploration of chemical space and synthesis of objects that would never exist anywhere else in the universe.

So I often, I'm not giving an origin of life talk today, but when I do talk about the origin of life, I usually talk about the fact that we wanna go out to space to look for life. And really what we should be doing is looking here on earth in chemical space, because where aliens exist is in the possibility space of things that our universe can construct, and we can do experiments on this planet to look for it without having to worry about

time to get there. And this is interesting to me because it makes it a tractable scientific problem because you're testing a hypothesis in the lab, and you're not just hoping life is common and you're going to happen to hit a planet that it exists on. You can actually start to make predictions about where it should exist. So this is sort of where I'm trying to go with thinking about origins of life is how do we actually look in chemical space for alien life. Now when you think about this space of possibilities, DNA is one molecule in that space.

Proteins are, you know, well it's not one molecule. It's many molecules in that space, depending on your sequence. Proteins are many molecules in that space, but they exist in this huge combinatorial space, and they're products of selection. And so I would say that information is necessary for those molecules to exist. They don't exist spontaneously for free.

They have to emerge as a product of selection and evolution. So some people think DNA can just, you know, be found on Mars. I don't agree, and I'll give arguments why later in my talk. But this idea that information is necessary for some things to exist is not intuitive to us when we're thinking about molecules. Because I think a lot of people think that any complex molecule, no matter how complex, should be able to form in the universe outside of life. And I would argue no.

Now the reason I want to argue no is for the same reason that I think that satellites will not arbitrarily start forming on a planet and launching themselves into space outside of an evolutionary process. So if you think about what's necessary for a planet to do what's called anti-accretion, so familiar physical process is accretion, right? Planets form by gravitational accretion of more mass onto objects that are already getting more massive in the early solar system. Anti-accretion is kind of a weird process.

It might happen by a fluke. For example, if a planet gets hit by an asteroid, it injects some stuff into space. But our planet is the only one we know of right now in the entire universe that has a reliable process of anti-accretion happening from the surface of the planet. That process is that there is a intelligent physical system on the surface of that planet that knows enough about the regularities of the laws of gravitation, some kind of information that's extracted from its physical world to build little boxes and launch them into space.

So that's the product of four billion years of evolution, acquisition of information and knowledge to constrain space possibilities to build satellites and launch them. And a feature of that is it's become reliable and regular. It's a law-like process on our planet. It can happen again and again. An asteroid hitting a planet and launching something

into space is obviously physically possible, but it's kind of a fluke spontaneous event. So it has a different quality to it. And so satellites in the way that we have them, I don't think would exist outside of evolution. Another example of just pushing on the laws of physics as they exist now are not sufficient to explain intelligent things that understand the laws of physics and what they can do. It's very David Deutchian kind of way of thinking about it.

Is thinking about elements on the periodic table, the elements. There are some that, all the way up to 118, Agnesium, which is the highest atomic element that we make in the lab on Earth right now, that are most reliably produced by technology. So if you look at all the physical processes that make the elements,

we might have some that were made just after the Big Bang, some are made in the deaths of stars, and some are made on planets that have evolved intelligent beings that can control all of the mechanisms and boundary conditions sufficient enough to actually synthesize these elements in high abundance. It's not that they're forbidden by the laws of physics,

it's just never gonna happen unless you constrain the space sufficiently far to actually produce these elements, which means that you need information to constrain the space to produce them. So the argument I like to make about life is that information, this property, that is very abstract and we don't know how to talk about associated to life, and its objects only arise as a product of evolution, learning, and selection,

and this is what we call the physics of life. So it's sort of like, you know, in the history of physics, we have discovered new domains of physics by kind of trying to think about things from new perspectives. And one of my favorite unifications in physics was the unification of celestial and terrestrial motion. Right? So before that in human history, no one knew those were governed by the same physics. And now we accept that gravitational gravity is a universal physics. It exists everywhere.

But there are some objects in our universe, like planets and black holes, where gravity is most intense and is like a really prominent physics that we need to use to study those objects. We don't really care about gravity and the atomic nucleus. Information, I think, is a universal physics, but I think the place where it becomes most relevant is when we study life, because that's where that process is most concentrated.

So we are the manifestation of whatever this physics is that we don't understand. Now this is contrary in many ways to sort of the standard way we talk about things in physics. So I was taught in my physics class about particles and boxes and also how brains can emerge from particles and boxes with very low probability. Has anybody heard this before that you're you know, we might be Boltzmann brains. Do you feel like a Boltzmann brain?

we might be Boltzmann brains. Do you feel like a Boltzmann brain? I felt one like a minute ago, but I'm not sure now. No, I'm here. Okay. So the argument is that brains can fluctuate into existence spontaneously because statistical physics allows it and quantum physics allows it. And I think there's a lot of problems with this argument for various reasons. One is that we've never observed it happen. So maybe you might want to ask questions about your laws of physics if they make predictions you can't observe. It is interesting cases where people try to use this as cosmological bounds on properties of the universe and stuff.

