
Loading summary
Commercial Narrator
If you work in university maintenance, Grainger considers you an MVP because your playbook ensures your arena is always ready for tip off. And Grainger is your trusted partner, offering the products you need all in one place, from H VAC and plumbing supplies to lighting and more. And all delivered with plenty of time left on the clock. So your team always gets the win. Call 1-800-GRAINGER visit grainger.com or just stop by Grainger for the ones who get it done.
Donald Hoffman
When you manage procurement for multiple facilities, every order matters. But when it's for a hospital system, they matter even more. Grainger gets it and knows there's no time for managing multiple suppliers and no room for shipping delays. That's why Grainger offers millions of products in fast, dependable delivery so you can keep your facility stocked, safe and running smoothly. Call 1-800-GRAINGER Click grainger.com or just stop by Granger for the ones who get it done.
Tom
Let me just ask one question. Is cause and effect real outside of the headset?
Donald Hoffman
I think inside the head. Inside the headset, no.
Tom
So we have a useful fiction inside the headset. You don't think there's cause and effect, right?
Donald Hoffman
Well, we have the appearance of cause and effect, Right. So if I hit the cue ball and. And it hits the eight ball into the corner pocket, it looks like the cue ball caused the eight ball to careen into the corner pocket.
Tom
Yes.
Donald Hoffman
And that all works, but it's a fiction of causality. In the old video game of Pong, you have these little paddles and this ball and you hit and it looks like the paddle is causing the ball. And the fiction of causality is good enough that you can play a game, you can actually figure out how to put maybe a little spin on it and so forth. Or in more advanced games like virtual reality or games, or you know, like Grand Theft Auto, you have a nice fiction of causality. I turn the wheel to the left, my car goes to the left, turn the wheel to the right. But it's all a fiction. The wheel has no causal powers. The gas pedal has no causal powers. But it's a useful fiction. Evolution gave us a useful fiction.
Tom
I don't know that that's true. So think about a system was created. Let's. Let's take the analogy of Grand Theft Auto. So a system is created such that it awaits input from your. Your control pad, whatever that control pad may be. Now, admittedly my control pad, I press it. An electrical signal is sent to something that turns on or off or opens or closes or whatever, and then a whole cascade of things happens. But it, it is. I mean, you could even trace back the causality to I ate something, that gave me the ability to create ATP, which gave me the ability to generate electricity, which gave me the ability to fire a muscle, which gave me the ability to press a button, which triggered this electrical chain reaction that caused something to happen on the screen. But it, it, There is a chain of causality in the headset. There is a chain of causality in the headset. Like even. Even though it is the perception is that, you know, I'm turning and the act of actually just moving that thing makes it turn. So I get that that isn't happening, but it does create this chain reaction that can be in. In the headset. Understood.
Donald Hoffman
Right.
And so there. I disagree. So if you look, for example, in a VR version of it, where you see your virtual hands holding a virtual steering wheel.
Tom
Yep.
Donald Hoffman
There's no feedback from the screen to the computer.
Tom
There's no feedback from the screen to the computer. But there is feedback. Like, if you think about VR, so
Donald Hoffman
there is a real cause and effect, but it's not what we see in the headset. On the headset, it looks like the wheel that I'm seeing in the headset, in the VR headset is what's causing the car.
Tom
Yes. But I am typically in VR, I would be mapping my real hands either by holding a controller or by having cameras. So I'm mapping my hands. My hands do a movement, and this program is programmed to wait for the input from the movement of my hands to then trigger that sequence that I just listed before.
Donald Hoffman
So there is a cause and effect.
Tom
Yep.
Donald Hoffman
But.
But notice it's not from the things that you're literally seeing in the headset.
Tom
I agree with that. But I'm saying. But there, there still is a ch of cause and effect.
Donald Hoffman
Absolutely.
Tom
Because when you say that it is an illusion of the eight ball hitting the thing, it's like, that doesn't seem true within the construct of the headset, I'm saying.
Donald Hoffman
Right, right. Well, so maybe another example might help on this. So when you, like, drag your icon, you've written a email and this icon is blue in the middle of your screen, you drag it to the trash can.
Tom
Yep.
Donald Hoffman
You are using, say, a mouse or a touchpad. And in. So in this analogy, that would be like the real cause and effect. But if you said, but it's really the motion of the icon on the screen to the trash can that's causing the file to Be deleted. That's just an illusion of causality, right? There's no feedback from the pixels of the screen into the computer. And that's what I mean. Everything inside space and time is like seeing the icons on your desktop. Things move around on your desktop, but it's due to something like your joystick or something else. But the headset, since the headset is all of space and time, what that means is that I'm saying that everything inside space time is part of that fictional cause.
Tom
I now get what you're saying. So in the headset, that perception is illusion, and I was dragging a sort of back and forth between.
Donald Hoffman
But it's good.
Tom
I mean, that's very, very good. Now take me outside. Does cause and effect exist if time doesn't exist?
Donald Hoffman
So two answers and I'll get to the death. I think in the realm of conscious agents, there is a notion of cause and effect, but it boils down to a notion of free will. That seems to be one of the key notions of cause in this. In quantum mechanics, by the way, I should point out that they. When you do quantum computations, normally in normal computations, there's a causal order. If you do this multiply and then an add, that order is important. If you do the add and then the multiply, you get a different answer, right? The order, the causal order. But it turns out in quantum mechanics, you can get rid of causal order. When you do your computation, you can have a superposition of, you know, multiply followed by add, with add followed by multiply both orders. So you have a superposition of the causal order. And it's a theorem that in general, you'll be faster, you'll be more efficient if you let go of causality in space time, and it turns out that it's actually been done. When you build these things and let go of causality in space time, you tap into greater efficiency. So when you let go of space time, you're also letting go. When space time is doomed means also that everything that we believed to be causality in space time was just a very useful fiction in our headset, just like the fictions in various VR games that we play. It's a useful fiction, lets us play the game. There is an underlying causal order, and we're arbitrarily ignorant about what that causal order is. Now, about death, I don't know. But here's an interesting idea. Suppose you go to a VR arcade with some friends to play virtual volleyball, and you put on your headset and bodysuit and you're on a, you know, a, like a beach volleyball scene. Palm trees and sand and the net and you're playing VR volleyball for a while. Then one of your friends, you know, says, I'm thirsty, I need a drink. He takes off his headset and body to go get a drink. His avatar sits lifeless on the sand. It collapses on the sand. It looks within the VR headset, within the game as though he's dead. But his consciousness has not ceased. He's merely stepped out of that interface.
Tom
All right, now we're in what I'll call the Phineas Gage problem. Okay, so once. So Phineas Gage, for those who don't know, it was Railroad Worker, one of the most famous examples in neuroscience. He's hitting a tamping rod. It's like a three foot rod thicker than your thumb, and it shoots up through his jaw and out the top of his head, taking, if I remember right, a teacups worth of brain matter, which seems impossible but never loses consciousness. But they say he's forever different.
Different.
And he used to be like super sweet and he was one of the best workers. And then he becomes this belligerent asshole and he can't hold a job. So I will say that's, that is the. So using this notion of the headset or the umvelt, it's like once you alter the way that his brain works, and I'm fully willing to accept that this is a problem only inside of the headset, but once you alter that function within the headset, he's fundamentally different. Now if the headset is, is what I'll call the umwelt. It's our interpretation of the stimuli. Once he goes takes the headset off the intern. Now we're into transistors, diodes, electrical gravity or. Oh God, I may have tripped us up with gravity, but everything's shy of that one. Now you have he. The experience of that person would be fundamentally different. I would be fundamentally different because I'm no longer experiencing things through the lens of my brain, essentially. So while it would seem to me that there's no way around the fact that in the game homie is dead and so, or lifeless, to use your example. So since that's the only thing that I want to relate to that person through my headset, that's what I'm used to, that's where the emotion lies. And then now once I step outside, even emotion would be called into question, almost certainly, in fact definitively in your explanation. The way that I process emotions, it would just be unrecognizable outside because I no longer have the brain. I no longer have all of those things. Interesting. You said quite possibly. Right. So there is a possibility to you that outside the headset, it's close to what we experience inside.
Donald Hoffman
Right. See, so for me, it's going to be a matter of following the math on this. So the theory of conscious agents itself doesn't require an agent to have a self, so.
Tom
Meaning it can shatter into a bunch of little pieces.
Donald Hoffman
Well, that a self is something that a network of conscious agents has to construct.
Mm.
So we, the Borg, it would be like an interface representation. Okay. So my. What I call myself is. Is on this kind of view, no less a construction than space and time. And so it's not clear to me how much of that construction will survive death.
Tom
Meaning taking off the headset. Taking off two possibly very different things.
Donald Hoffman
Well, it looks for people who still had the headset on, it would be interesting. It looks to them like death. But for the person who's actually having the headset taken off, maybe it looks like what some people describe with psychedelic experiences or near death experiences and so forth, or, you know, really extreme 5 DMT, Meo DMT experiences and so forth. These are things I want to explore. So at this point I'll have to say, you know, not only, you know, I'm likely to be wrong. I'm still trying to figure out what the ideas are that I would want to put on the table that are wrong. So. But the theory of consciousness that I have right now doesn't require conscious agents to have a self, to have memories, to have the ability to learn, or anything like that. So networks of conscious agents construct selves, they construct memories, they construct patterns and logics of emotions and so forth. It's going to be very interesting to ask the question about how much. My parents died recently, right. So it's very, very difficult. And so we all would like to think that there is some way to have contact with a person. Anyway, if you're a physicalist, of course that's out of the question, right. If your consciousness is nothing above and beyond brain activity, then when brain activity ceases, there's no consciousness. End of story. That's very, very clear in the theory of conscious agents. It's quite possible that your self dissolves, but the conscious agents, it seems to me, will still be conscious agents. They've just maybe dissolved this particular data structure that they created, the space time data structure with a particular personality and memories and so forth.
Tom
So basically you're saying If I can just put different words around that to make sure that I understand. The conscious ages coming together to form a person are a data structure.
Donald Hoffman
Well, they form the person as a data structure. So everything that I believe about me is very different than what I believed when I was five, and I didn't believe anything when I was maybe one. And so over the years, the decades of my life, I've put together this story myself. This is sort of an idea that some of my physicalist colleagues have said, like Dan Dennett, a self is a story that we weave. It's a narrative that we give. Maybe there's something to that. Whereas where I wouldn't go with Dan, you know, and he's a wonderful guy, he's brilliant, but, you know, it's okay to disagree. He would say that, you know, there's nothing but the brain activity and the narratives that it creates. And I'm saying, well, I like the idea of the narratives, but the consciousness is the fundamental reality. I have to let go of physicalism because of evolution and natural selection. And so maybe consciousnesses won't keep the narratives. Maybe they will. I have some of my colleagues who are working with me who think, you know, that we will keep the narratives, that we will keep that sense of self. But my colleague who thinks that doesn't have any mathematics to support it. And so for me, on the one hand, of course, I'm really open to all the different ideas, but as a scientist, if I can't put it precisely in math, I don't know what I'm talking about yet. I mean, you. And that's what you find is unless you can make it absolutely precise, most of the time we think we know what we're talking about. And when we make it precise, we realize, oh, okay, no, I was maybe in the neighborhood, but I didn't really know what I was talking about. That's what mathematics really comes back and teaches you. And that's the thing about a really good scientific theory. Once you write it down, you become a student of your theory. Like so Einstein, when he wrote down the equation of general relativity, he had the big idea. Falling in an elevator. You would feel weightless. Big, big idea. If it was free falling elevator. And he took him eight years to take that idea and what you call the equivalence principle and turn it into the equation of general relativity. He wrote, it took him eight years hard, hard work to take your intuition and go, oh no, my intuition. What do I really mean by that? It took him eight years and he was Einstein right Right. For the rest of us, if Einstein's ideas aren't that quite precise, and take him eight years to get it so precise that he knows exactly what he meant. That's what, that's what I'm talking about. So he writes down the equation. A year later, a guy named Schwarzfield who's in the front lines of World War I, is solving Einstein's equations on the front lines. And he. He solves equations and he discovers black holes. And he writes back to Einstein and says, your theory says there are black holes. Einstein didn't know that. He didn't believe it. He spent decades disbelieving it. His equations were right. Einstein was wrong. The equations become smarter than the genius who wrote them down. And that's another reason why we do these ma Mathematical models of science. We take our intuitions. It might take us a decade to take our intuitions and actually figure out what we really were thinking and get them so precise that we say, oh, that's the only logically consistent way of stating what I thought I was trying to think. And once you've put it down there, then all of a sudden you become a student. That thing that you've written down is going to teach you. And that's what I have with this theory of conscious agents. When I wrote it down, I had no idea about a number of things. I didn't know that agents could combine. It was a friend of mine who pointed that out to me. Agents can combine. And also that a weird thing that it predicts is that our free choices are not part of our conscious experiences. You can't directly experience your own free choices. You can experience that you chose, but you can never actually experience yourself choosing.
Tom
Why?
Donald Hoffman
Well, it's really quite interesting. You can experience, like if I go, you know, here's, there's chocolate and vanilla. I'm going to choose between chocolate and vanilla. Well, I just chose chocolate. But how did I do that? Well, I had some deliberation process, but when I finally. All I can do is see myself reaching for the chocolate or the vanilla. I can see my cogitation processes. But. But the.
Tom
But isn't that all in the headset?
Donald Hoffman
All that I'm seeing is in the headset. That's right. So I'm seeing. So, by the way, I only know my actions through my headset. I actually don't know what I'm doing when I reach out and grab something. I don't know what I'm doing in objective reality. In the realm of conscious agents, how
Tom
does all of this play out in your real life, and I've heard you talk about that there are moments where you have. I'm sort of putting words in your mouth, but almost a meditative experience where you transcend the notion of self. How so? How do you stay so enthusiastic about this for so long when it seems like, I mean really, really at a deep fucking level, man, as a human experience, this just all feels so real.
Donald Hoffman
Right? Right. Well, I wouldn't say that I transcend the self, but what I do get once in a while is a glimpse that, oh, this is just a headset. I actually feel it that I'm just rendering this. Most of us feel like space just exists. I'm stuck inside space. There's this big stage. I'm on the stage. It's very different. I think, by the way, the next generation will probably get this much easier. Those who have just been raised spending a lot of time in VRs that are as compelling and as immersive as everyday life. It's going to be just sort of obvious. You take your headset off and go, it's a no brainer to think, well, this is just a headset too. And to just sort of be there. So I think that it'll, it'll be for the next generation. The fact that I'm having a hard time about it, thinking about it this way and imagining it experientially will just be sort of an artifact of the technology I grew up with. If I grew up with VRs that were really good as opposed to the stuff that we grew up with, which is not that good, then it would just be sort of obvious. You do it when you're young enough. It's just this obvious that, yeah, I'm just seeing a VR headset too. Because by the way, here's one way to think about it. If you close your eyes, you just see sort of gray, right? Mottled gray in front of you. So it doesn't look like nothing. It looks like model gray. But what is it like backwards, back through your head when you close your eyes? Well, it's not modeled gray. It's nothing. And it's really the first time you really. If you close your eyes and experience that.
Yeah.
What is it like in front of me? Yeah, it was just gray, sort of model gray. What is it like behind me? Absolutely nothing. That's the headset. You only have a headset of space time in front. There is no headset behind. Now you have a. Not a visual headset. Now you have this. I can Put my hands back there and do stuff. So I have this. But it's all a creation.
It's all.
So I do get glimpses of that once in a while. But there were no. Now put on the natural selection language, right? So I have to pick the language of the science that I want to use, you know, because I don't have a better language in some sense, for. For discussing this evolution. There were no selection pressures for us to see the truth. And so there were no selection pressures for us to not take space time as the truth. And so we do. Piaget tells us, you know, when we begin to take objects as real, these aren't just like little data structures that you create that they really exist all the time. He called it object permanence. And Piaget said that, you know, when a kid is about 17 months, 16 or 17 months of age, they don't have object permanence. You take a little baby doll, put it in front of a child, they play with it. You put it behind the pillow. If they're 16 months old, they just doesn't exist. It doesn't exist, Piaget said. And then. But at 18 months now, they go and crawl around and try to get the object, the baby doll, out of the. Behind the pillow. Later experiments showed maybe down to three or four months. But the point is, these experiments show that we're programmed. Now I'm using the evolution language. We're programmed by natural selection to buy into the illusion that objects exist even when they're not. Perceived object permanence. When we're three or four months old, we're not rational. It's being done to us without our permission. And so by the time we come to the age of reason, it's the water that we don't know that we're wet. It's the water we've been swimming in all of our life. We just have been programmed to take this as the reality. I took it for the reality. It was only because I couldn't solve certain problems like the problem of consciousness. And it was only because when I looked at evolution, it began to tell me, space, time is not the reality, cannot be. It's the wrong language. It must be only like a headset that I was going, holy. I mean, I still remember the first time I realized this must be just a headset. I had to sit down. It was such. I mean, I was a grown adult. I was like. I was around 30 years old or something like that. The first time I realized this, it was such a shock. I had to Sit down. Everything that I believed all of a sudden disappeared. But of course, the next moment I was again visually believing. I'm in reality. I'm seeing the truth. So the programming is there. But ever since that moment when the math. It was the math that did it to me.
Tom
What do you think created math?
Donald Hoffman
Very interesting question. My own thinking in terms of this idea that consciousness is fundamental and conscious agents are fundamental. When we actually study consciousness. And there's been a scientific study of consciousness since 1860. It's called the field of psychophysics. There's a guy named Gustav Fechner who started the whole field. And a lot of my research has been in psychophysics, where we literally get mathematical models of conscious experiences and we test people very, very carefully in the lab. We find that mathematical experience, that conscious experiences are mathematically structured. My experience of this water bottle, the mathematics is unbelievable. There are. You can write down differential geometry, reflectance functions. I mean, the mathematics is incredible. It's true of all of our conscious experience, of everywhere we look. Conscious experience, it seems so squishy. It goes through your fingers. How can you. There's mathematics. So the way I think about math and experience is that mathematics is like the bones of the living conscious experience. They're not. They can't be divorced from each other. There's more to experience than just math, but there's not less than math. There's math and more. And so that mathematics and conscious experiences have a deep, intimate relationship that I'm still trying to understand. But.
But.
But the empirical evidence is quite strong. I mean, all the psychophysics that people have done, we just find mathematical structure everywhere. And that's why I came back to this Godel's incompleteness theorem, where that theorem is just saying no matter how many mathematical structures you discover you haven't started, there will be endless. More structures come from. Girdle figured out how to do this. So it came from.
Tom
And can I. Let me. Maybe Now I'm understanding it. So we once thought the atom was the smallest structure. And then we discovered there's something below that and something below that and something below that. Are you saying outside of the headset, there's just no end to the something below that?
Donald Hoffman
That's right. Interesting. That's right. That. And this is what Godel's result is, that there is no end to the mathematical structure.
Tom
So it's the incompleteness theorem that you will never be able to complete it. You'll never be able to get to sort of ba. Oh, God, I'm gonna put words in Your mouth base reality. I don't know another way to say it.
Donald Hoffman
That's right. That's what that Godel's theorem seems to be telling us is that no matter how complicated the mathematics is, that you know effectively you haven't even begun yet.
Tom
And when you say theorem, I assume that means that this is a math equation that Godel put forward and not. It's not Godel's theory, it's Godel's theorem.
Donald Hoffman
It's a theorem. In other words, it's true that we can't ever know all mathematical truths.
Tom
To me we're. My, my ignorance makes this a miracle. I'm willing to accept it just to not be abusive.
Donald Hoffman
Well, I'll just give you a clue about the kind of thing that he does. Right.
Tom
In the math.
Donald Hoffman
In the math there are these things called self referential statements that cause problems. So if I say this statement is false, now was that statement true or false? Well let's look. If the state the statement is, the statement is false. Well if it's false, then it's true.
Tom
True. Right.
Donald Hoffman
But, but. Well then it's.
Tom
If it's true then it's false.
Donald Hoffman
That's right, yeah. So you get these, when you have self reference, you get these problems that pop up, you know the, so the, the barber of Seville cuts all and only the hair of those who don't cut their own hair. Who cuts the hair of the barber of Seville.
Tom
Right.
Donald Hoffman
All these kinds of things. So Godel was able to take this kind of thing and make mathematical statements self referential. And he was able to create a sentence that said that a statement in mathematics that says this sentence is true within the system but can't be proved. So he was able to construct that. So he actually has a theorem. So he proves and then he shows that no matter. So even if you add that sentence in, there'll be a new sentence that's self referential. And so what he shows by this kind of structure, of course the true theorem and the true proof is incredible. There's girdle numbers. You have to be not just a mathematician, you have to be a brilliant mathematician, a logician to even understand it. So it's very, very few people who actually understand.
Tom
But every time you solve it, it's self referen.
Donald Hoffman
That's right, it's the self reference that's the key. So just so there's not in complete magic, I wanted to let you know that based on this notion of a self referential statement. But, but the bottom line is it shows that there's an endless, in principle, endless possibility of exploration of mathematical structure. And, and since I just mentioned this,
Tom
just to like almost give it a silly answer, it's the, the mathematical equation is what's below math. And the answer is math, it's structure
Donald Hoffman
all the way down.
Commercial Narrator
Forever.
Donald Hoffman
Forever.
Forever. But if consciousness is fundamental, that means that there's endless conscious structure, endless consciousness.
Tom
I feel like that's just something that says we don't fucking understand. Like that to me, quite frankly, is as sort of. I just have to accept it as turtles all the way down. Saying math all the way down or consciousness all the way down is the same as saying turtles all the way down. There's just literally no difference in my limited mind.
Donald Hoffman
Right, so this is where we're going to have to come with. Every scientific theory, right? Every scientific theory will have some set of miracles. And that's. That bothers me as much as it bothers you. Yeah, I don't like it, but I.
Tom
So when I bump up against that, I take a pretty. And I don't know, maybe this is stupid. Enlighten me. But here's, here's where I come up. So I will routinely be asked if I believe in God. And the answer is no, I don't believe in God in any of the ways that people mean when they say God. But there is so obviously something that I don't understand that it is just self evident to me. You would use the language, and maybe rightly so, that this is a headset. And so you have an intuitive sense that there is something beyond the headset and you have no idea what it is. And therefore you just say there is something I don't understand, right? So even here's the thing that used to fuck with me as a little kid. The universe is expanding. Expanding into what? When you build a house, you build it on land, right? So the land is there, the land is on the planet. So it's like there's this sense of to, for it to expand, it has to expand into something. Therefore something had to exist. It was very easy for me to just go, yep, there's something here. I don't understand. The fact that general relativity and quantum mechanics don't play well together. Yep. There's something I don't understand. Like I am, I am very okay with just going. And there's something I don't understand. It's, it's the, the part that gets hard for me is the, it's math all the way down where it's, it's no longer an acknowledgment of this is something we just don't understand. And it's saying, and now believe that it's math all the way down. And I don't know that it really matters, to be honest, but that, that's where I always bump and go.
