Lucas Perry: Welcome to the Future of Life Institute Podcast. I’m Lucas Perry. Recently we had a conversation between Max Tegmark and Yuval Noah Harari where in consideration of 21st century technological issues Yuval recommended “Get to know yourself better. It’s maybe the most important thing in life. We haven’t really progressed much in the last thousands of years and the reason is that yes, we keep getting this advice but we don’t really want to do it…. I mean, especially as technology will give us all, at least some of us, more and more power, the temptations of naive utopias are going to be more and more irresistible and I think the really most powerful check on these naive utopias is really getting to know yourself better.”

Drawing inspiration from this, our following podcast was with Andres Gomez Emillson and David Pearce on different views of identity, like open, closed, and empty individualism, and their importance in the world. Our conversation today with Anthony Aguirre follows up on and further explores the importance of questions of self and identity in the 21st century.

This episode focuses on exploring this question from a physics perspective where we discuss the view of reality as fundamentally consisting of information. This helps us to ground what actually exists, how we come to know that, and how this challenges our commonly held intuitions about there existing a concrete reality out there populated by conventionally accepted objects and things, like cups and people, that we often take for granted without challenging or looking into much. This conversation subverted many of my assumptions about science, physics, and the nature of reality, and if that sounds interesting to you, I think you’ll find it valuable as well. 

For those of you not familiar with Anthony Athony, he is a physicist that studies the formation, nature, and evolution of the universe, focusing primarily on the model of eternal inflation—the idea that inflation goes on forever in some regions of universe—and what it may mean for the ultimate beginning of the universe and time. He is the co-founder and associate scientific director of the Foundational Questions Institute and is also a co-founder of the Future of Life Institute. He also co-founded Metaculus, an effort to optimally aggregate predictions about scientific discoveries, technological breakthroughs, and other interesting issues.

The Future of Life Institute is a non-profit and this podcast is funded and supported by listeners like you. So if you find what we do on this podcast to be important and beneficial, please consider supporting the podcast by donating at futureoflife.org/donate. These contributions make it possible for us to bring you conversations like these and to develop the podcast further. You can also follow us on your preferred listening platform by searching for us directly or following the links on the page for this podcast found in the description.

And with that, let’s get into our conversation with Anthony Aguirre.

So the last time we had you on, we had a conversation on information. Could you take us through the history of how people have viewed fundamental reality and fundamental ontology over time from a kind of idealism to then materialism to then this new shift that's informed by quantum mechanics about seeing things as being constituted of information.

Anthony Aguirre: So, without being a historian of science, I can only give you the general impression that I have. And of course through history, many different people have viewed things very different ways. So, I would say in the history of humanity, there have obviously been many, many ways to think about the ultimate nature of reality, if you will, starting with a sense that the fundamental nature of external reality is one that's based on different substances and tendencies and some level of regularity in those things, but without a sense that there are firm or certainly not mathematical regularities and things. And that there are causes of events, but without a sense that those causes can be described in some mathematical way.

So that changed obviously in terms of Western science with the advent of mechanics by Galileo and Newton and others showing that there are not just regularities in the sense that the same result will happen from the same causes over and over again, that was appreciated for a long time, but that those could be accessed not just experimentally but modeled mathematically and that there could be a relatively small set of mathematical laws that could then be used to explain a very wide range of different physical phenomena. I think that sense was not there before, it was clear that things caused other things and events caused other events, but I suspect the thinking was that it was more in a one off way, like, "That's a complicated thing. It's caused by a whole bunch of other complicated things. In principle, those things are connected." But there wasn't a sense that you could get in there and understand what that connection was analytically or intellectually and certainly not in a way that had some dramatic economy in the sense that we now appreciate from Galileo and Newton and subsequent physics.

Once we had that change to mathematical laws, then there was a question of, what are those mathematical laws describing? And the answer there was essentially that those mathematical laws are describing particles and forces between particles. And at some level, a couple of other auxiliary things like space and time are sort of there in the backdrop, but essentially the nature of reality is a bunch of little bits of stuff that are moving around under mathematically specified forces.

That is a sort of complete-ish description. I mean certainly Newton would have and have not said that that's a complete description in the sense that, in Newton's view, there were particles and those particles made up things and the forces told them exactly what to do, but at the same time there were lots of other things in Newton's conception of reality like God and presumably other entities. So it's not exactly clear how materialist Newton or Galileo for example were, but as time went on that became a more entrenched idea among hardcore theoretical physicists at least, or physicists, that there was ultimately this truest, most fundamental, most base description of reality that was lots of particles moving around under mathematical forces.

Now, that I think is a conception that is very much still with us in many senses but has taken on a much deeper level of subtlety given the advent of modern physics including particularly quantum mechanics and also I think a sort of modern recognition or sort of higher level maybe of sophistication and thinking about the relation between different descriptions of natural phenomena. So, let's talk about quantum mechanics first. Quantum mechanics does say that there are particles in a sense, like you can say that there are particles but particles aren't really the thing. You can ask questions of reality that entail that reality is made of particles and you will get answers that look like answers about particles. But you can also ask questions about the same physical system about how it is as a wave and you will get answers about how it is as a wave.

And in general in quantum mechanics, there are all sorts of questions that you can ask and you will get answers about the physical system in the terms that you asked those questions about. So as long as it is a sort of well-defined physical experiment that you can do and that you can translate into a kind of mathematical form, what does it mean to do that experiment? Quantum mechanics gives you a way to compute predictions for how that experiment will turn out without really taking a particular view on what that physical system is, is it a particle? Is it a wave? Or is it something else? And I think this is important to note, it's not just that quantum mechanics says that things are particles and waves at the same time, it's that they're all sorts of things at the same time.

So you can ask how much of my phone is an elephant in quantum mechanics. A phone is totally not the same thing as an elephant, but a phone has a wave function, so if I knew the wave function of the phone and I knew a procedure for asking, "Is something an elephant?", then I could apply that procedure to the phone and the answer would not be, "No, the phone is definitely not an elephant." The answer would be, "The phone is a tiny, tiny, tiny, tiny, tiny bit an elephant." So this is very exaggerated because we're talking phones and elephants, all these numbers are so tiny. But the point is that I can interrogate reality in quantum mechanics in many different ways. I can formulate whatever questions I want and it will give me answers in terms of those questions.

And generally if my questions totally mismatched with what the system is, I'll get, "No, it's not really that." But the no is always a, "No, the probability is incredibly tiny that it's that." But in quantum mechanics, there's always some chance that if you look at your phone, you'll notice that it's an elephant. It's just that that number is so tiny that it never matters, but when you're talking about individual particles, you might find that that probability is significant, that the particle is somewhat different than you thought it was and that's part of the quantum uncertainty and weirdness.

Lucas Perry: Can you unpack a little bit that quantum uncertainty and weirdness that explains, when you ask questions to quantum mechanics, you don't ever get definite answers? Is that right?

Anthony Aguirre: Almost never. So there are occasions where you get definite answers. If you ask a question of a quantum system and it gives you an answer and then you ask that question immediately again, you'll get the same answer for sure.

Lucas Perry: What does immediately mean?

Anthony Aguirre: Really immediately. So formally, like immediately, immediately. If time goes by between the two measurements then the system can evolve a little bit and then you won't definitely get the same answer. That is if you have a quantum system, there is a particular set of questions that you can ask it that you will get definite answers to and the quantum state essentially is that set of questions. When you say an electron is here and it has this spin that is, it's rotating around this direction, what you really mean is that there are a particular set of questions like, "Where are you? And what is your spin?" That if you asked them of this electron, you would get a definite answer.

Now if you take that same electron that I was going to ask those questions to and I would get a definite answer because that's the state the electron is in, but you come along and ask a different question than one of the ones that is in that list, you will get an answer but it won't be a definite answer. So that's kind of the fundamental hallmark of quantum mechanics is that the list of questions you can ask to which you will get a definite answer is a finite one. And for a little particle it's a very short list, like an electron is a very short list.

Lucas Perry: Is this because the act of observation includes interaction with the particle in such a way that it is changed by the interaction?

Anthony Aguirre: I think that's a useful way to look at it in a sense, but it's slightly misleading in the sense that as I said, if you ask exactly the right question, then you will get a definite answer. So you haven't interfered with the system at all if you ask exactly the right question.

Lucas Perry: That means performing the kind of experiment that doesn't change what the particle will be doing or its nature? Is that what that means?

Anthony Aguirre: Yes. It's sort of like you've got a very, very particularly shaped net and you can cast it on something and if the thing happens to have exactly the right shape, your net just falls right over it and it doesn't affect the thing at all and you say, "Oh, it has that property." But if it has any other shape, then your net kind of messes it up, it gets perturbed and you catch something in your net. The net is your experiment, but you mess up the system while you're doing it, but it's not that you necessarily mess up the system, it's that you're asking it a question that it isn't ready to answer definitively, but rather some other question.

So this is always true, but it's kind of the crucial thing of reality. But the crucial thing about quantum mechanics is that that list is finite. We're used to asking any question that... I've got a mug, I can ask, "Is it brown? Is it here? Is it there? How heavy?" Whatever question I think of, I feel like I can answer. I can ask the question and there will be an answer to it because whatever question I ask, if it's a well-defined question before I ask it, the mug either has that property or it doesn't. But quantum mechanics tells us that is true. But there's only a finite number of answers there are built in to the object. And I can ask other questions, but I just can't expect the answer to already be there in the sense that I'll get a definite answer to it.