So I think that they are interesting as a conceptual tool, but I don't think it's a physical argument. And it's interesting how the Boltzmann brain argument has permeated logic of other arguments that people wanna make about the nature of life and mind and how they relate to physics. It's not like a culturally isolated phenomenon. It's like people really intrinsically think

the universe can generate anything anywhere for free. And I think that's tantamount to intelligent design because it basically says the universe has information for every object to spontaneously fluctuate into existence anywhere at any time. Where is all that information? Please tell me. OK.

People want to tell me in the audience I like this. OK. Got some arguments coming. I'm excited. All right. So where do brains come from? Well, if you deconstruct a brain and look at its past history,

brains are products of evolution, right? So they have a long lineage. So this is a very abstract representation of a brain. It's a human brain, but just imagine, you know, like it needs a body. So I'm not saying brains can be disembodied. So I'm just drawing a picture because I like drawing pictures, they're helpful.

It's really hard to get the art right though. So minds don't spontaneously fluctuate into existence. They are assembled across time through evolution and selection. So if you think about every step to make a mind, to build a brain, every step takes time. And you can't compress all of that time into an instantaneous, spontaneous fluctuation.

A mind is an object that requires time to be constructed. And evolution is the process we know that does that. And so part of the structure of the theory that we're building to explain life talks about objects as being assembled over time. There's a particular way we talk about assembly as a construction process, but the idea being that every evolved object has a structure across time as far as how it is built. So if you start from elementary building blocks, the way we think about it mostly is for molecules right now. So I'm doing a big reach right now by talking about brains, but it's supposed to be a

general theory of evolution and we're working our way toward universal features of the theory, and I have some promising stuff while we're in both labs going on right now. Okay, so if you think about a mind, and you actually look at how you can construct the mind, that is the physical structure of the mind. This is gonna be the argument of the whole talk. It's not the instantaneous object of the brain,

but the brain is an object that is very large in time. And I'm going to say this over and over again, but this is sort of my conception of evolved objects. They're informational, and you can't see the informational properties because they are ones that have a size and time, not in space. And okay, so this is a mind. Now I'll explain a little bit more in a few minutes

about how we think about how to construct this kind of space. So I'm doing it very conceptually here in a quantitative framework as far as the theory that we're building and the principles of the theory to kind of do assembly theory one-on-one. But one important feature of this is once you deconstruct an object in this way, you can see that it's been selected out of a very large space of possible objects. Right? So if I started, if I took a brain apart and, you know, I wanted to look at its pieces, the molecules that make it, those molecules could be assembled in an exponentially growing space. Every time I add two pieces of a molecule together,

every time I form a bond, the size of the space gets exponentially larger. If I get to the size of cells, the space is even larger still. And once I get up to brains and technological civilizations, that space is very large, and it is so large I don't even think that we can even imagine how big that space is.

So remember I said that we can't even estimate how many possible molecules there are with the known periodic table elements. So a brain actually is very deep as far as it's structured, but then as you're building up this space, you see what a huge space of possibilities that a mind exists in. And in some sense, the brain has carried with it that entire history is just rolled up in the present. So it has some capability of understanding

the structure of the full space because it's a combinatorial space and it has some remnants of it stuck in the brain. When you look at that combinatorial space, there could be other minds in that space. And here I'm showing two kinds of brains that exist on our planet, which exist in the same combinatorial space

and in fact share a lineage up to a certain point in their history before they diverged from each other. So if you look at the causal chains of events of operations or construction processes that assembled our brain and you looked at the one that assembled an octopus brain, there would be a lot of overlap. We're not that dissimilar.

But if I looked at my brain and somebody else's in the room, we're even more similar, which is one of the reasons we can share experiences with each other is because we're practically the same object when you look at us deep in time. Now the reason for taking this approach when we're thinking about origins of life, and I also think a lot about alien life,

is we need to generalize beyond life as we know it, I think, to understand a lot of the properties of life as an abstract feature of our universe, in the sense that we talk about in theoretical physics. So if you think about the sort of paradigm of theoretical physics, like the paradigm of the framework of what is theoretical physics about. It's about building abstractions that unify as many parts of reality as possible under a similar conceptual framing. Right, so motion is a good example.