Donald Hoffman
Well, yeah, there are two big camps on this. One is that we invent math and the other is we discover it. Right? So when a new theorem is published, did the person discover it or did they invent it? Right, that's, that's one of the big questions.
Tom
If, if isn't it self evident that it has to be discovered?
Donald Hoffman
Well, the, but if, if it is discovered, then girdle, Girdle's incompleteness theorem says it is math all the way down, right? So that's, that's, that's, that's the thing. But if it's discovered, I'm sorry, if, if it's invented, the, then that points at us and says, who is this discoverer? Who is this inventor that's doing all this? And it's really important for everybody to understand that every scientific theory, as we said earlier, stops, explanation stops. There's going to be some place where we say, grant me this, please. And if you grant me this, then I will explain everything else. Like Einstein says, grant me space and time. If you grant me space and time, I will write down these mathematics. And then it turns out there's black holes and they're all interesting stuff. We can do GPS because of Einstein's theory of special relativity, general relativity, it gives you all this. But we are granting space and time. Now someone else, like there's a guy named Seth Loy who says, okay, I'm not going to grant you space time. I'll start with quantum bits and quantum gates. Outside of space and time, it's just abstract quantum computational stuff. And I can show you how to boot up space time general relativity from quantum bits and quantum gates. The curvature of general relativistic space time has to do with the action of the gates and so forth. So he's no longer assuming space time, he's explaining it, but now he's asking for a different miracle. Please grant me quantum bits and quantum gates. Now you can imagine someone going, well, now I'm going to do better than Seth Lloyd. I'm going to explain to do something deeper than quantum bits and quantum gates. That's perhaps what Ninimar Arcani Hameda is doing at, you know, Princeton. So maybe he'll get quantum mechanics emerging from something deeper but he will then say, grant me. What he's asking for is the amplitude and some other structures like that. So if you grant me this amplitude, I can give you space, time and quantum mechanics and so forth. So that's the nature of explanation. And I don't. I'm still having to come to terms with that. I would like to have a theory of everything, and I can only have a theory of everything except these assumptions for my theory. And those assumptions are my miracles. And so at the foundation of every scientific theory, there's this moment of humility. Explanation stops here. It also stops one other remarkable place, whenever in our theory we have a probability that can't be reduced by greater knowledge. So in a Newtonian world, if I flip a coin, in principle, if I knew in detail the mass of the coin and its distribution and exactly how I flipped it, I could tell you heads or tails with probability one. But I don't know the initial conditions well enough. And so I have to give you a probability of a half. That's a subjective probability, epistemic probability. But suppose, no matter, there's a probability that no matter how much I know, the probability can't completely go away. Then that's no longer epistemic. There's something more interesting going on. This was the debate between Bohr and Einstein about quantum mechanics. There are probabilities that come up there. Einstein was saying those probabilities are just our lack of knowledge. God doesn't really play dice. There's no fundamental probability going on there. Bohr said, no, no, no, you don't tell God what to do. These are not epistemic probabilities. These probabilities cannot be reduced, period. They're probabilities for everyone, including God, metaphorical God.
Tom
One thing I wanted to ask you that came to me when I was reading is the double slit experiment is one of the weirdest thing in physics for me that it really messes with me. So for people not familiar with the double slit experiment, you take a single photon, you shoot it through a slit, and if you're not measuring it, as it happens on the sort of back wall, you would see like a bullet mark, right? So you fire a single bullet, it goes through the slit and it hits the wall. If you. Sorry, that's if you watch it, if you don't measure it, then it goes through like a wave and you get an interference pattern on the back. You can put another slit and if you're watching it, it goes through one slit. If you're not watching it, then it goes through both slits like a wave. None of us just seems so weird. But if we. Well, in fact, I'll, I'll ask. Does your hypothesis about consciousness address the issue of the double slit experiment?
Donald Hoffman
Yes. So it will come out. I believe that what those experiments are showing us is that if we assume that everything is happening in space and time and that space time is the fundamental reality, then we're going to be confused. What we're seeing is space time is just a visualization tool we're using for things that are happening outside of space and time. And so they're not constrained to travel through space and time. Absolutely not constrained. We're constrained to see them as though they're traveling through space and time. And that's why quantum mechanics looks so weird.
Tom
But why does observing or not observing change its state?
Donald Hoffman
Because we're creating reality as we render it.
Tom
That's what I was wondering if this is the look at the moon. Not look at the moon exactly right.
Donald Hoffman
The moon does. In fact, Einstein asked one of his colleagues when they were walking, said, do you really believe the moon doesn't exist or only exist when someone looks? It doesn't exist otherwise. He was talking about quantum mechanics. And my interpretation of quantum mechanics is exactly that, that, that space time itself doesn't exist when it's not observed and therefore the particles inside spacetime.
Tom
I hear you talk about the double slit experiment, doesn't that. So it potentially. Let me ask, is that potentially like a proof of your theory?
Donald Hoffman
Well, unfortunately there are no proofs in science. There you, every theory has lots of hypotheses and auxiliary facts and assumptions and so forth. And if, if your theory doesn't come out quite right, you don't know what went wrong in it. And also, even if every experiment that you've done is compatible with your theory, maybe you just haven't been smart enough to think of the experiment that will take it down. So, so, so no real serious scientists would say that any scientific theory has been proved.
Tom
Does it point in the right direction though? It's like you, you bring up a lot of examples, but that's not one. Is there a reason, is there a hole in that one already that you see or. Because to me that, that is some compelling shit like hey, you want to wonder or you want to know if the moon exists when you look at it or if it's garbage bin. I keep forgetting, forgetting your language.
Donald Hoffman
Garbage collected. Garbage collected.
Tom
Garbage collected when you look away. Boom. Double slit experiment.
Donald Hoffman
Well, so the double slit experiment is completely compatible with what I'M saying the reason I don't take it as a proof is because there are some physicalists who have the multiverse or many worlds interpretations of many worlds interpretations. So Hugh Everett, for example. So what these guys will say is that to really understand superposition and all these weird quantum things, you have to realize that whenever you make a measurement, whole new universes spin off and all possible states that are allowed by the quantum state function, the wave function, are true in some universe. And so many serious physicists believe in the many worlds interpretation and related but distinct thing of the multiverse, which is a different thing. They think that there are multiverses. And so what is true is that local realism is false. So local realism is the claim that objects in spacetime, like, say, a proton, have definite values of their properties, like position, momentum and spin. Definite values even when they're not observed. And there's two parts, that's the first part. And that they have influences that propagate no faster than the speed of light. We have very, very good evidence to say if we know anything, we know that local realism is false. But that leaves open whether it's the locality that's false. Things you could imagine things having influences faster than the speed of light. So guy named David Baum has a theory in which things have influences faster than the speed of light. Or whether it's realism that's false, which is what I'm claiming that realism is false. That a particle doesn't have a position or momentum or spin when it's not observed because you create it as a headset element when you observe. So that's why I say that we know that local realism is false. I claim that's the realism that's false. But there are some who can claim it's the locality, right?
Yeah, I would say that. Yeah, I would agree. I would say that there's a point in which a new portal in our interfaces is opened up into consciousness. So, for example, I have a portal into Tom's consciousness. It's not perfect, it's fallible, but it's real. I can guess if you're interested. I can guess if you're hurting or I can guess. But it's a genuine portal into your consciousness. But of course, my experience is not your consciousness. Your consciousness is separate. And so our interface does give us genuine portals into the consciousness of others. My cat, I have less access, my mouse, even less microbe. My interface has given up. My portal has become really dumb. It's not because there's no consciousness out there. It's that my portal has got too much noise or is just not opened up to it. So, yes, when we have babies, we are, in a way that we don't understand, opening new portals into the realm of consciousness. So I want to understand that it's very much like if someone is building something in Minecraft and they put things together. If you're really inquisitive, you'd like to know what was really going on inside the computer that looks like me building this thing that I see in my Minecraft world. And that's what I want to do. So in Minecraft, there are different things that are in some sense portals to a whole bunch of code that's going on in the computer that's allowing me to build this stuff.
Tom
Okay, let me start making some guesses here on where you're going. So I'm assuming that you're chasing this problem because ultimately you want to be able to figure out, hey, guys, guess what? Space time is fundamentally the wrong way to look at it. And if we get the right way to look at it, we will have all just an avalanche of insights may be usable in the way that quantum discoveries have given birth to modern life. Most people don't put two and two together. They just think the, you know, maybe the atomic bomb or something like that, but that there's real. Your cell phones, gps, like, all of it has to do with physics. So if we can get to the. The underlying physics of whatever gives birth to the. The simulation now we could have a whole new avalanche of insights that allow us greater manipulation of the world or whatever it is precisely that's driving you. Which brings me back to the collider. You're paying attention to that because you. If I understand how these things explode, then I can backtrack it and understand how they came together in the first place. Which in your prediction is going to take us outside of the simulation into the world of consciousness. Which do you imagine that world has physics and desires like does the. The consciousness must have something propelling it to do things,
Donald Hoffman
right? The.
The only idea I've got there is
this idea from Godel of just infinite exploration unbounded.
Tom
That is the. Okay, so this is going to bring us to God here because I feel like you've already said this is a magic moment or a miracle moment. So something has given the spark of desire to consciousness to explore Godel's infinite. Like, look at all the different manifestations that consciousness can take. And so we are all but one. Sort of pop up of a very Specific kind of consciousness. And who knows? There could be a Brazilian of these things all over the known and unknown universe. So as we're doing that, it begs the question of, in your mind, is there such a thing as God?
Donald Hoffman
Well, this gives us possibly a chance to have a language which for the first time ever, we might be able to formulate a precise hypothesis about what we mean by the word God and start to do science.
Tom
So, for example, do you have a hypothesis around that?
Donald Hoffman
Well, yes. So I have a precise definition of a conscious agent. And one definition I could give for God would be the agent, which is the combination of all the conscious agents, the single. So suppose it turns out that mathematically, when I look at the mathematics of conscious agents and how they combine to form new agents, suppose that the mathematics tells me there is in some sense always one largest agent that's being formed. I don't know if that will be the case. I don't know. But suppose then I could say that that's God. Suppose it turns out that, no, what's going on is that there are many, many quote, unquote, maximal agents, but there's no single maximal agent and they're not identical. Then that would. Then it might be a polytheism. But the nice thing about this is we could see the word God has been used in contradictory ways for thousands of years by various traditions. And we fight to the death over a term literally that's not well defined, because your God is not the same as my God. And it's a tragedy of human history that hundreds of millions of people have died because of differences over what we mean by the word God. And what I would like to see is sort of a humble approach to that term, saying, look, my tradition has its ideas, your tradition has its ideas. Clearly, since we all disagree, at most one of us is right and possibly all of us are wrong. So why don't we come together, take all of our best ideas and then try to sift through. And here's where science can help, where we take the ideas and try to make them precise, mathematically precise, because that's when you find out if things gel, if they're consistent or if they actually. You're talking nonsense, right? You know, if you say A is true and not A is true, well, you're talking nonsense, right? So. So can we do something like that and evolve ideas about the notion of God and have a scientific spirituality where God becomes something that we don't assume that we know? We're not dogmatic. I know what God is We're humble, we say these are the best ideas we've got so far. Let's try to make them precise so we can figure out precisely where we're wrong and then evolve it. So I see this direction going forward, the possibility of a scientific spirituality where we take a lot of the spiritual insights that the spiritual traditions, the mystical traditions have had for thousands of years and we take the tools that science has learned by studying our headset and we use those tools to take the first step outside the headset that the spiritual traditions have been telling us. Look, you guys are stuck in the headset. You're stuck in the headset. Well, they're right. But by studying the headset we got the tools that we needed to actually step outside the headset in a way that can be precise and we can figure out what we're doing and make precise theories. And so I see science and spirit, we're at this really interesting convergence point in human history where there's been a war between science and spirituality. In effect, they've been, these are non physicalists over here, the spiritual side, these are physicalists over here, the scientists, scientists have all the hard nosed tools, it's all woolly and imprecise over here on the spiritual side. And so they've been at odds at each other, right? But now let's take the best insights and I'm not saying that anybody's right or wrong, I'm just saying let's take the best insights from the spiritual traditions and be willing to call nonsense, nonsense and take the best tools from science and take and also call theories nonsense that are nonsense and try together to make precise theories about the spiritual realm and our position. You know, what are we here for? What is life about? These deep, deep questions that are all important to us and see where it goes. And you're right, I think that as we do this, we're going to understand if this idea that there's a network of conscious agents outside space time and we begin to really understand how in part it works and how it relates to space time. The technologies that we're going to get are stunning. The analogy I would think of is like this. Again, going back to Grand Theft Auto. You can imagine being someone who's a wizard at Grand Theft Auto. We all just look in awe at what this wizard can do and that's truly impressive. But now the software engineer comes along who actually knows the software of the game and he can take the gasoline out of the tank or let the tire out of the airs of the wizard's car. So now the wizard is nothing compared to someone who knows how the software works outside of Grand Theft Auto. Once we get outside of space time, we understand how this network of conscious agent works and how space time is just a visualization tool. We will be in the position of that software engineer that can actually just give the Grand Theft Auto Wizard a flat tire. We'll be able to he could change the road. He could change the dimensions of the road. He could make buildings disappear. He could make space and time shift in any way he wants. That's the kind of Pandora's box potentially that we're going to open up here. It's both as a scientist and human being. It's exhilarating and scary as all get out. That in the wrong hands is incredible power.
Commercial Narrator
Tired of overpaying with DirecTV, Dish offers a reliable low price every month without surprises. Get the TV you love and start watching live sports news and the latest movies, plus your favorite streaming apps all in one place. Switch to Dish today and lock in the lowest prices and satellite TV starting at $89.99 a month with our two year price guarantee. Call 888, add dish or visit dish.com today. If you work in university maintenance, Grainger considers you an MVP because your playbook ensures your arena is always ready for tip off. And Grainger is your trusted partner, offering the products you need all in one place, from H Vac and plumbing supplies to lighting and more. And all delivered with plenty of time left on the clock. So your team always gets the win. Call 1-800-GRAINGER visit granger.com or just stop by Grainger for the ones who get it done.
Tom
Yes, there's so many sort of layers of guesses between us and there. But if, yeah, if people have the ability to essentially edit the Matrix and anybody can do it, then that's going to certainly go into some pretty interesting places. Before we ponder that, though, I want to ask you, what do you think of Einstein's statement that I want to know God's thoughts, everything else is just details?
Donald Hoffman
Well, there's a sense in which I am very much on board with Einstein. In some sense. The attitude is who cares that there are so many kinds of quarks and who cares that there are so many kinds of gluons and that the mass of the electron is such and such and I mean that's not what it's really about. I mean, we need to understand those things, but not for their own sake. What Einstein and theoretical physicists are really after is they're trying, as Einstein said, I want to understand the mind of God. Right. I want to understand what are the big deep principles that would lead to the laws that have as their consequences. There are these kinds of quarks and these kinds of gluons and so forth. So it's the deep, deep insights. The same kind of thing that you and I are chasing here when we are talking about ideas like Godel's incompleteness theorem and Godel's candy store, like these, you know, even if that's not the right idea, it gives you a flavor for the deep kind of idea that we're after that we think of. This could be a deep enough idea that it could be a, a guiding principle for a lot of the details that we see. So it's in that sense that I would would agree completely with Einstein on that. Now, I think Einstein was very much attached to space and time.
Tom
Seems fair.
Donald Hoffman
Yeah.
He was perhaps the person with the most profound original insights ever about space and time. I don't know if he ever let go of space time. Maybe in his later years, maybe he did, I don't know. But most of us, it's really hard. The avant garde of physics right now are letting go of space time. If you just Google space time is doomed. Google that phrase, you can find videos by Nima Arkani Hamed, for example, and he can tell you exactly why state of the art physicists today are saying space time is doomed. So the mind of God, whatever it is, it's not space time itself, it's something deeper. But they don't know what that something deeper is, what the physicists are doing because they don't have any, they don't have any idea what's beyond space and time. What they're doing, which is really brilliant, they're saying, let's take the mathematics of the models that we do have in space time, quantum field theory, gravity, let's look at that mathematics and see that mathematics is like a light into the dark, beyond space time. Can we? And they're flashing that light into the dark as far as they can let that light go. So they're trying to see what the structure of our current theories might tell us, to guide us in our ideas beyond space time. But ultimately we're going to need to make a creative leap. We're gonna have to just. And this is the fun of science, you make a creative guess, you make it mathematically precise and then go back and say, okay, here's my guess about what's behind Spacetime, then you have to project it back into space time where you can test it. And so, so, so the physicists are doing great and I'm reading their work very, very carefully because I need to build on their work. I'm trying to plug my theory of conscious agents. I want to show that the, the long term behavior of conscious agents, so called asymptotic behavior. So this, this is a dynamics on graphs, it's a Markovian dynamics, and it has an asymptotic behavior. It turns out, I believe I can show that that asymptotic behavior plugs into structures that the physicists are finding called cosmological polytopes, amplituhedra, socahedra and so forth that have the same kind of permutation properties that the asymptotic behavior of consciousness does. So the asymptotic behavior, consciousness leads to certain permutations to classify it. Permutations turn out to be the fundamental thing that the physicists have found beyond space time. That's as deep as they've gotten. There are these things called permutations beyond space time that lead to what they call on shell processes, which they can then lead to predicting the scattering amplitudes. So that's the thread I want to pull. Start with a theory of consciousness where it's a vast social network of consciousnesses. Its asymptotic behavior is what the flashlight of physics has seen. The permutations, that's as far as they can go.
Tom
What's asymptotic mean?
Donald Hoffman
Long term behavior. So I'll give you very, very concrete with an example. Suppose that you're in your car on the freeway and you're driving and there's hundreds of people on the road with you.
Everybody.
It's a conscious person turning the steering wheel, pushing on the gas, hitting the brake, and so forth. Now imagine you're up in a helicopter, you know, half a mile up, and you're looking down. You're the Channel 7 news and what you see is a bunch of little specks, little dots moving on the freeway. You don't see any consciousness. You could think of it as particle flow, a bunch of particles flowing. Well, there's a little, the particles are all stopped here. There's a traffic jam and now the fluid is flowing. You could use like fluid dynamics, the kind of math that physicists use to describe that, because that's the long term behavior. You're looking at it from far, far away. You don't see the consciousness. So you're looking at long term behavior, not the little Turn of the wheel, the stepping on the brake of each person. You're looking from very far away. That's asymptotics, looking at the long term from a far away distance. So that's why physics has never seen consciousness. First, they're studying our headset, not consciousness. And second, the headset itself has only been designed to show us the asymptotic behavior of consciousness, because that's how we data compress, right? This is data compression. I can't see all the Twitter users and what they're tweeting. I see trends. Trends are the asymptotics. And so it turns out when you look at the trend behavior of consciousness, you get these things called that we mathematicians call permutations. And that's as far as the. So that's what I'm coming from my side. Consciousness asymptotics leads to permutations. The physicists pointing from the headset of space time into the dark, their flashlight gets to permutations. So we're meeting in the middle. So that's what I'm trying to do, is to bridge that gap. It'll take a couple years. I'm not a physicist, so I'm just learning as much as I can and I'm working with physicists. But, you know, there's a lot of conceptual work to do there. But then this will get back to the question you raised earlier about artificial intelligence, because once we understand how this whole thread gets pulled that I just described from consciousness through asymptotics into space time, we can reverse engineer space time and we can open new portals in the interface into conscious agents. The conscious agents are there. We'll be able to rejig our interface to open new portals and see new aspects of this social network of conscious agents that we've never seen before. And maybe some of the technology that we use to do that will look like circuits and software. I don't know. We'll see. And if it does, it would be. Then in some sense we might say, sure, sure, AI with circuits and software has opened. We now have consciousness in that AI. But it will be different. It will be not that the circuits and software and silicon was the founding thing and it created the consciousness as an emergent property. Instead, we just. Circuits and software are the language of the interface that we use to describe this much deeper thing we did that opened up a hole, opened up a portal in the interface that looks like circuits and software in our interface. But that's not the truth. That's just what it looks like in our interface, the deeper thing is something with conscious agents and how they operate. That we'll learn.
Tom
So let me make sure I understand that. So are you saying that we can create entities that meet some sort of requirement as to open a portal to conscious beings?
Donald Hoffman
That's what I.
Tom
The one that we know how to do now is sexual reproduction, and I guess asexual reproduction, depending on where you draw the line of consciousness. But we can make puppies, we can make humans and those. The act of that somehow, some way meets some criteria that then.
Donald Hoffman
Exactly.
Tom
Okay.
And that there is some configuration that we could do with robotics or AI or a combination thereof that would open a similar portal.
Donald Hoffman
You've got it exactly right, Tom. We have, as you just pointed out, we have proof that our interface does have the technology to open new portals. It's pretty low tech.
Tom
Where do we have that proof when
Donald Hoffman
you have a kid? Okay, it's a low tech. Right, right. It's low tech.
Tom
But we know it happens.
Donald Hoffman
We enjoy the process. But we have technology. In other words, our interface has proven that it's got the technology to open new portals.
Tom
Well said.
Donald Hoffman
Now the question is, if we can understand what's really going on there, we should be able to do it. And it may look like AI. And this may also be an answer to a very, very deep question about alien intelligences. Where are they? Well, we're looking at our headset, of course. We can't see them.
Tom
That's interesting.
Donald Hoffman
They're out there.
Tom
That's really interesting.
That's really interesting.