So this is a very subtle way that there's this interactive process between the observer and the thing that's observed. If we're talking about something that is maximally specified that it has a particular quantum state, there is some way that it is in a sense, but you can't ever find that out because as soon as you start asking questions of it, you change the thing unless you happen to ask exactly the right questions. But in order to ask exactly the right questions, you would already have to know what state it's in. And the only way you can do that is by actually creating the system effectively.

So if I create an electron in a particular state in my lab, then I know what state it's in and I know exactly what questions to ask it in order to get answers that are certain. But if I just come across an electron in the wild, I don't know exactly what questions to ask. And so I just have to ask whatever questions I will and chances are it won't be the right questions for that electron. And I won't ever know whether they were or not because I'll just get some set of answers and I won't know whether those were the properties that the electron actually had already or if they were the ones that it fell into by chance upon my asking those questions.

Lucas Perry: How much of this is actual properties and features about the particles in and of themselves and how much is it about the fact that we're like observers or agents that have to interact with the particles in some ways in order to get information about them? Such that we can't ask too many questions without perturbing the thing in and of itself and then not being able to get definitive answers to other questions?

Anthony Aguirre: Well, I'm not sure how to answer that because I think it's just that is the structure of quantum mechanics, which is the structure of reality. So it's explicitly posed in terms of quantum states of things and a structure of observations that can be made or observables that can be measured so you can see whether the system has a particular value of that observable or not. If you take out the observation part or the measurement part, you just have a quantum state which evolves according to some equation and that's fine, but that's not something you can actually compare in any sense to reality or to observation or use in any way. You need something that will connect that quantum state and evolution equation to something that you can actually do or observe.

And I think that is something that's a little bit different. You can say in Newtonian mechanics or classical physics, there's something arguably reasonable about saying, "Here is the system, it's these particles and they're moving around in this way." And that's saying something. I think you can argue about whether that's actually true, that that's saying something. But you can talk about the particles themselves in a fairly meaningful way without talking about the observer or the person who's measuring it or something like that. Whereas in quantum mechanics, it's really fairly useless to talk about the wave function of something without talking about the way that you measure things or the basis that you operate it on and so on.

That was a long sort of digression in a sense, but I think that's crucial because that I think is a major underlying change in the way that we think about reality, not as something that is purely out there, but understanding that even to the extent that there's something out there, any sense of our experiencing that is unavoidably an interactive one and in a way that you cannot ignore the interaction, that you might have this idea that there's an external objective reality that although it's inconvenient to know, although on an everyday basis you might mess with it a little bit when you interact with it, in principle it's out there and if you could just be careful enough, you could avoid that input from the observer. Quantum mechanics says, "No. That's a fundamental part of it. There's no avoiding that. It's a basic part of the theory that reality is made up of this combination of the measurer and the state."

I also think that once you admit, because you have to in this case that there is more to a useful or complete description of reality than just the kind of objective state of the physical system, then you notice that there are a bunch of other things that actually are there as well that you have to admit are part of reality. So, if you ask some quantum mechanical question, like if I ask, "Is my mug brown? And is it spinning? Where is it?" Those kinds of questions, you have to ask, what is the reality status of those questions or the categories that I'm defining and asking those questions? Like brownness, what is that? That's obviously something that I invented, not me personally, but I invented in this particular case. Brownness is something that biological creatures and humans and so on invented. The sensation of brown is something that biological creatures maybe devised, the calling something brown and the word brown are obviously human and English creations.

So those are things that are created through this process and are not there certainly in the quantum state. And yet if we say that the quantum state on its own is not a meaningful or useful description of reality, but we have to augment it with the sorts of questions that we ask and the sort of procedure of asking and getting questions answered, then those extra things that we have to put into the description entail a whole lot of different things. So there's not just the wave function. So in that simple example, there's a set of questions and possible answers to those questions that the mug could give me. And there are different ways of talking about how mathematically to define those questions.

One way is to call them course grained states or macro states, that is, there are lots of ways that reality can be, but I want to extract out certain features of reality. So if I take the set of possible ways that a mug can be, there's some tiny subset of all those different ways that the atoms in my mug could be that I would actually call a mug and a smaller subset of those that I would call a brown mug and a smaller subset of those that I would call a brown mug that's sitting still and so on. So they're kind of subsets of the set of all possible ways that a physical system with that many atoms and that mass and so on could be and when I'm asking questions about the mug, like are you brown? I'm asking, "Is the system in that particular subset of possibilities that I call a brown mug sitting on a table?"

I would say that at some level, almost all of what we do in interacting with reality is like that process. There's this huge set of possible realities that we could inhabit. What we do are to divvy up that reality into many, many possibilities corresponding to questions that we might ask and answers to those questions we might ask and then we go and ask those questions of reality and we get sort of yes or no answers to them. And quantum mechanics is sort of the enactment of that process with full exactness that applies to even the smallest systems, but we can think of that process just on a day to day level, like we can think of, what are all the possible ways that the system could be? And then ask certain questions. Is it this? Is it that?

So this is a conception of reality that's kind of like a big game of 20 questions. Every time we look out at reality, we're just asking different questions of it. Normally we're narrowing down the possibility space of how reality is by asking those questions, getting answers to it. To me a really interesting question is like, what is the ontological reality status of all those big sets of questions that we're asking? Your tendency as a theoretical physicist is to say, "Oh, the wave function is the thing that's real and that's what actually exists, and all these extra things are just extra things that we made up and our globbed onto the wave function." But I think that's kind of a very impoverished view of reality, not just impoverished, but completely useless and empty of any utility or meaning because quantum mechanics by its nature requires both parts. The questions and the state. If you cut out all the questions, you're just left with this very empty thing that has no applicability or meaning.

Lucas Perry: But doesn't that tell us how reality is in and of itself?

Anthony Aguirre: I don't think it tells you anything, honestly. It's almost impossible to even say what the wave function is except in some terms. Like if I just write down, "Okay, the wave function of the universe is psi." What did that tell me? Nothing. There's nothing there. There's no way that I could even communicate to you what the wave function is without reference to some set of questions because remember the wave function is a definite set of answers to a particular set of questions. So, I have to communicate to you the set of questions to which the wave function is the definite answer and those questions are things that have to do with macroscopic reality.

There's no way that I can tell you what the wave function is if I were to try to communicate it to you without reference to those questions. Like if I say, "Okay, I've got a thingie here and it's got a wave function," and you asked me, "Okay, what is the wave function?" I don't know how to tell you. I could tell you it's mass, but now what I'm really saying is, here's a set of energy measuring things that I might do and the amplitude for getting those different possible outcomes in that energy measuring thing is 0.1 for that one and 0.2 for that one and so on. But I have to tell you what those energy measuring things are in order to be able to tell you what the wave function is.

Lucas Perry: If you go back to the starting conditions of the universe, that initial state is a definite thing, right? Prior to any observers and defined coherently and exactly in and of itself. Right?

Anthony Aguirre: I don't know if I would say that.

Lucas Perry: I understand that for us to know anything we have to ask questions. I'm asking you about something that I know that has no utility because we're always going to be the observer standing in reference, right? But just to think about reality in and of itself.

Anthony Aguirre: Right. But you're assuming that there is such a thing and that's not entirely clear to me. So I recognize that there's a desire to feel like there is a sort of objective reality that is out there and that there's meaning to saying what that reality is, but that is not entirely clear to me that that's a safe assumption to make. So it is true that we can go back in time and attribute all kinds of pretty objective properties of the universe and it certainly is true that it can't be that we needed people and observers and things back at the beginning in order to be able to talk about those things. But it's a very thorny question to me, that it's meaningful to say that there was a quantum state that the universe had at the very beginning when I don't know what operationally that means. I wouldn't know how to describe that quantum state or make it meaningful other than in terms of measurable things which require adding a whole bunch of ingredients to the description of what the universe is.

To say that the universe started in this quantum state, to make that meaningful requires these extra ingredients. But we also recognize that those extra ingredients are themselves parts of the universe. So, either you have to take this view that there is a quantum state and somehow we're going to get out of that in this kind of circular self-consistent way, a bunch of measuring apparatuses that are hidden in that quantum state and make certain measurements and then define the quantum state in this bootstrapping way. Or you have to say that the quantum state, and I'm not sure how different these things are, that the quantum state is part of reality, but in order to actually specify what reality is, there's a whole bunch of extra ingredients that we have to define and we have to put in there.

And that's kind of the view that I take nowadays, that there is reality and then there's our description of reality. And as we describe reality, one of the things that we need to describe reality are quantum states and one of the things that we need to describe reality are coarse grainings or systems of measurement or bases and so on. There are all these extra things that we need to put in. And the quantum states are one of them and a very important one. And evolution equations are one of them in a very important one. But to identify reality with the state plus the fundamental laws that evolve that state, I just don't think is quite the right way to think about it.

Lucas Perry: Okay, so this is all very illuminating for this perspective here that we're trying to explore, which is the universe being simply constituted of information.

Anthony Aguirre: Yeah, so let's talk about that. Once you let go, I think of the idea that there is matter that is made of particles and then there are arrangements of that matter and there are things that that matter does, but the matter is this intrinsically existing stuff. Once you start to think of there being the state, which is a set of answers to questions, that set of answers to questions is a very informative thing. It's a kind of maximally informative thing, but it isn't a different kind of thing to other sets of answers to questions.