You know, I could look at all moving objects and I don't care about their color. I don't care about their shape. I care about their mass and their velocity. Right, so what abstractions matter for talking about unifying motion is mass and velocity. And those are inventions of our theories that correspond with measuring devices. What I'm saying here is when we're talking about evolved objects, the abstractions that matter are this kind of depth in time and also its relation to other objects. But I'll talk specifically about concrete observables in a minute. But the reason for doing this is because we don't think,

necessarily, at least in this day and age, and I hope is the topic of this conference, that the only minds that exist are biological minds. I'm not sure that I'm convinced any AI right now is a mind, but other people might be. But we do care about technological minds and we care about alien minds. What is the space of possible minds

our universe can construct? I think this is a really relevant question and an interesting question for reframing some of the ways that we talk about the nature of minds. And obviously that's an important, very pressing conversation that people are having now. So when we think about where we exist in the space of possibilities and how we interact with the world and

that other things have co-evolved with us on the same planet, there's a whole bunch of trajectories through this space that we don't know. And those are, you know, maybe technological minds that might be evolved from our same chain of events, our same causal chain constructed over time, or they might be alien minds

that have completely different histories than we do. So I've been putting on this axis here assembly time because I'm not really actually thinking about this as clock time. I think about time in a different way than standard conceptions of physics. It's much more like a causal time. And it's always interesting to me to look back at the history of physics trying to invent new physics because you see that every theory of physics invents its own concept of time, so why not? So actually it's interesting because every century invents a concept of time and they're

really strongly correlated with the cultural narratives of that period in time and technological inventions. Yeah. I'm actually more thinking about like clockwork universe and like Newtonian time is invented based on clocks, right? So what is causal time? I think it's because we live in a very complex world and realizing time is not exactly linear. It's combinatorial, which is fun.

Okay, so I mentioned at the beginning this assembly theory. So assembly theory is a theory that is structured to solve the problem of the origin of life, which I'll talk a little bit how we're thinking about that in a minute. But the way that is like sort of the base of assembly theory is to say that there are objects you observe, right? Those objects we observe are physical things that exist like you and me. And you can deconstruct those objects to their basic building blocks. So let's not talk about you and me because it's too complicated,

but let's talk about a molecule. A molecule is made out of bonds, right? So people think it's made out of atoms, but if you're actually talking about constructing a molecule, a molecule is made out of bonds. So bonds are the physical thing that build molecules. Molecules made out of bonds. So if I take a molecule apart and I look at its bonds, there's a whole space of molecules I could build, that combinatorial space of chemistry I was talking about, and it's exponentially large. And in fact, if I had no physical constraints, that space is super exponentially growing with every new operation I do in that space.

So that space is huge. That space is actually non-physical because I haven't imposed any of the laws of physics or operations we know that can happen in our real universe. So we might want to add some constraints and say we only want to do possible operations. We only want to make bonds that actually conform. So LEGO is maybe even better. So some of you are not chemists. I see people scratching their heads like thinking about how would you construct a human and

how would you construct a molecule? How many people have played with LEGO? LEGO, okay, we know LEGO. Okay. You know how they have little like little pieces on them? How many people have played with Lego? Lego, OK, we know Lego. OK. You know how they have little pieces on them? So what are the knobs called? Iron two?

Studs. Let's just call them studs. OK, so they have little studs on them, right? So assembly, the space of possible Lego structures, if there were no physics involved, no constraints imposed by how they have to fit together. Let's say you have superglue, it's like a really huge space, right? You can make all kinds of weird stuff at a LEGO, and every time

you add a new piece, the number of possibilities grows. If I am constraining myself to actually using those little studs to piece things together, I've constrained the space of possible objects to be ones that are physically consistent with the rules of LEGO universe. And then the next sort of constraint you might want to add is that you can't spontaneously assemble things for free. You have to build them out of things you built already. And this is what we call assembly clause or assembly contingent. This is an old label,

so it's actually called assembly contingent in the paper, but that's this one. So you've added more constraints. And this is the feature that objects are built along historically contingent trajectories, which means if I wanna build a Lego castle, I build a part of it first, and then I add that part to another part,

and then I add that part to another part. And so this actually constrains the sort of trajectories through what objects get built, because once you start constraining and building things, you can only build out of what you built before. And so this means that the sort of trajectories for building objects are historically contingent. So what we see when we take objects that exist

and we deconstruct them is a particular example of a contingent history, but it tells us something about the space of histories that could have been, that sort of space around it of what evolution could have done but didn't. And so assembly theory doesn't have initial conditions. It doesn't have laws of motion. It has objects and their causal histories and then constraining the future based on

the causation that you find in objects in the present. Now this gets at some interesting things because what you talk about is fundamental in this theory is not what we talk about as fundamental in current theories of physics. And one of the reasons for that I think is really important because, again, looking at the history of science, we have invented a lot of technology in the short history of science. And every time we invent new technologies, what we call fundamental changes. A good example is atoms, right?

So atoms are the things in the periodic table, and they're called atoms because we thought they were indivisible, like the Greek atom. And then we realized they had substructure. And now, sort of the most fundamental theories that people wanna propose are things like string theory, which is even more fundamental, but currently inaccessible to our technology.

We wanted to find fundamental as the smallest objects in the universe. The problem is, when you're an observer existing in the universe, what you call fundamental changes with your evolutionary history, how much technology you've evolved, how much you can measure. And so a theory of evolution that tells you about where you are in an evolutionary chain cannot have a fixed, finite object as an indivisible object as your fundamental particle. It can't be the fundamental structure of your theory.