Donald Hoffman
We may. I'm not saying we won't find some of them out there. Our headset may give us access to certain alien intelligences on planets around Alpha Centauri and so forth, but it's going to be trivial compared to all the consciousnesses that are out there. That's what a headset is for. It's there to hide all of the consciousnesses because it would be overwhelming to interact with them. So that's where they are. They're all around us. We're like ants that don't see the guy with the RAID can coming at them. They can't. They just can't see it. Right. So they're all around us, but we are stuck on our headset. And we think that we're the epitome of advancement and knowledge and so forth. It would be funny. And it may be the case that every. That that's a rookie mistake that every consciousness goes through in its simulator, that it thinks that because its headset is only going to give it the best access to itself and others like it and less access to others. And so, in its own eyes, it will always be the wisest in its environment. And so it may be a rookie mistake that every simulator that consciousness creates has that we all have to go through this phase of going, oh, we're the best, we're the greatest, we're the epitome of evolution. And then we slowly realize, oh, no, no, no, no, I'm in a simulator. And there's this infinite range of other possibilities of consciousness, Some so profound that if I could see them, I would be inclined to fall down and worship them. Who knows? But it's a really humbling position. So we go from being the top of creation, the smartest. The brain is the most incredibly complicated thing we've ever found in the entire universe, that point of view, to, oh, wow, we've made a rookie mistake here. There are consciousnesses out there that are far more mind blowing than anything I could even concretely imagine. So it's a very, very different kind of thing. And this also, by the way, puts a different spin on the question of life living versus non living.
Tom
How so? I struggled with this the first time I heard you bring it up.
Donald Hoffman
Dead.
Tom
Still feels dead to me. How does it? Right, because you're saying the very thing that I perceive as them is the consciousness coming through the portal. And if I could retract down that bit of consciousness, it would be just like dealing with Timmy.
Donald Hoffman
Well, in that direction. So that's the right direction. And by the way, you know, I don't know the answer, so you might be closer than me. So when I say the right direction, Close to what I'm thinking, and maybe you're right and better. Better than mine. An analogy, I think that's. That's useful. Here is again, a virtual reality. And I like, you know, say you go with a bunch of buddies to a virtual reality arcade and put on headset and bodysuit and to play virtual volleyball. And so you find yourself immersed in a beach scene with the beach volleyball court and so forth, and you start playing beach volleyball for a little while. Then your friend Joe says, excuse me for a minute, I'm thirsty. I'll get a drink. So he takes off his headset and bodysuit to get a drink, and his avatar collapses, motionless on the sand. From the point of view of the VR game, he's effectively dead. He's unplugged, but he's just unplugged from the headset. He's not dead. And so if space time isn't fundamental, death may be nothing more. What we call death is just unplugging from the headset. Now, what the consequences of that are for consciousness, for example, maybe consciousness survives what we call death. But maybe the notion of a self, of me, maybe that doesn't survive, I don't know. This is going to be. I don't want to be doctrinaire about this. I want to be open to, you know, first having bold ideas. You know, you have to be crazy, drink beer, have some ideas. And then when you're sober, you go back and you start, you know, okay, let's do that. Looked really great when I, you know, when a gin and tonic. But now it doesn't look so great when I'm sober. Right. So you have to do, you know, explore, but then go back. So we want our mathematical model of consciousness to give us insights. What is a self? What is this thing that I call a self? Will that survive detachment from an interface? Or is that itself somehow wedded to the interface? What is the relationship between my interface, my spacetime interface, and my notion of self? I'm really attached to this body, but this body is just a symbol, you know, I am not this body.
Tom
I want to get into something really fast. So one thing that I find very confusing, not confusing, I understand it. And there's some part of my brain is rejecting it. So what I like about the reductionist model of quantum mechanics and all of that is that there is something beyond what all shorthand to personality. So a gluon is a gluon is a gluon, right? It's just a particle, and it acts in a certain way according to certain laws. And I can get that. Things acting in a certain way according to certain laws, as they group up into more and more complex things. Those more and more complex things are now the amalgamation of all these different laws. And they look like humans. Oh, my gosh, look at that. But as I break them down, they die. The immune system stops fighting off bugs. They return to a different anatomical state. Let's call it dirt. And now they're reacting in a different way. But then those could be consumed by something that ends up becoming a piece of the egg. And then the egg is re fertilized. And now you actually become something again. And so we're just these set number of particles that sort of take different forms, break apart, take a different form again. And it makes sense to me largely because there are things beyond personality. Personality is only associated with things that Take a very particular form, the sort of living biological things, as we would call it, and the individual pieces exist outside of that. And that's reflected back to me because if I go and give somebody brain damage, they will act fundamentally different, right? Deny somebody oxygen for a certain period of time, but then don't let them die, and they will be fundamentally different. And so it's like everything in the simulation, I fully understand that. But having lived in the simulation, all of that makes a lot of sense. Now, when I try to imagine, come up with a metaphor, whatever, where consciousness is primary, and this may just be a failing of. My brain is so primed to think of consciousness as being an element of personality that it just falls apart as I try to imagine, okay, wait. While this consciousness is in this body, it has personality. And I can fuck up its personality and make it act fundamentally different. But hey, when you come back out, one of two things has to be true. Either its fundamental nature was different than its manifestation as Timmy. And so it was like, hey, you were Timmy for a while. Oh, you got damaged, but you're still you. And Timmy's messed up now, but I can pull you out and you'd be just like you were before. But if I, as little Tommy, come and re engage with the consciousness that was part of Timmy, I wouldn't recognize this person at all. So even if they're in Timmy normal, they come out of Timmy Timmy still normal. I still wouldn't get this person or the consciousness because. And this is where again, my brain just immediately starts grabbing a hold of metaphor. So my brain goes, oh, okay, cool. I get this. The consciousness is like a puppet master and the very act of taking on the body, right? So pregnancy, cool, you got a baby. At some point, the puppet master brings consciousness to that puppet. And now that that consciousness gets to experience the girdle's infinite candy store range of states of consciousness, of qualia, and goes, oh, wow, cool. What a neat ride. As I experience this. Oh, this gets damaged. It just changes my experience of what this looks. This is what Sally was like before the brain damage. This is what grapes taste like to her now that she's had catastrophic brain injury. How interesting. And that's where I'm like, one of those two interpretations is way off because I can't reconcile them both now because I've lived my entire life in the headset. It just makes so much more sense to me that individual parts act one way. When they come together, they act another way. I mean, it's just. That's the human body, right. If you're just looking at bacteria, it feels very different. If you're just looking at mitochondria, it's very different. If you're looking at me without bacteria, different me without mitochondria, I'm dead. First of all. And it's only as you bring all these together that I actually act the way that I act. So I understand how I am this I am, you know, a totality. I don't want to put words in your mouth, but it's very poetic and super interesting. A small G God, right? Something that is like the accumulation of trillions of things that have come together in this pretty amazing form, right. That is delicate and precious and would never want anything to happen to it, but obviously does not rule the universe. And, you know, just trying to get through the day. But, you know, very interesting is you, if you define God as a collection, little G. God as a collection of these individual consciousnesses that have come together to create this sort of symphony that we recognize as a cheetah or a person or a gorilla or whatever that I get that to me is very interesting. When I start thinking as consciousness as fundamental, I start. I make the mistake that I read that as personality as fundamental.
Donald Hoffman
Right. The first interpretation is the one that I agree with you about, that the consciousnesses come together to create new varieties of consciousness. And it's a very novel and creative process. And all I mean, when I say consciousness is fundamental is that it's not space and time and physical stuff. That's the fundamental thing. It's that these kinds of consciousnesses that can combine to create new forms of consciousness, that's the direction that science needs to go to try to understand what's really the fundamental dynamics of what's going on. How is it that two consciousnesses, when they interact, create a new consciousness? And there, by the way, we can get some information like from split brain patients that we talked about last time. Right. There are people that have the commissure. Yeah. The corpus callosum is cut. And in one case, the left hemisphere believed in God and the right hemisphere was an atheist. Left hemisphere wanted to be a draftsman, right hemisphere wanted to be a race car driver. Completely different personality. So what consciousness is doing in that case is what happens when I put an atheist together with a believer in God and make the believer in God also want to be a draftsman and the atheist also want to be a race car driver. What kind of new consciousness do I get when I put those two together and put a corpus closon between them and force them to live together in a lifetime. What new. How do they. So the adjudication of that. And we all sort of sometimes will feel like, you know, it's Friday night. Part of me wants to do this. I'd like to, you know, kick up my heels and have some fun. Part of me wants to be a geek and study some more physics, whatever it might be. Maybe those are really two separate consciousnesses, two separate conscious agents that are negotiating and in the process, a new consciousness exists.
Tom
This is where I come back to. I'm so into that idea from Einstein, and this may not be what he meant. And maybe I'm just doing a poetic interpretation. Interpretation. But I want to know God's thoughts or the mind of God. Everything else is just details. Is like, I want to know what, what imbued these conscious entities with any propellant. Like what makes them step through the portal. Are there times where they're like, fuck it. No, I don't want to step through the portal. I just want to do my version of a Friday night chilling over here, you know, because when you think about two conscious entities coming together, one it is. They are. You get very distinct and predictable types of consciousness depending on whether it's right hemisphere or left hemisphere. So it is obviously not. You can very quickly oversimplify. But they're different, right? One side is going to be prone to religiosity and the other side would be prone to not. Like that's predictable. You wouldn't get. I forget which is right or left. But like whichever has the propensity for religiosity. You're never going to get it that the other side has a propensity. Like if I told you, hey, Vs Ramachandran, who studied the hell out of this, like, hey, one of the sides is religious. He's not going to be like, oh my God, I have no idea. It's 50 50. He's going to be like, oh, that's the left or the right, whichever it is. So now it's like this is so tied to the physicality that exists in my headset that it's like teasing out what is the. The what is the thought of God? What is the thing? Like, what's the game being played here? Like why. Why is there a puppet master? You know what I mean? Like, it's exactly so interesting.
Donald Hoffman
Absolutely, Tom. And that's the spirit in which scientists are. And I'm approaching this, right? We're saying let's. We try to reverse engineer exactly these kinds of things. What do we find from the neuroscience of split brain patients? What clues can we get to what's really going on outside the headset? So. But the first thing you have to do is realize I have to think outside the headset. So I have to let go of the reductionist view and say, how are neurons doing this? And saying, instead, let's talk about consciousness on its own terms. What's consciousness doing? And the neural structures that I'm seeing, I have to think about them. Okay, this is just my virtual reality interface. How can I reinterpret all these neural networks and what they're doing in terms of what networks of conscious agents are really doing? In other words, it's much harder work conceptually.
Tom
You must spend a lot of time thinking about this. Do you have a method for getting breakthroughs? Is it really a gin and tonic? And then go organize your ideas? Is it go jogging? How do you do this?
Donald Hoffman
Well, I have several things I do on a daily basis and I'm grateful to my wife for giving me some space to do it. I meditate for several hours. Okay, three hours a day, typically.
Tom
Whoa.
And you are pursuing the lack of thought, or you're pursuing to let your mind chase a thought.
Donald Hoffman
Entire silence, nothing but silence and watching what happens there. The idea from my geeky point of view about it, is if I'm going to go beyond the interface, then I've got to let go of the interface entirely. Any concept, any thought that I have is by definition still in the language in the interface that I've got. So going into utter and total silence is not easy. And it's both exhilarating and terrifying because there are no security blankets when you let go of all thoughts. And if you. So one thing I do is I watch myself.
Tom
You say when I look at the moon and then I look away, that I. You haven't said this in this interview, but I've heard you say it many times. You trash bin it or something like that. You're no longer rendering it garbage collector. Garbage collector would. Another way to say that the moon doesn't exist, quote unquote. If there were no species to perceive it, it's really that the moon wouldn't be created. Essentially, it's like is requiring an umvelt of a particular species to create that shorthand over evolution. To say whatever that thing is, right, that the moon is meant to represent. It has been created by humans as a moon. Maybe bats see it entirely differently and fish don't have any sense that it's Real at all. Who knows? It's really that we.
Donald Hoffman
We.
Tom
Evolution. Maybe is. Is a better way to say it, that evolution has created this virtual reality. And therefore, if that species ceased to exist, its version of virtual reality would obviously cease to exist because it's being created.
Donald Hoffman
Exactly. Oh.
Tom
But I'm going to say it's being created by the brain. You're gonna say the brain's not real. Is that true? Right, the brain. Okay. Yeah. This is where the shit gets so complicated. Okay, so I'm gonna shut up now. Back to your theory of consciousness.
Donald Hoffman
Right. So. But I'll just point your VR thing is, is I think a good example for anybody who's spent time in VR. What I'm saying will be obvious. Right. If you're playing a VR game of, like, race cars, you see a red Corvette. When you turn your headset that way, you know that you're only seeing a Corvette that you're creating. When you turn your head that way, you turn your head to the other side. Now you're seeing a blue Mustang. The red Corvette is gone. It doesn't exist. There's no red Corvette Corvette in the computer that's running the game. The red Corvette is only in your mind. When you look over there now you're seeing a blue Mustang because you're making that. And so you. You're rendering these things and then destroying them. There is a reality, but it's not Corvettes and it's not Mustangs. It's the supercomputer that's running the game. And that's what I'm. All I'm saying is evolution gave us this headset, and it's no surprise. I see the moon. I render a moon. I turn away. I don't render a moon. So the moon doesn't exist. There is something, but it's just not like it's not the moon. It's nothing like the moon. Just like there's something. There's a super. Supercomputer in the VR analogy. But in the supercomputer, if you looked, you'll never find any, you know, green Mustangs or red Corvettes.
Tom
You give an example in your book that is so powerful. If you would take a second, you describe what's happening in your eye when you look at a scene that includes a red apple. And the way you describe it at the photoreceptor level, I was like, oh, my God. It gave me such an understanding of how terrifyingly complicated things actually. Do you remember the part that I'm talking about?
Donald Hoffman
Right. So this is now just normal physiology. And so for the moment I'll be talking as though, you know, I believe in brain science. And we have to do that.
Tom
Right? Right.
Donald Hoffman
So you have to bracket everything is within the framework of a theory. So I'm now using neurophysiology and physics right now for this to describe this. So when you look at a red apple, and suppose there really is a red apple, just for sake of this argument, it's got a real shape and light rays hit it and they have certain frequencies and they pass through the lens of your eye, which focuses it on the back of your eye just like a camera would. And on the back of your eye you've got a piece of brain called the retina. It's nervous tissue. So it's a piece of nervous tissue. It has 120 million photoreceptors, it's like 120 megapixel camera. And each photoreceptor is just reporting how many quanta of light, how many photons it catches. So I caught three, I caught 10, I caught 50. That's all. You've got a bunch of numbers. So you have 120 million numbers. There are no colors, there are no shapes, there are no motions, there's just 120 million colors. Numbers, not even colors. It's like if you look at the digital output from a video camera, you'll just see a stream of numbers. If you look at the stream of numbers, you'll see the problem that vision has. You can't tell from the stream of numbers what's going on. You have to create three dimensional objects and shapes and colors and so forth from all those numbers. And so that's the problem that we have in vision. You have all These photon, counts, 120 million photon counts on each eye. And from that you have to then create objects. See that it's a boy on a bicycle eating a hot dog. Know all of that is you. And that comes, that's not just theory. It becomes really an important problem when you're trying to build computer vision systems, right? You're trying to build a self driving car, say with, with passive vision systems. Well, so the vision systems are cameras, the video cameras, say they're taking in video. Maybe they have, you know, a few million pixels that they come in that each, each, you know, maybe 70 times a second or something like that. Well, well, those pixels are just numbers. You've got millions and millions of numbers coming in every second. There's nothing in there that says that's a boy That's a car, that's a stop sign. There's nothing in there that says that you have to have megabytes of software that's really intelligent, that takes all those numbers and starts computing with them to figure out three dimensional shapes, to figure out what the objects are, and then figure out, whoa, I'm about to hit a boy, I need to hit the brakes, and so forth. So this is not just abstract self driving. Cars have to solve the problem of starting with numbers that are unintelligent in some sense, just a bunch of numbers, and giving you an intelligent assay of what's happening in the world. And so that's why a third of the brain, literally a third of the cerebral cortex, the higher part of our brain, is involved just in visual perception. When you add the other senses, it's more like half the brain is involved in sensory perception because the senses are doing an incredibly complicated job. But from my point of view, what they're doing is they're building a VR world. And it takes a lot of processing power. You need supercomputers, you know, what would have been considered supercomputers to do VR in real time. And that's what we're doing. We open our eyes and it looks like we're just seeing a 3D world with objects and shapes and colors. It seems so real and so just we're seeing the truth.
But.
But because you have billions of neurons, trillions of synapses that are doing it all within about 100 milliseconds. And so you're so fast at it that you just think you're opening your eyes and seeing the truth. You're seeing a VR world that you're projecting out there in real time. Of course, now I'm going to rescind the brain part, right? So the brain itself is part of the space time interface. So the brain itself is just our VR symbol for something deeper that's doing the real work. And so the question will be, what deeper theory can we come up with? Right? It's going to be a theory outside of space and time. But this gets back to what we have to do in science all the time. We push our current theories to the limits till they break, and then we have to actually take a creative leap. We can't necessarily just use the language of our theories that we just broke. We have to come up with a deeper language. And that is a leap. When Einstein went past Newton, he took a deep leap. And when quantum mechanics went past Einstein and Newton, it took an even deeper leap. And the language was entirely different. But you can show, for example, that if you start with Einstein, you get back Newton roughly as the speed of light goes to infinity. The way Newton talks about space and time and matter, mass, those terms actually mean something different than what they mean in Einstein. In Newton, mass is mass, and you have the same mass, period. In Einstein, your mass depends on your velocity. Your length depends on your velocity. Distances depend on your velocity. None of that. Space and time and mass don't behave that way at all. In Newton, we use the same words, but they mean something very radically different. In Einstein and in quantum mechanics, it's even a deeper leap. But with quantum mechanics, you get back Newton as something called Planck's constant goes to zero. Again, roughly. I mean, this is for first approximation. So the leap I need to make here now is evolution by natural selection is telling us that the language of space and time, then objects in space and time, is the wrong language to describe objective reality. So the leap I need to make is to say, is there a deeper theoretical framework that I can come up
with
such that when I look at the dynamics of that deeper framework and project it back into our VR interface, which is space and time? So I've got this VR headset of space and time. That's what evolution has told us. This thing is just a headset. You've got to guess, right? I'm not telling you what's outside the headset. I just told you all I can tell you. It's a headset. Now it's up to you to take a stab at what's behind the headset. So that's what I'm up to. So I'm probably wrong, right? I mean, we weren't evolved. Evolution by itself doesn't say anything that makes me think I'm evolved to see the truth, but it at least is telling me that whatever the truth is out there, you're only seeing a headset. It's good enough to tell us that, but it's not good enough to tell us what's outside. That takes a leap outside of the theory of evolution. But the constraint on that theory is we need to show how our deeper theory could lead to me having a headset. And in that headset, it looks like things are evolving according to evolution by natural selection. In other words, my deeper theory has a strong constraint constraint on it. It better look like Einstein's theory of space time, it better look like quantum field theory, and it better look like evolution by natural selection. The three big pillars of modern science. If I Can't do that. Then I know I'm wrong.
Tom
So even though all of those things exist in the headset.
Donald Hoffman
That's right.
Tom
Okay, this is fascinating. So now, tall order. Do you have that idea?
Donald Hoffman
I have a proposal.
Tom
Let's hear it.
Donald Hoffman
I'm probably wrong.
Tom
Can, before you give us your proposal, I have. I want to acknowledge you. Dude, the way that you talk like that, I love so much. I can't remember if it was Plank or Bore that said anybody. This is a terrible paraphrase. Anybody that thinks that science advances because objective truth is presented is absolutely wrong. Science advances because the old guard dies and the new people grow up just believing it to be self evident. I hate that so much. The fact that people are not willing to be wrong drives me crazy. Makes me want to choke half the world out. So the fact that you talk from the perspective of, hey, look, I've got a theory, it's probably wrong. Like, oh, dude, I love that so much. I wish more people were. Were as hardcore as you to present you.
Donald Hoffman
You.
Tom
Oh, God, you call it bold and precise. I. I presented a theory hypothesis that's both bold and precise.
Donald Hoffman
Right, Right.
Tom
Love that. And that you're willing to be wrong. So anyway, I just had to take a second. Every time I hear you say that, I want to stand up and clap because so few people are willing to own that they're probably wrong, but they're not afraid to make a bold and precise prediction. So.
Donald Hoffman
Right.
And the thing about that is I just want to understand. And so if I'm stuck on my ideas and won't let go of them, then if I'm wrong, I'm going to not understand. So it's really stupid to think that you're. You have to have enough hope that your ideas have some promise, that you pursue them, but not be dogmatic about them. That's. That's a. It's a fine balance.
Tom
Very much so. All right, so now our bold and precise claim is.
Donald Hoffman
So I'm proposing. I'll say what I'm proposing and then I'll say why I went that direction. So I'm proposing that reality is a vast social network of interacting conscious agents, so that consciousness is fundamental. And think of it like the Twitterverse. Right? There's tens of millions of Twitter users, billions of tweets, lots of stuff trending. It's. Twitter users are tweeting and following. And so it's all a big social interaction. Right. So I'm proposing, and this is a mathematically precise proposal, that there are things called conscious agents. So conscious experiences like the taste of chocolate, the smell of garlic, are fundamental and limited choices based on those experiences. That's part of the whole structure. So experiences that inform choices, that's going to be the fundamental idea in a vast social network. And the idea then about our headset is as follows. If you are a Twitter user and you want to understand deeply what's going on in the Twitterverse, well, you can't engage with all 10 million users and the billion tweets. It's just overwhelming. You would die before you could even read all this stuff. So what do you do? Well, whenever we have big social media data, we have to have visualization tools. Those tools will necessarily ignore most of the data. And the part that they don't ignore, they're going to compress it down. They're going to digest it and compress it into some eye candy that we can understand. Some objects in three dimensions that have nice colors and move in certain ways. And using that visualization tool, I can maybe see what's trending in New York, what's happening in all, you know, so the big scale of Europe, what's happening in little scale, and Irvine and so forth. So I'll have a tool that lets me zoom in and out, and it'll be ignoring most of the stuff. And that's what I'm claiming. Space, time is, and physical objects. It's our headset. It's a visualization tool that certain conscious agents use to interact with this vast social network that would otherwise be completely overwhelming. And so we've made the rookie mistake. Rookie mistake of taking our headset for the final reality.
Tom
Okay.
Donald Hoffman
We have a tool, and we thought it was the thing we were visualizing.
Tom
Amazing. Very clear. Now let's back up and break these down part by part, because conscious agent, I'm familiar enough with your work that I kind of know what you mean, But I don't think people understand that, like, how small. You take that down because you're not talking, that they're. Oh, hey, this is all a bunch of people, which is probably what somebody hearing this for the first time thinks that you think there are invisible people that make up this social network. How far? Like, so turtles. All the way down, boys and girls. So we're talking about consciousness all the way down.
Donald Hoffman
Right.
Tom
How like our neutrinos, like, are they conscious?
Donald Hoffman
Right.
So no. So this is very different from.