That is to say that I've got information about something, kind of is saying that I've asked a bunch of questions and I've gotten answers about it so I know about it. If I keep asking enough incredibly detailed questions that maybe I've maximally specified the state of the cup and I have as much information as I can have about the cup. But in that process, as I ask more and more information, as I more and more specify what the cup is like, there's no particular place in which the cup changes its nature. So I start out asking questions and I get more and more and more information until I get the most information that I can. And then I call that, that's the most information I can get and now I've specified the quantum state of the cup.

But in that sense then a quantum state is like the sort of end state of a process of interrogating a physical system to get more and more information about it. So to me that suggests this interpretation that the nature of something like the quantum state of something is an informational thing. It's identified with a maximal set of information that you can have about something. But that's kind of one end of the spectrum, the maximal knowing about that thing end of the spectrum. But if we don't go that far, then we just have less information about the thing. And once you start to think that way, well what then isn't information? If the nature of things is to be a state and a set of questions and the state gives me answers to those questions, that's a set of information. But as I said, that sort of applies to all physical systems that's kind of what they are according to quantum mechanics.

So there used to be a sense, I think that there was a thing, it was a bunch of particles and then when I ask questions I could learn about that thing. The lesson to me of quantum mechanics is that there's no space between the answers to questions that I get when I ask questions of a thing and the thing itself. The thing is in a sense, the set of answers to the questions that I have or could ask of it. It comes much less of a kind of physical tangible thing made of stuff and much more of a thing made out of information and it's information that I can get by interacting with that thing, but there isn't a thing there that the information is about. That notion seems to be sort of absent. There's no need to think that there is a thing that the information is about. All we know is the information.

Lucas Perry: Is that true of the particles arranged cup wise or the cup thing that is there? Is it true of that thing in and of itself or is that basically just the truth of being an epistemic agent who's trying to interrogate the cup thing?

Anthony Aguirre: Suppose the fundamental nature of reality was a bunch of particles, then what I said is still true. I can imagine if things like observers exist, then they can ask questions and they can get answers and those will be answers about the physical system that kind of has this intrinsic nature of bits of stuff. And it would still, I think, be true that most of reality is made of everything but the little bits of stuff, the little bits of stuff are only there at the very end. If you ask the very most precise questions you get more and more a sense of, "Oh they're little bits of stuff." But I think what's interesting is that what quantum mechanics tells us is we keep getting more and more fine grained information about something, but then at the very end rather than little bits of stuff, it sort of disappears before our eyes. There aren't any little bits of stuff there, there's just the answers to the most refined sets of questions that we can ask.

So that's where I think there's sort of a difference is that there's this sense in classical physics that underlying all these questions and answers and information is this other thing of a different nature, that is matter and it has a different fundamental quality to it than the information. And in quantum mechanics it seems to me like there's no need to think that there is such a thing, that there is no need to think that there is some other different stuff that is non-informational that's out there that the information is about because the informational description is complete.

Lucas Perry: So I guess there's two questions here that come out of this. It'd be good if you could define and unpack what information exactly is and then if you could explore and get further into the idea of how this challenges our notion of what a macroscopic thing is or what a microscopic or what a quantum thing is, something that we believe to have identity. And then also how this impacts identity like cup identity or particle identity, what it means for people and galaxies and the universe to be constituted of information. So those two things.

Anthony Aguirre: Okay. So there are lots of ways to talk about information. There are also qualitative and quantitative ways to talk about it. So let me talk about the quantitative way first. So you can say that if I have a whole possibility space, like many different possibilities for the way something can be and then I restrict those possibilities to a smaller set of possibilities in some way. Either I say it's definitely in one of these, or maybe there's a higher probability that it's one of these than one of those. I, in some way restrict rather than every possibility is the same, I say that some possibilities are more than others, they're more likely or it's restricted to some subset. Then I have information about that system and that information is precisely the gap between everything being just equally likely and every possibility being equally good and knowing that some of them are more likely or valid or something than others.

So, information is that gap that says it's more this than some of those other things. So, that's a super general way of talking about it but that can be made very mathematically precise. So if I say there are four bits of information stored in my computer, exactly what I mean is that there are a bunch of registers and if I don't know whether they're ones or zeros, I say I have no information. If I know that these four are 1101, then I've restricted my full set of possibilities to this subset in which those are 1101 and I have those four bits of information. So I can be very mathematically precise about this. And I can even say if the first bit, well I don't know whether it's 01 but it's 75% chance that it's zero and 25% chance that it's one, that's still information. It's less than one bit of information.

People think of bits as being very discrete things, but you can have fractions of bits of information. There's nothing wrong with that. The very general definition as restrictions away from every possibility being equally likely to some being more likely than others. And that can be made mathematically precise and is exactly the sort of information we talk about when we say, "My hard drive is 80 gigabytes in size or I have 20 megabits per second of internet speed." It's exactly that sort of information that we're quantifying.

Now, when I think about a cup, I can think about the system in some way like, there are some number of atoms like 10 to the 25th or whatever, atoms or electrons and protons and neutrons or whatever, and there are then some huge, huge possible set of ways that those things can be and some tiny, tiny, tiny, tiny, tiny, tiny, almost infinitesimally tiny subset of those ways that can be are something that I would label a cup. So if I say, "Oh look, I have a cup", I'm actually specifying a vast amount of information by saying, "Look, I have a cup."

Now if I say, "Look, I have a cup and inside it is some dregs of coffee." I've got a huge amount more information. Now, it doesn't feel like a huge amount more of information. It's just like, "Yeah, what did I expect? Dregs of coffee." It's not that big of a deal but physically speaking, it's a huge amount of information that I've specified just by noticing that there are dregs of coffee in the cup instead of dregs of all kinds of other liquids and all kinds of other states and so on.

So that's the quantitative aspect, I can quantify how much information is in a description of a system and the description of it is important because you might come along and you can't see this cup. So I can tell you, there's some stuff on my desk. You know a lot less about what's on my desk than I do. So we have different descriptions of this same system and I've got a whole lot more information than you do about what's on my desk. So the information, and this is an important thing, is associated with somebody's description of the system, not necessarily a person's, but any way of specifying probabilities of the system being in a subset of all of its possibilities. Whether that's somebody describing it or whatever else, anything that defines probabilities over the states that the system could be in, that's defining an amount of information associated with those probabilities.

So there's that quantity. But there's also, when I say, what is a mug? So you can say that the mug is made of protons, electrons, and neutrons, but of course pretty much anything in our world is made of protons, neutrons, and electrons. So what makes this a mug rather than a phone or a little bit of an elephant or whatever, is the particular arrangement that those atoms have. To say that a mug is just protons, neutrons, and electrons, I think is totally misleading in the sense that the protons, neutrons, and electrons are the least informative part of what makes it a mug. So there's a quantity associated with that, the mug part of possibility space is very small compared to all of the possibilities. So that means that there's a lot of information in saying that it's a mug.

But there's also the quality of what that particular subset is and that that particular subset is connected in various ways with things in my description, like solidity and mass and brownness and hardness and hollowness. It is at the intersection of a whole bunch of other properties that a system might have. So each of those properties I can also think of as subsets of possibility space. Suppose I take all things that are a kilogram, that's how many protons, neutrons, and electrons they have. So, that's my system. There's a gazillion different ways that a kilogram of protons and neutrons and electrons can be where we could write down the very exponential numbers that it is.

Now, if I then say, "Okay, let me take a subset of that possibility space that are solid," that's a very small subset. There are lots of ways things can be gases and liquids. Okay, so I've made a small subset. Now let me take another property, which is hardness. So, that's another subset of all possibilities. And where hardness intersect solid, I have hard, solid things and so on. So I can keep adding properties on and when I've specified enough properties, it's something that I would give the label of a mug. So when I ask, what is a mug made of? In some sense it's made of protons, neutrons, and electrons, but I think in a more meaningful sense, it's made of the properties that make up it being a mug rather than some other thing. And those properties are these subsets or these ways of breaking up the state space of the mug into different possibilities.

In that sense, I kind of think of the mug as more made of properties with an associated amount of information with them and the sort of fundamental nature of the mug is that set of properties. And your reaction to that might be like, "Yes it has those properties but it is made of stuff." But then if you go back and ask, what is that stuff? Again, the stuff is a particular set of properties. As deep as you go, it's properties all the way down until you get to the properties of electrons, protons, and neutrons, which are just particular ways that those are and answers to those questions that you get by asking the right questions of those things.

And so that's what it means to me to take the view that everything is made up of information in some way, it's to take a view that there isn't a separation between the properties that we intersect to say that it is something and the thing itself that has those properties.

Lucas Perry: So in terms of identity here, there was a question about the identity status of the cup. I think that, from hearing your talks previously, you propose a spectrum of subjectivity and objectivity rather than it being a kind of binary thing, because the cup is a set of questions and properties. Can you expand a little bit about the identity of the cup and what the meaning of the cup, given that it is constituted from this quantum mechanical perspective of just information about the kinds of questions and properties we may ask of cup-like objects.

Anthony Aguirre: I think there are different ways in which the description of a system or what it is that we mean when we say it is this kind of thing. "It is a cup" or the laws of physics or like, "There is this theorem of mathematics" or "I feel itchy", are three fairly different statements. But my view is that we should not try to sort them into objective facts of the world and individual subjective or personal perspective kind of things.

But there's really this continuum in between them. So when I say that there's this thing on my desk that is a cup, there's my particular point of view that sees the cup and that has a whole bunch of personal associations with the cup. Like I really like this one. I like that it's made out of clay. I've had a lot of nice coffee out of it. And so I'm like ... So that's very personal stuff.