So assembly theory is a theory of objects. It's kind of funny because I got through a whole physics education without knowing what an object was. And then I started to think about it. And so we define objects as having several features. One is that they're finite and distinguishable. So you see a basketball and a soccer ball in the universe. You don't see a schmear of every possible kind of configuration of ball in between.

They're finite, distinguishable objects. Objects are breakable. I could decompose your brain into atoms. I can decompose a molecule into atoms. I could decompose your brain into atoms. I can decompose a molecule into atoms. I cannot decompose an electron into anything. It is not an object. It's a limit of your observations. Objects exist more than once.

This is a weird feature and probably one of the most unintuitive things, but one of the key features of assembly theory is this idea of copy number, that things that are created by evolution can be produced again. And that the copies of objects matter. So we are not identical objects as humans. So you say, like, I'm not an object in the theory. I don't exist. But actually, if you look at us and you look at our causal history, and you look at where

we exist in the space of possible, we're nearly identical. You know, we're one% of our genome is different. Right, so you can follow these causal trees building up these objects all the way to the branches, and we're the kind of the things that are alive now are just the branches of these trees. Objects are lineages. So this gets to the point about objects existing in time, right? So in order to get to something like you or I, we think, you know,

about four billion years of evolution has to happen on a planet. And my friend Michael Lockman is really great about saying when you ask him how old he is that he's 3.8 billion years old, because some of the information in his body is that old. And even in your lifetime, you're not the same Adam's you were when you were born.

So what is you? You are the lineage. You are the thing that keeps getting reconstructed. Objects form via selection. I love this website. Sometimes I put it on my slides. That just will generate a human. It's one of the neural networks that's just fun on the internet.

But you know, like the whole idea is this person does not exist, right? So the space of things that cannot exist is really big. And so you have to ask, well, why are we the ones that exist? Why are chairs the things that exist that we sit on? Because they were selected out of that possibility space to be finite, distinguishable physical objects. Selection is the mechanism that does that. So this depiction here is an assembly space. It is an assembly space for adenosine, which is a molecule that is used in DNA.

And I mentioned that molecules are built out of bonds. So the way we do it is to build the assembly space is you do a joining operation between two elementary building blocks. You build something else. Once that thing is built in the assembly space, you can use it again to build something else and again, and combine it with other features that exist in the causal pathways to build up the object.

But when I talk about assembly theory being a theory of objects, this is what we would usually call the object. But in assembly theory, the object is this pathway. It is an informational object extended in time. So, Adenos object is this pathway. It is an informational object extended in time. So, adenosine is this pathway. It has all of those features as sort of the compression of this object as a physical object

as it is constructed across time. Okay, so I mentioned, you know, thinking about theories of motion, we talk about very specific variables that appear in our abstract theories to describe those objects. And then we give those variables physicality. We talk about these being the real physical features of objects. The things that we assign as physical features of objects are features of our theories, not

necessarily the objects. They're features of our descriptions. They have to have a correspondence with the real properties of the objects. Otherwise, we're not doing physics. But in assembly theory, we care a lot about being able to go in the MATLAB and measure things because we actually wanna do the experiment to solve the origin of life in the lab

using digital chemistry to explore chemical space. And we need to be able to measure when life evolved from a chemical soup in the lab. So we have to measure it. And so the two observable properties that we have in assembly theory, the assembly index, this is how many steps did it take to construct your object? So this is what we mean by assembly time.

How many causal operations? What's the minimal set of causal operations to construct that object? And that set of causal operations has very specific properties that are unique to assembly theory. They're not features of other theories of compressibility, although they have some relationship

to them. Copy number is the second feature. How many copies of the object do you find? If an object emerges once in the universe, it's not necessarily evidence of selection and causation. It could be random. Right? So I said brains don't form randomly

because there's too many construction steps. I would never expect to find a brain as a single object. I will always expect to find brains as collections of objects that are related by a causal history that evolve on a planet in bodies. I don't think minds emerge alone. Okay. So I've already covered some of these,

but I think they're worth just going through a little bit more slowly, and I think I have a little bit of time here. So one thing I've already said is universe can build new things only if the pieces exist already to build them. So assembly spaces are recursive. So the universe in assembly theory is a recursive universe. Everything that you see is a recursively stacked object.

It's built out of things that were built before it. And actually, this feature of recursion is so prominent that I would not expect any evolved objects, anything selected to exist in our universe to not have that property. And I have really strong quantitative reasons for saying that that I don't, that we're writing in a paper right now.

But the structure of these paths is what I described already. So this is not showing molecules now, but this is a graphical representation. So if I have two elements and I stick them together, I can make a line. And if I stick those together and add an extra bond by the process, because I have to stick everything together here,

I can make a triangle, but I could also have made a square. And then what follows subsequently is contingent on that prior history. So you can see how the number of ways of building things as you construct objects becomes vastly untamable in some sense. But the feature of these paths that's important that I want to show you is that one, that they're combinatorial, right? So you take discrete elements and you stick them together to build things. And they're compositional. You can only build on things that you've already built.