Tom
Is that because it's in the. The headset?
Donald Hoffman
That's right. Neutrinos are particles inside spacetime.
Tom
Okay.
Sorry.
Damn it.
Donald Hoffman
Nothing inside spacetime is.
Tom
I fell for rookie mistake. Okay, so out, outside. So do you delineate between advanced cognition and consciousness?
Donald Hoffman
Not in principle, no.
Tom
Interesting.
So do you.
I'm really trying to get you to use other words. So define agents. Right, so we're conscious. We're. This is a collection of conscious agents having a social ex. Did you say social media specifically? No, you did not. Social network.
Donald Hoffman
Social network. They're like a vast social network.
Tom
Got it. Okay, so give me what a conscious agent is.
Donald Hoffman
The simplest example, the most trivial agent that the mathematics allows is an agent that has only maybe two experiences. Maybe like red and green. Okay, that's all experiences.
Tom
Why does it have to be two?
Donald Hoffman
Well, it could be even just none. You could have an agent that has none or one. But I tend to think about what I call a one bit agent as sort of fundamental. But I could have an agent that has only one experience, like nothing or red has an experience.
Tom
Makes me feel like I'm not, I'm not interpreting what you say in the same way that you mean it. So you, you said something and it went by so fast, which was the taste of garlic and chocolate are fundamental mental, if I remember correct experiences. That's right. Okay, so literally the taste of garlic is a conscious agent.
Donald Hoffman
No, it's an experience.
Tom
Okay, so that thing, it's not like, hey Bob, you're the taste of garlic.
Donald Hoffman
No, no.
Tom
Okay, so sorry to use overly crass language, but that sort of. That was my initial interpretation of what you said. Okay, so that makes sense. So I will then push and say if that's the case, is this not. Would not garlic and chocolate be tied to the umvelt of the species?
Donald Hoffman
Absolutely. So certain agents will.
Tom
Does that mean it's in the headset then?
Donald Hoffman
Well, those experiences. So the headset is created out of your experiences? Uh huh. You. So what an agent does is uses some of its experiences as a format for a headset.
Tom
But.
Okay, so what I'm trying to get to. And I think this is what you're proposing is, hey, we take the headset off and we see. Like what do we see? And I get that we have the problem of perception and that's the. That's all back to the umvelt and stuff. So C is a stand in for. Obviously, I don't know how we would be interpreting this world. In the movie the Matrix, it is green code. Right. So when he explodes the agents, they explode into code. So code is their fundamental element. What is your fundamental element? Consciousness. I get that, but now I'm trying to understand, like, how if it's consciousness all the way down, what is consciousness like? Is it a physical substrate or is it not? And we have to let go of the very notion of physicality.
Donald Hoffman
Yeah, I'm letting go of physicality in the sense I'm completely, completely letting go of space and time and. And particles, electrons, protons and neutrons. Those are only headset entities.
Tom
But is it fair to say that you have no idea then exactly what consciousness is? You just.
Donald Hoffman
Well, so. So what we do. So here's. This gets at the fundamental way we build scientific theories. And this is what we talked about a little bit earlier, which is about every theory has miracles, right? So every scientific theory has certain assumptions
Tom
that it makes that we just have to grant.
Donald Hoffman
You. We just have to grant. No, there's no theory of everything. We hear about a theory of everything. There is none.
Tom
So your miracle is consciousness.
Donald Hoffman
That's right. Okay, so there. So I'm saying that there are entities that I'll call conscious agents. These agents themselves are not conscious experiences.
Tom
Why do they have to be networked?
Donald Hoffman
Well, it turns out that when agents interact, they form new agents. So when I didn't know this, when I wrote down the math, I was just writing down what. What could I possibly mean by consciousness being fundamental? I wrote down the minimal structure I could think of that could have some set of experiences. So there's some set of experiences that this creature, this entity could have and a small set of actions that it could take. And that was all I wanted to write down. But then when I had them interact with other agents. Right, because the actions are to affect the experiences of other agents, it turned out the interactions satisfy the definition of a conscious agent. So when agents interact, they form new agents.
Tom
Let me ask a question that I think is gonna explode this apart and help us all understand what you mean. Do I exist?
Donald Hoffman
Yes.
Tom
In what way?
Donald Hoffman
So you. Your conscious experiences exist. So all I can see is your skin, hair and eyes. Yep, I see. But that's just my interface symbol. If you look at yourself in the mirror, all you see in the mirror is skin, hair and eyes. But you know firsthand that what you don't see in the mirror, your hopes, your dreams, your aspirations, your headache, all the rich world of your conscious experiences that is not visible in the headset.
Tom
So my conscious experience is both in the headset and outside of the headset?
Donald Hoffman
That's right. That's right.
Tom
So it exists outside of space and time.
Donald Hoffman
That's right.
Tom
In fact, this is where the shit starts to get right.
Donald Hoffman
Real weird. It's pretty interesting because all I can see of you are the experiences in my own headset. And so my headset is made up of my own experiences that I've put in a particular format of space and time. And so. So it's just a format. In a VR, for example, you have a certain format in which the VR is presented. That 3D VR format has nothing to do with the shape of the supercomputer. It's just the format of that. And so this, what I see of Tom right here, is just what my headset allows me. But I believe, and you know firsthand, that what I can't see in my headset is this rich world of your conscious experiences. When I look at my cat, my headset only shows me fur and something really cute. And I believe that behind that cute little icon in my interface, my headset, there is a real consciousness. But I have less insight into the consciousness than I have into a human with a mouse, even less with an ant, even less. And then when I get to things that I call rocks and, you know, protons and neutrons, I have no insight at all. No surprise. The whole point of a headset is to simplify things.
Tom
Does this conscious entity need to eat?
Donald Hoffman
Very interesting, because that gets to what? See, this is now outside of space and time. It's not evolution, it's not food and so forth. So the question is, what are these agents up to? What are they doing? What's the. Why are they having any kind of actions at all? And the answer is, I don't know yet. I've got a mathematical definition of conscious agent. We're starting to play with dynamical systems of them. And as to the question of why they would have any dynamic why would consciousness do something as opposed to nothing? What kind of answer could be deep enough? Or at least what kind of proposal could be deep enough? And I've only had one idea ever that I've heard that seems deep enough to at least be on the table,
Tom
and that's consciousness all the way down.
Donald Hoffman
Well, no, I'm saying if we assume it's consciousness all the way down, what are all those consciousnesses up to? What is the social network doing? Why?
Tom
Okay, so the idea is the social network. Well, what's the one idea?
Donald Hoffman
So. So the one idea is the girdle's incompleteness theorem, right? So if consciousness is all there is, conscious agents are all there is.
Tom
Yep.
Donald Hoffman
Then mathematical structure is only about consciousness and conscious Agents, because that's all there is. And that means Godel telling us that there's endless exploration of mathematical structure means there's endless exploration of the possible kinds and varieties of cosmic consciousness and conscious agents. And what consciousness is up to is. Is what I call the kid in the candy store theory. Godel tells us there's an infinite candy store of exploration of possible conscious structures.
Tom
And the candy is all the variations of conscious experience.
Donald Hoffman
Consciousness and conscious experiences. That's right. And that's. I'm not saying it's right, but at least it's deep enough.
Tom
All right.
Donald Hoffman
That it could be, you know, it's on the table.
Commercial Narrator
DISH has been connecting communities like yours for the last 45 years, providing the TV you love at a price you can trust. Watch live sports news and the latest movies, plus your favorite streaming apps, all in one place. Switch to DISH today and lock in the lowest price in satellite TV starting at 89.99amonth with our two year price guarantee. Call 8- 888@dish or visit dish.com today.
Tom
Okay, so here's the good news. I am the guy that's dumb enough to, like, need everything explained, and hopefully that will be useful to the audience. Okay, so I've got this infinite candy store of consciousness. The thing I cannot get past is there's some. There is utility in creating the headset, otherwise why the fuck would it exist? So if there is utility in the creation of the headset, first of all, I get it probably begs questions that you don't know, and it's beyond the scope. And you've already asked me to just accept that the miracle is consciousness, and you're not going to tell me anything beyond that because nobody has a theory of everything, respect. Get that? But I want to poke in the spirit of fun partly, and then just to see, like, where the sort of edge cases are.
Donald Hoffman
Sure.
Tom
All right. So we don't know if conscious entities need to eat. And the reason that I asked that question was because all of this starts with you looking at fitness. And so if there is fitness into taking multiple bites of the apple, therefore, logic is going to be born out of that so that I know to keep eating the apple. But then it begs a question of, well, what am I underneath the visor, underneath the headset? Just to see.
Donald Hoffman
Right.
Tom
To keep our nomenclature consistent, what am I beneath the headset? And why does it matter that I have a representation that that is based in the idea of fitness of eating an apple? Right. So that's where I'm like, what. What is that a representation of? Why does this representation need to exist? Like, have you. I'm sure you've daydreamed about this, even if. Okay, let's hear it.
Donald Hoffman
Those are the fun questions, right? That's. That's what we're really interested in, is answering those questions. And so we're going at it in two different directions. One is, you know, sitting back in our armchairs and trying to think like Godel's Incompleteness theorem and so forth and saying, could this be.
Tom
Can I repeat what Godel's incompleteness theorem is? It's the one part of this. I'm nodding and smiling, but I'm like, the fuck is he talking about? So it goes something like this. And you will tell me where I'm wrong. So, in fact, God, can I even articulate what I think. I'm going to fumble through this. Okay. I don't even. I feel like I understand a middle piece of it. I don't know where it begins or where it ends, but that we have some sort of mathematical equation that says you're never going to find the limit to which you can explore one thing. But I don't get why that thing is. Godel's Incompleteness theorem has nothing to do specifically with consciousness. Right?
No, no, no.
Somebody leveraging that to explain. Why is it, then, that consciousness is the only thing that you've seen put on the table that ties into that?
Donald Hoffman
Oh, well. So, okay. The reason I went after consciousness was I've been trying to solve what's called the hard problem of consciousness. How is consciousness related to brain activity?
Tom
Yes.
Donald Hoffman
And so people have been trying to show how consciousness can be booted up from brain activity, and we've utterly failed to do that or how the illusion of consciousness could be rooted up from brain activity. And there are no mathematic, absolutely no mathematically precise theories after decades of effort, that could explain even one specific conscious experience or one specific illusion of conscious experience, like the taste of vanilla, or why we. Why this kind of brain activity must be the illusion of the taste of vanilla, why it could not be the illusion of the taste of chocolate. There's nothing on the table. I mean, there's no science that can predict even one specific conscious experience or one specific illusion of conscious experience. And so the reason I went after consciousness being fundamental, the reason I went after that that's different from Godel's Incompleteness there was that I didn't want to be a dualist. Right. So as scientists, we try to create a Theory based on as few assumptions as possible. And we only want one kind of assumption. We don't want to have, like I want this physical stuff and I want this consciousness st stuff. You have to choose. Pick either physical stuff that's unconscious or pick consciousness stuff that's not physical. But don't do both. If you do both, that's dualism and it's not as clean. Maybe we'll have to. But we don't want to go there. So maybe dualism will end up being where we have to go. But I'm not going to start there. I'm going to start with physicalism doesn't seem to be working out. It seems to be principledism is that
Tom
you can stack enough neurons together that they suddenly become conscious.
Donald Hoffman
That's right. Space and time are fundamental. And that's where this, what evolution is basically saying physicalism is false. That's why I went after evolution. Natural selection says the language of space and time is not the language of objective reality. That means physicalism as we currently conceive of it is false. We can't boot up consciousness from neural activity because neural activity just a data structure in your headset. Neurons do not exist when they're not perceived. They couldn't possibly create consciousness. They're not even there. Consciousness, I'm proposing, creates neurons when we look inside skulls. But neurons could not possibly create consciousness. And I was forced to that by looking at natural selection. So I'm proposing that if I believe that I really do have headaches and I really do taste chocolate and I really do have conscious experience, and again, I could be wrong about that. It could be an illusion.
Tom
And just to be clear, in fact, we never finished this because I'm still. I'm just paralyzed by Godel's incompleteness theorem. But we're going to move on from that to not browbeat the poor audience. So going back to this network of conscious agents, you're saying, and this gets into the part about the brain I wanted to talk about. So split brain patients can have an experience of. Of being literally two separate people. So you go in, you cut the corpus callosum. So everybody to your point for a minute. I'm going to assume that the brain actually exists.
Donald Hoffman
Sure, sure.
Tom
So you go and you cut the corpus callosum, which allows for communication between the two hemispheres of the brain. And suddenly you realize that two personalities will emerge within the same head.
Donald Hoffman
Do they?
Tom
They don't both have internal dialogue, do they?
Donald Hoffman
They have different likes and dislikes.
Tom
But do they have Internal. Because one side handles language. So they both do have.
Donald Hoffman
Both have language.
Perfect.
Actually, the right hemisphere is very adept at language.
Tom
Fascinating. So they both have language. One could be an atheist, the other devout, which is so crazy. And that's a real example right. From the literature.
Donald Hoffman
My friend Vs. Ramachandran has a video online. You can Google that and say, you know, split brain patient Ramachandran. This patient has, you know, believe atheist and a believer. And one, you know, I think in his case the right hemisphere didn't believe in the left hemisphere was a believer.
Tom
It's crazy. I've never seen the video, but that's.
Donald Hoffman
Oh, it's great. One case, the right hemisphere wanted to be a race car driver. The left hemisphere wants to be a draftsman. They have just completely different personalities. They can play 20 questions with each other. You can give a word to the right hemisphere, and the left hemisphere will sometimes fail in 20 questions, they can't figure out what's in the right hemisphere's head. Whoa. So there, so there are separate contents of consciousness. So separate that you can lose at the game of 20 questions with your other hemisphere.
Tom
Your. That would be a very sort of simplistic example of two different conscious agents that have come together.
Donald Hoffman
Yes, that's. And that's the idea. So you are one conscious agent.
Tom
I. Well, the way that I perceive myself is as one conscious agent, but secretly I just a whole bunch of conscious agents.
Donald Hoffman
Well, that's, that's, that's the thing. You are. It's. Both are true. So you are one conscious agent, but you're also two. And you're also probably a huge lattice of interacting conscious agents.
Tom
You know the microbiome.
Donald Hoffman
Yeah, a little bit. Yeah.
Tom
Okay, so in your gut you have like trillions of bacteria. It's crazy. You have more foreign cells in your body than you have your own cells. Would they be conscious agents?
Donald Hoffman
Nothing in space time, strictly speaking, is a conscious agent.
Tom
But do they not represent something? That's another thing. Okay, I'm sure resisted asking you this question because I think I know how you will answer it. The very like, no one knew the microbiome existed. So if this is all in my headset, how is something so detailed that once somebody looks, it's like, oh my God, it's proliferating. There's all this stuff. How can we discover something new if it's literally just made up? Like the first time that somebody cracked open a skull, why did everyone look at it and see the same brain? The first time we looked into the microbiome why do we all see the same thing? Shouldn't a novel thing that evolution did not prepare us to see and understand, like, how do we all see the same thing Now? I think I actually know how you're
Donald Hoffman
going to answer it, right? Because once you have a good visualization tool, it depends on where you take it. So I could take my visualization tool for the Twitterverse, right? And I can zoom in on Irvine to look very, very close, and I can zoom back and look at the whole United States and see what that looks like. So a good visualization tool, tool for a social network lets you zoom in and out.
Tom
But doesn't that assume that everything. Oh God, let's see if I can articulate this. Doesn't that assume that tweets are a substrate that is universal, so that the, the, the first image that came to mind was why the hell, the first time somebody got their head bashed open, did everyone see brain the same way? And I thought, okay, well, he's gonna say that the reason they did is because you have a facsimile in the headset for tissue that's made from atoms. And so, like that, that visual structure, it doesn't care like what it is. It's photoreceptors taking in. I see this much light on this receptor, this on that, all that. So it's like, hey, I have a system for dealing with visuals and therefore when photons bounce off of this thing, it's going to construct something. And I understand things about 3, 3D and mushiness and texture and all that. And so my brain is programmed for that. So no matter what you put in front of it, it's, it's going to see that. I assume that is correct, roughly.
Donald Hoffman
But this is the, it's more like if you are trying to look at the Twitterverse and you have a, if you design a really good tool and it lets you zoom into what's happening just in your block versus in your city versus in your county versus in your state versus your country versus the whole world versus Europe and so forth. If the tool is really good, it's going to let you, you will see different kinds of structures as you move in. Maybe it's very, very all the same in my block where, you know, we all have similar ideas and we do the same thing, or in my, my county, but it'll be very different. And so, so the reason we see, right when you look inside of a brain, you are not just. And you see all these neurons and so forth, you're not just one agent, you're two. You're a whole lattice of conscious agents. What we're doing is using our visualization tool to look at the whole list of conscious agents that are together forming you. So we're using that visualization tool to look in finer, finer detail at agents that perhaps are having smaller and smaller sets of conscious experience.
Tom
But we don't really know that it's a one to one relationship between a cell or whatever and what else you
Donald Hoffman
might say, yeah, it's going to be many to one. That's right. So for example, I see someone I call Tom in front of me, but that's, I see one Tom. But my theory is proposing that, that there are countless conscious agents that I'm interacting with. There's one highest level conscious agent, but immediately below it there are two that I associate with what I call the left and right hemisphere. And then below each of those there's countless more. And there's more than one personality in Tom. The right and left hemisphere agents probably have very, very different personalities. It seems to be a general trend. They're very, very different. And who knows what goes among all those agents all the way down. So my visualization tool, of course all I see right now is skin, hair and eyes. That is pretty simple compared to what I'm claiming.
Tom
You are really good looking skin, hair and eyes.
Donald Hoffman
Very, very good skin. And it's a really complicated and intelligent network of conscious agents. But I see just very, very little. But when I look, when someone looks inside and sees a brain, the 86 billion neurons they're seeing there is my visualization of tool telling me there is a lot of conscious agents in a really complex social network going on here. That's what I'm seeing is 86 billion neurons. And then when we get down to chemistry, which is, you know, explode, that's even more, more complicated now, you know, saying, well, my interface is starting to give up because you're not seeing much about consciousness with, with neurons you might be getting some notion of networking and exchanging information. Maybe at the biochemistry, you're not seeing that quite as well. When you get down to quarks and gluons, you may be giving up, but there's tons and tons of quarks and gluons. And that's my interface telling me, look, I'm showing you a lot about Tom. I'm not showing you too much about his two hemispheres, the two agents. I'm showing you very, very little. But I'll show it to you what you call 86 billion neurons. And eventually my interface is going to have to just give up. Because the whole point of the interface is the network of conscious agents is too complicated for you to grok. You can't grok it. So I'll give it to you one agent at a time. Here's Tom. Not even Tom's left and right hemisphere, just Tom and then all the agents. If you want to, you know, you can get out. You know, if Tom will let you, we can go in there and look in his brain and we can.
Tom
Tom will not let you.
Donald Hoffman
Yeah, I don't think he will. That's right. I wouldn't. That's right. So that's why the tool is showing us more complexity all the way down. It's a vast social network. And each agent isn't just a standalone. We're a combination of many, many other agents. That's going to be one part of the theory that's really interesting is mathematically precisely looking at all the ways that agents can combine. But agents do combine. We'll be going to some new mathematics, I think something called infinite categories in
Tom
topoi, where infinite categories, what.
Donald Hoffman
Infinite cat category theory, and it's called infinity categories and also topoi theory. Some. Some T, O P O I. Right.
Commercial Narrator
The hell is that?
Donald Hoffman
It's some fairly abstract mathematics that allows you to economically start to describe countless other ways of interaction. And so what I want to. I mean, I have a few ways that my current mathematical model, which just uses things called Markovian kernels and measurable spaces, obviously plotting for what? I mean, most mathematicians would say it's fairly plotting, but it works and it's real math. We can start with that and then go to these category theories and so forth to actually get the full richness of all the kinds of ways that they could connect and interact. And so I want to go after that. But the one question you asked earlier was why does it look like we have to eat? And do the conscious agents themselves have to eat? And what are they up to and how can we know? And here's, here's my answer to that. First, I don't know the answer.
Tom
So.
Donald Hoffman
So here's how I'm going to go after it. First, I. I am thinking like Girdle's incomplete and certain we can come back to it. That might be a deep motivation for the whole dynamics. But suppose I'm. That's wrong. And I realize it's wrong. And I realize I'm just not smart enough and my team is just not smart enough to figure out what conscious agents are up to and why they're doing it, and so forth. Here's what we're going to have to do. We're going to have to propose a mathematical projection from the dynamics of conscious agents into space and time and propose how that dynamics gets represented in terms of, for example, quarks and gluons interacting. Right now I'm studying scattering amplitudes at the large Hadron Collider. I want to show how I can predict scattering amplitudes when two gluons hit four gluons go spraying out from the dynamics of consciousness. The, the reason for doing that is that if I'm not. Well, a couple reasons, but one is if I'm not smart enough to figure out what consciousness is about, what I'll then do is say, here's a mapping in that gives me back space time. Now, now that I know what, what's happening in space time, I'm going to pull it backwards and say, what does that tell me about the dynamics of consciousness? And I'll go, oh, I never thought about that. So conscious agents have to be interacting this way for it to look like this in my headset. So if I'm too stupid to figure it out, and it's very, very likely, I will have to take my theory of conscious agents, project it into my headset, and say, what would it look like? Oh, I'm getting the wrong answer. Okay, so I need to change my dynamics of consciousness this way so it really looks like the scattering amplitudes of quarks and gluons in the Large Hadron Collider. And when I get that match, then I go, okay, this is at least one dynamics of consciousness that gives me the right answer in my headset. Now, what is it telling me about what consciousness is up to? So if I'm not. If we're not smart enough to do it from first principles, and I want to do it from first principles, but if I can't, I'll try to reverse engineer relativity theory, evolution by natural selection, quantum mechanics, reverse engineer all of those, pull them back to the realm of conscious agents, look what they're saying about the social dynamics, and then try to get a clue about what that's about and answer questions like, am I forced to think that they need to eat? Or is there some deeper principle? Is it, for example, that I can only send an experience to another agent if an agent sends an experience to me? Is there that kind of dynamics? And is it going to be like the small world networks we see, for example, in the Internet where you get big hubs like Google and Apple which get lots of hits, and then tiny little guys like Kaufman who gets almost no hits, and then a few in between. But, but, but you get this. And, and there's in some sense the number of social connections you have is some sense your notion of fitness. The creators of Google are billionaires. Hoffman's not a billionaire. There's a correlation between the number of hits that Google gets, the number of hits Hoffman gets, and the difference in richness. Now, is that a deep property of these conscious agent networks or not? See, these are the. And this is what was found fun about science. I don't know the answer. Evolution by natural selection has told me it's not in space and time.