There's cupness which is obviously not there in the fires of the Big Bang. It's something that has evolved socially and via biological utility and all the processes that have led to our technological society and our culture having things that we store stuff in and liquids and-

Lucas Perry: That cupness though is kind of like the platonic idealism that we experience imbued upon the object, right? Because of our conventional experience of reality. We can forget the cupness experience is there like that and we identify it and like reify it, right? And then we're like, "Oh, there's just cupness there."

Anthony Aguirre: We get this sense that there is an objectively speaking cup out there, but we forget the level of creation and formulation that has gone on historically and socially and so on to create this notion, this shared collective notion of cupness that is a creation of humanity and that we all carry around with us as part of our mental apparatus.

And then we say, "Oh, cupness is an objective thing and we all agree that this is a cup and the cup is out there." But really it's not. It's somewhere in this spectrum, in the sense that there will certainly be cups, that it's ambiguous whether it's a cup or not. There will be people who don't know what a cup is and so on.

It's not like every possible person will agree even whether this is a brown cup. Some people may say, "Well actually I'd call that grayish." It feels fairly objective, but obviously there's this intersubjective component to it of all these ingredients that we invented going into making that a cup.

Now there are other things that feel more objective than that in a sense, like the laws of physics or some things about mathematics where you say like, "Oh, the ratio of the circumference to the diameter of a circle." We didn't make that up. That was there at the beginning of the universe. And that's a longer conversation, but certainly that feels more objective than the cup.

Once it's understood what the terms are, there's sort of no disagreeing with that statement as long as we're in flat space and so on. And there's no sense in which we feel like that statement has a large human input. We certainly feel like that ratio was what it was and that we can express it as this series of fractions and so on. Long before there were people, that was true. So there's a feeling that that is a much more objective thing. And I think that's fair to say. It has more of that objectivity than a cup. But what I disagree with and find kind of not useful is the notion that there is a demarcation between things that are and aren't objective.

I sort of feel like you will never find that bright line between an actually objective thing and a not actually objective thing. It will always be somewhere on this continuum and it's probably not even a one dimensional continuum, but somewhere in this spectrum between things that are quite objective and things that are very, very subjective will be somewhere in that region, kind of everything that makes up our world that we experience.

Lucas Perry: Right. So I guess you could just kind of boil that down by saying that is true because all of the things are just constituted of the kinds of properties and questions that you're interested in asking about the thing and the questions about the mathematical properties feel and seem more objective because they're derived from primitive self-intuitive axioms. And then it's just question wormholes from there, you know? That stand upon bedrock of slightly more and more dubious and relativistic and subjective questions and properties that one may or may not be interested in.

Anthony Aguirre: Yeah. So there are a couple of things I would say to that. One is that there's a tendency among some people to feel like more objective is more true or more real or something like that. Whereas I think it's different. And with more true and more real tends to come a normative sense of better. Like more true things are better things. There are two steps there from more objective to more true and from more true to better, both of which are kind of ones that we shouldn't necessarily just swallow because I think it's more complicated than that.

So more objective is different and might be more useful for certain purposes. Like it's really great that the laws of physics are in the very objective side of the spectrum in that we feel like once we've found some, lots of different people can use them for all kinds of different things without having to refigure them out. And we can kind of agree on them. And we can also feel like they were true a long time ago and use them for all kinds of things that happened long ago and far away. So there are all these great things about the fact that they are on this sort of objective side of things.

At the same time, the things that actually matter to us in and that are like the most important things in the world to us are a totally subjective thing.

Lucas Perry: Love and human rights and the fact that other humans exist.

Anthony Aguirre: Right. Like all value at some level ... I certainly see value as very connected with the subjective experience of things that are experiencing things and that's purely subjective. Nobody would tell you that the subjective experience of beings is unimportant, I think.

Lucas Perry: But there's the objectivity of the subjectivity, right? One might argue that the valence of the conscious experience is objective and that that is the objective ground.

Anthony Aguirre: So this was just to say that it's not that objective is better or more valuable or something like that. It's just different. And important in different ways. The laws of physics are super important and useful in certain ways, but if someone only knew and applied the laws of physics and held no regard or importance for the subjective experience of beings, I would be very worried about the sorts of things that they would do.

I think there's some way in which people think dismissively of things that are less objective or that are subjective, like, "Oh, that's just a subjective feeling of something." Or, "That's not like the true objective reality. Like I'm superior because I'm talking about the true objective reality" and I just don't think that's a useful way to think about it.

Lucas Perry: Yeah. These deflationary memes or jokes or arguments that love is an absurd reduction of a bunch of chemicals or whatever, that's this kind of reduction of the supposed value of something which is subjective. But all of the things that we care about most in life, we talked about this last time that like hold together the fabric of reality and provide a ton of meaning, are subjective things. What are these kinds of things? I guess from the perspective of this conversation, it's like they're the kinds of questions that you can ask about systems and like how they will interact with each other and the kinds of properties that they have. Right?

Why are these particular questions and properties important? Well, I mean historically and evolutionarily speaking, they have particular functions, right? So it seems clearer and that I would agree with you that there's the space of all possible questions and properties we can ask about things. And because of historical reasons, we care about a particularly arbitrary subset of those questions and properties that have functional use. And that is constituted of all of these subjective things like cups and houses and like love and like marriage and like rights.

Anthony Aguirre: I'm only, I think, objecting to the notion that those are somehow less real or sort of derivative of a description in terms of particles or fields or mathematics.

Lucas Perry: So the sense in which they're less real is the sense in which we'll get confused by the cupness being like a thing in the world. So that's why I wanted to highlight that phenomenological sense of cupness before where the platonic idealism we see of the cupness is there in and of itself.

Anthony Aguirre: Yeah, I think I agree with that.

Lucas Perry: So what is it that defines whether or not something falls more on the objective side or more on the subjective side? Aren't all the questions that we ask about macroscopic and fuzzy concepts like love and human rights and cups and houses and human beings ... Don't all those questions have definitive answers as long as the categories are coherent and properly defined?

Anthony Aguirre: I guess the way I see it is that there's kind of a sense of how broadly shared through agents and through space and time are those categorizations or those sets of properties. Cupness is pretty widespread. It doesn't go further back in time than humanity. Protozoa don't use cups. So cupness is fairly objective in that sense. It's tricky because there exists a subjectivity objectivity axis of how widely shared are the sets of properties and then there's a different subjective objective axis of experience of my individual phenomenological experience of subjectivity versus an objective view of the world. And I think those are connected but they're not quite the same sense of the subjective and objective.

Lucas Perry: I think that to put it on that axis is actually a little bit confusing. I understand that the more functional that a meme or a idea or concept is, the more widely shared it's going to be. But I don't think that just because more and more agents are agreeing to use some kind of concept like money, that that is becoming more objective. I think it's just becoming more shared.

Anthony Aguirre: Yeah, that's fine. I guess I would ask you what does more and less objective mean, if it's not that?

Lucas Perry: Yeah, I mean I don't know.

Anthony Aguirre: I'm not sure how to say something is more or less objective without referring to some sense like that, that it is more widespread in some way or that there are more sort of subjective views of the world that share that set of descriptions.

If we go back to the thinking about the probabilities in whatever sense you're defining the probabilities and the properties, the more perspectives are using a shared set of properties, the more objectively defined are the things that are defined by those properties. Now, how to say that precisely like is this objectivity level 12 because 12 people share that set of properties and 50 people share these, so it's objectivity level ... I wouldn't want to quantify it that way necessarily.

But I think there is some sort of sense of that, that the more different perspectives on the world use that same set of descriptions in order to interact with the world, the more kind of objective that set of descriptions is. Again, I don't think that captures everything. Like I still think there was a sense in which the laws of physics were objective before anyone was talking about them and using them. It's quite difficult. I mean when you think about mathematics-

Lucas Perry: Yeah, I was going to bring that up.

Anthony Aguirre: You know, if you think of mathematics as you've got a set of axioms and a set of rules for generating true statements out of those axioms. Even if you pick a particular set of rules, there are a huge number of sets of possible axioms and then each set of axioms, if you just grind those rules on those axioms, will produce just an infinite number of true statements. But grinding axioms into true statements is not doing mathematics, I would say.

So it is true that every true mathematical statement should have a sequence of steps that goes from the axioms to that true mathematical statement. But for every thing that we read in a math textbook, there's an exponentially large number of other consequences of axioms that just nobody cares about because they're totally uninteresting.

Lucas Perry: Yeah, there's no utility to them. So this is again finding spaces of mathematics that have utility.

Anthony Aguirre: What makes certain ones more useful than others? So it seems like you know, e, Euler's number is a very special number. It's useful for all kinds of stuff. Obviously there are a continuous infinity of other numbers that are just as valid as that one. Right? But there's something very special about that one because it shows up all the time, it's really useful for all these different things.

So we've picked out that particular number as being special. And I would say there's a lot of information associated with that pointing to e and saying, "Oh look, this number", we've done something by that pointing. There's a whole bunch of information and interesting stuff associated with pointing out that that number is special. So that pointing is something that we humans have done at some level. There wasn't a symbol e or the notion of e or anything like that before humans were around.

Nonetheless, there's some sense in which once we find e and see how cool it is and how useful it is, we say, "It was always true that e^ix = cos(x) + i sin(x). Like that was always true even though we just proved it a couple of centuries ago and so on. How could that have not been true? And it was always true, but it wasn't always true that we knew that it was interesting.