The steps must be time and operations that are physical and observable. So the assembly index is a physical observable of molecules. We think it probably is of other objects too, but we have to figure out how to construct the assembly spaces in physically meaningful ways. But this is one example of an assembly pathway. But where it gets interesting is that this sort of idea of building a molecule by taking this sort of minimal path through the space of construction histories

for the molecule is actually a feature that's measurable for real molecules. So this is showing a mass spec as you're adding bonds to a molecule and how the fragmentation pattern in a mass spec actually corresponds with the assembly index. So my collaborator Lee Cronin is rather ingenious on these things. And so he actually developed assembly theory originally

to go in the lab with a mass spec and try to figure out if a system was evolved or not. That was sort of the measurement challenge. And so he took the features of what mass spec measures of molecules to try to construct a way of talking about how evolved the molecule was. And actually, it turns out it's not unique to mass spec. You can measure it with NMR and infrared.

So it's not a feature of a mass spec measurement, but assembly index is measurable. Okay, so I've already mentioned this, but I want to state it more concretely. Complex objects can only exist where there are physical systems with the information to construct them. This is how complex objects are products of evolution. They require objects making more complex objects, making more complex objects, making more complex

objects. Unless you have that whole stack of causation, you cannot form them because you need all the informational constraints along those chains to actually produce that specific object out of that huge possibility space of things that could have existed but don't. That's the only way to get the information. So we expect above a threshold assembly threshold we expect to observe any object unless it was selected to exist. So this is one of the reasons that we're trying to use assembly theory for alien life detection because we don't care about what molecules they are. All we care is are they a product of evolution or not?

Are they deep enough in this construction history that they're signatures of evolution, that selection had to happen? And so in this community, you're probably familiar with the neural correlates of consciousness, right? And that people really think consciousness is not about the neural correlates or something deeper going on there. In my field, the origin of life, people are always fixated on the chemical correlates of life.

Can we find DNA? Is it a protein? How do I make amino acids from a prebiotic soup? That is not the question to ask. The question to ask is, is this system informational or not? Is it evidence that information was acquired over time? And how do we actually talk about that property? So we want to go after the actual physics, not the objects that we see.

What is the abstraction that explains them? So these are two of my favorite quotes on material reality, one from Roger Penrose, who we heard about this morning. I don't like the word materialist because it suggests we know what the material is. And Madonna knows she's a material girl and we're living in a material world. Materials coincidentally are defined by our theories. And so, I think I live in a material world, but I think the things that I think are material are more what other people call informational, because that's actually what we need to talk about

when we talk about the physicality of objects like us, as I've talked about. So this is kind of the image that emerges from these ideas about the nature of life. Life is about these lineages propagating information about these assembly spaces. Original life happens in chemistry. It's actually an abrupt transition. We have a really concrete way of talking about phase transitions in assembly spaces and when they happen, which will be a paper that I hope comes out later this year, which quantifies the original life in different spaces.

New physics emerges in high dimensional combinatorial spaces. What I mean by that is the physics of selection is only apparent when you have so many combinations, the universe cannot exhaust them all, so it has to choose. I say choose in an active sense because we do choosing, but not in a primordial suit. It doesn't choose, but selection does happen. So we can use mass spec to measure for molecules and validate that such a threshold exists.

So this is a paper that came out in 2021 in Nature Communications, which took a whole bunch of samples of living dead, samples blinded by NASA, put them through a mass spec, measured their assembly, and showed that biology was the only thing that produced molecules with high assembly. So we can use it for alien life detection potentially. So we're at a conference about minds, not life. So I wanna just talk a little bit about maybe some new philosophical structures emerging from this theory that I think are fun to play with

when we're thinking about what a theory for minds might entail. The first is that minds do not exist as instantaneous objects. They exist within a historically contingent combinatorial space, right? They are this sort of assembly. They, you know, if assembly theory's on the right track in explaining the physics of life

and we can generalize to other kinds of physical things besides molecules, like the things molecules are built into, then minds should have a size and time just like molecules do. So I like this idea of thinking about hyperobjects, things that are too big for us to conceive. I think that we're not used to thinking about physical objects as having sizes and time.

We see them as having size and space. When I look at a chair, I see the chair is a little more than a foot across. I don't see the feature that it took a billion years to construct the chair. But I also don't see that space time is curved around me. So these abstractions we build from our theories are really non-intuitive,

but the point is are they explanatory or not? I think one of the things that this would explain, which I rather like, is the virtualization of reality. Because things that are informational, why do they look abstract? They look abstract because we're looking at objects that are really deep in time. They're really deeply recursively stacked objects. They compress a lot of time in a small volume of space.