It's saying it's not.
This is all a headset. You're going to have to think as a scientist out of the headset, but I can't tell you what's outside the headset. So we have to be very imaginative and we have to choose what we're going to go after because of course we're probably going to be wrong. So you need to choose what you think is going to be interesting. I'm going after consciousness because I'm trying to solve this hard problem of how is consciousness related to our brain activity. And the mathematics is forcing me to see this vast social network. If I bring it to consciousness, there's this whole network, but I don't know what it's about. All the science that we've ever done, all the science that's ever been done so far has only been in our headset. Quantum field theory assumes spacetime. The fields are defined over spacetime. Relativity is space time. And evolution by natural selection has been about what evolves in space time. All of our science, which is very, very good science, has been in our headset. We've never really stepped out of our headset. But science has the tools. There is some initial work. There's a guy named Nima Arkani Hamed at Princeton Institute for Advanced Study at Princeton who I think is taking some really important steps beyond the headset. But Keys, he's already realized spacetime is doomed. And so he is already being very adventurous and stepping outside of space time, looking for mathematical structures in which space and time and quantum mechanics do not appear.
Tom
Okay, this is where things start to break a little for me. So I'm going to articulate what I think you have said about. Because all of this was born for you, you basically said you have two choices. You can either keep space Time and get rid of evolution. Or you can keep evolution and get rid of space time. And your argument goes like this, and obviously stop me where I go wrong, that all evolution cares about is fitness payoffs. There are certain behaviors that must be taken in order for you to have children that have children. And there's so much data coming in that, like looking at a desk or a table or something would simply be a string of numbers that dictate the number of photons reflecting off that coming back. And you could interpret that as different colors and brightness and all that, but in reality, it is just the math of that situation. And what you see around you is something to simplify that so that you only have to render what you're looking at. When you look away, it ceases to exist. There's just the overwhelming complexity. It would paralyze any animal. So the evolutionary strategy that was chosen is this crazy simplification that we exist in, which is very convincing, but it's ultimately a convincing lie and isn't real. So I'm like, okay, cool. I can actually get behind that. Where it breaks down for me is what's actually evolving. Like something needed this simplification. What is that something?
Donald Hoffman
Right? So there's. So there's got to be a deeper dynamics, right? There's some deeper dynamics that in space, time looks like evolution, but natural selection. That's what it looks like inside of our headset. But. But there has to be some deeper dynamics. And I. I've only had one idea that I've run across that seems deep enough. I'm not saying it's right, but it's. There's at least. But. But this idea will tell you how deep we have to go to at least get a contender. So if consciousness is doing something, if it's quote unquote evolving, but not from a natural selection point of view, just. So by evolving, I just mean it has some kind of dynamics. And the question is why? What is it up to you? What is consciousness up to? Right? Because now the typical stuff of trying to get food and procreation, that doesn't apply outside of space and time, presumably, it might not apply. So what's the deeper thing? The only idea that I've seen that's deep enough comes from something called Godel's Incompleteness theorem.
Tom
Yeah, I struggled with this in the last interview. Now I think I have it mastered.
Donald Hoffman
Oh, okay, well, then you're better than me because Godel's theorem is unbelievably deep. But basically, what Godel said without Going into the details is there is no end in principle, no end to the exploration of novel mathematical structure, Even quote unquote, God could never know at all.
There's always going to be an unbounded
inquiry ahead, no matter how much you know, there's unbounded structure left to explore. And if we make the assumption, and now I'm just speaking as a scientist, like saying, look, I don't know what the truth is. We put some ideas on the table and see where they go, right? That's how you do science. So maybe everything I'm saying is wrong, but I'm trying to be precise so we can figure out where it's wrong. So if we assume that consciousness is all, is really the fundamental reality, a bunch of these conscious agents, the social network, and if that's all that there really is, then the only thing that mathematical structure can be about is consciousness. That's all it's about because that's all there is. So Godel's theorem then would say because there's an unbounded and never ending possibility for exploration of mathematical structure. That means there is a never ending and unbounded possibility for the exploration of varieties of conscious experience. In other words, consciousness is like a kid in a candy store, but it's an infinite candy store. I call it Godel's Candy Store. It's a never ending exploration that is in principle never ending. It's not like you can ever come to the end of it. And so that is deep. I'm not saying it's right, but it's deep enough to be a real contender. Why should consciousness be doing anything at all? Well, here's one reason. Because no matter how much consciousness explores its possibilities, it could never come to the end of its own possibilities. And so it's in a never ending self exploration. And what we call Tom and Don are just parts of this overall exploration of consciousness and all of its possibilities. And you know, our little bit that we're exploring right now, as rich as it seems to us, is trivial, literally trivial compared to all the possibilities that Godel says are out there. And we get some feeling of the triviality. But I can imagine a square. I can imagine a cube. Now go up one more dimension. Imagine a cube in four dimensions. My brain halts and my mind catches on fire. Nothing happens. That's only four dimensions. I can't even go to four dimensions. I mean, how that. That's terrible. It's just an incredible limitation. I can only see three dimensions of color, red, sort of a red Green dimension, blue, yellow and so forth, Light and dark. We have three cone systems and so forth. There are some pigeons that have four color receptors. Presumably, they're in an extra dimension of color that I can't even imagine what it's like. Can you imagine a specific color you've never seen before? Try to imagine one specific color you've never seen before. So as rich as our world seem, we know that there's a rich possibility of conscious experiences that we can't even concretely imagine. But consciousness itself, on this theory, is exploring all these possibilities. And right now we're sort of stuck on this little headset. Three dimensions, small amount of color that we can see and so forth. Just we thought it was the whole world. No, it's just. It's a little headset. It's a prism, right? It's. For me, I realized I'm sort of stuck. I mean, this is like my imagination is stuck in only three dimensions. My colors are stuck in a certain range. Consciousness itself is exploring the vast possibilities that's going on. So that. Now, here's the challenge. Suppose we chase that down. We say, okay, there's going to be this ongoing dynamics of consciousness. Constantly going beyond what it knows, exploring the candy store further and further. Well, there's several things to say about that. One is, that's not the candy store. Makes it sound like all fun and games, right? Oh, but there's going to be a dark side to it, because going into the unknown means letting go of what you know. And that can be terrifying. So this suggests that consciousness is going to have this thing going on all the time, of the exhilaration and the terror of going literally into the unknown, where literally you don't have concepts. All of your current concepts are inadequate. And this is what Godel is saying. These are going to be new structures that transcend. So for those who meditate, when you go into silence, it's both healing and terrifying, right? If you really let go of all thoughts and go into the void, it's. It's sort of. It can be scary. Like you want to go back into. You grab back on your. It's a life jacket, right? You grab onto your teddy bear, your thoughts. So there's going to be this. To really go into new conceptual frontiers, you have to let go of old concepts. So there's going to be that aspect to it which is sort of a spiritual side.
Tom
Is it that aspect to making progress?
Donald Hoffman
I think so. I think that.
Tom
What's the central question for you?
So.
And I'll prime you by telling you my central question.
Donald Hoffman
Sure.
Tom
The central question in all of this that I find interesting is what is the correlate?
Donald Hoffman
So
Tom
I'm in this VR experience to the advantage of something, and I want to know what that something is. It isn't me as I perceive me, but it's something. Right, so what's your central question?
Donald Hoffman
Right, right. The.
Well, the answer is I don't know.
Tom
But there's no specific thing that you're chasing.
Donald Hoffman
Well, what I'm chasing is the idea that we're chasing, that the consciousness itself is only about. Exploring new possibilities of experience.
Tom
Okay, so I'm going to push on that. Let's go into the way that AI researchers think and work. And I know that was your degree, so this ought to be. I'm taking you right home. You have to imbue the robot with directives, desires. Right. So to put it in human terms, it's like there are certain things that humans want to do. We're an active species, so we tend to go and explore floor. We try to take control of our environment. We certainly eat, procreate, like all those. We have the drive to quench thirst, hunger, sex, so on and so forth. So, like, those things are pushing us to go do things even if we don't want to. We find ourselves compelled to seek pleasure, to move away from pain. All right, by default, AI wouldn't have those, so now you've got to give those. Those things. And I love that concept of minus a desire. Any entity just sits or is blown around by physics. Right. So obviously we're in the headset now, but like, you've got wind and so, okay, something lighter than the wind is going to move, but otherwise everything just sits. So what is it? Like what. How has this conscious entity, in that question that you just posed, been given some sort of impetus to explore the infinite candies in the infinite candy shop?
Donald Hoffman
Well, by the way, I'm on board with what you're saying. I think that that's really a good way of thinking about this and that you could think about our. Your life. And my life right now is we're in a simulator, a space time simulator. And we've been given like an AI system. We've been given certain intrinsic desires that we then find. And so. So, so my wife is an artist. That's how she's exploring. I am still stuck at third grade stick figures. So that's not how I'm exploring. I'm exploring in a different way. So, so, so there's billions of humans. And there are billions of different ways that we explore in music, in art, in literature, in science, various kinds of science, meditation. So where it sports. There are all sorts of ways that consciousness is exploring through us, and there is not like one is the best
Tom
or that you are what makes consciousness want to go to all the assay to create this VR experience, to run each of us as sort of an individual experience in consciousness.
Donald Hoffman
The idea would be that that is what consciousness is about, is the exploration of all of its possibilities.
Tom
So is that just the miracle? Like, I've now hit the part where you're like, I don't fucking know, this is the miracle.
Donald Hoffman
Well, yeah. And when we say that, that's something that we have to say is true of any explanation, right? Any explanation. At some point we say, if you grant me these assumptions, then with those assumptions, I can explain everything else.
Tom
And I totally buy into that. Here is the basic assumption I've always made about the miracle. That the whole point is to say, hey, everybody, I've got this theory. Here's my miracle. Check it out. Theory works, right? Yeah.
Yeah.
Wow. This is better than everything. Okay. The only reason I've told you about this thing is so you can tell me how to move the miracle back even farther. Or are you just like, cool, I'm good with that miracle. And now I just want to explore within that where that miracle is.
Donald Hoffman
Or, well, so I want to do both. So. So this, this miracle is so new that I, you know, I need to explore it a little bit. Right. So. So I want to push on this one. But the miracles I'm proposing are. Consciousness is fundamental. So experiences themselves don't come from physics. They are fundamental. So that's a miracle. And second, this other miracle that Godel proved, where he proved it, but it's still a miracle, right? The incompleteness of mathematics that you can. That there's endless exploration, that's just a fact. Which is a remarkable fact. We proved it. But it's nevertheless this. Why should the universe be that way? Why should logic demand that that's. So that's. But it's a satisfying kind of endpoint where we say, that's the way the world is. Here's the theorem that says that's the way the universe is. And that. That one theorem tells us why. If you're up to exploration, then have fun, because you will always. There's no getting there. That's the key. There's no arriving.
Tom
Can I say Godel's incompleteness theorem in a real world example because this one, anybody that watches the first episode, I must ask you 75 questions about this theorem. And I just could not get my head around it. And now the punchline is, as far as I can tell, it's turtles all the way down. So Godel's incompleteness theorem, as I understand it, says no matter how much you use math to try to find base reality, you will never find it. There is no end. There will always be another layer to discover on and on and on and on and on. Literally for infinity, right?
Donald Hoffman
So I can say a little bit more about what Godel was up to on that and how it relates. So Godel was the question they were asking. It was a mathematical question. That is if you write down a bunch of axioms, right, some these things are true, these things are, give me two points, define a line, two lines intersect, give me a few things like that. And then you ask what are all the theorems that you could prove from those assumptions? And so, you know, so we did all this in high school algebra and so forth. We would prove different things. But Godel, the big question that mathematicians had, David Hilbert really put this out there was can we, with just grinding through these theorems and proofs, get all mathematical truth is that there's some finite
Tom
number, there's maybe it's 6.7 million, but
Donald Hoffman
we can do it even if it's infinite in principle, if we could just grind forever, would we get all the truths? And that was. So Hilbert asked that question and Godel showed that the answer is no, that no matter how big your axiom system is, how many axioms you've got, he found a way to show here's something that's true, but you can't prove it within your axioms. So now, so you can take that new truth, stick it as one of your axioms. And then he showed, well, there'll be another truth that now pops up. And so this never ends. And so no matter how. So you can't just mechanically grind through and get all the truths. And the key thing about this, which is really interesting for consciousness now is Godel used self reference. It's when you get a mathematical system to talk about itself that the problems start to arise. And that's really interesting because what I'm suggesting here is that consciousness is trying to understand itself and all of its possibilities. And because of that self reference, it's a never ending enterprise.
Tom
And you can never sort of quietly draw a parallel between math and consciousness. Are they separable or inseparable entities?
Donald Hoffman
That's a really deep and important question. I don't know. Here's just my thoughts. I study psychophysics, which is a branch of science which we experimentally and mathematically study conscious experiences, your perceptions of shape and color and so forth. We write down math models and we test people in the lab. Every conscious experience has structure and we can write down mathematical structure. I've never seen a conscious experience that doesn't have mathematical structure.
Tom
And the structure is derived from what, what neurons are firing and things like that.
Donald Hoffman
Well, sometimes it has nothing to do with neurons at all. For example, in color space you can talk about. It seems like red and orange are closer to each other than red and green. That's a structural notion. And we can write down what we call color spaces and they turn out to be these non Euclidean color spaces. Taro Endow and others who did the mathematics on this come up with these really interesting complicated mathematical spaces which model all the relationships of colors to each other. So we have color experience and their structure. When you see a shape like a ball, we can write down equations that model the 3D shape that you're experiencing. So there's an intimate connection that's been studied scientifically between conscious experience and mathematics. The way I think of it is that consciousness is to mathematics like the living organism is to its bones. For if you're a vertebrate, you need the bones, but you're more than the bones. And so that's the way I think about consciousness can't be reduced to mathematics. I don't think there are others like Tegmark who would say that everything can be reduced to mathematics. Max Tegmark says that kind of thing. You know, he may be right, but that's not what I'm saying. I'm saying that mathematics is the structural aspect of conscious experience, but there's more to the conscious experience than just the structure, in part because so far as we know, different kinds of conscious experiences can be structurally identical. So, and if that's the case, meaning
Tom
something with the same structure could have different qualia feel.
Donald Hoffman
Exactly right. That's right.
Tom
So is that context dependent?
Donald Hoffman
How see, well, that's now where I'm getting to the edge of my understanding here. It's, it's. I did actually prove a theorem. It's called the Scrambling theorem, and I published it in 2006 where I proved that. That your colors could be scrambled with respect to mine, but we would be functionally identical in every single experiment we could do. So there was no structural thing that I could do to show that your color experiences were different from my color experiences. So your experiences aren't devoid of mathematical structure, but the mathematical structure isn't the whole story about your conscious experiences. That's the best I can say right now. And I agree with you. If that feels like we need to press further on that. I couldn't agree more. We do need to press further on that, and I look forward to. Because I think there some profound insights are yet to be revealed.
Tom
So where this all to me, I'm so curious. Like, I'm always trying to find how is something usable, right? How does this come back? So if it reveals something about myself and how I can move through the world in a way that's more interesting, more fulfilling, more whatever. If it actually. The big thing between last one and this one was the idea that space time is as malleable as video game code and that we really could, even though it would only be an experience in the headset, that suddenly things within the headset, like faster than light travel, suddenly become very real. We just have to figure out what the real substance of the headset code is so that we can get in there and begin to make changes. But that to me, forces the question of the me that I perceive as me is inherently an object of the. The headset. So there is. There is no me outside of the headset, but there is something outside of the headset. And I don't know if. If our minds maybe just work so fundamentally differently, but as we talk about this is what I start imagining. Okay. Hey, Tom. This is a collection of different conscious beings. And consciousness is the, you know, the driving force. And it's running through all these experimentations. And so I'm like, okay, cool. There's this thing underneath the 3D and it's, you know, essentially popping up all these different. Well, let me see what the color blue would be like for a kid that grew up in Tacoma, Washington, and went to Harvard elementary. You know what I mean? And like, oh, that's interesting. And then. But like, needed my entire sort of life and universe to go for it, to truly be an infinite exploration of what these different things. So there's a me that they spun up just to know the color blue at the age of 44, you know what I mean? Have with this exact history. And then there's. Oh, over here, I'm going to do that same test, but I want to know what salmon tastes like. When you had the early bad experience with salmon as a kid, but later you fall in love with this girl who's really into salmon? What does salmon taste like then? And so that's like where my mind goes, right? And so now I'm imagining like this consciousness of some kind. Maybe it too is the Borg. And, you know, maybe the Borg is what's feeding its drivers and all of that. But I'm still trying to imagine, like my brain immediately goes to, there's a thing. And that thing in my mind is basically the most classic looking alien ever. But they have like these long fingers that, like, each finger is an entity that becomes this, a conscious entity living in the headset just to go, oh, cool. That's what that experience is like. Do you have any sort of analogy metaphor that you think about this stuff in? Or is this just because I have a limited brain? I'm not sure.
Donald Hoffman
No, I love your metaphor. And I don't have a better one. And there's a couple things to say about that metaphor, though, which is one, in some sense, you know, a lot of spiritual teachers tell us to relax and enjoy the ride, right? Don't get all anxious, don't get all bent out of shape and so forth, rich or poor, whatever you might be, enjoy the ride. And in some sense when you do that, it feels right. I mean, it feels like, I mean, of course I'm exploring, I'm not being lazy, but there's something about, hey, you know, smell the roses, enjoy this thing. You know, you don't want to come to the end of your life and not have really enjoyed the ride. And so the idea that it's about not arriving anywhere, but the joy of exploring for its own sake, that's one thing that's on the table here. Is that what it's about? Now there's another thing that I've put on the table though, which is, of course, we should be very modest about all of our claims here. I'm probably wrong, right, about all of this, right? But as a scientist, what I want to do then is to say, well, how can we go about trying to figure out where we're wrong as fast as possible? And for me, what I'm doing while I'm exploring these ideas that we're talking about, I'm also trying to say, how would consciousness, precisely this vast social network, project mathematically into space time? Clearly there's a projection. I'm interacting with Tom's consciousness. I'm not seeing that consciousness. I'm seeing skin, hair and eyes. I'm seeing a space time projection. I'm not actually seeing your emotions. I'm not seeing Your mood, but I am genuinely interacting with your experiences. It's a genuine interaction. And so there is a projection from this conscious realm into space time. I want to get a mathematically precise model of that projection. The thing about gluons in this large hadron collider, predicting that is just not because I'm so interested in gluons, it's just to make sure that I'm not BSing myself. Did I get the mapping right? If I can't predict gluon interactions, then I still have my homework to do. But now, once I get that, here's the kicker. We can take what we understand in the headset and pull it backwards. If we can project from consciousness to the headset, then we can try at least to go from the headset and pull backwards. It's a fallible enterprise, but it may help us to open our minds to the possibilities for deeper theories of what's going on outside the headset. So. So the reason I'm doing this is because I can't even imagine a specific color that I've never seen before. I can't imagine in four dimensions. In other words, I take it as a given that I'm deeply, deeply limited in my imagination, and I need all the tools I can get to help me step outside of my headset and try to guess the unfathomable outside of there. So that's why we want to get this mathematically precise mapping into space time and then pull it back to give us a guide that can help us to not just be wildly speculating about what's going on in the realm of consciousness, but to have more tailored ideas that we can test in a loop. I get ideas in my headset. I pull them back to what they might mean outside the headset, take those ideas, play with them, project them back into the headset to get new predictions. So, in other words, if we don't want to BS ourselves, we have to figure out a way to make this an experimental scientific enterprise. And that's the loop. But this is a bigger scientific loop. It goes outside the space time headset to a theory of what's going on in consciousness, projecting back into space time, which is the only place where we know how to do experiments, and going through that loop. So science is up to the task. But here's the kicker. For me as a scientist, this approach says there will be no theory of everything.
Tom
Why not?
Donald Hoffman
Because Godel's theorem says.
Tom
You just said there's too much. You can't bring it all together.
Donald Hoffman
Any theorist is gonna have a finite set of assumptions, and the implications of that will be finite.
Tom
Well, could have such a. A deep theory of so many things that, I mean, it's. There's always a new sort of theorem to discover. But does that in any way, shape or form, break? Because here's. Here's my understanding of why people are searching for a theory of everything. You've got Einstein, Newton dealing with the macro world, and then you've got the quantum dealing with the microscopic world. And for whatever reason, all of your predictive models, when things are big, they work just fine. All of your modeling, when it's small works just fine. But when you try to make either one of them work together, then it falls apart. But we are. I could see sort of closing the loop on that. That becomes an everything, right? Even though it's actually just a very small subset, it's the headset. And so it's not the, you know, the true theory of everything. But if it allows you to unify sort of the known universe, that seems like it would be incredibly useful. And I'm assuming there's some. There is something for us to understand, like that loop you were talking about that goes and takes in consciousness, comes back into the headset that we could create a loop. So I'll make this more concrete again, thinking maybe more as a writer than as a physicist, for sure. But. So when I was thinking about, okay, the Big Bang and things expand, what are they expanding into? And the idea that a universe could sort of pop in and out of existence, like, how's all this possible? And one idea that's put forth is that basically the. The everything. So I'll hesitate to call it the universe, but the everything, whatever that is, is like foam. And every bubble is a universe. And the bubbles sort of come and go, and you're in the bubble. All you see is the bubble. The bubble, to you seems like everything. But if you were to actually puncture through that bubble. Oh, now I'm in the, you know, the bigger matrix of stuff. And so you. You begin to understand how you get a temporary limitation. But actually, then once you see what's beyond you, like, oh, I get it. And then Godel's incompleteness theorem answers the question of, okay, but what's that goop sitting in? Oh, it's sitting in this space. What's that space sitting. Oh, it's sitting in this. Oh, fucking girdle, right? And you just, like. It just. It goes on and on forever. But you would like, at some point being able to Predict seven, you know, bubbles on. It's like ah, do I really need to go to the infinite? There's just as long as if the, you poke through one bubble and all of a sudden all physics are backwards and it doesn't make sense. Okay, that would be problematic. But assuming you're just puncturing through to the next thing that's exactly like where you currently are, seems like you could come up with something unifying.
Ish.