So it's kind of the interesting-ness and the pointing to that particular theorem as being an interesting one out of all the possible consequences that you could grind out of a set of axioms, that's what was created by humanity. Now why the process by which we noticed that that was an interesting thing, much more interesting than many other things, how much objectivity there is to that is an interesting question.

Surely some other species that we encountered, almost surely, they would have noticed that that was a particularly interesting mathematical fact like we did. Why? That's a really hard question to answer. So there is a subjective or non-objective part of it and that we as a species developed that thing. The interesting-ness of it wasn't always there. We kind of created that interesting-ness of it, but we probably noticed its interesting-ness for some reason and that reason seems to go above and beyond the sort of human processes that noticed it. So there's no easy answer to this, I think.

Lucas Perry: My layman's easy answer would be just that it helps you describe and make the formalization and development of mathematical fields, right?

Anthony Aguirre: Sure. But is that helpfulness a fact of the world or a contingent thing that we've noticed as we've developed mathematics? How, among all species that ever could be imagined that exist, would almost all of them identify that as being useful and interesting or would only some of them and other ones have a very different concept of what's useful and interesting? That's really hard to know. And is it more or less objective in that sort of sense?

Lucas Perry: I guess, part of my intuition here is just that it has to do with the way that our universe is constituted. Calculus is useful for like modeling and following velocities and accelerations and objects in Newtonian physics. So like this calculus thing has utility because of this.

Anthony Aguirre: Right. But that which makes it useful, that feels like it's something more objective, right? Like calculus is inheriting it objectiveness from the objective nature of the universe that makes calculus useful.

Lucas Perry: So the objectiveness is born of its relationship to the real world?

Anthony Aguirre: Yes, but again, what does that mean? It's hard to put your finger at all on what that thing is that the real world has that makes calculus useful for describing it other than saying the real world is well-described by calculus, right? It feels very circular to say that.

Lucas Perry: Okay, so I'm thoroughly confused then about subjectivity and objectivity, so this is good.

Anthony Aguirre: I think we all have this intense desire to feel like we understand what's going on. We don't really understand how reality works or is constituted. We can nonetheless learn more about how it's constituted and sitting on that razor's edge between feeling pride and like, "Yes, we figured a bunch of stuff out and we really can predict the world and we can do technology and all these things", all of which is true, while also feeling the humility that when we really go into it, reality is fundamentally very mysterious, I think is right, but difficult.

My frustration is when I see people purporting to fully understand things like, "Oh, I get it. This is the way that the world is." And taking a very dismissive attitude toward thinking the world is not the way that they particularly see it. And that's not as uncommon an attitude as one would like. Right? That is a lot of people's tendency because there's a great desire and safety in feeling like you understand this is the way that the world is and if only these poor benighted other souls could see it the way I do, they would be better off. That's hard because we genuinely do understand much, much, much more about the world than we ever did.

So much so that there is a temptation to feel like we really understand it and I think at some level that's more the notion that I feel like it's important to push back against the notion that we get it all. Like you know, we more or less understand how the world is and how it works and how it fundamentally operates. Among some circles that's more of the hubristic danger of falling into that then there is falling into the, "We don't know anything." Although there are other parts of society where there's the other end too, the anti intellectual stances that like my conception of reality is just as good as yours that I just made up yesterday and we're all equally good at understanding what the world is really like. Also quite dangerous.

Lucas Perry: The core draw away here for me is just this essential confusion about how to navigate this space of what it means for something to be more subjective and objective and the perspective of analyzing it through the kinds of questions and properties we would ask or be interested in. What you were just saying also had me reflecting a lot on people whose identity is extremely caught up in nationalism or like a team sport. It would seem to be trivial questions or properties you could ask. Like where did you happen to be born? Which city do you particularly have fondness towards? The identity of really being like an American or like really being a fan of the Patriots, people become just completely enthralled and engrossed by that. Your consciousness and ego just gets obliterated into identification with, "I am an American Patriot fan" and like there's just no perspective. There is no context. When one goes way too far towards the objective, when one is mistaking the nature of things.

Anthony Aguirre: Yeah, there are all sorts of mistakes that we all make all the time and it's interesting to see pathologies in all directions in terms of how we think about the world and our relation to it. And there are certain cases where you feel like if we could just all take a little bit more of an objective view of this, everyone would be so much better off and kind of vice versa. It takes a lot of very difficult skill to approach our complex world and reality in a way that we're thinking about it in a useful way in this wide variety of different circumstances where sometimes it's more useful to think about it more objectively and sometimes more subjectively or along all sorts of other different axes.

It's a real challenge. I mean that's part of what it is to be human and to engage in a worthy way with other people and with the world and so on, is to have to understand the more and less useful and skillful ways and lenses through which to look at those things.

At one time, almost everything we do is in error, but you also have to be forgiven because almost everything that you could do would be an error in some way from some standpoint. And sometimes thinking that the cup is objectively real is an error. Thinking that you made up the cup and invented it all on your own is also an error. So like the cup is real and isn't real and is made up and isn't made up. Any way you think about it is kind of wrong, but it's also all kind of okay because you can still pick up the cup and take a drink.

So it's very tricky. It's a tricky reality we reside in, but that's good. I think if everything was straightforward and obvious, that would be a boring world.

Lucas Perry: If everything were straightforward and obvious, then I would reprogram everyone to not find straightforward and obvious things boring and then we would not have this requirement to be in a complicated, un-understandable world.

Anthony Aguirre: I think there's a Douglas Adams line that, "If you figure it all out, then immediately it all stops and starts again in a more complicated way that becomes more and more difficult. And of course this is something that's happened many, many times before."

Lucas Perry: I don't know how useful it is now, but is talking about emergence here, is that something that's useful, you think, for talking about identity?

Anthony Aguirre: Maybe. There's a question of identity of what makes something one thing rather than another and then there's another question of personal identity and sort of my particular perspective or view of the world, like what I identify as my awareness, my consciousness, my phenomenal experience of the world and that identity and how it persists through time. That identity and how it does or doesn't connect with other ones. Like, is it truly its own island or should I take a more expansive view of it and is it something that persists over time?

Is there a core thing that persist over time or is it succession of things that are loosely identified or tightly identified with each other? I'm not sure whether all of the stuff that we've been talking about in terms of properties and questions and answers and states and things applies to that, but I'm not sure that it doesn't either.

Lucas Perry: I think it does. Wouldn't the self or like questions on personal identity be arbitrary questions in a very large state that we would be interested in asking particular questions about what constitutes the person? Is there a self? The self is like a squishy fuzzy concept like love. Does the self exist? Does love exist? Where do they fall on the subjective objective scale?

Anthony Aguirre: Well there are many different questions we could think about, but if I think of my identity through time, I could maybe talk about how similar some physical system is to the physical system I identify as me right now. And I could say I've sort of identified through time with the physical system that is really much like me and physics makes that easy because physical systems are very stable and this body kind of evolves slowly. But once you get to the really hard questions like suppose I duplicate this physical system in some way, is my identity one of those or two of those and what happens if you destroy the original one and, you know, those are genuinely confusing questions that I'm not sure that the sort of niceties of understanding emergence and the properties and so on, I'm not sure how much it has to say about it. I'm not sure that it doesn't, but having thought a lot about the earlier identity questions, I feel no less confused.

Lucas Perry: The way in which emergence is helpful or interesting to me is the way in which ... the levels of reality at which human beings conceptualize, which would be like quantum mechanics and then atomic science and then chemistry and then biology and so on.

We imagine them as being sort of stacked up on each other and that if reductionism is attractive to one, you would think that all the top layers supervene upon the nature of the very bottom layer, quantum mechanics. Which is true to some sense and you would want to say that there is fundamental brute identity facts about the like very, very, very base layer.

So you could say that there are such things as irreducible quantum atoms like maybe they reduce into other things but that's an open question for now. And if we are confident about the identity of those things, there's at least a starting place, you know from which we would have true answers about identity. Does that make sense?

Anthony Aguirre: Well the sentences make sense but I just largely don't agree with them. And for all the reasons that we've talked about. I think there needs to be a word that is the opposite of emergence, like distillation or something, because I think it's useful to think both directions.

Like I think it is certainly useful to be able to think about, I have a whole bunch of particles that do these things and then I have another description of them that glosses over say the individual actions of the particles, but creates some very reliable regularity that I can call a law like thermodynamics or like some chemical laws and so on.

So I think that is true, but it's also useful to think of the other direction, which is we have complicated physical systems and by making very particular simplifications and carving away a lot of the complexity, we create systems that are simple enough to have very simple laws describe them. I would call that a sort of distillation process, which is one that we do. So we go through this process when we encounter new phenomena. We kind of look for ways that we can cut away lots of the complexity, cut away a lot of the properties, try to create a system that's simple enough to describe in some mathematical way, using some simple attenuated set of concepts and so on.

And then often we take that set and then we try to work our way back up by using those laws and kind of having things that emerge from that lower level description. But I think both processes are quite important and it's a little bit intellectually dangerous to think of what I'd call the distillation process as a truth-finding process. Like I'm finding these laws that were all already there rather than I'm finding some regularities that are left when I remove all this extra stuff and then forget that you've removed all the extra stuff and that when you go back from the so-called more fundamental description, to the emerged description, that you're secretly sticking a lot of that stuff back in without noticing that you're doing it.