That's why they look virtual. And this means that other minds might be possible in the future of the construction process. So this brings the question of technologies we're building now. Are they deeper in time than we are? Are they flat? I think the kind of technologies we have are more indicative of social minds than human minds, and therefore I don't really qualify them as creative as we are. But I'm happy to discuss all of those things.

But I want to go back to my initial question after kind of bringing all this framing in of how we're thinking about things, which is this idea of like, are there things that can exist because minds exist? And they would be evidence of all of the features that we care about understanding in a theory of mind. And to me, the most interesting thing that we do as humans is imagination. We can imagine things in our minds, we can share them between us, and then we can build them over centuries. And so an object that I like to use as an example

is rockets, right? Humans were imagining sending things to space long before we had the theories or the technology to actually do it. But now rockets are not just imagined objects that we might talk about, they're real physical objects in the real physical universe, with real physical properties.

Where does this come from? Well, I mentioned that I think minds are small in space but big in time, which means that all those combinatorial possibilities exist in that structure that is a mind and the future is constructed by that combinatorial past making possible futures. So in assembly theory, the future is constructed combinatorially from the assembly space that objects exist now. So if you don't think an object is instantaneous, but you think it's a causal graph of these assembly pathways, you have a lot of structure to build futures on.

The future is not linear. The future is bigger than the past in assembly theory. It's deterministic, but it's not determined because you don't have enough stuff now to determine what exactly will happen. And novelty is therefore possible in this theory, which is super important for a theory of life, I think, because life is mostly about novelty generation.

And so the future is more open, more selection is possible for objects that are deeper in the assembly space. And so if you look at a structure like a mind, it's not just that the assembly space might define the object, but the combinatorial future horizon is larger for objects that are deeper in the assembly space. And this is one of the reasons that our imagination is

pretty expansive and why technology is increasing the size of that. So two horizons and phase transitions might be important. I should say might be important in assembly theory. I think the life one, you know, we're pretty close to like really solidifying. Like we have a really concrete formalism about life emerging in molecules and are working out various features of the theory. The phase transition part is worked out.

Mines, I think, are interesting because they pack so much selection in time in a small volume of space that now they actually become active in the selection. It's not just like selection builds up, but it gets select, you know, like all that information gets condensed in an object and they're able to do things, imagine things that weren't possible before

and actually construct them. And I think this is the physics of minds that I find interesting because it's actually embedding us in an evolutionary history and a sort of physical structure for the reality that we live in. I'm going to end by just saying I have a book coming out later this year, which is like

first time I've ever been able to say that. I'm really excited. Actually, this is the first time I've ever said that. In a talk, yay! Life is no one knows, but thanks! Took me forever to write it. Cool. I'm super excited.

It took me a really long time. And then I also want to thank my lab and my collaborators because they're amazing. I work with the most fantastic group of people and they're all crazy and really rigorous, which is a great combination. Yeah, very. Thank you. Do we have time for any questions? Mark.

Thank you. Thank you. It's hard to explain abstraction, but I hope you enjoyed it. Thank you. Thank you. It's hard to explain abstractions that don't exist in a lot of minds yet. Absolutely. And I have a ton of questions I want to ask you later.

Sure. When you talk about combinatorial time and emergence in that quasi-temporal dimension, it makes me think a lot about diachronic emergence. Okay, I'm not familiar with that. So like, you know, probably other people aren't so that's true.

Yeah. So, so like, you know, synchronic emergence would be like emergence that occurs at the same time. Oh, I see. And diachronic emergence would be emergence that occurs sort of across time. So like, if you think about an economy and like the emergence of markets in an economy, like that would be like a diachronic emergent phenomena. Right. Do you, would you see diachronic emergence as being like ontologically distinct from

this sort of combinatorial emergence that you're talking about? Not necessarily. You'd have to know more about it. But my, my sort of philosophy going into these things is all emergent things are temporal. Like I think the reason an emergence looks weird in standard physics is because we think emergence happens in space and not in time. And it's just because the object is constructed over time. So you're looking at an object that's big in time and when you like try to look at objects

that are small in time to explain it, you missed all the time. You coarse-grained out time. You lost all of the physics that you care about. And when you say temporal, do you mean like temporal in the way that we perceive time or more like a quasi-temporal kind of... It's not a clock time. It's a causal time. Okay.

So it's about ordering of events. Very interesting. Yeah, I knew you were going to ask that. Hyper objects, the second time I've heard it is you. Talking to Mark about it before. Cool. Any more questions? Great talk by the way. I'm going to keep it quick.

Thank you. You know, I often think about the way we actually perceive and go about solving issues through cognition and language. So your theory of mind is very interesting to me. And I'm wondering how much of getting the language right in the way you're actually describing things is going to affect the way you solve the problem. Because I know you talk about hyper-object.

Oh, I think absolutely 100%. I use different words all the time and it confuses people, but it's because I don't want to become semantically closed. Because I think the words matter. I think they really matter. And I think when you're building a new abstraction, you're building a new language. And so you need to be really careful about what words you choose, for sure. Perfect.