Donald Hoffman
Right. So instead of turtles all the way down, it's girdles all the way down. Right. It's just endless exploration. And Godel's theorem could be like the deep meta theory. It's not the theory of everything, it's sort of a meta theory that tells us why we will constantly be exploring for new theories. So if you were looking for the deepest theory from this point of view, maybe I'd have to think about it. But maybe Godel's theorem would be this meta theory that explains why all the concrete theories will never be the final theory of everything. And so in that sense maybe with Godel we have a candidate for this meta theory of everything, which explains why the joy of exploration will never cease. Which is an interesting. But I agree this is a really interesting thing to try to understand as scientists. And we can use some of the tools that we've been learning from deep learning, so deep neural networks and so forth, where we get the agent based models of interaction for exploration and learning. So those, those tools that like DeepMind at Google has some papers recently where they, they've come to the same conclusion that my team has come from evolution. They found that instead of giving deep neural networks a strict model of the world that they want them to make reasons in or to make guesses in or to live in, they just give them, as you were saying, some, some payoffs, some you know, positive and negative rewards for their behaviors. And they don't tell them about the world, they just let the, the deep neural network explore, explore and they build their own worlds. And it turns out they don't build worlds which mimic the truth, they build virtual realities which are a dumbed down user interface that let them get the most rewards and they don't see the truth. So we actually, so there's some papers I can give you a link to Some papers from DeepMind in the last couple years where they've stumbled onto the same thing that evolution has stumbled onto, that the exploration never leads to the truth, that leads to the uneasy interface which sort of doesn't maximize, but which satisfies it gives you what you need in terms of the rewards. And I would like to say one thing, that an objection that people will have at this point, some people will go, this is crazy. Of course we see the truth. I know enough truth not to get hit by that car. And we know enough truth to build computers and to send rockets to the moon. So this is completely nuts to say that we don't know the truth from what we've said. Now people might be able to understand what I'm about to say, or guess what I'm about to say. It's sort of like if you're in one of the really neat little programs right now that, where you can build stuff. So they're building games like, you know, Roblox or Trove or Fortnite or Minecraft and you know, you can, you can become a master in one of those worlds, right? And so you might say, you can imagine a person saying, well, what do you mean? I'm not in reality. I can build these wonderful worlds. I can build all these tools, I can do all this great stuff in Minecraft or in these various worlds. And yeah, within that quote unquote reality of Minecraft, yes, you're seeing that reality and you are building things in that reality and it requires intelligence and so forth, but you're not seeing again the circuits and the software and the voltages. That's the real reality outside your Minecraft world. And so that's, that's, that's the answer to someone who says, well, I know enough about reality not to get hit by a car and to build a computer and to send your rockets to the moon. Yes. Well, you're in your Minecraft world and we've become very, very good at that Minecraft world.
Tom
I'm obsessed with the movie the Matrix.
Donald Hoffman
Yes.
Tom
And in your dedication, you. I forget the names of the three people, but you said, I offer you the red pill. That's right. Which I thought was pretty interesting. So what in your world, what is the red pill? What is the red pill waking us up to?
Donald Hoffman
Right.
Most of us believe that we see reality as it is. When we look up and see the moon, it's because there really is a moon and it would exist even if there were no observers to see the moon, it would still exist. And we don't believe that we see all of reality. No one thinks that we see everything that there is to see. But we do believe that we've been shaped by natural selection to see those aspects of the truth that we need to see, to Stay alive. And so that our perceptions of space and time are giving us a genuine insight into a real space and time that would be there even if there were no observers to perceive it. And also our perception of objects like tables and chairs, the moon, quarks and leptons and so forth, that these things would also exist and have roughly the properties that we see even if there were no creatures, no observers to see them at all. And the red pill that I'm offering is to say that if you believe evolution by natural selection, then the mathematics of natural selection makes it very, very clear that the probability is zero that any of the language that we use in our perceptions, the language of space and time, the language of shapes and objects and position and momentum and colors and so forth, is the wrong language to describe objective reality, whatever that reality might be. It's not that we're getting the shape of this table a little bit wrong or the color is a little bit off. It's that no description in the language of space and time and objects and colors could ever be done be true. Reality, whatever it is, can't be described in that language. If we buy evolution by natural selection. So we have a choice between taking one of our best confirmed scientific theories seriously, namely evolution by natural selection, or taking our intuitions that space, time is fundamental and objects are fundamental. Taking that intuition seriously. And I decided to side with this science on this one.
Tom
All right, so a lot of this stuff is going to be really difficult to tease out, but I think we have to sort of break it into points. I don't want to spend our whole time describing the theory. We may get lost for a fair amount of time in that. But I. What I hope we can do is sort of lay out the thinking and then get into some of the implications, why it matters, and all that good stuff. So if, I guess, let's back up. So tell me how natural selection gives us the mathematical model that invalidates all that. That's probably the right place to start, right?
Donald Hoffman
Most of us think of natural selection and evolution as a biological theory. That's sort of intuitive, right? You know, if you are better adapted to the environment, you're more likely to pass on your genes and so forth. But it turns out that in the 1970s, a brilliant guy named John Maynard Smith was able to use the ideas of evolution by natural selection and the tools of mathematic mathematics, in particular, game theory, to create a new field, evolutionary game. Can you.
Tom
Can you explain game theory?
So the.
When I was diving into your world, when I was reading the book. When you talk about miracles, and I don't want to do this out of order, but you talk about like every theory has sort of a base assumption. It's like we can't, we can't explain that one. So just please accept that this is the, the miracle, as you call it. And then from here, the rest of my argument is going to make sense. Sense, because I don't understand the math well enough.
The math.
Every time you mention this and nobody pushes you on it, I don't understand the math well enough. It feels like the miracle to me, but it isn't to you. So what actually, what is the math? If we can sort of say it at a, at a lay person's level, what is the math? How does it show that there's a zero probability that we're describing reality as it actually exists?
Donald Hoffman
Right.
So, so the intuition behind using games theory for evolution is very much like if you play a video game, right? If you're playing a video game, you have to focus on getting points as fast as you can. If you get enough in a short enough time, you get to the next level, otherwise you die. And an evolution is very much like that. There's points, they're called fitness payoffs, and you have to gather them as quickly as you can. And if you get enough roughly in a short enough period of time, you don't go to the next level, but your genes in your children go to the next level. And so it's, that's roughly why we want to think about game theoretic kinds of things. But it's, of course, it's more detailed
Tom
than what game theory is. Literally that what survives the next round.
Donald Hoffman
Yeah. So game theory is literally a mathematical theory of strategies in games and how different strategies may be better or worse in certain circumstances. So for example, if you. One game that our species plays is a social cooperation game and we were hunter gatherers, if we all went out and cooperated, I went out and hunted wildebeests and you went out and hunted wildebeests and others gathered berries and so forth. We all worked hard and came back at the end of the day. If I didn't get enough and you got more, then you might share with me and tomorrow I might share with you. So that's cooperation. But. And so that works very, very well. That's, that's a strategy that works very well.
Tom
How would we define this game? So I was tempted to say, okay, so it's about the people who have the strategy to win this game called Life. But I'm not sure that's actually how you define it. Would it be the game called procreation or.
Donald Hoffman
Okay, so in evolution, fitness is all, all about procreation. In fact, basically that's how it's defined. It's almost, it almost sounds like it's tautology that fitness being more fit means that you're having more offspring. So whatever you do that lets you have more offspring is by definition giving you greater fitness. And so, so in the case of cooperation, I mean, that's not the only strategy by which a species could be successful at having offspring in the next generation. But for Homo sapiens, that was a strategy that we did use. And in many social species do that, ants and bees and so forth. But as soon as you have that strategy of cooperation, then it turns out there's another strategy that could be very fit and that's cheating. So I could pretend I'm a loafer. Now, I don't want to go out there and put my life on the line in front of a wildebeest. And so I just go down to the river, I hide out and relax and take it easy and come back and go, oh, I worked really hard today. I couldn't find anything. Could I share some of yours? Well, and, and so it turns out that that strategy is very, very fit in the sense that I didn't put my life on the line. If you're willing to share with me, I'm going to survive, so I'm going to have. So that strategy will actually proliferate. I'll have kids, I'll have more loafers. But it turns out if you have too many loafers, too many cheaters, then no one's bringing home the bacon and the whole thing collapses. And so you get the idea that the fitness of a strategy depends on the other strategies that are around. If everybody else is cooperating, then being a cheater is very, very effective. It's a fit strategy. If everybody's a cheater, then it's not fit to be a cheater because everybody's going to die. And so that's what we call frequency dependent selection, that the frequency of a different, of a strategy will affect its fitness. And so that makes it more complicated. And that's why we have to do mathematics. You can actually, with mathematical precision, predict exactly what proportion of the population will be cheaters and what proportion will be cooperators once you know certain things about the strategies and their payoffs. And so that's why we use evolutionary game theory. It allows us to go Away from just intuitive notions of evolution and selection and so forth, to precise definitions of strategies, of their fitness, how they interact with. When 3 or 4 or 5 or n strategies interact, what happens? It gets very, very complicated. Then our intuitions just give up, but the math can still carry on where our intuitions give up.
Tom
Okay, so you partner with a mathematician and did you have a theory that you wanted to see if he could write the sort of algorithm for? Or how did that work? How did we end up focusing on math?
Donald Hoffman
Right. So in about 2008, I decided I really wanted to go after this because I had a suspicion that evolution would not favor creatures that see reality as it is. I figured that it would be too complicated and take too much time. And it turned out that was correct, but it was more interesting. So I got a couple of my graduate students working on this. I work with them. We learned evolutionary game theory and we started. They wrote simulations, and so we just simulated creatures with different strategies and we let them see all of the reality or none of the reality. We let them just see fitness payoffs.
Tom
Give me some of the data points. So one of them is going to be fitness payoffs.
Donald Hoffman
Right.
Tom
When you're writing in math, like, are you making this from a human perspective? Are you taking this from an ants perspective? I've got to imagine the math looks different depending on the species strategy that's already evolved sort of over evolutionary time. So. Oh, God, I can only imagine how complicated this is. But give me a. Give me a couple of the variables so that this stops being one of the miracles for me and I can really start to understand. Sure.
Donald Hoffman
So one kind of game that we had them play was a foraging game. So you could think of a big
Tom
checkerboard or big chessboard and you had them play. What do you mean them?
Donald Hoffman
So my graduate students then had these simulated creatures and we just plop them down at random in this big, huge checkerboard.
Tom
Okay.
Donald Hoffman
And we had also planted resources around on the checkerboard.
Tom
And did you have to give them incentives?
Donald Hoffman
Well, in. In some cases we would just. We would either allow them to evolve incentives. So that was what we called a genetic algorithm. So we started off, and they were all stupid. They, they all had random genes for how they moved and how they would try to feed and how they would perceived.
Tom
But there had to be some sense of score. Right. So if we're using game theory, there has to be some sense of. You have an objective. That objective is to gobble the resources or whatever.
Donald Hoffman
Yeah. So certain resources that were just randomly distributed on the checkerboard. Gave you high fitness payoffs, some gave you low fitness payoffs and so forth.
Tom
So we just placed so just defined as fitness payoff without anything beyond that.
Donald Hoffman
Exactly.
Tom
They had, they had a driver if you will, to put it in human terms. They had a drive to acquire fitness. Fitness payoffs.
Donald Hoffman
Right, right. At least in when we, when we programmed them. Now when we just had the genetic algorithm, there was no drive at all. The, they, they just what would make them move? So they, they had to do something. But the genes were random so they, they would do stupid stuff like they would stay in one spot and try to eat and, and they would get nothing and they just keep trying to eat there the whole time until that whole session and they would just die off.
Tom
Right.
Donald Hoffman
So in the first generation, right we
Tom
did they have the desire to eat and procreate? Like were they looking to mate or
Donald Hoffman
how did you all the only, the only desire was we, we told them they had to do something every step of the game. That's all we did.
Tom
And was there a list of items for them to choose from? Eat, move.
Donald Hoffman
Yeah, yeah, that's right. Eat and move. And that was pretty much it. And step. Yeah, eat and move.
Tom
That's right. And then you had something on the back end that said if they score this many number of points, then they essentially procreate and they move on to the next generation. If they don't, they're dead. That's exactly. Okay, got it.
Donald Hoffman
So the first generation, pretty much all of them just acted very stupidly. They didn't get any fitness points or very, very few. Some bumped against walls and just kept bumping into the walls. And so we would just breed the ones that were a little bit less stupid than the rest and then make a new generation of them. And we did that for four or 500 generations. And then by the time we, we'd done that 500 generations we had creatures that looked very purposeful, they were foraging in an almost optimal pattern. They weren't wasting any time and they were getting high fitness payoffs. And we could look at their perceptions and we found that they evolved perceptions that didn't show them the truth. It showed them only fitness payoffs. Which was no surprise to me. But what was a surprise was the reason why, why it wasn't just that it was too expensive to, to see the truth. It was that seeing the truth and seeing what you need to survive are very, very different things. And, and so, so I went to a mathematicians once. I had the simulations. And I realized that this looked like it was a real result and I gotten some new ideas about what was really going on. So the simulations really taught me it wasn't just that it was too time and resource expensive to see the truth. It was that in some sense the truth is completely irrelevant. And also that the fitness payoffs themselves don't tell you anything about the truth. They're just independent. So I went and worked with a mathematician named Chetan Prakash and we've worked together. He's the mathematician, I'm not. And we have. And I've gotten some other collaborators as well, some very, very good collaborators. And we have a couple papers that we've now written where with two different angles on, on the theorem. And the bottom line is, is this. It's, it's straightforward to prove that. Well, straightforward. We have a mathematician working with you. But to prove that the. An organism that sees reality as it is in whole or in part part is never more fit than an organism of equal complexity that sees none of reality, whose senses actually don't have the right language to see reality and is just instead tuned to the fitness payoffs.
Tom
Okay, so this is sort of the first thing that I trip over. So I get where I could see that it is more advantageous to be optimized for fitness payoffs than it is for reality. And this might be a good time to give people your sort of VR explanation so that we, we can bring this into something they can visualize. But first let me finish what I'm bumping on. So I get how if you're optimized for fitness payoffs, that makes more sense than being optimized for reality. Reality could have too much just complexity. The processing power that it would take to understand is crazy. Which is exactly why I think your desktop analogy is probably the better one to hit right now. But what I don't understand is why it is necessarily true that you need to hide reality. Like why that would be part of it. That seems right. Sort of a bridge too far for my, my simplistic mind.
Donald Hoffman
So that is a bit technical, but I, but the top level idea is that fitness payoffs depend on the state of the world. Right. So to be concrete.
Tom
And the organism.
Donald Hoffman
And the organism.
Right.
They depend on the world and the organism and the state of the organism and the action. So I mean, one example is, you know, if I have a T bone steak and you know, the fitness payoff of that T bone steak for a hungry lion is pretty high. But for the same lion that Wants to mate, it's very, very low. And for, you know, a cow in any state, and for any action, the t bone steak offers no fitness payoffs.
And.
And another example is if I'm 5,000 meters underwater, that's pretty bad. Bad for me. But for a benthic fish, it's perfect for a benthic fish. Right. So evolution by natural selection has this idea that there is an objective reality, and fitness payoffs do depend on that reality. But the payoff for the same state of reality could be very, very different for a benthic fish than for me. For a benthic fish, 5,000 meters underwater is the same state of reality as it would be for me, but the payoff is very, very different for the benthic fish.
Tom
Okay.
Donald Hoffman
So it would kill me.
Tom
This.
This may not be the thing to dive into, but I'm going to push on this a little bit and see if we get somewhere fruitful. What where I sort of come to in your theory is that they're lurking under this. It is a reality. Right. The moon is there to describe something. It's my shorthand. It's not. It is not a meaningless shorthand. It is a shorthand to help me get my. My fitness payoffs.
Donald Hoffman
Fair.
Tom
But it. It represents something. Okay? So if that is true, understanding that I am 5,000ft underwater still seems relevant, even though it's not. It doesn't have a fitness payoff. So obscuring that or obfuscating that for me so that I can't understand that I'm 5,000ft underwater is not helpful. Okay.
Donald Hoffman
Right. Right.
Tom
I can tell that if we chase that wind, we won't get where we want to go. But if you give us the desktop analogy, because this, I think, will give us the anchor that we need to keep exploring.
Donald Hoffman
Right. So if evolution didn't shape us to see the truth, what did it shape us to see? And I think the good analogy is that it gave us, like, a desktop interface. So if you're writing an email and the icon for the email you're writing is blue and rectangular in the middle of your screen, does that mean the email in your computer is blue, rectangular, in the middle of the screen? Middle of the computer, of course. Not that. I mean, anybody who thought that misunderstands the point of the interface. It's there not to show you the truth, which in this metaphor would be the circuits and software, the diodes and resistors, magnetic fields. You don't want to deal with that. If you had to deal with magnetic fields to write an email Good luck. You would never. No one would hear from you. And so that's what evolution did for us. It gave us a desktop interface that's there to hide the truth. Right. The desktop interface on your computer is there explicitly to hide the circuits and software. You don't want to see that stuff. That would. Seeing the truth would get in the way.
Tom
But isn't evolution somewhat of a blind watchmaker? I mean, to just steal from my man, Mr. Dawkins. So if it is blind, it's not hiding anything. So it's just. It's optimizing you for something. So that's where I get into the, like the, the punchline of your theory. Not to get too far ahead, but for people to understand why I'm stopping you. So the, the punchline of the theory is we are so. The far, like, so far off from what is real as to like, not even be able to. To conceive of what our world really is. And we will definitely get into space. Time is doomed. It's one of the most fascinating, fascinating things to come out of your theory. But it's like, it doesn't feel like a blind watchmaker is going to hide something from me. It just, it, it has no sense of it. It's just. Here is the shortest path across the checkerboard to get to this thing. It's not trying to trick me into thinking that there is no checkerboard. It's just like, take this path, go that way. It's the shortest path. I'm essentially lazy, right? If you think about caloric realities, right? So the way that I look at a human, I want to write a book called the Physics of Being Human. And to do that, I would really. It's the, like you write the book that you need. So to understand in myself why I am both driven and lazy is so fucking weird. So I both have these huge ambitions. I want to do this crazy shit, but I also want to sit on the couch and eat chips, right? So it's like. And they're both real, man. I am not like, when people think that I'm being humble or whatever. No, no. I really have like a hardcore drive to sit around and do nothing. But I have this like sort of weird balancing thing. So for me to understand it from a. A caloric utilization standpoint, right? As humans, we have these massive brains that are calorically just ravenous. And so for me not to have to forage all the time, I take a strategy where I'm conserving calories. Again, blind watchmaker not, not somebody going, hey, you know, it'd be really smart. It just realized the ones that didn't conserve calories died and, you know, they didn't survive a famine or whatever. And the ones that sort of balance this, like, I'm gonna go down and I'm gonna face the wildebeest and I'm not gonna hide all the time. You know, I'm gonna put myself at risk and all that stuff. The ones that found that balance, they were the ones that procreate and they pass on their genes. Okay, so part of the physics of being human is to both be adventurous and lazy at the same time. Where, where I'm trying to like, figure all this out then is, okay, I have this layer, I have my interface, but it, it doesn't feel like it's that, like radically far from the, the truth. So I'll give you my example. So, all right, it's mapping this room. You're going to tell me, fuck, dude, you don't even understand, like, space time isn't real. None of this shit is real. I look away, the moon doesn't even exist.
Donald Hoffman
Right, Right.
Tom
That seems like it would be more problematic for me to navigate the world if all of that were true. So what I want to understand is how much of what you say about the computer interface is really that divorced from reality. Oh, man. Stick with me. I'm going to see if I can actually articulate this. So the computer analogy is super profound. It's easy for me to understand. I don't want to have to deal with electrical fields and all the other things, the diodes or whatever, all that stuff. But when I think about whacking into a rock, that seems like a way closer thing to my reality. I've mapped something over that. Same with the moon. I've mapped something over it. It isn't actually that thing, right? But it feels like it would be mapping over some gravitational object that rotates around the Earth. But I think you're going to say that's not true.
Donald Hoffman
Right? So what analogy? Because it. Our intuitions rebel at the idea we're not seeing the truth. Because what we're doing here works so well. Right? And I'm open to that.
Tom
We're not seeing the truth. Where, where I get lost is how fucking divorced can it really be and still be useful?
Donald Hoffman
Right, well, so there's two aspects to that. One is just that, you know, the mathematics of evolution is quite clear.
Tom
Meaning the probability that you see the truth is zero is zero.
Donald Hoffman
That's right. And. And that we have a paper that we just submitted on Monday where we, we show that the fitness payoffs erase all evidence of world structure. Almost surely with what the does that
Tom
mean when you say world structure? What do you mean?
Donald Hoffman
So the world, presumably there is some world, some reality.
Tom
Yes. Like atoms.
Donald Hoffman
Well, you don't need to postulate exactly what that.
Tom
I need to know what you mean by the word, the word structure.
Donald Hoffman
So it could be. And, and what we show is it doesn't matter what the structure is. Pretty much the result holds. So it could be, for example, a structure like a distance relationship, what we call a metric or a topology or a measurable structure.
Tom
Okay.
Donald Hoffman
Or a total order. Like one is smaller than two is smaller than three. That's a total order, Right. So whatever the. So what you can show is that no matter what structure you might think the world has, you can prove that the fitness payoff functions that govern our evolution. So you're right. There's no of course, goal, right? It's not a goal directed kind of thing. Evolution is not trying to do anything. But the point is that the fitness payoffs which govern evolution, the probability that they will actually preserve the structure in the world, so that by being tuned to the fitness, you're tuned to the structure in the world, that probability is precisely zero. That's what we prove.
Tom
So the only thing that we know is what you think is real is the only thing we know is not real. Is that accurate?
Donald Hoffman
Well, I would say according to theory of evolution, now we can step back and ask ourselves what the theory of, you know, what we think about the theory of evolution. Right?
Tom
But before we get to that, because you, you do, you, you have a pretty death challenge to how. Because my first, literally the first note I took about you was my whole problem is you're laying out, hey, evolution mathematically proves that essentially evolution, evolution isn't true. So I was like, if, if everything I think evolution is true, how can it also be true that everything I know and think is false? Right. Which is essentially what I just heard you say. So mathematically I can prove this. The quote unquote structure of everything I think is absolutely not true. And yet I'm using evolution to base that math on.
Donald Hoffman
Right?
Tom
So walk people through how that isn't a contradiction.
Donald Hoffman
Right? So it's what we show is that the fitness payoff functions erase any information about the structure of the world, so that the structure of our perceptions is just unrelated to the structure in the objective world. But the argument that I just gave does not apply to our math and logic. So just perception, just perceptions. So you have to be very, we have to look at all of our cognitive capacities.