So that's sort of my point of view, that the notion that we can go from this description in terms of particles and fields and that we could derive all these emerged layers from it, I think it's just not true in practice for sure, but also not really true in principle. There's stuff that we have to add to the system in order to describe those other levels that we sort of pretend that we're not adding. We say, "Oh, I'm just assuming this extra little thing" but really you're adding concepts and quantities and all kinds of other apparatus to the thing that you started with.

Lucas Perry: Does that actually describe reality then or does that give you an approximation, the emergent levels?

Anthony Aguirre: Sure. It just gives you answers to different questions than the particle and field level does.

Lucas Perry: But given that the particle and field level stuff is still there, doesn't that higher order thing still have the capacity for like strange quantum things to happen and that would not be accounted for in the emergent level understanding and therefore it would not always be true if there was some like entanglement or like quantum tunneling business going on?

Anthony Aguirre: Yeah, I think there's more latitude perhaps. The statistical laws and statistical mechanics are statistical laws. They're totally exact, but the things that they make are statistical descriptions of the world that are approximate in some way. So it's like they're approximate but they're approximate in a very, very well defined way. I mean it's certainly true that the different descriptions should not contradict each other. If you have a description of a macroscopic phenomenon that doesn't conserve energy, then that's a sort of wrongheaded way to look at that system.

Lucas Perry: But what if that macroscopic system does something quantum? Then the macroscopic description fails. So then it's like not true or it's not predictive.

Anthony Aguirre: Yeah, not true I think is not quite the right, like that description let you down in that circumstance. Everything will let you down sometimes.

Lucas Perry: I understand what you're saying. The things are like functional at the perspective and scales at which you're interested. And this goes back to kind of this more epistemological agent centered view of science and like engaging in the world that we were talking about earlier. I guess, for a very long time the way that I viewed science as explaining the intrinsic nature of the physical, but really it's not doing that because all of these things are going to fail at different times. They just have strong predictive power. And maybe it was very wrong of me early on to ever think that science was describing the intrinsic nature of the physical.

Anthony Aguirre: I don't think it's entirely wrong. You do get something through distilling more and going more toward the particle and field level in that once you specify something that the quantum mechanics and the standard model of particle physics say gives you a well-defined answer to, then you feel really sure that you're going to get that result. You do get a dramatically higher level of confidence from doing that distilling process and idealizing a system enough that you can actually do the mathematics to figure out what should happen according to the fundamental physical laws, as we describe them in terms of particles and fields and so on.

So I think that's the sense in which they're extra true or real or fundamental, is that you get that higher level of confidence. But at the cost that you had to shoehorn your physical system, either add in assumptions or cutaway things in order to make it something that is describable using that level of description.

You know, not everyone will agree with the way that I'm characterizing this. I think you'll talk to other physicists and they would say, "Yes they are approximations, but really there's this objective description and you know, there's this fundamental description in terms of particles and fields and we're just making different approximations to it when we talk about these other levels."

I don't think there's much of a difference operationally in terms of that way of talking about it and mine. But I think this is a more true-to-life description of reality, I guess.

Lucas Perry: Right. So I mean there are the fundamental forces and the fundamental forces are what evolve everything. And you're saying that the emergent things have to do with adding and cutting away things so that you can like simplify the whole process, extract out these other rules and laws which are still highly predictive. Is that all true to say so far?

Anthony Aguirre: Somewhat. I think it's just that we don't actually do any of that. We very, very, very, very rarely take a more fundamental set of rules and derive.

Lucas Perry: Yeah, yeah, yeah. That's not how science works.

Anthony Aguirre: Right. We think that there is such a process in principle.

Lucas Perry: Right.

Anthony Aguirre: But not in practice.

Lucas Perry: But yeah, understanding it in principle would give us more information about how reality is.

Anthony Aguirre: I don't believe that there is in principle that process. I think the going from the more fundamental level to the "emerged" can't be done without taking input that comes from the emerged level. Like I don't think you're going to find the emerged level in the fundamental description in and of itself without unavoidably taking information from the emerged level.

Lucas Perry: Yeah. To modify the-

Anthony Aguirre: Not modifying but augmenting. Augmenting in the sense that you're adding things like brownness that you will never find, as far as you will ever look, you will never find brownness in the wave function. It just isn't there.

Lucas Perry: It's like you wouldn't find some kind of chemical law or property in the wave function.

Anthony Aguirre: Any more than you'll find here or now in the state of the universe. Like they're just not there. Those are things, incredibly useful things, important things like here and now are pretty central to my description of the world. I'm not going to do much without those, but they're not in the wave function and they're not in the boundary conditions of the universe and it's okay that I have to add those. There's nothing evil in that doing that.

Like I can just accept that I have to have some input from the reality that I'm trying to describe in order to use that fundamental description. It's fine. But like, there's nothing to be worried about, there's nothing anti-scientific about that. It's just the idea that someone's going to hand you the wave function and you'll derive that the cup is brown here and now is crazy. It just doesn't work that way. Not in there. That's my view anyway.

Lucas Perry: But the cup being brown here and now is a consequence of the wave function evolving an agent who then specifies that information, right?

Anthony Aguirre: Again, I don't know what that would look like. Here's the wave function. Here's Schrodinger's equation and the Hamiltonian. Now tell me is the brown cup in front of or in back of the tape measure? It's not in there. There's all colored cups and all colored tape measures and all kinds of configurations. They're all there in the wave function. To get an answer to that question, you have to put in more information which is like which cup and where and when.

That's just information you have to put in, in order to get an answer. The answer is not there to begin with and that's okay. It doesn't mean that there's something wrong with the wave function description or that you've got the wrong Hamiltonian or the wrong Schrodinger's equation. It just means that to call that a complete description of reality, I think that's just very misleading. I understand what people intend by saying that everything is just the wave function and the Schrodinger equation. I just think that's not the right way to look at it.

Lucas Perry: I understand what you're saying, like the question only makes sense if say that wave function has evolved to a point that it has created human beings who would specify that information, right?

Anthony Aguirre: None of those things are in there.

Lucas Perry: They're not in the primordial state but they're born later.

Anthony Aguirre: Later is no different from the beginning thing. It's just a wave function. There's really no difference in quality between the wave function now and at the beginning. It's exactly the same sort of entity. There's no more, no less in it than there was then. Everything that we ascribe to being now in the universe that wasn't there at the beginning are additional ingredients that we have to specify from our position, things like now and here and all those properties of thing.

Lucas Perry: Does the wave function just evolve the initial conditions? Are the initial conditions contained within the wave function?

Anthony Aguirre: Well, both in the sense that if there's such a thing as the wave function of the universe, and that's a whole nother topic as to whether that's a right-minded thing to say, but say that there is, then there's exactly the same information content to that wave function at anytime and that given the wave function at a time, and the Schrodinger equation, we can say what the wave function is at any other time. There's nothing added or subtracted.

One is just as good as the other. In that sense, there's no more stuff in the wave function "now" than there was at the beginning. It's just the same. All of the sense in which there's more in the universe now than there was at the Big Bang has to do with things that we specify in addition to the wave function, I would say, that constitute the other levels of reality that we interact with. They're extra information that we've added to the wave function from our actual experience of reality.

If you take a timeline of all possible times, without pointing to any particular one, there's no time information in that system, but when I say, "Oh look, I declare that I'm now 13.8 billion years from the big bang," you're pointing to a particular time by associating with my experience now. By doing that pointing, I'm creating information in just the same way that we've described it before. I'm making information by picking out a particular time. That's something new that I've added to what was a barren timeline before I've added now something.

There's more information than there was before by the fact of my pointing to it. I think most of the world is of that nature that it is made of information created by our pointing to it from our particular perspective here and now in the universe seeing this and that and having measured this and that and the other thing. Most of the universe I contend is made of that sort of stuff, information that comes from our pointing to it by seeing it, not information that was there intrinsically in the universe, which is, I think, radical in a sense, but I think is just the way reality is, and that none of that stuff is there in the wave function.

Lucas Perry: At least the capacity is there for it because the wave function will produce us to then specify that information.

Anthony Aguirre: Right, but it produces all kinds of other stuff. It's like if I create a random number generator, and it just generates a whole list of random numbers, if I look at that list and find, "Oh look, there's one, one, one, one, one, one, one, one, one," that's interesting. I didn't see that before. By pointing to that, you've now created information. The information wasn't there before. That's largely what I see the universe as, and in large part, it's low information in a sense.

I'm hemming and hawing because there are ways in which it's very high information too, but I think most of the information that we see about the world is information of that type that exists because we very collectively as beings that have evolved and had culture and all the stuff that we've gone through historically we are pointing to it.

Lucas Perry: So connecting this back to the spectrum of objectivity and subjectivity, as we were talking for a long time about cups and as we talked about on the last podcast about human rights for example as being a myth or kinds of properties which we're interested in ascribing to all people, which people actually intrinsically lack. People are numerically distinct over time. They're qualitatively distinct, very often. There's nothing in the heart of physics which gives us the kinds of properties.

Human rights, for example, are supposed to be instantiating in us. Rather, it's a functional convention that is very useful for producing value. We've specified this information that all human beings share unalienable rights, but as we enter the 21st century, the way that things are changing is that the numerical and qualitative facts about being a human being that have held for thousands of years are going to begin to be perturbed.

Anthony Aguirre: Yes.