Yeah, thank you for a very refreshing line of thought. Thank you. Since you use the word object a lot. Yeah. It's a new one for me. I like it. Thank you for a very refreshing line of thought. Since you use the word object a lot, where does an object end? Where does it end? Boundary in space or space-time? Yeah, I think the idea that they have to have finite boundaries. So for molecules, it's actually pretty easy to define the boundary because of like the sort of, you know, like molecules are like have an energy potential

and things that actually give them physical structure. One of the reasons we're having a hard time developing assembly theory for minerals is because it's really hard to define what an object is in a mineral. There's like a crystal unit cell, but the way it repeats is not exact copies because it defects. So it's super hard to define an object in a mineral. It's also hard to define objects in some classes of strings, although in other classes of strings it gets a little easier because there's some semantics about what bounds a string. There's a couple things that are really hard about assembly theory.

One is constructing mathematics of it for new spaces. One is, what is an object? And two is, what are the physical mechanisms of constructing the object? And if you know those two things, you can construct an assembly theory for anything, but those things are actually super hard to define in a way that's rigorous and measurable. And this is one of the reasons that we're taking our time trying to develop assembly theory properly for other kinds of systems besides molecules. Molecules were just like a freebie.

They have these properties one off the bat. It's really weird. Yeah. Isha? Sorry, I just have one question. If an object consists of the evolutionary time history, what about the things that are no longer part of it? So, for an example, is a human mind part of the object because it's part of the history that created it, of a rocket.

Yeah, it's super interesting. Like I, you know, I have to kind of compress things to like be able to explain them, but this idea of lineage as a propagating information is really something that we've been developing with Michael Walkman at Santa Fe Institute. And Michael, you know, has this really nice like way of talking about it, which I really resonate with, is

like, we are bundles of lineages. So you're not like – you're a particular configuration that has come together, but you're parts. Like, all the things you create or all the things that are parts of you can also become parts of other objects. So you're not – like, even when – so the example I like to give is like, Einstein's biological lineage wasn't very creative in the universe. He doesn't is like, you know, Einstein's biological lineage wasn't very creative in the universe.

He doesn't have a lot of ancestors, right? But his creative lineage, like the ideas he created, have propagated through a lot of minds. So that lineage has actually been much more persistent than his biological lineage. But they're both lineages. And they both should be talked about as objects equivalent in assembly theory. It doesn't make a distinction between abstract objects and things that we would traditionally consider physical objects.

Well, I really like how you looked at it in the Noamah piece I brought before. Thank you. Looking at technology as an extension of that propagating lineage. Yeah. Where do you think that's going down the road? Oh, technology. I mean, how big does that space get? Yeah.

Artificial intelligence. How big does life get? Yeah. I think, I mean, the projection I have is if we can solve the origin of life. So a technosphere, you have to emerge a technosphere to solve the origin of life, right? So a technosphere understands where its deep biosphere came from, and then the technosphere

is a planetary scale living structure can reproduce on another planet. Thinking with, you know, once you get AI able to maybe increase that directed selection like we're doing right now with AI. It's like, how quickly can that expand and how quickly can things adapt and evolve?

Yeah, fun to think about. The new technologies are kind of deeper or flat, not as creative as we are. Is it because the assumption into AI, it's not keeping this new proposition that you have in mind? If they adopted that the models will change? I think it's possible to build things, but I just think that we're in the earliest phases of what these things are. And I, I, the way I kind of see like, you know, large language

models to me are super interesting because it's like you took language, which was a technology that was distributed across many, many minds, right? And you compressed it in a small device so that an individual human can now interact with language as a physical object. That's what a large language model is to me. It's not so, but if you look at the ecosystems that our technologies were creating,

they might have more of my like functionality than any individual does. So I tend to think about more like global emergence of technology and like how are these things getting integrated together? Which I did write about in this piece called AI's Life in Naweema. Yeah, it's fabulous.

Where am I looking? Sorry. Hi. Hi. Hi. Sorry. I like to put a face with a question. Thank you for the speech. My question is, when you're talking about mind, you use brain as an illustration, isolated from the rest neural system, neural system. So I wanted to put a nervous nervous system.

I just couldn't get a good visual. So I actually I use my. Yeah. So this is one of the things about the semantics, like which word do I use? Brain, mind, nervous system. Like what do I actually mean? I don't know. Okay. Perfect. So I'm agnostic. Next question. The question is related.

Yeah. What about congenitin, congenitin twins like that have. Oh, right. Yeah. The next question is related. What about conjoined twins? Oh, right. So they technically are the same object. They share the same body. But if they have two brains, they realize themselves as two separate minds. Yeah, I think that's interesting why you even recognize yourself as separate minds? Yeah. I think that's a, yeah. I mean, it's interesting why you even recognize yourself as separate from the world.