Tom
One, this whole thing in oomvel problem.
Donald Hoffman
Well, it, what it's saying is that your umvelt is like your user interface, that every creature, you know, Homo sapiens has one, um, one user interface. We have the Apple interface and someone else, you know, some other creatures have the Mac or the PC interface or, or whatever.
Tom
Yep.
Donald Hoffman
And, and, and there's going to be a wide variety of interfaces that evolution evolves. Every species has its own class of interfaces. And in each case the interface never shows any species the truth at all according to evolution. But there are selection. The selection pressures that erase information about the structure of the world in perception do not also apply to math and logic. The reason is that we do have to have some elementary ability to reason about fitness payoffs. Two bites of an apple give me roughly twice the fitness payoffs of one. So not reasoning about objective reality, just reasoning about fitness payoffs and the logic of fitness payoffs.
Right.
So that's why there's no selection pressures necessarily to be geniuses of math and logic. But at least the selection pressures are not uniformly against any capacity in math and logic. Whereas in the case of perception, one can show the pressures are uniformly against any access to the structure of the world in terms of the structure of what we perceive in our senses. So that's why we have to be very, very careful. Certain, for example, Christian philosophers, Alvin Plantinga, for example, has argued from, not mathematically, but informally from evolution, saying that it would make all of our cognitive capacities unreliable. And there, therefore evolution by itself was unreliable theory. And therefore we should not, you know, not believe it. And I'm not saying anything like that at all. I'm saying that the theory of evolution has a core that John Maynard Smith found evolutionary game theory. When we look at that core, we find that there are certain peripheral assumptions like DNA exists whether or not it's perceived space and time exist. Those peripheral assumptions turn out to contradict the mathematical core of the theory. And one can prove that. But math and logic, our ability with math and logic does not contradict the evolutionary, the core of evolutionary theory. So we have to be very, very careful. That's why when you do this, you know, it's not just hand wave anymore. You really have to look at the replicator equation. You really have to look at the fitness payoff functions and do combinatorial analyses and so forth. This is very, very careful work. But that's what we do with our best scientific theories, we take them very, very seriously. We look at their equations and say, okay, if the equation entails the probability zero that we see reality as it is, then we've got a choice. We can agree that we don't see reality as it is, or we can say we need to revise the theory. Now, we don't have an alternative to evolution by natural selection. So if someone wants to propose one, they've got a lot of work to, to. Because evolution with natural selection is an incredibly successful theory.
Tom
So space time, though. And you're, you're ready to ditch that one?
Donald Hoffman
Sure.
Tom
So, okay, so I'm, I'm getting, I think what you're saying about our umvelt. And because we haven't defined that, I'll define that. So umvelt is your senses take in the world in a certain way. And we have different senses in a bat. So a bat can do echolocation. We cannot. And therefore the way that a bat interprets the world is very different than the way that we interpret the world. And every sort of species has a different umvelt. You can even have humans that have a different, a slightly different umvelt one that's colorblind. There's a million other examples. But you, you get these variations that are massive between species. You get like a dog, the, the amount that they can smell is crazy. They can smell a seizure coming, which is. Is absolutely bananas. Whereas of course, you're not going to get a human that does that because of the number of send receptors in the nasal canal. And so cool. All right, that, that's an umvelt. So I get perception. I get that this is a, a problem of perception. I think I understand the math part. I prom. Because I'm going to. I am making a layman's assumption that math is a universal language. I've heard that repeated a million times. I am so bad at math, I don't even understand that. But I accept it out of ignorance. The logic part. Let me define logic based on what I think you're saying and you tell me if I'm getting it correct. So logic, the way that I understand it, from the way that you just use it. Because I would have said it's human reasoning. So I would have gotten tripped up on like natural selection or, oh God, the one. You just use it as an example. And you said that, that we think space time. A great example.
Right.
We have, in my layman's view, we have logic our way to space time. But that's failed me. So tying it back to fitness payoffs, I think, is your definition of logic that we have to be able to reason that one bite of an apple is not as good as two bites of an apple. It's. Is it really that basic, that it's entirely tied to fitness payoffs?
Donald Hoffman
That's well, from an, from the evolutionary arguments that I'm giving. Right, so the arguments that.
Tom
But do you think that logic reaches deeper than that when you say that it's untouched by this false interpretation? Oh God, I'm putting words in your mouth.
Donald Hoffman
Right, so. So that gets to the bigger picture of what I'm up to here. Right, so, and this is how science progresses. What we do is we take our current best theories and we try to push them to their limits and find out where they break, where they fall apart. And when we do that, that's when we break out the champagne. Because the whole point in science is to push our best theories to the limit to find out where they break down and then get some clue about a deeper theoretical framework. And the constraint on that deeper theoretical framework is it better agree with our current theory where our current theory is?
Tom
Right, but only if you're saying that that theory is logic. This is another one of those times where I bumped against what you're saying. So, and this may be that simple, is that what you mean? That like where we are using true logic as you define it, tied to fitness payoffs? I think then it must agree because you've said if you work backwards, I'm not trying to get rid of evolution, I'm not trying to get rid of the things that we know. Like we can launch a satellite into space and geo target you, but that requires relativity. So whatever we get to, whatever answer we come to better be backwards compatible with the ideas that essentially work. Can you give me a definition though of work? And is it tied to logic as you define it? And thusly fitness payoffs.
Donald Hoffman
Right, so. So the idea would be that scientists are doing reasoning, right? Science is about reasoning.
Tom
Is reasoning a synonym for logic to you?
Donald Hoffman
Pretty much that's right. And if we, We can't. Logic is in some sense non negotiable in the sense that if I let go of reason and logic, then there's nothing left. Right.
Tom
Is that true?
Donald Hoffman
Because all we can do if we, if we're having a conversation, we can talk informally, but we can't make very good progress unless we are absolutely precise in what we say. Imagine that.
Tom
But let me tell you how that struck me so does somebody who's schizophrenic, do they have reason and logic?
Donald Hoffman
Well, in the sense that if we're trying to understand our situation in the universe, humans will make stories.
Tom
Yes.
Donald Hoffman
If the stories are internally self contradicting, you can be pretty sure you're going to be in trouble. Right. So any internal contradictions are going to destroy your theory. So the first thing you have to do is make sure that what you're saying doesn't contradict what you're saying, because that just means you're saying nonsense. So one reason we use math and logic is to make sure that what we're saying isn't just flat out nonsense. But once we're past that criterion, and much of what we've said is nonsense, but once we get past that criterion, then it turns out we may have been using our terms very imprecisely. We might use the word space, we might use the word time, and we think we know what we're talking about. But when you actually push, you find out, oh no, I would say this in this situation, that and that situation. Oh, and they contradict. I've contradicted myself again. So once again we find that if we just use intuitive concepts, we get trapped in self contradiction, namely nonsense. So the whole point of being mathematically precise, one point is to make sure that we're not doing nonsense. The second is to take our ideas and find, force us to put our ideas very, very precisely so we know exactly what we're saying.
Tom
I get the math part though. The part that I'm really trying to wrap my head around and maybe we should just move the fuck off this. But I'll take one more swing at it. Math. I get if, if math truly is a universal language that just sort of gets at the substrate of what this, whatever this is, it actually gets to that. Cool. I can see how the fact that we, that chasing fitness payoffs manipulates our umvelt. I get that it doesn't touch math. Cool. Now what I'm trying to understand is you said specifically math and logic. Now if you just said math, we'd be done and we'd move on. But you say math and logic. So then I immediately go to schizophrenics. Think they're being logical.
Donald Hoffman
Right.
Tom
What on earth makes us think that we are not just the way that our, our reality is essentially this total abstraction.
Donald Hoffman
Right.
Tom
In your own theory. And, and I would agree, man. I, I routinely think in myself and, and express other people. Your brain is creating a virtual reality Now I never, ever, ever conceived of it as sort of radically different as you. But I get that we're in sort of this, this huge abstraction, but what makes us think that that are the way that we reason, the way that we logic outside of math, the way that we reason and logic outside of math isn't the same as a schizophrenic. Where it is, it is so delusional. Is it just the internal consistency or. And couldn't that itself be a delusion that our fear of internal contradiction is problematic? Is, is actually a delusion in and of itself.
Donald Hoffman
Oh, very, very good. So I see your question now. So yes, there are deep issues here in terms of, of logic. And there is something called Godel's Incompleteness theorem that basically give it to me. You familiar with that?
Tom
No, not at all.
Donald Hoffman
So. So Godel is probably arguably one of the most profound results ever in human thought. Godel proved it's called Godel's Incompleteness Theorem. And he proved that any axiomatic logical or mathematical system that's rich enough to do arithmetic, there will be statements that are true but can't be proven within that system, unprovable truths. The notion of truth. Truth goes deeper than the notion of proof in that system. You might say, well okay, I'll take that truth and put it into my axioms. Then he said, well, but then there'll be new truths that you can't prove. And this goes on forever. What he showed was that the exploration of mathematical structure is in principle endless. It's unbounded. And in fact, if we get to it later on my theory of consciousness, I'll argue that that's what consciousness is up to. This unbounded exploration of all the possible possibilities of consciousness that comes from Godel's theorem. But you're absolutely right.
Tom
We can choose Godel's theorem one more time, right. As simply as you can.
Donald Hoffman
The simple bottom line is there is no end to the exploration of mathematical structure.
Tom
For us to be able to dive into that though, give people the like 3 minute primer on why your take is so different than everybody else who takes a really sort of physicality based approach to consciousness.
Donald Hoffman
Right. So most people who are looking at how consciousness is related to the physical world, like our brain, are assuming that somehow the physical world is primary, it's the source of all cause and effect in reality. And therefore consciousness is an effect of physical causes, presumably neural activity in the brain, for example, but maybe also computer activity in artificial intelligence that's sufficiently complicated. And my take is that our best science today, quantum field theory, Einstein's theory of gravity, and evolution by natural selection. All three of our best scientific theories today are telling us that space time is not fundamental and that physical objects in space time are not fundamental. Reality and science is good enough. Our best scientific theories are good enough to tell us where they stop. But they're not good enough, of course, to tell us what's beyond. That's up to us as creative scientists to try to guess what. What is a deeper theory of reality that goes beyond space and time. But when we project it back into space and time, that deeper reality needs to look like gravity and quantum field theory and evolution by natural selection. And so basically what I'm saying is that my brilliant colleagues, and these are my friends, they're brilliant. I'm not putting them down anyway. But when they assume that physical objects like neurons or computing systems cause consciousness, I think that they're running against what our best scientific theories are telling us. They're telling us space time is doomed. We need a deeper notion of reality. And for most science, it's no problem to work in space time. But when we're dealing with consciousness, the fiction that physical objects have genuine causal powers, which is a useful fiction most of the time, for everyday neuroscience, it's a very useful fiction. I use it myself. But for consciousness, that's when the fiction comes back to bite you. And just a quick idea. If you're playing a virtual reality game like Grand Theft Auto, but in a VR souped up multiplayer version, it's a perfectly useful fiction to think that the steering wheel causes the car to move around to the left or to the right and so forth, that's perfectly harmless fiction. But if you're a software engineer looking to upgrade the actual software that runs the program, that fiction, if you were stuck in that fiction, you would not be equipped to do the upgrade that you needed. In other words, if you're trying to look at the reality outside the headset, you can't just live with the notion of cause and effect inside the headset. You actually have to expand your horizons to understand a deeper notion of cause and effect. And that's what I claim we have to do with consciousness. It's useful most of the time to talk about physical objects as though they have causal powers. Perfectly harmless. But when it comes to consciousness, it's the single big obstacle that's stopping progress in its tracks.
Tom
Okay. Whew, you did it perfectly. Now I want to like, really drill into that quickly so that people understand where we're going to be going from here. So we spent almost two and a half hours together last time with me doing. With my p. Brain, doing my best to grapple with these topics. And finally I was able to understand a couple things that I'm going to lay out now that open up what this interview is going to be about. And it's going to be very interesting. We're going to cover things like AI and whether that becomes conscious faster than light travel, which I think your theory predicts all kinds of very interesting things. So to say that analogy in my own words, maybe from a lay perspective, so people get what you're saying about Grand Theft Auto because this is the key to understanding the consequences of your theory. So when you think about what's really going on as a computer programmer is trying to make this game called Grand Theft Auto. You're dealing with algorithms and mathematics and moving electrical currents around a Xbox or whatever. And so ultimately your goal is to move electrons and things like that around this machine that then creates this supposed experience right. On a TV that I can interact with.
Donald Hoffman
With.
Tom
Okay, when you understand that the difference between the reality that that programmer has to deal with, which looks nothing like, it's literally numbers. It's all code, all math. Math designed to move electrons around. Okay, that's not exactly right, but that's close enough. And that reality versus what you see, which is that, oh, look, I'm driving with the steering wheel. What you're saying is what you see when you walk around the room and interact with stuff and you think about space time and Einstein and relativity is as divorced from reality as the computer programmer is from the gamer who plays his game. Now, it took me a very long time to understand that's what you're trying to get people to understand. And then it was one thing that was like a real key in the lock for me was somebody asked you a question. It was. Was so great. And they said, donald, this is wonderful.
I love it.
But now predict we have the same conversation in 10,000 years. What are we going to know then? And you said, oh, that space time isn't real and so we can create space time ourselves. Because space time is. Space time is the Grand Theft Auto game. It's not the code underneath. So once you understand, Einstein just figured out a way to describe Grand Theft Auto. He did not figure out a way to describe the actual underlying code. And so if. If the space time is just Grand Theft Auto and we know we can manipulate Grand Theft Auto, we can now manipulate space time. And you Said, we'll have this conversation from Alpha Centauri. And I was like, oh, my God. Even if you just get excited about the chills, even if you just get excited about our ability to manipulate what's in the headset, right? Manipulating space, time traveling faster than light, changing fundamental physics, things like that, then even that would be interesting, let alone the other part that you're saying, which is, oh, you might actually be able to figure out what the underlying code is.
Donald Hoffman
That's. That's exactly right. And. And so one way to think about it as science, for all of its incredible breakthroughs and wonderful theories, has only been studying our headset. Science has not yet been studying objective reality outside of our space time. Virtual reality that from an evolutionary point of view, was just evolved as a way for us to play the game of life and stay alive long enough to reproduce, not to show us the truth. So we have really, in the, in the last four or five centuries, really gotten to be wizards of our headset, wizards of the Grand Theft Auto game. But just because you're a wizard of Grand Theft Auto does not mean that you know anything about the circuits and software that are running the game. And for someone who thinks that they know everything when all they know is Grand Theft Auto means they. They're still stuck in the headset. Science, I want to claim, has the tools. We have learned the right tools as we've studied our headset. We are ready to use those same tools to venture outside the headset, outside of space and time. So science has the right tools. We just have to open up, you know, open our minds to the fact that we're just playing a game inside space time. It's just a virtual reality. There is an entire world out now. The best physicists are already there, like Nima Arkani Hamed and many of his collaborators who are saying space time is doomed. We're looking for the reality that's beyond space time, in which, in fact, the very language of quantum theory and relativity theory don't hold. There are no, what they call Hilbert spaces. So there's no Hilbert space there. In what they're finding. There's no quantum theory. You need Hilbert space.
Tom
What is Hilbert space? I've never heard of that.
Donald Hoffman
Oh, it's. It's a mathematical structure that is used to describe quantum states and their evolution. It's called Hilbert space. And so. So what? They're looking.
Tom
You're saying they don't exist?
Donald Hoffman
That's right. So these physicists are saying, look, we're finding structures beyond space. Time and these structures don't care about Hilbert space and they don't care about locality and space and time. They don't care about so called unitarity, which is important for quantum theory. These are deeper structures. They project into space time and give us unitary and locality, unitarity and locality in space time. In other words, they project to the right things in our headset, but they're deeper and they have symmetries that you cannot see in space time. And when physicists find symmetries that are in the data but can't be captured in space time, they're very, very interested. So the best and brightest physicists are boldly stepping outside of the headset. But the question is, where do you look? What kinds of ideas? I mean, if it's not space and time, how do you come up with an idea about what reality is? So we just have to try ideas, make them rigorous, make them precise, and then project them back into our headset. That's the key. We know how to test things within the headset. So we're going to make theories of reality outside of space and time with mathematically precise models of how they map into space and time. And then we can test. So this is not just waving your hands and blowing steam, you know, this is not that. This is, we want rigorous theories outside of space time that have mathematical projections into space time that lead to testable predictions. I right now am studying the physics of scattering amplitudes for the like colliders. They smash protons together and they have gluons smash into each other and gluons go flying out and you can predict the probability of these particle events. So my goal is to start with a theory of consciousness outside of space time, Mathematically precise, show exactly how it maps into space time, and then hopefully predict those scattering amplitudes to 10 decimal predictions to, you know, 10 decimal places of accuracy. Then that doesn't mean I'm right. But at least now I'm doing the right thing to be testable.
Tom
Okay, I want to ground everybody in why I've gotten so hyped on this. So do you read science fiction much?
Donald Hoffman
A little bit. I'm mostly just very saddened by that right now.
Tom
Very saddened to hear that. But. So one of the things that I find so enjoyable about science fiction is they really play with big ideas. And I'm reading a book, it's the second book in the Three Body Problem series. And in that they're basically alien invasion is happening and they're trying to figure out like what they do. And the way that the aliens stop them from making progress. Spoiler alert for anybody who plans to read the series is that they, they send these photons that can basically interrupt the large colliders to give them impossible to interpret data. And so they shoot these things to the colliders and they, they just give them completely nonsensical data. And the reason they said that is they're like, if you can't make the. The breakthroughs at the fundamental layers of science, you can't progress. So for anybody who is feeling like, okay, these guys are drifting off into the ether, why is this interesting? How does it relate to my life? First of all, I'll say that you're a cognitive scientist. That is your background. So this starts with, hey, I'm trying to understand consciousness and the hard problem of consciousness. And it's leading me to this place that challenges the very substance of reality. And that matters because, by the way, we might be able to manipulate more of this stuff than we think, but we're still in this presentation layer, the headset. So now, as I begin to unwind some of this stuff, it's when we think about the fact that, okay, I'm guessing the scientists, the quantum scientists, the reason that they're now looking elsewhere is they cannot make what they see mathematically make sense with the models that they have of space time. And for them to make that leap outside is going to be brutally difficult in terms of. Science doesn't progress as the truth is revealed. Science progress. I think it was Niels Bohr, we talked about this last summer, Planck, that said basically, the old people have to die because they just, they can't update their model of the world. And then the next generation grows up just believing it to be true. So how are we holding out hope for people to step outside of that? And what methods are they using to discover this stuff? Is it going to be in the collider? Because that's all headset, right?
Donald Hoffman
So I can tell you what the physicists are up to a little bit on this and why I'm interested. As you said, I'm not interested in particle physics for its own sake. I'm a cognitive scientist. I'm looking for the simplest thing that I could possibly do to test a theory of consciousness in which consciousness is fundamental and space time is just a headset.
Tom
I think you have to explain that. We talked about it in the last
one, but it's where.
When you say that consciousness is fundamental, what do you mean?
Donald Hoffman
All right, so by consciousness, I just mean simple things like having a headache or smelling garlic or hearing the sound of a trumpet or having an itch or feeling a mood like love or hate or something like that. So simple feelings or sensory experiences that I could imagine even a mouse might have. Like a mouse might smell cheese and it might enjoy the taste of cheese.
Tom
I think everybody can get behind that, though. But aren't you taking it way farther down to, like, there is an entity, I don't know what word to put around it. Its entire experience is either green or nothing.
Donald Hoffman
Right. So I actually then, when I try to get a model of consciousness on its own terms, as a scientist, I have to write down a mathematical description of what I mean by having conscious experiences and acting on them, and entertain the possibility of very simple consciousnesses that might have, as you say, just one or two conscious experiences like green or nothing or something like that. And look at how these kinds of consciousnesses might interact and so forth. And so I've been with my colleagues Chetan Prakashan and Chris Fields and Manish Singh and others, working on mathematical models of dynamics of consciousness. It's much like a vast social network like the Twitterverse, where you have lots of social entities that are interacting, tweeting and following and so forth. It would be like that there's not just one consciousness, there are many, what I call conscious agents that are interacting. And there's so many of them that it's overwhelming. It's too much for any one conscious agent to talk to all or interact with all the other conscious agents. And just like with social data, if you're, you know, like with the Twitterverse, if you're trying to understand what's happening there, well, there's tens of millions of users, literally billions of tweets. You can't read the tweet, all the tweets, or interact with all the users. If you want to understand what's going on, you need a visualization tool, maybe a VR headset where you can sort of see what's trending in the United States and then zoom in to what's happening in Moscow, and then zoom out to what's happening over in China, and then zoom down to a particular street. And so what you want is simple graphics, simple eye candy, little objects that are colored that are doing different things, that somehow let you grasp what the billions of tweets and tens of millions of Twitter users are doing, but in a way that you can. So it's got to dumb things down, put it in a colorful, simple eye candy format that you can understand. And that's the big idea. I'm claiming that that's what space time is. Space time, the sun and the moon, physical objects. Everything that we see inside of space and time is just our visualization tool. The reality we're interacting with is nothing like the visualization tool. It's nothing like space and time. It's a whole network of interacting conscious agents outside of space time, a vast social network. And we've made the rookie mistake of assuming that our headset VR, our visualization tool, is the final reality. It's just a rookie mistake. It's so. It's like someone who's played Grand Theft Auto so long, they have no idea there's a reality besides Grand Theft Auto. We're like that right now.
Tom
All right, so let's go back to
this idea of conscious entities, which I find interesting.
It took me a while to grasp this one. So the easiest one is inside the human brain right now. There are two hemispheres. If you split the corpus callosum we discussed this last time, you get two different personalities. You've also got the microbiome, which conceivably all the microbes are conscious. And so when you inside of all of our cells, there's hundreds or even thousands of mitochondria, which are living organisms inside of our cells. So it is very objectively true to look at a human and say you're not a single entity. You are a collection of trillions of entities that we just sum up as, like, that's Donald, right? But in reality, it's not. And if I go in and mess with the consciousness, like, if I go in and mess with your microbiome, it'll change your personality, which is crazy. And I can't believe. Believe it's true, but nonetheless is true. I can affect your energy output by going and messing with your mitochondria, which, again, have their own DNA. It's not like it's just a cell in your body. This is its own entity that happens to be inside of your cells. And as we begin to recognize that humans already are just a collection of these much smaller consciousness, but they come together. I get where you're going now. I don't yet know the implications of that or what we're going to take away from it. But my relationship with you has been like, hey, this crazy statement that I can't fathom is true. I really want to write you off as a nut job, but I can't because, like, all of these things, like, do make sense. It's just, I so can't imagine what life outside the headset is. But the more time I have spent with the ideas, the more it's like, okay, these are starting to coalesce for me. So cool. I just wanted people to understand that consciousness thing. I certainly don't understand it yet, how it exists outside of the headset. But I can't deny that already, even in the headset, we're a collection of smaller consciousnesses.