Lucas Perry: You brought this up by saying... You could either duplicate yourself arbitrarily, whether you do that physically via scans and instantiating actual molecular duplicates of yourself. You could be mind uploaded, and then you could have that duplicated arbitrarily. For hundreds of thousands of years, your atoms would cycle out every seven years or so, and that's how you would be numerically distinct, and qualitatively, you would just change over your whole lifetime until you became thermodynamically very uninteresting and spread out and died.

Now, there's this duplication stuff. There is your ability to qualitatively change yourself very arbitrarily. So at first, it will be through bioengineering like designer babies. There's all these interesting things and lots of thought experiments that go along with it. What about people who have their corpus callosum cut? You have the sense of phenomenological self, which is associated with that. You feel like you're a unitary subject of experience.

What happens to your first person phenomenological perspective if you do something like that? What about if you create a corpus callosum bridge to another person's brain, what happens to the phenomenological self or identity? Science and AI and increasing intelligence and power over the universe will increasingly give us this power to radically change and subvert our commonly held intuitions about identity, which are constituted about the kinds of questions and properties which we're interested in.

Then also the phenomenological experience, which is whether or not you have a strong sense of self, whether or not you are empty of a sense of self or whether or not you feel identified with all of consciousness and the whole world. There's spectrums and degrees and all kinds of things around here. That is an introduction to the kind of problem that this is.

Anthony Aguirre: I agree with everything you said, but you're very unhelpfully asking all the super interesting questions-

Lucas Perry: At once.

Anthony Aguirre: ... which are all totally impossible to solve. No, I totally agree. We've had this enviable situation of one mind equals one self equals one brain equals one body that has made it much easier to accord to that whole set of things, all of which are identified with each other a set of rights and moral values and things like that.

Lucas Perry: Which all rest on these intuitions, right? That are all going to change.

Anthony Aguirre: Right.

Lucas Perry: Property and rights and value and relationships and phenomenological self, et cetera.

Anthony Aguirre: Right, so we either have a choice of trying to maintain that identity, and remove any possibility of breaking some of those identities because it's really important to keep all those things identified, or we have to understand some other way to accord value and rights and all those things given that the one-to-one correspondence can break. Both of those are going to be very hard, I think. As a practical matter, it's simply going to happen that those identifications are going to get broken sooner or later.

As you say, if we have a sufficient communication bandwidth between two different brains, for example, one can easily imagine that they'll start to have a single identity just as the two hemispheres of our brain are connected enough that they generally have what feels like a single identity. Even though if you cut it, it seems fairly clear that there are in some sense two different identities. At minimum, technologically, we ought to be able to do that.

It seems very likely that we'll have machine intelligence systems whose phenomenological awareness of the world is unclear but at least have a concept of self and a history and agency and will be easily duplicatable. They at least will have to face the question of what it means when they get duplicated because that's going to happen to them, and they're going to have to have a way of dealing with that reality because it's going to be their everyday reality that they can be copied, ad infinitum, and reset and so on.

If they're functioning is it all like a current digital computer. There are also going to be even bigger gulfs than there are now between levels of capability and awareness and knowledge and perhaps consciousness. We already have those, and we gloss over them, and I think that's a good thing in according people fundamental human rights. We don't give people at least explicitly legally more rights when they're better educated and wealthier and so on, even if in practice they do get more.

Legally, we don't, even though that range is pretty big, but if it gets dramatically bigger, it may get harder and harder to maintain even that principle. I find it both exciting and incredibly daunting because the questions are so hard to think of how we're going to deal with that set of ethical questions and identity questions, and yet we're going to have to somehow. I don't think we can avoid them. One possibility is to decide that we're going to attempt to never break those sets of identities.

I sometimes think about Star Wars. They've got all this amazing technology, right? They can zip across the universe, but then it's incredibly primitive in others. Their computers suck and all of their AI is in robots. One robot, one brain, one consciousness, they're all identical. So I have this theory of Star Wars that behind the scenes, there's some vast intelligence that's maybe baked into the midi-chlorians or whatever, that prevents more weird, complicated things like powerful AI or powerful software systems.

It's like overseer that keeps everything just nicely embodied in individual physical agents that do stuff. Obviously, that's not part of the Star Wars canon, but that's how it plays out, right? Even though there's all this high tech, they've neatly avoided all of these annoying questions and difficult questions by just maintaining that one-to-one correspondence. That is in some level an option. That is something that we could try to do because we might decide that not doing that leads to such a big open can of worms that we will never be able to deal with, that we better maintain that one-to-one correspondence.

My guess is that even if that was a good idea, we wouldn't be coordinated enough or foresightful enough to maintain that.

Lucas Perry: There would be optimization pressures to do otherwise.

Anthony Aguirre: There would. It would take some almost God-like entity to keep it from happening. Then we have to ask, "Where is the theory of what to value and how do we value individual people? Where is that next going to come from?" That last time, at least in the West, it was born out of enlightenment philosophy and coming out of, honestly, I think Judeo Christian religion. That's very tied together. Is there something that is going to come out of some other major philosophical work? I'm not sure that I see that project happening and unfolding.

Lucas Perry: Formally right now?

Anthony Aguirre: Yes. Do you?

Lucas Perry: No, I don't see that, but I think that there are the beginnings of that. I think that I would propose and others, and I don't know how others would feel, but that foundation instead of enlightenment philosophy about rights based off the immutable rights that beings have given their identity class, it would be in the future a sufficiently advanced science of consciousness would just value all of the different agents based off the understanding of the degrees and kinds of experience and awareness and causal implications that it could have in the world.

I would just do a kind of consequentialism and so far as it would be possible. Then I guess the interesting part would be where consequentialism fails because it's computationally intractable. You would want to invent other kinds of things that would stand in the way, but I feel optimistic that the very very smart things in the future could do something like that. I would ground it on consciousness.

Anthony Aguirre: I mean, there are so many questions even if you take the view that you're trying to maximize high quality phenomenological experience moments or whatever, I think there's so many things that that leaves either problematic or unanswered.

Lucas Perry: Like what?

Anthony Aguirre: What about beings that may have super high levels of awareness and consciousness but not positive or negative valence? Do they count or not? Does it mean anything that experiences are connected through time in some large set of personal identity or is a bunch of disconnected experiences just as good as other ones? There may be a positive valence to experience that comes out of its aggregation over time and its development and evolution over time that is absent from any individual one of those moments, all of which may be less good than a drug trip or just eating a candy bar, but like a life of eating candy bars versus a less pleasurable but more fulfilling life. How do we quantify those things against each other?

Lucas Perry: The repugnant conclusion, what do we think about the repugnant conclusion that's like kind of that. A quick definition, the repugnant conclusion is how you would compare a very small, limited number of amazing experiences against an astronomically large number of experiences which are just barely better than non-existence, very, very, very, very slightly better than a valence of zero. If all of those added up to be just like a fraction of a hair larger than the few really, really good experiences, which world should you pick? Hedonic consequentialism would argue that you should pick the astronomically large number of experiences that are barely worth living and that to some is repugnant.

Anthony Aguirre: I think it's safe to say that there is no proposal on the table that everyone feels like, "Oh yeah, that's the way to do it." I'd be profoundly suspicious of anything that claimed to be that. So I don't think there are going to be easy answers, but it may be that there's at least a framework from which we can stand to get into some of the complexities. That may be a very different framework than the one that we have now.

Where that will come from and how we would transition to it and what that would mean and what kind of terrible and wonderful consequences that might have, I think, certainly nobody knows. It's not even clear that anybody has a sense of what that will look like.

Lucas Perry: I think that one of the last questions here and perspectives that I'd like to get from you are how this perspective on how human perspectives on identity changes what we want. So this one-to-one correspondence, one body, one brain, one phenomenological self that feels like its consciousness is its own and is like an Island, how that experience changes what human beings want in the 21st century with regards to upgrading or merging with AI and technology or with cryonics.

If everything and everyone is numerically and quantitatively completely impermanent such that no matter what kind of technological intervention we do in 100 to 200 years, everyone will either be thermodynamically scattered or so completely and fundamentally changed that you won't be able to recognize yourself and the ethical implications of this and how it changes what kinds of futures people want. I'm curious to know if you have any thoughts of this holding in the perspective in your head of Max's book Life 3.0 and the kinds of world trajectories that people are interested in from there.

Anthony Aguirre: That's a big question. That's hard to know how to approach. I think there are many genuinely qualitatively different possible futures, so I don't think there is a way that things are going to turn out in terms of all these questions. I think it's going to be historically contingent and there are going to be real choices that we make. I'm of two minds in this, and that I do believe in something like moral progress and that I feel like there's an agreed sense that we feel now that things that we did in the past were morally incorrect, and that we've learned new moral truths that allow us to live in a better way than we used to.

At the same time, I feel like there are ways that society has turned out. It could have been that the world became much more dominated by Eastern philosophy than Western philosophy say. I think we would probably still feel like we had made moral progress through that somewhat different history as we've made moral progress through this history that we did take. I'm torn between a feeling that there is real moral progress, but that progress is not toward some predefined optimal moral system that we're going to progress towards and find, but that the progress will also have a whole bunch of contingent things that occur through our society's evolution through chance or through choice that we make, and that there genuinely are very different paths that we have ahead of us.

No small part of that will be our current way of thinking in our current values and how we tried to keep things aligned with those current values. I think there will be a strong desire to maintain this one-to-one connection between identity and moral value and mind and so on, and that things that violate that, I think, are going to be seen as threats. They are profound threats to our current moral system. How that will play out is really unforeseeable.

Will those be seen as threats that we eventually just say actually, they weren't that scary after all and we just have to adjust? Will they be threats that are just pushed aside by the tide of reality and technology? Will they be threats that we decide are so threatening that we want to hold on and really solidify and codify this relation? I think those are all possibilities, and it's also possible that I'm wrong and that there will just be this smooth evolution where our connection between our phones will become brain interfaces, and we'll just get more and more dr-individualized in some smooth way, and that people will sound an alarm that that's happening and no one will care. That's also quite possible, whether that alarm is appropriate or not.

Lucas Perry: They just look at the guy sounding the alarm, and then stick the plug in their head.

Anthony Aguirre: Right. So it's good for us all to think deeply about this and think about what preferences we have, because where we go will end up being some combination of where the technology goes and what preferences we choose and how we express them. Part of the direction will be determined by those that express their preferences convincingly and loudly and have some arguments for them and that defend them and so on. That's how our progress happens.

Some set of ideas prevails, and we can hope that it's a good one. I have my own personal prejudices and preferences about some of the questions that are, for example, asked in Max's book about what futures are most preferred. At some point, I may put more time into developing those into arguments and see if I still feel those preferences or believe them. I'm not sure that I'm ready to do that at the moment, but I think that's something that we all have to do.

I mean, I think, I do feel a little bit like step one was to identify some of the thorny questions that we're going to have to answer and talk about how we have to have a conversation about those things and how difficult those questions are going to be, but at some point, we're actually going to have to start taking positions on some of those questions. I think that's something that largely nobody is doing now, but it's not clear how much time we have before we need to have thought about them and actually taking a position on them and argued it out and had some positions prevail.

The alternative to that is this random process driven by the technology and the other social forces that are at work like moneyed interests and social imperatives and all those sorts of things. Having these questions decided by those forces rather than reflection and thinking and debate among people who are trying really hard to think about these questions, that seems like not such a great idea.

Lucas Perry: I agree. That's my felt sense too. We went from talking to information about emergence to identity. I think it would be really helpful if you could tie together in particular the information discussion with how that information perspective and discussion can inform these questions about identity in the 21st and 22nd centuries.

Anthony Aguirre: I guess one way that the identity and the information parts are connected is I made this argument that a lot of what the world is is information that is associated with a particular vantage point and a particular set of pointings to things that we have as an agent, as a prospective in the world. I think that there's a question as to whether there is moral value in that. There's a real sense that every person views the world from their own perspective, but I think it's more real than that and that when you identify a view of the world and all that comes with that, it really is creating a world in a sense.

There's some of the world that's objective at various different levels, but a lot of what the world is is what is created by an individual standpoint and vantage point that is seeing that world and interacting with it. I do wonder is there some sense of grounding some level of value on that creative act? On the fact that as a individual agent that understands and exists over time and assembles this whole sophisticated, complicated view of the world that has all this information content to it, should we not accord some high level of normative value to that, that it's not just a way to describe how the world is made, but what is valuable in the world be connected with that creation process by the individual standpoint?

That may be a seed for developing some bridge between the view of reality as information, information as something that is largely connected with a vantage point and a vantage point as something that is personal self identity and as connected now with individual consciousness and mind and brain and so on. Is there a way to inhere value in that ability to create lots of sophisticated information through interaction with the world that would bring value to also not just individuals but sets of individuals that together create large amounts of information?

That's something that develop further, I think. That link that view of how the world is constituted is this interaction between the agent of the world. Maybe there's something there in terms of a seed for how to ground moral value in a way that's distinct from the identification that we do now.

Lucas Perry: I guess there's also this facet where this process of agents asking particular questions and specifying certain kinds of properties that they care about and pointing to specific things, that that process is the same process of construction of the self or the egocentric phenomenal experience and conceptual experience of self. This is all just information that you specify as part of this identification process and the reification process of self.

It would be very good if everyone were mindful enough about thinking about where on the spectrum of objectivity and subjectivity these things they take to ultimately be part of self actually fall, and what are the questions and properties and features they're actually constituted of? Then what will happen is likely, your commonly held intuitions will largely be subverted. Maybe you'll still be interested in being a strong nationalist, but maybe you'll have a better understanding of what it's actually constituted of.

That's the Buddhist perspective. I'm just articulating it, I think, through the language and concepts that you've provided, where one begins seeing conventional reality as how it's actually being formulated and no longer confuses the conventional as the ultimate.

Anthony Aguirre: There's a lot of sophistication, I think, to Buddhist moral thinking, but a lot of it is based around this notion of avoiding suffering and sentient beings. I think there's so many different sorts suffering and there's so many different levels that just avoiding suffering ends up implying a lot of stuff, because we're very good at suffering when our needs are not met. Avoiding suffering is very, very complicated because our unmet needs are very, very complicated.

The view that I was just pointing to is pointing towards some level of value that is rather distinct from suffering because one can imagine a super sophisticated system that has this incredibly rich identity and incredibly rich view of the world and may suffer or not. It's not clear how closely connected those things are. It's always dangerous when you think about how to ground value because you realize that any answer you have to that question leave certain things out.

If we try to ground value in sophistication of worldview or something like that, then do we really not value the young kids? I mean, that seems monstrous. Even though they have a pretty simple minded worldview, that seems wrong. I think there are no easy answers to this, but that's just a sense in which I think I do feel instinctively that there ought to be some level of moral value accorded to beautifully complex, self-aware systems in the world that have created this sophisticated universe through there being experience and existence and interaction with the world.

That ought to count for something. Certainly, it's not something we want to just blindly destroy, but exactly why we don't want to destroy it. The deep reason, I think, needs to be investigated. That seems true to me, but I can't necessarily defend why.

Lucas Perry: That's really good and, I think, an excellent place to wrap up concluding thoughts. My ethics is so sentience focused that that is an open question, and I would want to pursue deeply why that seems intrinsically valuable for me. Just the obvious direct answer would be because it allows or does not allow for certain kinds of conscious experiences, which is what matters. That is not intrinsically valuable, but it is valuable based off of its relationship to consciousness obviously.

Of course, that's up for debate and to be argued about. Given uncertainty about consciousness, the view which you propose may be very skillful for dealing with the uncertainty. This is one of the most interesting conversations for me. Like you said, I think it's very neglected. There's no one working on it formally. Maybe it's just too early. I think that it seems like there's a really big role for popular media and communication to explore these issues.

There are so many good thought experiments in philosophy of personal identity and elsewhere that could be excellent and fun for the public. It's not just that it's philosophy that is becoming increasingly needed, but it's also fun and interesting philosophy. Much of it like the teleportation machines and severing the corpus callosum, it's perfect stuff for Black Mirror episodes and popular science things which are increasingly becoming interesting, but it's also I feel existentially very important and interesting.

I think I have a pretty big fear of death. I feel like a lot of that fear is born of those individualism, where you identify it with your own personal consciousness and qualitative nature and some of your numerical nature perhaps, and there's this great attachment to it. There's the question in journey of further and always investigating this question of identity and who am I or what am I? That process, I think, also has a lot of important implications for people's existential anxiety.

That also feeds into and informs how people wish to relate and deal with these technological changes in the 21st century and the kinds of futures they would or would not be excited about. I think those are generally my feelings about this. I hope that it doesn't just come down to what you were talking about, the socioeconomic and social forces just determining how the whole process unfolds, but there's actually a philosophical and moral reflection and idealization that happens there, so we can decide how consciousness ever evolves into the deep future.

Anthony Aguirre: I think I agree with a lot of what you said. I think we've had this very esoteric discussion about the nature of reality and self and all these things that obviously a lot of people in the world are not going to be that into, but at the same time, I think as you said, some will and some of the questions when framed in evocative ways are super just intrinsically interesting. I think it's also important to realize how large an affect some of this pretty esoteric philosophical thinking about the nature of reality has had that we had our moral system and legal system and governmental system were largely created in response to careful philosophical thinking and long treatises in the 17th and 18th and 19th centuries.

We need more of those now. We need brilliant works that are not just asking these questions, but actually compellingly arguing for ways to think about them, and putting it out there and saying, "This is the way that we ought to value things, or this is the ground for valuing this or that, or this is the way that we should consider reality and what it means for us." We don't have to accept any one of those views, but I fear that in the lack of daringly trying to deeply develop those ideas and push for them and argue for them that we will end up, as you say, just randomly meandering around to where the social forces pushes.

If we really want a development of real ideas on which to found our long-term future, we better start really developing them and valuing them and putting them out there and taking them seriously rather than thinking, "Oh, this is weird esoteric conversation off in the corner of philosophy academia, blah, blah, blah." De-valuing it in that way, I think, is not just not useful, but really misunderstanding how things have happened historically. Those discussions in the right way and published and pushed in the right ways have had huge influence on the course of humanity. So they shouldn't be underestimated, and let's keep going. You can write the book, and we'll read it.

Lucas Perry: Wonderful. Well, the last point I think is very useful is what you're saying is very true in terms of the pragmatics and illustrating that. In particular, the enlightenment treatises have very particular views on personal identity. The personal identity of people of color over time has shifted in terms of slavery. The way in which Western colonial powers conceptualize the West Africans for example, was in very particular way.

Even today with gender issues in general, that is also a mainstream discourse on the nature of personal identity. It's already been a part of the formation of society and culture and civilization, and it will only continue to do so. With that, thanks so much, Anthony. I appreciate it.