I think that's more mysterious than the fact that we have minds. Like, I have such a, because of the way I think about the physics of life, I hardly think of myself as an individual. Because I recognize that I'm a lineage and like everything on this planet is part of the same structure I am. So that part's interesting to me, but I think, you know, like our minds are in part constructed because of our social realities.

So I don't know if I can decouple thinking about consciousness from sociality. And that part's more interesting to me than the actual physical embodiment of like, what are the individuals? Most of my, I mean, all of my conscious thoughts are associated with my social, like communicating and like thinking about things that I like, you know, all the non-conscious stuff you guys don't get to know about and it's crazy down there. Thank you. Thank you.

That was Sarah, that was great. I take it you're going to disagree with my contention that consciousness came first. My question is, what's the role, I also believe that life, even very simple life, requires quantum coherence. At the micelle level, the cellular level and so forth. Do you disagree with that also? My philosophy on quantum physics and its role in life is kind of similar to my philosophy on thermodynamics

and also gravity. I think these are theories that, you know, they describe features of physical reality, but I don't think they're capturing the regularities relevant to what life is doing. And so things that life builds will have those properties because life is a physical system. But the thing, the abstraction we want to call life

is not embedded in those theories. Why is it an abstraction? But I'll just leave it at that. Well, because theories are abstractions. If you want a theory of life, it will be an abstract description of the regularities you associate to life. That's what a theory is. Well, like Schrodinger's theory of life as a quantum coherent state. I think he was after negentropy, which I don't agree with his theory, though. So I think I like after neg entropy, which I don't, I mean, I don't agree with his theory though. So I think, you know, I always, I like to quip that, like, you know, he didn't use mathematical deduction in what is life because he knew he couldn't actually mathematical life.

But, but, but I like, I think, I think it's a hard problem. It's a really hard problem. It's why we haven't solved the origin of life. We haven't designed the right experiments. Okay. Thank you. designed the right experiments. Okay, thank you. Yeah. All right, questions? Great talk, thank you. So I have a very basic question.

So what fundamental principles do you have in a simplifedia that distinguish conscious from non-conscious entities? You know, it's interesting because people always want to... I made a joke a few minutes ago, but I'm like way more interested in what's happening to the unconscious than my consciousness, because all these things just like pop in the surface sometimes and I'm like, where did that come from? And so I think it's interesting that we have experience in things, but I'm not sure it's

the most interesting thing going on in our brains. But that being said, I think the main difference is in the only way I know how to codify it in English language is this idea of imagination. I think for most of biological history, objects were built on things that existed in this much smaller combinatorial space because the objects don't have as large a history.

And there's something about the structure of the brain that I think the history is actually larger than the space. So it actually seems like it's pulling counterfactuals into existence, but it's because it exists in a larger space than is like, I don't know how to articulate it yet. But that's the closest answer I can give. Trying to work toward actually formalizing that, but it's really hard. Wonderful talk.

I'm sorry I didn't catch the whole thing, but the questions that you answered, I thought were very interesting because you talk about you're interested more in what is unconscious rather than what is conscious. It's the stuff that will be conscious in the future. That's why I'm interested in it. Right. Well, I'm interested Having a background cybernetics in kung fu. I'm interested in I look at it I look at it kind of like the you know, you have DOS and then you have windows running on top of it

Yeah, and if you get slapped in the head you get right back to DOS. Yeah so in in kind of the organization of the central nervous system into the mind, do you believe that that is part of the assembly blocks, I guess? Yeah. I guess in some sense, I think consciousness is really interesting because it exists in the present, right? So there is something very distinct about the present. Like we're hallucinating the present right now because our touch operates at.04 seconds,

so we feel the ground. Yeah, but I think that this is also interesting because people think the present is instantaneous. I think the present is thick. So like a present could take a second or one. Yeah. Nicholas Jeasant has a really nice way of talking about this in intuitionist mathematics and the fact that it takes time to compute numbers and like the present actually has a thickness associated with it if you don't if you don't assume numbers or

exists as a continuum. Right I always looked at dancing as an example of how they can play with the thickness of time. Yeah yeah. Awesome. Yeah psychedelics do it I guess I don't know. Thank you, Sarah. And these are great talks. Firstly, thank you for watching. Thank you for listening. There's now a website, curtjymongle.org, and that has a mailing list. The reason being that large platforms like YouTube, like Patreon, they can disable you for whatever reason, whenever they like. That's just part of the terms of service. Now a direct mailing list ensures that I have an untrammeled communication with you. Plus, soon I'll be releasing a one-page PDF of my top 10 toes.

It's not as Quentin Tarantino as it sounds like. Secondly, 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 Kurt 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 hey people are talking about

this content outside of YouTube which in turn greatly aids the distribution on YouTube. Thirdly there's a remarkably active Discord and subreddit for theories of everything where people explicate toes, they disagree respectfully about theories and build as a community our own toe. Links to both are in the description. Fourthly 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 rewatching lectures and podcasts.

I also read in the comments that hey, T.O. 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 Kurtjmungle 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

toe 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.