Donald Hoffman
Right? Maybe one help for thinking about your consciousness outside the headset. I might have mentioned this last time. If you just look at your face in the mirror, if you look, what you see directly is just skin, hair, and eyes. That's all you see. And if you looked inside, if someone opened your skull up and looked, you just see neurons and so forth. But what you know firsthand that you cannot see is your hopes, your dreams, your aspirations, your love of music, your mood, all the rich world of your conscious experiences. Compared to that rich world, this face is unbelievably simple and nothing like the world of your experiences. If I smile or you smile, I can guess that you're feeling joy. But a smile in no way resembles joy. A smile is a twitch of a face.
Tom
A joy is that. That's where it's like the things we see happening in the quantum realm that make no sense. That to you is the thing that tells you, your brain, the physical tissues, all of that, they are entirely in the VR experience.
Donald Hoffman
It's all VR experience. That's right. And so, strictly speaking, no physical object, including my body, is conscious. Strictly speaking, my brain isn't conscious, because my brain, in fact, doesn't even exist unless I render it right. So if you're playing Grand Theft Auto, I'm playing a VR version of I got the steering wheel in front of me, I'm holding the steering wheel. If I look to the side, I no longer render the steering wheel. And there is no steering wheel. When I look here, I render it, and now there is a steering wheel. The same thing is true, strictly speaking, of neurons and brains. They're there when you render them. They're not there when you don't. It's a VR system that you render objects in space and time as you need them because they're part of your visualization tool. And then you garbage collect them, you delete them when you don't need them. So strictly speaking, no physical object is conscious. So human bodies aren't conscious, rocks aren't conscious. So I'm not a panpsychist. A panpsychist is someone who at least one version of panpsychism is that there are physical objects that obey the laws of physics, and the laws of physics are in some sense fundamental, but consciousness is the reality that's inside the laws. And I'm saying something entirely different. I'm saying that the very laws of physics themselves are really no restraint to consciousness whatsoever. They're just a visualization tool that certain consciousnesses happen to use, but they're no restriction on the nature of consciousness itself at all.
Tom
And so before we move off that. I know you don't know, but I want you. You must have flashes of images or favorite metaphors or whatever. So if I buy that this really is a simulation. True, true, true. Like as. As simulated as, it's going to get down to the fact that space time is just a part of that simulation, and it is consciousness that is fundamental. Meaning that the complexity of my brain will never give birth to consciousness. My consciousness, which is outside of the headset, it's outside the VR world. It's outside the Grand Theft Auto game. That thing has a need to, from a evolutionary standpoint, based on the gains it gets from certain behaviors. It needed to create a virtual reality so that it wasn't overwhelmed with data. So what is that conscious being that was evolving as it sits outside of this? Where can I scrape through and find the conscious thing?
Donald Hoffman
Right. So, well, the consciousness won't be inside space and time.
Tom
Understood.
Donald Hoffman
Instead, space and time is inside consciousness.
Tom
Yep, I get that. But I'm still imagining a little person outside of the VR. Is that just fundamentally broken?
Donald Hoffman
No, I would say that that's probably. Given the current state of human imagination, including mine, that's probably the best thing that we can do of, is to think about these entities. And of course, as soon as we think about entities, we place them in space and time. That's how we think. So it's harmless to do that as long as you realize that the space that you're thinking about isn't. They're not embedded in space. They're the creators of space. That's. That's the interesting thing.
Tom
So it's hard for outside space and time. I am eating things, right?
Donald Hoffman
Well, so. So inside space. And so I don't know what we're doing outside of space and time. That's part of what I want to understand is what are we actually doing? We don't see. I don't know what I'm actually doing. I know what I'm doing under a description like I'm moving my hand right now. And if I grab a steering wheel in my car, I know what I'm doing under a space time description, but I don't know in ultimate reality what I'm really doing. It's just like the VR player. When they turn the steering wheel in Grand Theft Auto, they know what they're doing. In the language of the game, I'm turning a steering wheel. But what they're really doing in terms of the supercomputer. Right. Which in that metaphor would be the deeper reality. They're toggling voltages and magnetic fields and circuits that they have no idea. There's. There's probably trillions of voltage togglings going on for one turn of the wheel. All they see is a turn of the wheel. That's their notion of cause and effect. It's trivial. The real cause and effect is trillions of voltages getting toggled in some, you know, in fractions of a second. It's much, much more complicated. So from an evolutionary. So when I, when we talk about evolution, as you did, you have to be from this point of view, it's very, very important to be careful because now we have to talk about what part of the framework we're talking about. So when I talk evolution, I'm only going to be talking about assuming the headset. I'm within the framework of the headset because evolutionary theory is only a headset theory. It's not a theory of consciousness beyond space and time.
Tom
I'm going to ask you a super random question because I'm so curious. You have a wedding ring, so I'm assuming you're married. I'm assuming you love someone. Somebody.
Donald Hoffman
Yes.
Tom
That's. That's such a headset experience.
Donald Hoffman
Yes.
Tom
How often are you sort of in sort of a. This really is real, and I'm gonna treat it as such. And like, does your wife think this is all, like, does she find this decidedly unromantic?
Donald Hoffman
Well, my wife is an artist. She's a very talented artist. Her name is Geralyn. I call her Jerry. And she's, you know, she's not a scientist and she appreciates my science. And even though she can't really understand any of the math, and I appreciate her art even though I'm a monkey looking at Mozart. Right. You know, because I'm no artist. And so we have mutual respect for the talents, the complementary talents, and we sort of complement each other and.
Tom
But do you just have to turn that part of your brain off that's
Donald Hoffman
like a lot of time with her? Well, with her, yes, because she's happy for me to talk with her about it up to a point. And. And then we need to do something different. Right. That's. That's healthy for me. But.
Yeah.
Tom
Do you want to. So one thing that I find interesting is people that believe that this is all a simulation and that if we just find the right equation, we could essentially exit the simulation.
Donald Hoffman
Right.
Tom
And that's always struck me as intuitively false that there would be no way to exit the simulation. Again, this goes back to what I was saying earlier about the person who takes their headset off. They would just be so fundamentally different the way that you would be processing data. There is no sort of core you. I think, like, I can't conceptualize it feels to me unless the. I mean, God. Is this what you're saying? In fact, let me ask, is this what you're saying, that there is this. This conscious agent that I. In the headset, I, in fact, know there is a conscious agent that is me. And the way that I present to you right now is me with the headset on. But the conscious agent could actually possibly remain intact when it pulls off the headset. You addressed this a little bit earlier, but it's now becoming more concrete in my mind.
Donald Hoffman
Right.
It seems to me quite possible that conscious agents will continue to exist even when they step out of this.
Tom
So I could take the headset, it off, effectively exit the simulation, be like, what the. Is all this around me, but still have a sense of emotion and attachment and love and
Donald Hoffman
vision, quite possibly in the sense that. That this isn't the only headset. So maybe by taking this headset would
Tom
essentially be my brain. Right. So I'd be. I'd be out of this headset and would I necessarily be putting on another.
Donald Hoffman
Well, the headset. Doesn't your brain remember? The. The headset isn't your brain. The brain is just one of the symbols in your headset for what I'm using for all the agents. For all the agents that are combining to form you. So I. I imagine that it's possible for an agent to go to a different headset.
Tom
Could they go to no headset, or would they be then existing in a realm of pure math?
Donald Hoffman
Well, and that. That I don't know. So in the. In the following sense.
So.
So it's.
It.
So the part that I think I'm confident about, but again, we'll see. Agents could get new headsets and a wide variety of them, so we could really explore and we could. And it could be Very headset we
Tom
can shorthand to umvelt.
Donald Hoffman
That's right.
Tom
Different way of perceiving and it may not bat headset. I can jump in a dolphin headset
Donald Hoffman
and maybe let go of space and time and do something different than space and time. My notion of self may migrate in the process. Right. Some of my emotions maybe, not others, who knows?
Tom
Do you think of taking the headset off as an event horizon beyond which we just cannot possibly even guess?
Donald Hoffman
The mathematics that I've got says that a conscious agent always can have awareness without experience. There is the awareness without experience. That's right. The math is very clear about that. So when I write down the set, the space of possible conscious experiences of conscious Asian, I have to write down what's called a probability space, which is a set of possible conscious experiences with.
Tom
I think of awareness as I exist.
Donald Hoffman
Yeah, no, it's not. But no, I, this just awareness without an eye. So this would, so, and by the way, I make no claim to be, you know, expert in any like, mystical spiritual traditions like Buddhism or Hinduism or so forth, but I've been told that they do have this notion. And so I'm not speaking as an expert, but I've heard that they have this notion of awareness without content and that certain meditators claim to be there. And, you know, I meditate a little bit and I might get little glimpses of a notion of awareness without self, without any particular content. That's the closest I can get conceptually to thinking about a conscious agent without a headset. It's a field of pure awareness, but it transcends in any emotion any notion of self, any specific conscious experience. Experience.
Tom
But you've never done psychedelics because the number one punchline that people say is, yo, you have no sense of self. Like, you dissolve into oneness.
Donald Hoffman
Right? Yeah, no, I, I, I, Why don't you. I haven't, I may at some point, you know, where it's legal and so forth.
Tom
So far.
Donald Hoffman
Well, what stopped me so far is. Well, I'll put this way. I've been reading the experiences of people, so I'm benefiting from their experiences. And I talk with, with people extensively, actually some very, very extensively about their experiences. And so I've studied them. There is a price to pay.
I.
Tom
But only in the headset.
Donald Hoffman
But only in the headset. Well, well. And while I'm in the headset, see, I really, I'm so eager to pursue these ideas. I think I've gotten. I know people say, well, if you don't do it for yourself. You can't know what it's really like to be without a self. Well, but I can look at the math and the math actually says, yeah, I'm seeing that in my math. So I've gotten the insight that I need there. Did you. But maybe I will do at some point, you know.
Tom
Did you listen to Sam Harris's podcast about taking 5 grams of psilocybin?
Donald Hoffman
I think I haven't heard him do that.
Tom
It's very interesting because I know you know Sam, you must listen to. To that podcast. It is fascinating. And he was talking about the fact that. And I haven't done it either, by the way. So. And I have a very clear. I am a physicalist. Is that what you call.
Donald Hoffman
Right. Physical.
Tom
So because of that I'm like, I am not with my brain, like I'm super paranoid. I do. Part of me wants to do it. So if you cut my corpus callosum, one hemisphere wants to do psychedelics really badly. The other side is like, get the out of here. We're not doing psychedelics. Psychedelics is a very bad idea. Absolutely. So I'm. I'm sort of stuck in this like go, no go scenario anyway.
Donald Hoffman
I agree. That's when I said there's a price to pay. That's exactly what I was. That's. That's the.
Tom
I am super paranoid. So Sam said in, in the experience he had a moment where he. He forgot that he did drugs. Like he had no sense of, oh, I have taken a drug to be in the state and I'm simply in the state and I'm going to be in the state forever. And this is what life is.
Donald Hoffman
Is.
Tom
And he was like, it, it is a type of hell where it is just. It's. Now I'm putting words in his mouth. The feeling I got, and he may have actually said this was that it was like a form of just never ending terror.
Donald Hoffman
Right.
Tom
And so you sort of pass through that and he was like, there are other moments where it's. It's never ending bliss. And you're like, it's going to be blissful forever.
And it was just, ooh.
The thought of where you have no concept I have done something to myself or I have taken a drug and this is a consequence of that. And it was. Will ultimately wear off that. It's interesting. It is interesting, man. And I get how that can really shape people's perception of what is real and like really shake you loose. Because as I read this stuff and for anybody that's Made it this far. First of all, congratulations. This shit is so deep and so heady. But the more time that I've spent with your ideas, the more I actually. I feel like someone is sort of filing off a kid callus on the bottom of my foot, and it's like, oh, whoa. We're getting to, like, a different sensation here. I never realized that I had a perception in. You know, when you've got the calluses that are, like, half an inch thick, it's like you just forget that that can actually feel something. So the more time I've spent with your ideas, the more I'm like, man, there really is something here. And so I. The first note I took on you was so arrogant and so aggressive, and I was like, like, dude, does this guy not realize that he invalidates his own theory by saying, like, this is about natural selection? Because I didn't understand your whole thing about math and reason.
Donald Hoffman
Right?
Tom
So.
But then you, like, you start spending more time with me. Like, I can't just discredit this. Then you start exploring it. Like, I definitely, before I started thinking about your stuff. Die hard physicalist. I will probably regression of the mean, Right. So I'm gonna slide back to that unless I really spend time. Time on this.
Because it is such a compelling illusion.
Donald Hoffman
Absolutely.
Tom
But it. It does. I can feel things pushing me at my back to experiment with psychedelics for reasons like this, where I. Yes, I could meditate for the next 40 years and maybe get to the point where I could have one of those experiences of what it means to be aware without a sense of. I literally can't imagine what that is right now. I can imagine blanking out. I can't imagine having a sense that there is awareness without me being inserted somehow into that.
Donald Hoffman
That's right.
Tom
So it would be really fascinating to very quickly be ejected out of, you know, my normal state of consciousness and into this. And part of. Part of what I promised myself I would do in this interview is to really figure out, like, why the fuck does this all matter?
Donald Hoffman
Like you.
Tom
And that's why I'm like.
I'm dancing on the same.
This idea of, like, you love your wife. Like, there. There is a realm in which you're like, the headset is pretty rad, and it's given me amazing. And I'm really having a hard time, like, actually stepping outside of the headset. And so so much of my life is predicated on the headset.
Donald Hoffman
It.
Tom
It may not even be possible to retain myself, which I value. I'm Assuming, right. So I value myself, I value this experience. I fucking value my wife. Right? That's the one that really, really scares me.
Donald Hoffman
Like, right?
Tom
I get so much out of my wife.
Like I've imagined my wife and I actually had this conversation one time, like, hey, a magic genie shows up. What do you wish for? And she was like, please don't wish for super intelligence. And I was like, why? Like that's such a rad option, right? To be the smartest human that ever existed. She said, if you did that, you'd no longer be in love with me. And I was like, you're actually right. Because if you're, you wouldn't love any human anymore. Like, I love my dogs, but not like I love my wife because there's such a gap in how we can relate to each other. So when I think about like stepping outside of the headset, man, you're giving up everything that you value. And that's like really, really trippy. Now I admit if you told me, hey, Tom, here's like a pill you can take and it's going to give you a little peek outside the headset, I'd be like, right, I'd have to do it. I'd have to take a peek. I'd have to see like what it was. Because I don't believe that drugs are necessarily punching through to some truth. I don't find myself like super compelled to do it. But the idea of looking beyond is both exhilarating and terrifying because of that loss of the things that I, I am so invested in.
Donald Hoffman
Yes, I, I'm on the same page with you. And it's, you know, interaction with my wife is one of the greatest pleasures, you know, makes life meaningful. And my, my daughter and I've got three grandkids and, and you know, my son in law and you know, my students.
Tom
Son in law made the list. Not bad.
Donald Hoffman
Yeah, yeah, Absolutely.
Yeah.
Yeah. Jay is a great guy. And, and in this case, I'm interacting with other conscious agents and I'm benefiting. So I'm now going into my theory. This is now, it seems impersonal, but I'm going back to the theory. Within the framework of the theory, there's a dynamics of conscious agents. As conscious agents interact that they learn, they get new comprehensions and they create new agents. So what we're doing in this headset, in terms like we're interacting, there's something new. We're both growing, we're both learning from this experience. We come away from it different. And that seems to Be part of it. If this kid in the candy store theory of consciousness is on the right track, then we're experiencing it right now. We're kids in candy stores, we're exploring and we're wondering what's on the next shelf of candy, right? That's what we're saying here when we say, well, I'd love to see if I take five or what happens, what's going on there. So that may be what it's really about, is that it's exploration. And maybe in meditation, one thing that does happen is that you get less and less. Grasping of things that you used to be grasping about. I find that I'm able to let go of things that fears, for example, but pictures of myself, it's really in some sense a dismantling. The best metaphor I can come up with is I read sometime that when a caterpillar goes through metamorphosis, right, goes into a cocoon, the immune cells of the caterpillar try to kill and they do kill the cells that are trying to begin the process of transformation into a butterfly. And for a while the immune cells of the caterpillar, it's a battle. But eventually the immune cells of the caterpillar get overwhelmed and. And then much of the caterpillar liquefies. Now that can't be fun, right? Liquefaction, meaning that all the structures that were everything that you knew as a caterpillar are turning into goo. And no wonder your immune system is fighting that tooth and nail until your immune system gets overwhelmed. But finally the immune system gets overwhelmed. Most of the structure of the caterpillar turns into to goo and then the transformation happens. That's what meditation feels like to me. Which means it's a double edged kind of thing. It's both extremely painful because everything that I know and have been connected to and addicted to is dissolving. But on the other hand, I'm realizing, wait, wait, well, that wasn't absolutely that essential. I thought it was essential, but it's not. And there's a new kind of structure that's being built that I have no idea. A caterpillar can't figure out what the butterfly is going to be, presumably. And so maybe that's what it's like to be starting to change headsets.
Maybe that's.
Maybe meditation is a way of letting go of some of the restrictive trappings of one headset and upgrading, right? And you're getting the 3.0. Now you just had the 2.0, now it's 3.0. I don't know. But these are the kinds of things I do want to explore within the mathematics. And that's why I'm sort of. It all fits. It doesn't mean it's right, but it all fits. The kid in the candy store, my own kid. Anticipation of seeing what's next. It does fit. Maybe that's why I like the kid in the candy store theory. That's just me and you know, for other people, that's not what it's about. But I think all of us do wonder about what's next and why are we here and what is it about. That's why I'm. I mean, that's one reason I do. This is it. Life is very, very short. I want to explore and things. Most of what I've believed very deeply has been very, deeply wrong. Most of what humanity has believed very deeply has been very deeply wrong. We have a very good. We're almost about 100% consistent in being deeply wrong. We've believed that spacetime is fundamental. Almost everybody believes that spacetime is fundamental. We all believe the Earth is flat. Now, a few very advanced physicists, Ed Witten has said spacetime is doomed. David Gross has said it's doomed. Nima Arkani Hamed is saying space time is doomed. And these guys, especially Nima, are now really being adventurous, very, very brave and saying, let's go outside of space and time into a world where we can't think. Just imagine what they're trying to do. We're trying to think entirely outside space and time. Like as you said as a kid, you're going, what could possibly be on the other side of space and time? These guys are saying, not only is there something on the other side, I need to think deeply about it. And here's a mathematical structure in which space and time, time, quantum mechanics and unitarity don't even appear in the language. And then I'll show you how our headset. They don't call it headset. That's now me ad living for them. But so how space and time, which I'm calling headset, how space and time and quantum mechanics and general relativity appear from these deeper structures in which there's no space and no time. So that fits perfectly with what, what I'm saying now. They have no idea what this deeper structure is about. And what I'm up to is I'm actually. Nima gave a class at Harvard last fall, more than 20 lectures on, for graduate students on these deeper structures outside of space and time. I am taking his Class on my own. There's all on YouTube. So I'm just studying it. I'm transcribing his lectures, studying them because I believe that I can show with my team, I'm mathematicians. So they'll show that the long term behavior of this dynamics of consciousness that we're working on, what we call the asymptotic behavior will give rise to the structures he's seeing like his amplitude and so the reason why and then the amplitude. He already shows how to build up space time from that. That way I'll be able to go all the way from conscious agents through the long term behavior, conscious agents to through the amplitude Hedron to space time. I can show you how the headset is built. That's my goal. So I'm. I'm really quite excited. Once I get, you know, I know enough to be worth his time. I may talk with him but. But I'm not going to waste his time until I know enough. So I'm trying to figure out how the headset is built. Once we, if we succeed, we'll be able to reverse engineer that headset and the technologies we'll, we'll be able to play with the parameters of space time. So it's like suppose you're a wizard at Grand Theft Auto and you can play within the game. Do all sorts of things that people find amazing. That's great. But imagine someone who actually knows the source code. They can take the wizard and they can give him a flat tire. They can take all the gas out of his tank. They can make the road infinitely long. They can do whatever they want to. They can play with the very parameters. So the wizard is nothing. All of our science right now has made us wizards. We're eventually going to get the source code of the game.
Podcast: Tom Bilyeu’s Impact Theory
Guest: Donald Hoffman, Cognitive Scientist and Author
Release Date: April 24th, 2023
Duration Analyzed: [01:00]–[235:19]
This episode delves into the provocative claim that everything we experience—space, time, causality, even the self—is not ultimate reality, but rather a virtual "headset" our minds use to interface with a truer, deeper reality. Cognitive scientist Donald Hoffman joins host Tom Bilyeu for a deep dive into radical ideas about perception, consciousness, the illusion of the physical world, and the scientific and philosophical evidence suggesting that what we call reality may function more like a simulation.
Hoffman compares our sensory perception to the experience of wearing a VR headset or using a computer desktop—what we see is a highly simplified, adaptive interface, not the underlying reality.
Using evolutionary game theory and simulations, Hoffman argues that evolution shaped us to perceive only what's useful for survival (fitness payoffs), not what's true.
[01:00–07:43]
"Everything inside spacetime is part of that fictional cause. Everything moves around on your desktop, but it's due to something else."
— Donald Hoffman [04:34]
[07:43–12:30]
"When you take the headset off... maybe it looks like what some people describe with psychedelic or near-death experiences."
— Donald Hoffman [10:44]
[22:13–27:37]
[33:33–38:58]
[58:24–62:08]
[29:44–33:33; 126:51–131:16]
"Every scientific theory will have some set of miracles... If you grant me this, then I will explain everything else."
— Donald Hoffman [29:44]
[194:20–200:46]
"We're eventually going to get the source code of the game."
— Donald Hoffman [230:34]
This episode gives a meticulously argued, mathematically grounded case for why everything we think of as reality—space, time, objects, even our sense of self—is a simplification, a user interface rather than direct access to the truth. If Hoffman’s ideas are right, the next revolutions in science will go past the VR headset, and for the first time, scientific tools may allow us to rigorously explore what has previously only been the subject of spirituality or speculative fiction.
For Listeners Who Want More: