Audio Player
Starting at:
Tim Maudlin on Quantum Realism, Bell's Theorem, Pilot Wave, and Interpretations of Quantum Mechanics
October 11, 2022
•
3:01:00
•
undefined
Audio:
Download MP3
ℹ️ Timestamps visible: Timestamps may be inaccurate if the MP3 has dynamically injected ads. Hide timestamps.
Transcript
Enhanced with Timestamps
416 sentences
27,497 words
Method: api-polled
Transcription time: 176m 57s
The Economist covers math, physics, philosophy, and AI in a manner that shows how different countries perceive developments and how they impact markets. They recently published a piece on China's new neutrino detector. They cover extending life via mitochondrial transplants, creating an entirely new field of medicine. But it's also not just science they analyze.
Culture, they analyze finance, economics, business, international affairs across every region. I'm particularly liking their new insider feature. It was just launched this month. It gives you, it gives me, a front row access to The Economist's internal editorial debates.
Where senior editors argue through the news with world leaders and policy makers in twice weekly long format shows. Basically an extremely high quality podcast. Whether it's scientific innovation or shifting global politics, The Economist provides comprehensive coverage beyond headlines. As a toe listener, you get a special discount. Head over to economist.com slash TOE to subscribe. That's economist.com slash TOE for your discount.
Tim Modelin is a professor at NYU, a visiting professor at Harvard and Carnegie Mellon, as well as an influential philosopher of science specializing in the metaphysical foundations of physics and logic. Today we discuss various interpretations of quantum mechanics, including Bohmian mechanics, which is also known as pilot wave theory. We talk about super determinism, as well as how quantum mechanics can be formulated without observers. Also reformulations of the foundations of physics without numbers,
and also how to place time as primal rather than a derived notion or rather than spatializing time, which is what's generally done. Next time with Tim, we'll also get to the liar's paradox and computation and consciousness. Write down your questions below because Tim will most likely be coming on again for a part two. Today's sponsor is Brilliant. If you're familiar with Toe, you're familiar with Brilliant, but for those who don't know, Brilliant is a place where you go to learn math, science, and engineering
Through these bite-sized interactive learning experiences, for example, and I keep saying this, I would like to do a podcast on information theory, particularly Chiara Marletal, which is David Deutsch's student, has a theory of everything that she puts forward called constructor theory, which is heavily contingent on information theory. So I took their course on random variable distributions and knowledge and uncertainty.
in order to learn a bit more about entropy. Now there's this formula for entropy, essentially hammered into you as an undergraduate, which seems to have fallen from the sky. However, when you take Brilliant's course, it was the first time that I could see that it's an extremely clear and intuitive formula. That is to say that
It would be unnatural to define it in any other manner. Visit brilliant.org slash toe, that is T-O-E, to get 20% off the annual subscription, and I recommend that you don't stop before four lessons. I think you'll be greatly surprised at the ease at which you can now comprehend subjects you previously had a difficult time grokking. At some point, I'll also go through the courses and give a recommendation in order. Thank you and enjoy.
Professor, I should let you know that I'm extremely, extremely excited to speak with you. You're one of the brightest minds and most clear minds, and those don't go hand in hand usually. Someone else that has those two qualities is John Bias, who's down to earth but extremely bright at the same time. You learn a lot reading his stuff, yeah. Usually when I interview someone, they have one or two distinctive ideas, whereas for you, I feel like I flip through your papers and there's such a
A multitude is like Douglas Hofstadter. I'm going to have to use this as an introductory and hopefully you'll come back on to speak about all of them. What is it about your work that's most exciting? Parenthetically, you can talk about what that work is. Well, because I'm in a philosophy department, I have this amazing freedom to pretty much look into whatever I feel like looking into, which is why if you look through my work, it covers
a variety of different fields. Most of my work has been in foundations of physics, although I have some other stuff in semantics, a paper on consciousness and various other philosophy of science topics. At the moment, what's most exciting is some work I've been doing for the last, gosh, five years.
which is going in a direction I never really anticipated, which is working on discrete geometry and ways to understand space and time as having a discrete rather than continuous structure. And just within the last six months or so, I got to the point of actually having a structure
and doing some calculations and suddenly finding relativity kind of falling out into my lap from foundations literature about as unrelativistic as you could possibly imagine. So that to me has been very exciting. Whether it will go anywhere at the end of the day or not, I don't know. But if nothing else, it's an illustration that relativistic structure may emerge
in a very natural way from a foundation that looks completely unrelativistic. So that to me has been quite exciting. So that's what I've been doing most recently. I mean, there's some other things trying to get numbers out of physics. I don't like having numbers in physics, so I've been trying to get rid of them and I made some progress on that. A variety of things. Okay, you're going to have to elaborate on that last point.
Well, okay, so I can say some things very simply. If you go back to ancient Greece, which is always a nice place to go back to, and ancient Greek mathematics was divided into two sub-disciplines, arithmetic and geometry, and arithmetic was the theory of numbers, and of course for the Greeks the numbers are, well, what we would normally say, what we would call the counting numbers, right, the positive integers, although in a way,
The numbers for the Greeks started with two, not with one, which is, as I like to point out, also the way we talk, right? Why? Because if someone says, do you have any pets? And I say, yes, I have a number of pets.
The number is one. That's like a joke, right? If I say I have a number of pets, you think I have at least two. If I say I have a number of pets and the number is zero, then it's really a bad joke, right? Then you say that just doesn't count as a number of anything. And we don't normally count one as a number of things.
So, you know, if you just follow that language, you'd say, yeah, when you're when you're talking about the numbers, you're talking about beginning with two. That's why one was very special for the Greeks. One. We all have a number of PhDs. Yeah, exactly. You know, so it's not so strange anyway. So there was number theory and then there was there was geometry, which was the theory of of continuous magnitudes of space, basically. I mean, and between the two of them,
The balance was very much on the side of geometry. I mean, you had all of Euclid, 13 books of Euclid, with amazing proofs. Okay. You know, proofs of things you would never imagine would even be true, much less provable. And Euclid, everybody knows Euclid, not that many people have heard of Nicomachus of Rhodes, who was sort of the top guy in arithmetic.
And that was the balance for a couple millennia was that if you look at math, it was the queen of math was geometry and number theory was kind of lower down. And if you look at Newton and Galileo and so on, all of their proofs are basically use the tools of geometry.
And what you notice is that if I tell you the physical world has a geometrical structure, that seems like a very natural thing to say. I mean, geometry was thought to be about the structure of space itself, and space itself is part of the physical world. If I tell you the world has a numerical structure, it's not quite
Immediately so clear what I could mean I mean people it's reported that the Pythagoreans who love numbers said things like a horse is the number 10 and you just go like what in the world are you talking about? How can a horse be the number 10? The connection between numbers I think it's a very good therapy
Whenever you're looking at mathematical physics and you've got algebraic structure, numerical structure, to ask yourself where it came from, because usually it's introduced by means of some convention, like by means of, say, putting down a coordinate system and assigning numbers as coordinates. And because that's not intrinsic to the subject,
There are an infinite number of ways to do that, and then you have to be very careful about what in your representation is a mere artifact of the conventions and what corresponds to something that's really there. This is the kind of thing Einstein says. I mean, the reason it took him 10 years to get from special relativity to general relativity, and he says this, I think, in the autobiography, in the Shope volume, is that he kept thinking that the
coordinates had to have some physical significance. And it took them a long time to understand that they don't. And in order to have a mathematical structure that's capable of handling curved spacetime, it has to be, there are no special coordinate systems, right? In a flat space, Cartesian coordinates are special, and in a flat spacetime, Lorentz coordinates are special.
And you tend to only use them and they're very convenient. In a curved space, in a variably curved space, there are no special coordinate systems. And you're really confronted with the fact that the coordinate values don't mean anything by themselves. And then you have to think hard about what does mean something, right? What are the quantities that really characterize the space itself? So, you know,
There's an interesting history here. It starts actually with Descartes among people in La Geométrie, where he introduces not what we would call Cartesian coordinates, but he's very clear because... I'm sorry to be going on so long, but you started... No, it's fine. Please. You know, it was known from the Middle Ages, and especially the Arabic scholars, that if you're doing number theory,
Then you can use algebraic methods to solve problems and that the algebraic methods were really powerful. And so what Descartes asked himself is, huh, I wonder if there's a way that I can somehow use algebra to solve geometry problems. And what he was interested in was doing that. He's very clear about what he's doing. And in fact,
One of the things he realizes is that in numbers, when you're dealing with numbers, one is a special number, right? It's the only number that squares to itself besides zero. It's the only number that when you multiply it by anything else, you get the same thing back. So one is just picked out.
from the algebraic structure as really special. But when you're doing geometry, there's no unit that's special. There's no length that's special. So Descartes realized that in order to convert geometrical problems into algebraic ones, he had to arbitrarily choose a unit, which is really important. Is this overcome with affine spaces rather than just a vector space?
Well, look, he's dealing in a continuum. I mean, his interest in geometry are in a continuum, and no length is special in a continuum. I mean, you've got all these things. The other thing that is worth noting is that if you're dealing with numbers, part of what makes something a number field, why we even call a set of objects numbers at all, is that you can multiply them. And when I multiply two numbers, I get another number.
but there's no there's no corresponding geometrical operation if i give you two magnitudes if i give you two line segments say there's no operation corresponding to multiplication that will hand me back another line segment the best you can do in a euclidean space you can make an area right you can make a rectangle
But the rectangle is a different kind of thing. I mean the units in which you characterize it are different than the lengths. So there's no natural object that corresponds to multiplication in geometry.
But we use multiplication all the time in doing physics because we've converted everything into numerical representations, right? Just to be clear, let me just see if I'm breaking it down correctly, if you don't mind. So when you say, hey, if we have two lengths or two line segments, it's not obvious how to multiply them.
right so someone may say well can you add them you can add the lengths okay you can add the lengths yeah you can wait and then they may think well no i can multiply well you think you have some way of multiplying let me tell you this first line segment was actually the number one so your way of multiplying actually doesn't change it so you think hey i did some complex
and then I can just say no this first line was number two so that's why you say that there needs to be a unit picked out in order to multiply when you're thinking about line segments well it's not it's not just which comes first I mean certainly multiplication should be an operation that commutes it shouldn't matter what order I give them to you the real problem is the number one is the multiplicative unit so when I multiply any other number by one I get the same number back again
and there's no geometrical length that's picked out as special. That's why Descartes says in order to convert
a geometrical problem into an algebraic one, I have to make an arbitrary choice of unit. I'm going to say, oh, this length counts as one. If I do that, then I can define something that you would call multiplication, right? Now, suppose I have this special length that I've just picked out of the air. Oh, that's length one. Now I give you two other lengths, a and b, and then you make a rectangle out of a and b.
which has an area. And now you say, well, if I want another rectangle with the same area, but one of the sides is the unit, this arbitrary unit, how long does the other side have to be? And then that will give me a length. But it's only relative to the choice of that unit. If I choose a different unit, I'll get a different one. So how are you going to create all of this
richness of standard multiplication and addition using geometry? You don't try to replicate. The thing about multiplication is that again it's a function that takes two numbers as input and gives you a number back as output. So I can just iterate and multiply and re-multiply and cube and so on. The proper operation in geometry if I give you two lengths
Is it, I can construct something out of it, but what I construct out of it is not a length, it's a surface. And if I give you a surface and another length, I can construct a volume. And if I only have three dimensions, I'm done, right? Because if you now give me another length, there's nothing I can do. If it's a four dimensional space, okay, I could make a hyper volume. And if it's five dimensional space and so on. But a lot of what happens in physics,
You don't go up above the dimensionality of the space. And so basically what I've been playing with, which is, I think, I don't know if it's an original idea, but it's certainly not one that's been much investigated, is something I call full discrete geometry. And the full means you have zero dimensional things points. You have one dimensional atomic things. You can think of the lines. You have two dimensional atomic things.
Atomic surfaces, you have three dimensional atomic things, atomic volumes. And if it's a three dimensional space, then you've got those things to play with. And so you can't just keep, whereas if you have numbers, again, I can just take a number and multiply it by self and then again and then again and then again and then again as many times as I want and I'll still have a number. I never, you know, the multiplication operation
can be is a binary operation but it extended to any to multiplying any any number of numbers right any finite number there's no way that when you multiply outside of the dimension so it should jump out but instead it goes back to the beginning no no it's just not it's just not a defined operation because the operation again the the the operation is one that for example can take two one-dimensional elements and give you back a two-dimensional element
And it can take a one dimensional element and a two dimensional element and give you back a three dimensional element. But if you give it something else, it can't give you anything back because there are no four dimensional elements or five dimensional, right? So it just doesn't have, you know, it's not that it recycles back or anything. It's just not a defined operation. Now, when you say that there are no four dimensional nor five dimensional, what you mean is in the theory that you're constructing, what you would like would to be in this manifest image of three dimensional space, you want to get back to that rather than say,
something that's a bit more outrageous or unproven? The formal theory can handle four dimensional spaces, five dimensional spaces, a million dimensional spaces, 11 dimensional spaces if you want to do string theory. It's not that the formal apparatus is limited, but you make a decision, okay, I'm going to look at two dimensional
Think Verizon, the best 5G network is expensive? Think again. Bring in your AT&T or T-Mobile bill to a Verizon store today.
Jokes aside, Verizon has the most ways to save on phones and plans where everyone in the family can choose their own plan and save. So bring in your bill to your local Miami Verizon store today and we'll give you a better deal.
Let me just lay out some thoughts. So firstly, you make a reference to a number there by saying three-dimensional or four-dimensional. And this is strange because you're trying to build a world without numbers, but then we need a number in order to specify this world. So how does that work? So that works because the Greek numbers, the counting numbers, are perfectly fine if you've got something to count, right? It's not that I hate all numbers. I can say there are two chairs in the room because I'm just counting one, two.
But what happened historically, of course, if you want to convert geometry into arithmetic, you need more than the counting numbers. And if you look back at the history of it, it's very interesting. The Greeks really only recognized officially the positive integers as numbers. Then you can push a little and try and convince people there are these things that you call rational numbers. And that had a little bit of resistance to it, but not too bad.
Then, well, there are two things you have to do. You eventually have to go to the irrational numbers if you want to do Euclidean geometry. That had a lot of resistance to it, right? There are people who said, no, those don't make any sense because if I try and express them in decimal form, it goes on forever. It gets lost in infinity.
and and then you also need the negative numbers which you know there were people who were very upset about that and you can see why right i mean you say look okay i understand what you mean when you say you have two sheep i understand what you mean even when you say you have zero sheep but what would it be to have negative two sheep right right um there there can't be less than zero so negative numbers and it's just in the record in the historical record there was a lot of resistance to negative numbers
so i don't have any problem with integers if you have something to count and and the counting measure in this discrete geometry you use counting measure because you have a finite number of points or a finite number of lines or a finite number of surfaces in a certain situation and you can just count them and the counting is important so it's not that i hate all numbers
I just, when you use numbers, you should ask yourself, okay, where are the numbers coming from? If they're coming from counting something discrete, okay, that's a good answer. If they're not, then you need a fancier story about why you're using them at all. Let's get quickly philosophical, if you don't mind. And then I want to get to your views on quantum mechanics. There's different theories of truth. So there's correspondence, coherence, and pragmatism are the main three.
For the correspondence, this just made me think, well, in the correspondence theory of truth, could there be minus two because minus two wouldn't correspond to something of the world? But then it seems strange because many people who say, I believe in the correspondence theory of truth are perfectly fine using numbers, real numbers. I never thought about that. What does the correspondence theory of truth have to say about real numbers or even complex numbers? Okay.
I think the difficulty you're having is that you're touching on an extremely, extremely difficult subject. And one about which I don't have much to say because it's too deep for me. And that's the ontology of mathematics. It's mathematical reality. So if I'm in some intuitive sense of correspondence theorist, I say, look, at least for certain sentences,
What makes that sentence either true or false is whether it correctly or incorrectly describes some non-linguistic thing, the subject matter. And for physics, the subject matter is the physical world, so we kind of know at least what the target is. But mathematical items, as Plato pointed out, are not physical items. So if you want to be basically a correspondence theorist, as I do,
You're naturally going to be a Platonist in mathematics, which I am, right? That is, I think they're mathematical truths. And if you ask what makes them true, the answer is, well, mathematical reality. And if you ask me what mathematical reality is, I'm going to say, well, it isn't physical reality for sure. It has its own kind of existence, very much the way Plato would say. And there are facts about it. And some of the facts
We can discover, and some of them will never discover, as Gödel assures us. Okay, that's fine. There's a world that has more facts about it than we're able to. It's kind of amazing we can discover anything about it. I shouldn't be very surprised I can't discover everything about it. So what you're really asking, from my point of view, is not about the physical world at all. It's about the nature of purely mathematical reality.
And again, I very firmly believe there is such a thing, and it's independent of physical reality, right? I mean, physics could have come out all kinds of different ways, but however physics came out, it wouldn't change the mathematical facts at all. So physics is dependent on the mathematical reality, but not vice versa? Well, physics is not dependent on mathematical reality in a way. See, mathematical physics is a discipline that uses mathematical items
to attempt to describe the physical world. And the mathematical items exist. They are what they are, and they are what they are independently of what we think. And we can find out about them by, you know, in a math department, you can investigate the structure of the mathematical items. The deep question, which is always there, is why should these mathematical items serve as good representations of physical reality?
Now, this is where I would say, again, for many numbers, I'm puzzled. Why am I using numbers anyway? For certain geometrical things, again, purely abstract mathematical geometrical things, it seems to me perfectly plausible to say, well, they're good because the physical world literally has a geometrical structure. You know, Newton thought that space was a three-dimensional Euclidean. How do you three-dimensional Euclidean space structure?
And it could have had. It turns out not to have, but it could have had. And if it did have, then you'd perfectly well understand why Euclidean geometry, which remains a perfectly good mathematical investigation into Euclidean space, you would understand why it has so much utility for physics. Three dimensional Euclidean geometry, right? I mean, 10 dimensional Euclidean geometry exists just as well abstractly
When I think about Platonism, I think of it as being extremely mystically religious, and I don't scorn it
For being so, and I think Plato thought of the forms of something mystical as well, something related to God. I can ask your thoughts on that. But what I wanted to say was more about it's strange because you were saying that the physical reality doesn't depend on the mathematical. However, the fact that we can use it, it gives some visual analogy. It's as if we're these physical beings and we can reach into the closet of mathematics and pull out something and then use it as a tool here.
It doesn't seem like the mathematics reaches into us and then uses it there, although perhaps it does and we don't know. So it seems like that's a world that we pull from. Is there an interaction between these two worlds? There's a huge turn in Plato, the turn between the early and the middle dialogues, which is indicated in Mino. And in Mino is where you have a kind of standard Socratic
dialogue cross-examining Nino and showing that he doesn't really know what virtue is, what Arete is. And then he gets all messed up and says, I have nothing more to say. And Socrates says, well, why don't we keep talking?
And Meno quite reasonably says, wait a minute, you tell me you don't know what virtue is. And I now don't know what virtue is. Why should we keep discovering? We're two ignoramuses. Why should we keep discussing with each other? And Socrates says, well, you know, and then he kind of indicates the theory of reminiscence. But the important point is he then brings out this mathematical question. If I have a square. Say two units on a side, so it's four square, its area is four square units.
I want to build a double square that is one whose area is twice as big, eight units, along what line do I build it? So that's a perfectly good question and he brings this slave over who's not learned any geometry and the slave starts out
saying oh you double the side and then he points out they do some counting and say no that's not right that gives me 16 and he says well let's try three well that doesn't work that gives me nine and then the slave says i don't know right and then socrates draws in you know the diagonals and and again they just count and easily show that the square on the diagonal has twice the area now i don't think the thing in the world mystical about is the opposite of mystical right mystical suggests some kind of
inarticulable presentation that you believe without reason, right? And this is just the opposite. The reason you believe that the square on the diagonal has twice the area is you've got a proof, right? That's what Euclid gives you. He's not a mystic, right? Euclid's not a mystic. He's a guy who can prove things.
The thing that interested Plato is, look, I can actually literally know things about geometrical figures, know them, have proofs of them, have demonstrations, know what the truth is. And he says in Timaeus, which is the closest he comes to giving us the physics, Timaeus does
He says, look, don't expect that kind of knowledge of the physical world. The best you can expect of the physical world is what he calls an echos logos, which is a plausible story, right? A likely account of plausible story. And I'll give you a plausible story, but I can't prove that it's right. So Plato just recognized correctly that we have much better epistemic access, more secure epistemic access,
to mathematics than we do to physics. And that's not because he's a mystic. It's the opposite. He's kind of a hyper rationalist. Okay, so two thoughts. And I don't know if this is apocryphal or not. But Pythagoras, at least reputedly said that there was something ill about the square root of two. No, he didn't. I'm sorry. Okay, so at least infamously, infamously, or reputedly. This is a this is a fake story. interested in the truth here. There's a wonderful book called The Mathematics of Plato's Academy.
When we say there was an issue about the square root of two,
We mean by the square root of 2 a number. Let's go back again. 1.414... The Greeks did not recognize the existence of any such number. They didn't believe there was any number which if you multiplied it by itself you got 2. Why? Because they thought the numbers were the counting numbers and obviously you can't multiply any counting number by itself and get 2.
Now what they did recognize by a very interesting geometrical proof is that if you have a square and you have its side and you have its diagonal, that those are incommensurable. That is, that the length of one cannot be expressed as a ratio in terms of the length of the other.
Now that doesn't pick out the diagonal as the bad guy or the side as the square bad guy. Incommenciability of that sort is a relation between two lengths. It doesn't say, oh, the side is good and the diagonal is bad or anything like that. The ratio of lengths between the side and the diagonal does not correspond to the ratio between any two integers. That's what they knew. But it didn't seem to bother them.
In fact, they thought it was quite interesting. There was a whole terminology about links that are allogon, which would be like these that can't be expressed in terms of ratios, and they had further subdivisions among them, and nobody was upset about it. It's just an interesting fact. What led you down this route of trying to make geometry primal, like underive elements of reality?
Just mostly the thought that I could understand, I could see how a geometrical representation could in principle more or less directly correspond to the structure of the world, that the physical world could literally have a geometrical structure. And I don't understand how the physical world could literally have an algebraic structure.
So I like to get along as much as I can with geometry. This word geometry is interesting. When I was in university, I always thought of geometry as I forgot the exact name, but I took some course where it's almost like Euclidean geometry or it is Euclidean geometry and there was inversions. I had to do some problem with inversions. This is all I remember. What is that called? Analytic geometry. Is there some word for it?
Like planar geometry, something like that. I mean, I'm not exactly sure. Well, anyway, it's with what you construct with lines and circles. And then there's differential geometry. And to me, I see that as extremely algebraic. Firstly, it's also contingent on the real numbers. Differential geometry is basically what you use when the geometrical structure
changes from place to place. So in Euclidean geometry, you have all these nice symmetries, right? The structure of Euclidean space is the same everywhere, so there's kind of translational symmetries, and it's the same in all directions, so they're rotational symmetries, right? It's actually the same even under expansions and contractions, which is really unusual. That's not true in even a space of uniform curvature.
So it has all these nice symmetries but you could be living in a world where the geometry just doesn't have these nice symmetries and then the geometrical structure changes from place to place and then you have to use differential geometry which just means that you characterize the geometry local in some little region and then in the next one and then in the next one with all these little overlapping maps that cover the different regions and you don't have a constraint that the geometry looks the same
in all the regions. Now, none of that suggests using numbers, and you don't need to use numbers. I mean, numbers tend to get in again with you when you introduce coordinates, but they're coordinate-free characterizations of all of this geometry. So, as I said, if the space is curved, then no coordinate systems will be particularly great
The nice thing about spaces of constant curvature and especially zero curvature is that there are these nice coordinate systems like Cartesian coordinates.
So forgive me, because it's been a while since I've been in university. The way that I recall it is that you define a smooth structure by saying that so-and-so is infinitely differentiable. To define differentiability, you have to say that it goes from the reals to the reals. I'm sure you have some chart in between, it's like Rn, but somehow you have a curve, and that's R to the M, so R to the manifold, and then from the manifold down to the chart. So you require the reals here, then the reals on the chart, and then you say it's differentiable based on
I don't know if that's correct. Please forgive me if I'm incorrect. That's the normal way it's presented. I mean, what you said is perfectly accurate is the way it's taught, that you throw in functions from the space to the reals or the R to the N as part of the story. Yeah. And then you talk about whether a function from Rn to Rn is continuous and so on.
By appealing to the structure of the real one that is the usual way it's presented it doesn't have to be done that way and It should be obvious that there's something quite artificial about it because if i just have a space in front of me Where's this function into the real numbers coming from well it only comes by somehow laying down coordinates which could be done in infinitely many different ways and choosing some
arbitrarily and what you would prefer from as a pure as a calculational matter what they're doing is fine because it's easier to calculate with numbers as a conceptual matter you're better off avoiding this use of numbers where you can because they're only introduced in a kind of artificial way
So yeah, I mean, your memory of that is perfectly correct, but people teaching math are not. I mean, especially unless you really go deep into the foundations of mathematics, they're teaching you a tool. And when you learn calculus and vector calculus and so on, they're giving you a tool for doing calculations and grinding out answers.
that's a very effective tool. What happens if you're a philosopher like me is that it's not so interesting learning how to use the tool. I mean, when I was doing my PhD in history and philosophy of science, because I was specializing in physics, I had to do graduate level work in physics. And I took one course, and I remember a perfectly fine physics course,
Interesting, for my purposes, interesting things were said on day one, the first day, because it was on the first day that the professor was trying to set up, what are we doing? And then the rest of the course, which was like, how do you use Green's functions to solve differential equations? Okay, it didn't matter to me. I mean, it just it made no difference for my interests.
Now, if I needed to get numbers, if I needed to actually grind numbers out, I really needed to understand all the rest of that course. But that's not what I do, right? I'm more interested in, well, why am I using numbers anyway? You know, not how exactly do I use them or how do I solve these equations, right? Only rarely, once or twice in my life have I had to actually solve an equation to make progress.
Can you give me an instance of them, of the one or two times where you had to use an equation? Oh yeah, I mean, I remember this very well. My very first book, which is Quantum Nonlocality and Relativity, there's a chapter which is about, part of it is about trying to mimic
Non locality violations bells inequality through a certain trick which involves having particles or objects that you're observing that just don't show up disappear that never show up in the data okay and if you're very clever. You can use this. Gap in the data. So the data that does actually get recorded will look like it violates balance inequality even though something nothing non local is happening.
And then there was a question that arose, which is, well, how efficient can you be? That is, you want to say, I can rule out this kind of trick if my detectors have a certain efficiency. If my detectors are 100% efficient, if every time I throw a particle at a detector, it registers, then I can't use this trick because the trick really depends upon there being these missing data.
But then you say, okay, well, how exactly how much efficiency do I need before I can rule out this loophole? And to answer that, I had to actually solve a differential equation, which I offhand, I didn't quite know how to solve, I could write it down. And I was actually, I was in Croatia off the peach, I had a big yellow legal pad, and I had a little hand calculator.
and I swear to God I sat there and and numerically simulated an integral by hand with this calculator until I got it close enough to say hey that looks like pi over four and then once I saw that I said oh yeah it is pi over four then I could work out right how it goes. How long ago was that? Oh that was
30 years ago. Okay, so this is before Wolfram Alpha. Oh, yeah. Oh, yeah. This is this is before you were born. Okay. Oh, yeah. Okay. So just to be clear, efficiency, because I haven't heard that term efficiency Houston. I'm not an experimentalist. Yeah, the worst course in physics for me was the experimental part. I'm much more of a mathematical person. Well, let me see. Let me give an example. When we see the double slit experiment, you see these dots, dots, dots, dots, dots, dots, dots. Well, sometimes
You throw out a photon and it just doesn't register. Right. A dot doesn't click. Yeah. And then you think, okay, well, well, if it's not going to click, it's not going to click in some uniform manner. So the pattern that I eventually get this oscillating pattern, it is characteristic of the photon. But you're saying there are some ways of designing an experiment such that the missing data gives you a pattern that's actually not there.
There's a way, if you're an experimentalist, there's just a straightforward question, how efficient are my detectors? I have a beam and I know how many particles on average are coming through this beam and then I know how many particles on average are being detected by my detectors. And there's going to be a gap, right? I mean, stuff gets lost. Nothing is 100% efficient. In my book, I describe a certain trick
It's a very conspiratorial trick. It's not anything that would be physically natural. I mean, the whole thing is kind of crazy, but you're just doing logically possible. There's a trick you can use if there's high enough inefficiency in the detectors so that the registered data would violate the Bell inequality
without anything non-local and the trick involves a very judicious and special use of cases where data disappears right where no data shows up so it never gets into the recorded data and you can do that i mean it's not conceptually it's not really hard to see how it works at all it's just
There's a price to pay. You're simulating, by use of this trick, sending information where really no information is being sent. But you're simulating it by use of this trick. But the more information you have to simulate sending, the more inefficient the thing has to become. You have to more and more often
Use this trick of saying, well, one of the particles is just going to refuse to answer, right? I mean, normally you say I take a particle and I send it through a spin device and it's either going to come out up or down. But what if the particle has a third option, which is to just go mute and not give you any answer at all and then get that run gets deleted from the data. And so there's just a question of what those numbers look like. What is the most
efficient way you can run that trick, because if your detectors then have higher efficiency than that, then you can say, well, they can't be doing it, right? It's a loophole. There were these kind of loopholes that people talked about in tests of Bell's inequality, and the detector inefficiency loophole was a big one for a long time. But generation after generation of experimentalists just did it better and better and better and got higher and higher efficiencies, and all those loopholes have now been closed.
It's like the electrons are out to trick us. The physical story
It is not a natural one that anyone would ever come up with. You only come up with it if you're sitting here saying, I hate non-locality, I hate non-locality. I want some way to keep my physics local, yet consistent with the data that we have. And this is a trick, which as long as the data which you have, there were certain levels of detector inefficiencies,
It's a trick you could use. But there's never a reason to believe it's completely implausible, physically. There's no physical story behind it. It's just kind of looking, well, you know, this is possible. This is mathematically possible if I have enough inefficiency to play with. OK, now let's talk about quantum mechanics. What I noticed about you initially was I see you as bringing some sober reality to this whole realm of quantum confusion and mysticism.
that generally sells in magazines, spookiness, strangeness, other abnormal qualities. And then Bell's inequality is generally sold as a trade-off between non-locality and statistical independence or realism. So firstly, is that the case? And can you tell the audience your view of what the heck is going on quantum mechanically? Okay, so yeah, let me
Let me start a little backwards here. This term realism has snuck into this discussion, and it really should just be expunged. Realism has nothing to do with anything here. In fact, from a philosophical point of view, scientific realism has to do with epistemology. It doesn't have to do with ontology, with what there exists. I mean, to say, gosh,
i can save i can save locality by just giving up realism the way people say it i often say oh yeah nothing actually exists but thank god it's local i mean it doesn't make any sense if you try and trace down people say oh bel made some assumption of realism he made no such assumption
There's no assumption made that you can just deny and then you're okay. The actual theorem of Bell, which I think is the most important and astonishing result, I mean there's the mathematical theorem and then the fact in the lab that these inequalities are violated, this is the most astonishing thing in the history of physics. And the theorem as the piece of mathematics, this is just a comment about it as mathematics,
Every theorem you're deriving something from some suppositions. There are two suppositions. There's one that's called Bell Locality. It's the way that he expresses mathematically what it would be for a theory to be local, relativistically local, meaning that there's no causation or effects that go faster than light. That all of the events
that can count as a cause of something happening lie in or on its past light cone, right? Nothing that would require going faster than light. And in addition, it will be relevant for where this goes. Nothing coming from the future either. I mean, Bell's locality condition also rules out reverse causation that the future could could somehow cause things in the present. So there's the locality condition
And then as a mathematical fact there's a second condition that he uses in the derivation which is called statistical independence. And from these two assumptions he derives an inequality and says okay if any theory satisfies locality and satisfies statistical independence then there are certain constraints on the sorts of correlations you could possibly see
in situations where Alice and Bob are in their labs doing experiments so far away from each other that light doesn't have time to get from one lab to the other. So there couldn't be any effects, if all effects had to go at most at the speed of light, there couldn't be any effects from what happens in Alice's lab on what happens in Bob's or vice versa. So that's the theorem.
And then you go in and you see this inequality. I mean, first of all, you notice that quantum mechanics predicts violations of this inequality. And then you notice that you go in the lab and check and quantum mechanics, quantum mechanical predictions are correct. So you only had two suppositions from which this follows locality and statistical independence. So as a purely logical matter, you say, well, I have to at least deny one or the other. Now,
There are people who, for reasons I do not fathom, I just literally don't understand their reasons, are so deeply committed to locality that they're looking for any escape route. I mean, that's more or less what we were talking about with these loopholes, with these detection loopholes, right? They don't like the idea of non-locality and they're looking for any escape they can have.
from accepting it. And these detection loopholes for many years were one until the experimentalists just closed those loopholes. They're gone. So the only other loopholes, they look at this and they say, well, I don't want to die locality. I guess I have to deny statistical independence. And the problem with denying statistical independence is that it's kind of crazy and it's crazy and conspiratorial. And on top of it, as I've said and other people have said, it undermines
All scientific method. I mean, if in order to get out of accepting non locality, you deny what's required assumptions that are required to do science. That's not a good deal. Right? That's a really bad deal. Speaking of that,
I sent you a video. I'm unsure if you had a chance to Yeah, yeah, I did. Okay, so as a preface for those who are watching the video is by Sabine Haassenfelder, and it's on super determinism. Either way, I'll give a brief description in the first 20 seconds or so. She says, some people think that if we do away with statistical independence, then all the science is undermined. And she says her attitude is no,
So, I'm curious to know your thoughts on that. Firstly, on super determinism and perhaps what her specific comments were. I mean, let me walk through this, although let me before, because I knew you were going to ask this, let me just point out to people, because it's much better than listening to me. If you're interested in this, be sure you take this book, John Bell's Speakable and Unspeakable in Quantum Mechanics, and very carefully read
Chapter 12, which is called Free Variables and Local Causality, because Bell nails it completely. And any questions you have, Bell is an extraordinarily precise writer.
He's not so good, by the way, if you watch Sabina, she talks about some interview he gave, and he's not so good in interviews. He sometimes slips up, I would say, when he's talking off the top of his head, and says things which I think are just a bit sloppy and uncharacteristic, but when he writes, there are only a couple places in his entire book where I think he put a foot wrong in what he writes.
That's everyone. If you're interested in this topic.
Discussions with many people who want to talk about super determinism who've never read as far, you know, really have never read this particular thing, which is the thing that Bell wrote about. It's a free variables and local causality. And what I'm going to tell you is basically what Bell says. It has and let's just start off by saying this has nothing to do with free will. I mean, all this talk about free will has gotten into this discussion from the beginning. It was irrelevant.
The term super determinism is bad because even as Sabina says you can't get more deterministic than deterministic. The issue is the mathematical assumption of statistical independence that's used in the theorem. Now what is that assumption? Let me just give you an example and why it's important and why I say if you deny it
empirical science kind of goes out the window. So suppose we're going to do a test and there are different experimental conditions. You might think of them as experimental in control or in our case, you're going to either set up a spin measuring device in the X direction or the Y direction or in some other direction. We have these different conditions.
And I'm doing tests and I'm doing tests on a large number of items, a large number of rats or a large number of particles or whatever. And I'm looking at statistics. I'm looking at percentages of the ones I experiment on with these different conditions, whether they react this way or that way. And when I'm doing experiments on pairs on the correlations between the two sides. Okay. Now,
What i do if i want to do just to take the simple case everybody understands if i want to check whether smoking causes cancer in rats. I started with a big group of rats. And then i want to subdivide it into an experimental group in a control group. But i want to subdivide it in a way that there's no bias between the two groups that the two groups.
before i run the experiment are statistically similar to each other okay now there's some ways i can check for that like if there are male rats and female rats i can just kind of count okay what percentage of male rats are in this group and what percentage are in this group and i can make sure by distributing them that they have the same percentages sure the thing is i want the groups to be similar in all respects even respects i'm completely unaware of i mean maybe there's some genes i don't know about
How do I do that? Well, I randomly assign the rats to the two groups. This word random is important here? The word random here is absolutely important. What the word random there means, it is the condition that Bell uses of statistical independence. It means that the sorting mechanism
Does not is not biased or does not respond or is not sensitive to the rats. So if I you can imagine flipping a coin, for example, I mean, that's a physical kind of what Bell would call a physical randomizer.
There are other randomizers. I mean, he uses this example, which is very nice because it takes all issues about free will out of it. He says, look, you could use a pseudo random number generator, which works like this. Start at the millionth digit of pi. And then output a one or a zero, depending upon whether the next digit is even or odd. All right.
So that'll give you a sequence of ones and zeros that is a pseudo random sequence. We call it random because it will pass every statistical test for randomness you can come up with. It'll be about the same number of ones and zeros, but there will be no observable pattern. And you could use that and say, okay, if a one comes up, then the rat goes this way. If a zero comes up, it goes that way.
Now, why do I do that? Because if the sorting is independent, if whether the rat went this way or that way is independent of the nature of the rats, then with overwhelmingly high probability that's calculable, each subgroup will be statistically like the original group. If 60% of the rats had some unknown feature in the original group, then 60% in each of the subgroups.
will have that feature. And therefore the subgroups will be like one another. The really important thing you want is you want the statistically the subgroups to be like one another. And that's why we use randomized procedures for making these groups.
Now, if somebody says, now, suppose I do this experiment with the rats, I take this big group, I randomly divide them by one of these sorting mechanisms, right, by throwing dice or by this... Sure, sure. ...or whatever into these two groups, and then I expose one to smoke and I treat the other the same, except I don't expose it to smoke, and then I see a lot more cancer in this group, then we infer that the smoke caused cancer.
Now if somebody says no no no really what's going on is this some of the rats were predisposed to get cancer no matter what. And other the rats weren't and it just turned out that the ones that were predisposed got sorted over into the ones into the smoke group and the ones that weren't got sorted over into the other group.
And so the smoking had nothing to do with it. Now, this is the kind of argument that actual lawyers for the tobacco industry used to make when they were trying to argue that scientists had not that smoking causes cancer. And obviously, as a purely logical fact, yes, if if the sorting didn't work to make statistically similar subgroups, then that could be an explanation. But then you have to go on and say, I don't care how you sort sort any way you want.
right? Sort based on on pi, sort based on the number of shares of stock sold on the shot stock exchange, sort based on anything you want, no matter how you do it, somehow, the cancer prone rats are going to go this way and the non cancer rats are going to go that way. Now, that will, as it were, as a purely logical matter, count as a as something that would make the prediction that this group will get more cancer than that group. But we would just say this is lunacy.
Because you have no mechanism. You just don't like the result. You don't like the result, and you don't like what the result is telling you, so you're just grasping at straws. That's all that's going on here. That's what super determines. Somebody who just says, I'm just going to blankly deny statistical independence, no matter how you do this sorting, and in this case, you're doing these experiments
Alice is doing an experiment and Bob's doing an experiment. Each one is randomly choosing how to set their device. And it's there where the randomness comes in. How does the device get set? And they could be flipping coins or they could be running off pie or doing whatever the heck they want. And Bell's only assumption is that statistically, the group of particles that meet condition one
will be the same as the group of particles that meet condition two, where condition one and condition two are chosen by one of these pseudo random kind of processes. Now, Bell gives an even more extensive discussion of why this is true, but it's clear that when we say it would undermine science, experimental method depends upon, in all cases, using this. Now, I should say one more thing.
There's another idea, which is retrocausation, which is a different idea. It's not just a blank denial of statistical independence. It says, oh, statistical independence fails because the nature of the particles when I made them is somehow affected by
At the time I made them, it's somehow affected by what experimental condition they would be exposed to in the future. So that's the future affecting the past. That's retrocausation. And all of this is to save locality? Well, that doesn't save locality. So the first point I want to make is that retrocausation violates Bell's locality condition.
It would be kind of crazy to think look i have bob and alice and they're doing their experiments far apart and i absolutely will not believe that what alice does can have a causal effect on bob but i believe that stuff that happens to the future can have a causal effect on bob i mean that's that's even worse.
And it's even worse because if you retro causation, then because you have causation from the, you certainly have causation from the past to the future. If you also have causation from the future to the past, then you're going to be involved in causal loops. And then it's very hard to even write down a coherent theory. I mean, Wheeler and Feynman tried to do it. It's not clear. They wrote down some equations. It's not clear if they have any solutions. If you just have non-locality,
The obvious ways to implement it will not involve any causal loops, you won't have that problem. But the fact is, to say I want to deny non-locality and therefore embrace retrocausation, well first of all, retrocausation violates Bell's locality condition, so you haven't gotten out of it anyway, and you've gotten yourself even to a worse problem.
Forgive me because I'm uncertain here and I'll just fumble over my words and then you can perhaps cohere what I'm saying and say it better than I have.
How far does this correlation have to be? So when I hear statistical independence, the opposite of statistical dependence, which I assume is a correlation. And it seems to me clear that even when we sort rats, if we try to do so randomly by picking the one millionth digit of pi and so on, going forward from there, that there is still some some dependence, it's extremely, extremely small. Now, how? Well, how do we choose the number one million? Well, okay, this person said, Okay, well, let's choose a random number generator to choose that one million or to choose some number of the digit of pi.
Okay, well then when we press on the random number generator, why do we choose to press it five seconds from now versus 10 seconds from now because it would give you a different number. So essentially what I'm saying is that the world that we chose the rats from every single thing here is influencing our decision.
So every single thing is in the sort of rats temperature is influencing our decision, the fact that the rats have a certain amount of hair is influences our decision and so on. So just just there's a there's a extremely tiny, extremely tiny amount 0.000000002. But I don't know, is that the fact that it's non zero? Does that have any implication? Or does it need to be non zero sufficiently large?
Okay, so let me try and address this. There are two parts of this, I think. Let me try and address both parts of it. Let me just take my simple case so everybody can follow. I have a big group of rats, and some percentage of them, say, have a certain gene. I don't even know about the gene, but you say 60% in the big group have it.
And what i want to do is now partition this into two groups of rats and i want the two groups to be statistically like each other and therefore statistically like the original group which had sixty percent now if i just flip a coin. It just standard statistics.
will tell you, all right, there'll be little fluctuation. It doesn't mean that you'll get exactly 60% in each of your subgroups, right? One subgroup might end up with 61% and the other subgroup with 59%. Those are just statistical fluctuations. Using standard statistical methods, you calculate the likelihood of these fluctuations. If you're worried about them,
Then you use a bigger group. The probability of the fluctuations is smaller and smaller without limit. You can't drive them to zero, but you can drive them below any epsilon.
Just to be clear, what I'm wondering is, is the presence of any epsilon, does that undermine the... No, no, because the violations of Bell's inequality says a certain observed correlation in the lab cannot be greater than a given number. I see, I see. And the actual observed violation of that is so far, it's not close, right? It's not like
Gosh, it's just a tiny bit over. And if I just have a little epsilonic thing somewhere, no, no, no, the violation is radically far away from the limit. So if you say, oh, maybe the subgroups aren't exactly the same, of course, they're not exactly the same. But the chances of them being sufficiently different
to account for what's observed and furthermore of course what's observed is exactly what's predicted by quantum mechanics and if if what you were seeing was just the result of some random fluctuation then you wouldn't be able to predict it i mean that's the point about random fluctuations is you can't predict them at all but what you observe in the lab is precisely what you get out of a quantum mechanical calculation yeah there's one other thing i just want to if i can find it quickly from um
This thing of bells because it's relevant to what you said. Well, we could do this. We could do this. He says Actually, here's a relevant a relevant thing to what you just said He says He's talking about you've got some Things you treat as free variables in this case the choice of
of whether to do experiment a or experiment be the choice of allison bob nick he says of course there's an infinite infamous ambiguity here. About just what and where the free elements are.
The fields of Stern-Gerlach magnets could be treated as external. That's then treating those as free variables. So I can just imagine those. I set them when I do the calculation however I want. Or fields and magnets could be included in the quantum mechanical system with external agents acting only on external knobs and switches. Or the external agents could be located in the brain of the experimenter.
In the latter case, the setting of the instrument is not itself a free variable. It's only more or less closely correlated with one, depending on how accurately the experimenter affects his intention. As he puts out his hand to the knob, his hand may shake and may shake in a way influenced by the variables V. Remember, now this is now exactly to your point. This is Bell I'm reading, right? Remember, however, that the disagreement between locality and quantum mechanics is large, up to a factor of root two in certain sense.
So some hand-trembling can be tolerated without much change in the conclusion. Quantification of this would require careful epsilonics, okay? And, I mean, there's another thing that he talks about, which I just... Alright, again, let me just jump to the end. He's talking about, this is the part that I wanted.
Consider the extreme case. He's now talking about random generators, so we want something that's setting this equipment either this way or that way. For Alice, something that's setting it this way or that way. For Bob, it would be a bad idea to have it due to the, as it were, rim of the experimenter because the experimenter would have to be just sitting there
in a very boring way going oh now up now down and probably as we know people are not even good at creating random sequences right human beings are not good random number generators right right here's what he said he says consider the extreme case of a random generator which is in fact perfectly deterministic in nature and for simplicity perfectly isolated in such a device the complete final state right what it chooses
perfectly determines the complete initial state. Nothing is forgotten. And yet, for many purposes, such a device is precisely a forgetting machine. A particular output is the result of combining so many factors.
of such a lengthy and complicated dynamical chain and this is going what you were saying well what if it they do it a little later if i have you know suppose i have like when they do lotteries they have these ping pong balls that are bouncing up and down in this cage okay that could be a deterministic system so that some set of causes account for why this ball came up but that set of causes is huge it involves all the other collisions with all the other balls and all this other stuff going on so he says
Yet for many purposes such a device is precisely a forgetting machine. A particular output is the result of combining so many factors of such lengthy and complicated dynamical chain that it is quite extraordinarily sensitive to minute variations of any one of many initial conditions. It is the familiar paradox of classical statistical mechanics
that such exquisite sensitivity to initial conditions is practically equivalent to complete forgetfulness of them. To illustrate this point, suppose that the choice between two possible outputs corresponding to A and A' depended on the oddness or evenness of the digit in the millionth decimal place of some input variable. Then fixing A or A' indeed fixes something about the input.
i.e. whether the millionth digit is even or odd, but this peculiar piece of information is unlikely to be the vital piece for any other distinctively different purpose, i.e. it is otherwise rather useless. With a physical shuffling machine, we're unable to perform the analysis to the point of saying just what particular feature of the input is remembered in the output, but we can quite reasonably assume that it is not relevant for other purposes.
In this sense, the output of such a device is indeed a sufficiently free variable for the purposes at hand. For this purpose, the assumption one, which is the statistical independence assumption in the proof, is then true enough and the theorem follows. Now, if you just read that paragraph carefully and understand it, it puts this whole thing about, I'm going to deny statistical independence to bed.
I mean, this is just Bell. Okay, he understood this objection. He understood that logically, yes, he has two premises. So logically, you can deny either one. But then he says, look, denying statistical independence, that's just not physically the kind of thing a physicist could ever do and remain a physicist. Here's what's interesting. I respect you. I respect Sabin. I think you're both extremely bright people.
Why is it then that extremely bright, respectable people believe in super determinism if it's so obviously incorrect? So here's one answer. Nima Arkani Hamed said that there's this mistake that the public has about physics, which is that almost every moment in time, we need some radically new paradigm that we have to be revolutionary. He said, no, what is required is conservative revolutionaryism or revolutionary conservatism, whichever. And what he meant by that is
Well, let's take a look at Einstein. People say that with special relativity, what he did was extremely revolutionary. He said, no, he was revolutionary about one aspect and was extremely conservative about others. He said, these are certain aspects that I want to hold on to and here's what I'm willing to give up. And then he came up with this theory. Do you see it as some people are extremely conservative about locality and they're willing to be extremely revolutionary in other respects, like let's dismiss with
Statistical independence. Do you see it as that sort of issue or is it something else? I think it's something else. Look, you have to remember physicists, philosophers, everybody are human beings and then human beings have their foibles. I'm going to illustrate this.
One of the physicists who early on recognized and advertised Bell's result, because you have to bear in mind that if not for some kind of very lucky things happening, Bell's result might have gone entirely buried and nobody would have even known about it. I mean, it was published in the first volume of a brand new journal that went out of business within a couple years.
So people wouldn't even know about the journal. It happened to fall in the hands of Abner Shimony, and Shimony read it and understood how important it was, and Shimony happened to know John Clauser, who was looking for an experimental project. Okay, there's a whole history here, but it's a very iffy thing, right? That Bell's result could have just completely disappeared. But one of the physicists who early on
understood the importance of Bell's result was David Merman, and Merman was a good popularizer and he wrote some articles. He also wrote articles that were published even in philosophy journals explaining what Bell did. And the original Bell result, the one that Bell proves,
is a statistical result it says if you there's a certain kind of experiment if you do it over and over and over and over again you then accumulate statistics about correlations between outcomes on these two sides and this bell inequality puts mathematical constraints on how strong those correlations can be and that's what's violated and merman said he always had a suspicion
that the probabilistic nature of these predictions was really essential. The fact that you were dealing not with strict 100% predictions but only percentage predictions was somehow very deep to this result. He said he could never quite articulate why.
And then some years later, Greenberger, Horn, and Zeilinger came up with this very nice example. It's the one I use all the time now instead of Bell's original example, which instead of involving pairs of particles involves triples of particles. This is in my book if anybody wants to read it. It's a beautiful thing to be explained in five minutes.
But the nice thing about it is that the predictions there are not statistical. They're 100% predictions. They say, according to quantum mechanics, you should get this kind of result every time. So if it even fails once, quantum mechanics has failed, unless they're experimental error. And when Merman saw that,
he appreciated he wrote a very nice article about it and he said at the end i used to think that the probabilities here were really important i was wrong okay i think you know this just shows i was wrong because here you have a locality result that doesn't involve probabilities they're all one in zero and you know you can't ask for more than that from somebody okay he had his suspicions he stated them publicly when he got
Reason to see that those were wrong. He retracted them publicly and he wasn't upset about it. I mean, why would you be upset about it? I mean, I often say the nicest feeling you can have when you fall asleep at night is that you understand something you didn't understand in the morning or you've corrected an error in your thinking. I mean, people don't like to admit errors to different degrees. Right. And that's just human beings. Right. I mean, that's true of all human beings. You have to
Try to get familiar enough with at least some of the content that you can really make an independent judgment of your own about who's right and who's wrong. And then start more trusting the people who are right and being more leery about the people who are wrong. There's nothing else you can do. It's unfortunate. It's hard work, right? And anybody trying to pick up
physics from the kind of public presentation is not going to be in a position very well to do that. As I say, if you're interested in super determinism, get Bell's book, read chapter, it's only four pages, read chapter 12, read it carefully and think, think for yourself.
And ask yourself whether you think denying the statistical independence assumption that he uses is something that has any physical plausibility. And if you're wondering, that's the best thing to do. I'm not asking you to take it on my word.
Sure, sure, sure. So let's talk about the PVR theorem. Do you see it as one of the landmark theorems on par with Bell or second or even perhaps greater than Bell's? I would say I think it's in a certain way as important as Bell's result, but not as surprising. Not as surprising given Bell that we already know about? No, no, no, not as surprising even before Bell.
Let me say why. The PBR result says we use this thing, the wave function, this mathematical object at the center of the mathematical apparatus of quantum mechanics. Everybody uses it to calculate.
And again, one question you should ask yourself is, why is that? Why are we using this thing? One possible answer is, well, because it's a pretty good representation of some actual physical reality in an individual system. It represents some physical characteristic of the individual system in front of me and a relevant physical characteristic.
Now, even apart from Bell, if you ask, well, should I take this wave function seriously in that sense, or should I rather think
it's something merely epistemic that so the wave function is kind of for a single particle let's just talk about a single particle because it's easier although it's a little misleading single particle the wave function is defined over all of space so you can sort of think like a field and it's spread out and and the schrodinger equation tells you how it evolves and it evolves by a wave equation good okay is there anything in it if i send a single particle through a two slit experiment
I represent that single particle in part using a wave function and the wave function is spread out and the wave equation carries that wave function through both slits and from there each slit produces its own, as it were, little wave and then they interfere like water waves and you get this interference pattern.
Physically what we see if I do this two-sit experiment over and over again with individual single particles is what shows up on the screen is this interference pattern with these bands through it, light and dark bands. Now as soon as you see that, you say, gosh, okay, well that wave function must really represent something physical that does interact with both slits. That's how interference works.
Right? If each individual particle merely went through one slit and nothing interacted with the other slit, then you couldn't possibly get this kind of interference. So the idea that you should take the wave function seriously as representing a real physical feature of individual systems
So nothing to do with anybody's knowledge, nothing to do with epistemology, nothing to do with statistical characteristics of large groups of systems. It represents some physical reality that pertains to an individual system. That's just from two-slit interference. Everybody should believe that. And everybody I know does believe it. This was not something the people that I know that work in foundations, nobody ever questioned that.
Because you just say just the two, now two slit interference does not violate Bell's inequality. So this is not non locality. This is a different thing. And this is again another place where, you know, Feynman makes a mistake when he says, well, he wrote this before Bell, but he says all the mysteries of quantum mechanics are somehow packaged into two slit. No, they're not because Bell is
But if you just think about those kinds of familiar interference experiments, done with single particles and where you accumulate the data over time, anybody looking at that is going to say, gosh, that wave function must represent a real physical characteristic of individual systems. Now, what the PBR theorem does is it gives you something even sharper.
It says if despite all that, you want to think that there's something epistemic about the wave function, that it doesn't really represent physical reality of the individual system, it somehow reflects your information or something like that, then the PBR... This is also called cubism or no? Well, that's sort of what the cubists say. Cubism is a very difficult doctrine to even understand what it's claiming. Let's leave it aside for the moment.
But someone, you know, a common idea was that no, the wave function is kind of spread out, not because anything physical was spread out, but because we're ignorant about where something is, right? And the spread outness reflects us not knowing where it is, not that there is something in these different regions.
So I think anybody who just thought about two slit would say, no, that can't be right. It can't be merely epistemic. And again, David Merman makes this point in one article, he made the point very nice with a dialogue, a kind of dialogue with somebody who's trying to be epistemic. And they say, oh, you know, the wave function spreading out is just a matter of our ignorance. And then in response, somebody says, so it's our ignorance that goes through the two slits.
Right. And that makes no sense, right? I mean, something physical is interacting with both slits. It has to be. So what PBR did was they took the observation that I made, which is not a proof, and they gave a sharp definition of what it is to regard the wave function as representing something real about the individual system as opposed to
representing something about merely our knowledge of the system. And then they said, well, if you believe it's epistemic, by our definition, here's an experiment that you'll make the wrong predictions about. I mean, you'll make predictions, you'll have to make predictions that violate quantum mechanical predictions.
And by the way, they also use statistical independence assumption in their proof, but that's fine because the statistical independence function is fine. They use one as well. And it's a nice proof. I mean, and people who reacted at the time who said this is the most important proof since Bell, I think David Wallace said that. I think he's right. I think if you're doing foundations,
it just i mean you could say it was the last nail in the coffin and the coffin was already shut that you shouldn't have been an epistemicist in the first place for you know straightforward physical reasons right how do you explain these interference bands um but if somebody again wanted to be very recalcitrant and and and stubborn and say no no no i just really think there's something epistemic about
This wave function, the PVR theorem kind of puts that to rest and so you can move on. So yeah, I think it's a very important. I think it's an extremely important there. I think as I say, it's not as astonishing in what it's claiming as Bell.
There is a word, a phrase that you use called foundations of quantum mechanics or foundations, but that's the standard for foundations of quantum mechanics. Infamously, when you go to a physics course, if you get a physics degree and you ask about, well, what does this so and so mean, especially when it comes to quantum field theory, quantum mechanics, they say, hey, go to a philosophy department. Essentially, what I'm asking is what textbooks should one read? Let's say they're at the graduate level in order to become familiar with quantum foundations.
Is there a set of five books that you feel like, hey, read these in this order that gives you a great introduction? Well, I mean, the thing I was once did a thing at the end, they ask you name five books in your field that you would recommend.
And I said, and I was only half joking, I would say, I said, okay, get this book and read it five times. I mean, you know, this is a collection of articles. There are different levels, but it's not a single presentation. But Bell is really on target. Really, really beautifully on target. You have to be careful when you read it because he's so precise in the way he writes that he'll say something once and he won't beat you over the head with it a hundred times and you have to pay attention.
I beat people over the head. My technique, as I say, is to use a two by four and whack you over the head a hundred times and say the same thing. Bell's using a scalpel and sort of just like that. It's like Dirac. Yeah. I mean, it's just, I mean, it's, it's beautiful. And he's really, again, if you're worried about super determinism, just read this four page paper of Bell's. It's all there. So you should read that. If you're, if you're used to a physics, if you're a physics major and you're used to a physics presentation,
There's this nice book by Travis Norsen that came out, I don't know, five years ago. I think it's called Foundations of Quantum Mechanics, but it's really written for physics students as an introduction to quantum theory, but with attention to the historical background and conceptual detail that's not in a normal physics book.
And he goes over interpretations as well? Yeah. Then, you know, David Albert, who people may not know, got his degree in theoretical physics from Rockefeller University, but was interested in foundations and therefore switched to a philosophy department. So he's in the philosophy department at Columbia. You know, he wrote a book, which is more philosophical and a little more out there in certain ways called quantum mechanics and experience.
And it's not like Norsen's book, which is aimed at physics students. David's trying to take as much math out of it as possible and kind of look at the conceptual issues. I mean, you have my book, which is Philosophy of Physics, Quantum Theory, which is more math than David has and less math than Travis has.
Alright, it's in between the two. And every one of the books you mentioned, for those who are watching, that'll be in the description. I think, you know, between those four books, you certainly have everything you need. And the book by you, what was it called? Philosophy of Physics, Colon, Quantum Theory.
There's a little set of books. It's a series of books that are supposed to be introductory books for philosophers to different issues. And I did the philosophy of physics one, but there are two volumes. One is space and time, which is an introduction to space time theory and relativity. And the other is quantum theory, which is about quantum theory. And someday I'm going to do a third part or put them all together and do something on statistical explanation and thermodynamics and entropy and stuff. But that's somewhere in the future.
Hi, I'm here to pick up my son Milo. There's no Milo here. Who picked up my son from his school? Streaming only on Peacock. I'm gonna need the name of everyone that could have a connection. You don't understand. It was just the five of us. So this was all planned? What are you gonna do? I will do whatever it takes to get my son back. I honestly didn't see this coming. These nice people killing each other. All Her Fault, a new series streaming now only on Peacock.
What the heck is a be-able? A be-able. Be-able. Yeah, so what the heck is a be-able? And why did Bell invent this terminology? Because to me, as an outsider, it just sounds like something that exists, some ontological object. That's all it is. But that's, you see, the word ontology, I can testify from personal experience, it scares physicists, right? You say, what's the ontology of their theory? And they're like,
Ontology, because they didn't learn Greek, right? And ontology just means in Greek, ta onta are the things that exist, and logos is an account. So the ontology of your theory is, tell me what your theory postulates to exist. And
Bell wanted a word for the same thing. Now, he could have used ontology, but he didn't. I don't know why. I mean, he doesn't have a philosophy background. Maybe it didn't occur to him. And he even says in a paper, he says in the theory of local beables, he said, well, yeah, he wanted a word whose root was to be or to exist as opposed to observable. I mean, what he was doing was setting a contrast with the term observable.
Because so many physicists would say, oh, physics is about observables, observables, observables. And Bell said, no, no, no, it should be about be-ables, things that exist. I mean, he said he considered be-er, B-E hyphen E-R, but then that looks like beer, right? And then on the model of observable, he put be-able,
I would just call it the ontology of the theory, but philosophers would just call it the ontology of the theory and physicists would just get used to it. But you can call it the beables, it's fine because Bell used that. It's just what is your physical theory postulate to physically exist? Now this is a question which should be the first question, it's not a philosophical question.
It's not, oh, philosophers are interested in that. No, physicists, right? That's what your physics is actually postulating to physically exist, right? It's the first question you should ask of any physical theory if you want to understand it. What does it postulate to actually exist? And unfortunately, because foundations of physics is not part of a standard physics education, which it isn't,
This is get very nervous around questions like this and sometimes they they they suggest answers that don't kind of make any sense so there's a physicist who i will remain unnamed who was asked this question at one of our conferences. What's that what you know what are the vehicles of your theory what is your theory possibly and came up with the answer matrices.
No, that's not even in the right ballpark. A matrix is a mathematical object. A matrix is a set of numbers in an array. You're not going to tell me the physical world is made up of sets of numbers in an array.
The world just has a mathematical reality and
The physical world has no reality that's any different than any other mathematical object. So every mathematical object is physically, I mean, again, there's a certain point at which quite honestly, you just say, I don't really want to spend my time with this.
It's just, you know, no, there's a physical world, right? It's true that we can make really good predictions about it using some mathematical representations and that's a good thing to think about and why those mathematical representations are so effective. That's a really good thing to think. But what you're thinking about when you think about that is, okay, what might the physical world be like such that this mathematics makes such good predictions?
And that's why I said that's why I like geometry because one in the case of geometry. At least a possible answer is this abstract geometry is such a good representation of the physical world because the world literally has that geometrical structure, right?
then of course it's going to be pretty good at making predictions because you've got it right you know you've got a good representation of the way it really is and in the case of geometry i can understand that as a possibility in the case of you know pure number theory i don't understand what it would be to say look the world really is is has this numerical structure i don't know i just don't understand it i can't get i can't wrap my head around what would even mean
Going back to that geometrical structure that you were referencing in the beginning, is that something that you're currently working on or is there a paper that you can point people to to learn more about? I wrote a book called New Foundations for Physical Geometry, The Theory of Linear Structures. That came out a while ago and that was a way of
trying to cover the same content or conceptual territory that's covered by standard topology, point set topology, but in a different way, using a different foundation. And there was always supposed to be a second volume to that book, so the first volume was just pure math and the second volume was supposed to be application to physics. And I intended to write that and I thought it was going to be very straightforward.
and i was writing away at it and then i was teaching and then i got interested because one of the things that happens in the first book is i develop a way of thinking about topology which applies both for continual and for discrete spaces because the standard point set topology doesn't really work well in a discrete setting it just gives you kind of trivial answers anyway i got interested in this idea of discrete geometry and then i started developing this thing i call full discrete geometry and so this will eventually come out
as the second volume of New Foundations of Physical Geometry. I can't say exactly when, I mean I've got six chapters that are all about spatial structure and a few chapters now about adding time in for space-time and I have some really exciting recent results which I was talking about which are not published but if you're interested I've given the talk
Lots of times, in fact, I just gave the talk for the International Space Science Institute and they've put it up on the web within the last couple of years. So yeah, I've got more work to do to finish that book. I kind of see how it's going to go. There's some calculations and things and some examples that I need to think through.
Eventually that will come out. There's a phrase that I heard you say with Einstein. The general view is that he spatialized time, but instead what we should do is temporalize space. Right. Okay, so please expand on that because that sounds poetic. I don't know what it means. Okay, right. So one thing that people say that is just flat incorrect is that in special relativity somehow
Time is just another spatial dimension. Or that they're placed on equal footing. They're placed on, quote, as you say, equal footing. That's just a phrase also that you never hear anywhere else except in special relativity. So then that makes me think that people are just copying that phrase because you never use the word equal footing in math or physics except here when talking about space and time. Everyone repeats it like the powerhouse of the cells, mitochondria. Good, good, good. You've put your finger on something which is really smart and really perceptive.
And you should use it all the time. When a whole bunch of different people are using identically the same language, it's an indication of a problem. Because if you really understand something, when you speak about it, you won't always use the same phrases and different people will use slightly different phrases or produce different analogies and so on, because they understand it. And then they're filtering their understanding through language to try and
Convey that understanding. When everybody's repeating verbatim the same phrase, they've just memorized that phrase. Right. It's an indication of lack of really understand. I mean, think about it. I agree wholeheartedly about that. Think about the term equal footing. If by equal footing you mean equal, then why don't they say as they say, why don't they say, oh, he made everything into time?
But they say, no, somehow he made everything into space. Right. Now we know what it would be. We know what it is to add an extra spatial dimension. So we start out thinking space has three dimensions. You think it might have four? Fine. We know the tool to represent that. It's four dimensional Riemannian geometry. Well defined mathematical subject. That's not relative.
The geometry of special and general relativity is not Riemannian geometry. It's Lorentzian, and because it's Lorentzian, it contains a light cone structure that is pseudo. A purely spatial structure has nothing that corresponds to a light cone.
To say they're on equal footing, I mean, again, it's kind of also a little vague what that means, but it certainly doesn't mean that you've just turned time into space. And the real problem of people saying that you've turned time into space is that space doesn't have a directionality. That is, if I were to just go to somebody in the street and say, you know, the United States runs from north to south and not south to north, they'd look at me like, what in the world are you even saying?
What do you mean? There's a north-south dimension, if you will, but it doesn't go one way rather than another. Whereas, if I tell you time goes from past to future, and not from the future to the past, everybody says, yeah, of course. I mean, that's a triviality. Everybody knows time has a direction. We're constantly getting older. We wake up when we're older.
There's some that we'll get to that say time is an illusion, so I'll get your thoughts on that. Okay, but anyway, time, intuitively time has a direction and space doesn't. And if by equal footing, people took the lesson, which many of them do, is they say, oh, first of all, by equal footing, you mean you've spatialized time, somehow time has now just become more space, but space doesn't have a direction. So gee, I guess time doesn't have a direction. Now you're in a complete mess.
Now when I say you're temporalizing space, what I mean is something very, when I said that, is something very precise. Take a relativistic space time and blind yourself to everything except just the temporal structure, the proper time structure.
from that you can recover the whole thing okay sorry repeat that one more time from the time structure from the proper time structure you can recover the entire structure right information about the the whole space time geometry is contained within if you just give me all the causal curves all the time like or light like curves and their proper lengths in right in proper time
That nails down the entire space-time geometry. So to that extent, you could say time settles everything. Time has a direction. We're going to save that. Yep. But we agree space doesn't appear to have a preferred direction. Right. Because it doesn't. Somehow we can generate space from a preferred time direction. You can generate the spatial structure. Yeah. Yeah.
There's only one spatial structure that's consistent. There's only one full relativistic metric that's going to be consistent with just the temporal part. Now is this written somewhere in the paper or is this an idea that you're toying with? Here's some stuff that's well known. This is not original to me.
Just forget about, just give me the light cone structure. So we all know, I hope, that in a relativistic setting, space-time has a light cone structure. So any event, there's a double cone, a future light cone and a past light. And let me just comment on that. It's important that it's a double cone. That is, if I have an arrow pointing into the upper part of this light,
I can't continuously rotate it down into the bottom so it's pointing the other direction and always have it point timelike, that is, have it point within the cone. The timelike directions are divided into two disconnected groups, the future pointing ones and the past pointing ones. But the space-like directions, which are the directions that as it were point outside the light cone, they're continuously connected. I can get from any space-like direction to any other one just always remaining space-like.
Right. So that's the sense in which time has a direction. There's a fundamental distinction between the future directed things and the past directed things. And there's no corresponding distinction in space. Now, what's well known is if I just give you the light cone structure, just the light cone structure, to specify a relativistic metric, I need 10 numbers. If I give you just the light cone structure, that nails down nine of them. Okay?
That's well known. This was proven a long time ago. So there's just one other number that I need. And if you give me not just the light cone structure, but the length, so these timelike curves have a length that's called the proper time. If you tell me that, that'll nail down the tenth number. Interesting. Done. Okay. There's no other wiggle room. You've got everything. This is well known. This is not anything special to me.
And so just to be clear when we were talking about that if you're in the upper part of a light cone, then you can't go to the bottom part. We're in an orientable manifold or does this still apply? And it's always the assumption when you do relativity, always, always, always the assumption that the manifold is temporally orientable. So it's not like a Mobius strip where things get twisted up. And that's because time has two directions.
I would like to know your thoughts on Occam's Razor, whether it's a good principle, what are its pros and cons, and this relates to the metaphors, but we'll get to that. So what are your thoughts, what are the pros and cons of Occam's Razor? Well, again, part of the problem is when people use the term Occam's Razor, they often have different principles in mind. So what Occam said was something like,
Don't postulate entities without necessity. The problem with that is, what do you mean by necessity? If you take that too strongly, then it's kind of a crazy view because you're never really necessitated to postulate things. So then you weaken it and you say, look, don't postulate entities without good reason. That sounds good.
But that's kind of trivial. I mean, you should also not delete entities without good reason. Okay, you shouldn't do anything. You just use good reason. Okay, it becomes good advice, but kind of trivial. Yeah, sometimes people say, well, Occam's razor says, if I have two rival theories, and one postulates fewer things than the other. So again, we go to the idea of ontology or beables, then you should prefer the one with fewer.
Many times people say, oh, that's Occam's race. Take the one with fewer. That's certainly not advice that physicists follow, and I think for good reason. Why? For example, take a classical electromagnetic field of the sort that Maxwell did. So it's got charged particles, that's part of the ontology, and it's got fields, it's got electromagnetic fields.
Now you can imagine a theory that gets rid of the fields, a kind of action and a distance theory. This is what Wheeler and Feynman were doing. You just throw away the fields and you say, no, no, it's not that this charged particle affects that one by the effect of an intermediate field. It's that this one just directly affects that one, maybe with a time lag with nothing between them.
Well clearly that second theory has less ontology than the first one because it's you know they both have the particles and one has the fields and the other doesn't doesn't mean it's a better theory in fact it's it seems like a much worse theory because you give up certain principles of continuous action and so on and in favor of this
I mean, no physicist would take that seriously. They'd say, look, wait, but I have a whole set of equations for the electric and magnetic fields. I understand how they behave. I understand how jiggling this charged particle creates an electromagnetic field that goes through space and time and then hits this charged particle in an antenna and makes it go up and down. And then I turn on my radio.
Okay. And if somebody wants to explain that taking away the intermediary fields, that's just a very strange thing to do. I certainly shouldn't think that, you know, less is more in ontology. So again, if it's just, if the advice is, look, just think about the ontology when you're comparing two theories, compare their ontology and ask, what is the justification?
where one has more if one strictly has more what's the justification think about that justification that's good advice but it's not good advice to say no don't even think about that just just compare them and then and announce that the one with less ontology is better because you don't you have to think about what price you're paying
Yeah, get rid of it. With the Feynman case with Wheeler, was their mathematical treatment more complex than the original electromagnetic from Maxwell? Because the Wheeler-Feynman theory tried to get rid of the electromagnetic fields in favor of, and now this comes back to the stuff about retrocausation, in favor of having direct action at a distance
both to the past and to the future. So the way a charged particle behaves right now was determined by looking both on its past light cone and seeing where there are other charged particles on the past light cone and what they're doing, and on the future light cone and where there are other particles on the future light cone and what they're doing.
and so you had instead of what we normally have which is a boundary condition a single boundary condition from which you generate the future in the past you had a boundary condition to the past and a boundary condition to the future and to the future they use this thing called an absorber a perfect absorber condition blah blah blah but just mathematically that's a much more difficult problem much more difficult problem to know you can write down some equations to know whether they even have any solutions
It's just mathematically very intractable. Okay, so then it's intractable, which means that they weren't able to produce or reproduce the results? Yeah, as far as I know. I mean, I knew some people who were interested in this theory.
uh in munich and you know recent i mean within the last 20 years and we're working on it and to my knowledge they never exactly got a precise solution either but i haven't looked at it in much detail it's okay it's just clear on mathematical grounds when you have something like newton's equations if you give me an initial condition and tell me the positions and moment of all the particles
Then you can prove in certain circumstances there's a unique solution to the future, there's a unique solution to the past, and you can actually get your hands on it mathematically. But when you say, no, now I have a situation where I have one boundary condition to the past and one boundary condition to the future and I'm trying to solve in between,
it just mathematically i mean you can imagine how do you go about it i mean in the case of gluten literally you imagine on a computer you set some data here and then the computer computes okay then a little later this then a little later this then it just builds it up or a little earlier this a little earlier builds it down but if i if i give you data here and here and i have to somehow
fill in the middle in a way that's consistent with it, it's not even quite clear how you proceed. If Wheeler and Feynman were able to reproduce it, you would still have the same objection. It's complex. I'm not sure. I mean, you would really have to ask yourself, look, if the theory that postulates electromagnetic fields has this simple Newtonian presentation, which it does,
Right. I just give you an initial condition. Simplicity needs to factor in too. Sorry, I don't mean to interrupt. I apologize. So simplicity has to factor in as well. It's not just less assumptions. Is there another quality like aesthetics? What are the factors that need to be in place? Because Occam is the way it's ordinarily stated. You have two competing theories, whichever one has less assumptions but produces the same results. You choose that one. Yeah. I mean, again,
Even if you're a logician, if you're like a philosopher and you're used to these logical tricks, you say, well, what do you mean fewer assumptions? If I have a theory with 10 assumptions, then I just put a conjunction between them and I have a single sentence. That's where I was getting at. Well, that's my objection with Occam's razor. To me, it's unclear what counts as an assumption. So some people say the simplest is God did it. But I don't know if God is one assumption or if it's 10. I just pause at space time. Is that one assumption?
Space-time has a huge structure. So it's like a manifold is like second countable Hausdorff and then there's atlases and then there's orientations So is that one assumption? Yeah, and I think quite honestly It's a fool's errand to think that you can come up with some abstract statement of a methodological principle That'll be the right one to apply in all cases. I mean start with the problem in front of you The problem in front of you is going to be here's one theory. Here's another theory
And let's assume that they make the same empirical predictions or where they differ you can't check or something. So you can't say, I'm just going to do an experiment to try and decide between them. Then people will start producing justifications for thinking one is more plausible than the other. And your job is to think through those justifications. Now, I don't think any single sentence in Latin
is going to give you very helpful advice, universal advice about how to do that. It might be that in some sense one of the theories is simpler, is clearly simpler than the other with no penalty. Then you'd say, okay, the simpler one just seems more plausible. But usually there are trade-offs. How important are they? How much of this will you give up for that? There's not going to be a general calculus of this. You should just
Try to be as explicit as you can in the accounting.
If on grounds other than empirical grounds I prefer A to B, let me say why and then just listen. The kind of justification you might get may be very specific to the particular mathematics being used. I mean, God knows what. I mean, don't expect some guy in the Middle Ages wrote down something that is going to give you the royal key to deciding between these things.
And maybe there's no good way to decide between them. I mean, we should be prepared to have alternative physical accounts of the world that we don't actually have any particularly good reason to decide between. And if that's the situation, that's the situation. Speaking of what's tricky to decide between the illusion of time. So there are certain theories and certain interpretations of time.
Yeah, so you did give me some warning you were going to ask that question. And I'm afraid I just have to say I don't understand relational quantum mechanics. I looked at it a bit. I talked to Carlo about it a bit. I just don't understand it.
I know some people who looked at it more carefully and they were unable to find what they considered a coherent formulation of it. If you feel like you understand it, then you can explain it to me and I can try to respond. If you feel like you don't really understand it and can just repeat what other people say, we're a bit stuck. I just don't understand what the theory claims, even apart from time. My issue here isn't about time.
I don't understand what it means to say that all that's real are relations. I mean, sometimes there's this tag phrase that people say, oh, relations without Rolada. And this is, you know, you're just contradicting it. How can you have a relation without Rolada? Or things only exist when they interact and then you want to say, but wait, if they weren't already there, how could they interact? I mean, how can interaction occur between two things that don't exist?
I literally don't understand what's being claimed. And again, if you feel like you want to explain it to me, I'm all ears. Again, one little data point was once I was puzzled about this, I was talking to Carla and we were talking about non-locality, right? So in the non-locality case, Alice does an experiment over here and Bob does an experiment very far away. And the issue is
the correlations between the outcomes of the experiments they do, right? That's what Bell's inequality puts constraints on in these correlations. And as far as I could understand, Carlos seemed to be saying that there is no outcome for Bob
just, you can't just talk about what happened in Bob's lab, right? You normally think you could. You say, okay, Bob did an experiment. Did he get a dot here or a dot there? You know, was it an up or a down outcome? And it's like, no, no, no, he only had an outcome relative to Alice. And furthermore, somehow that outcome only occurs when Alice and Bob interact and Alice and Bob only interact when after the experiment, they come back together and share data or something like that.
And I said, but I don't understand, Carlos, so if I'm Bob, and I do my experiment, right, and I have my data, and then I'm waiting for Alice, Alice is to come meet me to share her data. And Alice comes and she says, you know, 10 minutes ago, I just got this result. Do I not believe her? I mean, do I literally just not believe? Yes, you got that result 15 minutes ago, long before I ever found out about it and long before we ever interacted.
And Carlo just looked at me like I'd asked a question that never occurred to him. I mean, he didn't have a ready answer at the time. I don't know if he does now. And at that point I said, I just don't understand. I mean, I think Alice is there doing her experiments and getting her outcomes, and Bob is there doing his experiments and getting his outcomes, even though they're not interacting with each other in a normal sense of interaction.
And those outcomes are what they are and they do or they do not violate Bell's inequality. I don't understand what the relational theory says about that case. I have some puzzlement because I asked Carlo, well, what's being related? And then I believe he said certain relations. Then I said, well, what are those relating? And is it not just some infinite regress of relations? And then he said, yes.
This is a real good story about Bronx and his dad Ryan, real United Airlines customers. We were returning home and one of the flight attendants asked Bronx if he wanted to see the flight deck and meet Kath and Andrew. I got to sit in the driver's seat. I grew up in an aviation family and seeing Bronx kind of reminded me of myself when I was that age. That's Andrew, a real United pilot. These small interactions can shape a kid's future. It felt like I was the captain. Allowing my son to see the flight deck will stick with us forever. That's how good leads the way.
There's a Dr. Seuss story about the lazy bees in Hotch Hotch.
So, you know, they find that when the bee watcher watches the bees, they're busier, so they hire a bee watcher. But the bee watcher was lazy, so they had to hire a bee watcher watcher, and then a bee watcher watcher to watch the bee watcher. And, you know, it's the same thing people say, well, you know, reality doesn't exist until an observer observes something. And you say, but wait, did the observer already exist?
Oh, no, no, the observer only exists because some super observer was observing the observer. But did the super observer already exist? Oh, no, there had to be a super, super observer. And then you said, this is just silly. I mean, you know, why are you doing this? And if someone said, well, is it turtles all the way down? Yes, there's an infinite hierarchy of observers observing other observers, bringing them into existence. You just say, look, you can say this stuff, but why are you doing it? I mean, we have perfectly good accounts of the physics here that don't involve this kind of ridiculous gymnastics.
What's motivating you to say things like that? Yeah. So the way that I view it, and again, just so you were clear, Carlo, another huge admiration for Carlo, huge intellectual giants. So I'm Switzerland when it comes to this. I think that many people, they believe that and I'm not categorizing Carlo as this, I'm referring to those people about the observers and infinite regress of super observers. So not regress, but superset. I think that people
Including myself, including everyone. I think that people reason backward and we believe that we're reasoning from founded. We say first principles, almost no one is a first principles thinker. We have certain values that we're trying to preserve and we reason backward and we say, well, you know how a implies B is the exact same as not B implies not a, let me give you a story.
I was driving with someone who is of extremely liberal temperament, and there's nothing wrong with that. And I said, oh, that dog breed is known for being extremely intelligent. And she's like, no. I'm like, what do you mean? No, it's just known for being intelligent. It has a higher IQ. She's like, what are you talking about higher IQ in dogs?
And then I realized what's happening is that she believed that what could be done is I could then infer onto humans and say, oh, well, there are different species here. So because that is not she wants to preserve that she reasons backward. OK, so I believe that that's what's occurring in many of these cases. Well, look, there's a there's a tradition. There's a tradition going back to boar and is kind of associated with the Copenhagen school or the Copenhagen spirit or whatever.
that thinks that the great revolution in quantum theory is because the observer unlike in classical physics, unlike in Newtonian physics, the observer somehow plays a central role in quantum theory.
And the problem is that on no actual existing precise understanding of quantum theory that we have, is that true? It's just that, I mean, for many years, for decades, there's been a intermittent series of meetings and the title of the meetings is quantum theory without the observer. And it's just
formulations, theories that would would account for the quantum phenomena, where in their formulation, they never mentioned observers. And there's a wonderful phrase that Bell has. At one point, he says, this is a theory that is neither requires nor is embarrassed by the existence of an observer, right? I mean, it's not that the physics is going to have trouble modeling observers, their physical systems, but
Gee, you know, I mean, we all think, don't we, that shortly after the Big Bang, a whole bunch of physical stuff happened and there was nobody there observing it? I mean, it's kind of, again, it kind of blows your mind when you try and see what are you even thinking if you think that physics needs observers to go on? It's just the opposite. Observers are physical systems of a certain kind, and you want the physics
to describe how they behave as by pure analysis of them as physical systems. And it's the same thing for measurement, right? A measurement, whatever you call a measurement is just a physical interaction between two things that's going to be described by some physical equations that don't mention the word measurement. And that's again, go read Bell's against measurement against quote measurement.
He says the term measurement should not appear in the formulation of any theory that pretends to be foundational. Because measurement isn't the right kind of concept to appear in the foundations of physics. Measurement is a derivative concept. Observer is a derivative concept. They shouldn't appear in the foundations of physics.
Lots of people convince themselves that, again, here is where I think there's something semi-mystical. There are people who like the idea that observers should become central to the universe.
It annoys them that observers seem to be this tiny little special class of humans in particular, some tiny little specks on something not playing a very central role in physics. They have an emotional attachment of some sort, but it seems to be contrary to physics as we understand it.
And certainly nobody's ever made progress doing that. Nobody's ever made progress saying, oh, here, I'm going to write the observer into the foundations of physics. And people have made progress doing the opposite. Okay, so two notes here. Number one, there exists perfectly coherent formulations, theories of quantum mechanics that don't involve observers. Sure. Okay.
So that's interesting. The pilot wave theory, the GRW collapse theory, I mean the many worlds theory in so far as you can make sense of it certainly does not make any reference to observers.
I mean, these are theories where they're written in the equations. I mean, again, Bell says somewhere, I want a theory where the physics is in the equations and not in the surrounding talk, right? These are all theories where you postulate a quantum state, maybe you postulate some local variables in addition to that, and you write down some clean equations about how they behave. And that's the theory.
It doesn't mention observers, it doesn't mention measurements, it doesn't mention anything like that. There's some physical stuff that you postulate, it behaves in a certain way that you postulate, and then you see what happens. So pilot wave, many worlds, and then what were the other? The collapse theory, Jardhi-Ramini-Weber objective collapse theories. I mean, there are a class of so-called objective reduction theories. I mean, Penrose has a kind of sketch of a theory for many years. He's had a sketch of a theory
Which ties the collapses to gravity, the GRW theory doesn't. What do you make of Penrose's theory? I mean, I understand he takes the foundational questions seriously. And foundationally, you only have three options. Okay, that's why you can kind of have a nice taxonomy of solutions to the measurement problem.
You can say that what you do is just logically you get into trouble if you try to maintain three principles. Principle one, that the wave function that's used in all these quantum theories provides a complete description of a physical system. And this was the EPR paper in 1935, the Einstein-Podolsky-Rosen paper, the title of it, is the quantum mechanical description of reality complete.
Meaning, if I give you the wave function, can you somehow derive from it all the physical facts about the system? They were arguing you can't. They were arguing it's incomplete. Okay. Many, Bohr and Heisenberg wanted to say it was complete. So you have one principle. Is the wave function complete? Second question. Does it always evolve by Schrodinger's equation? Schrodinger's equation is linear. That's a very important feature of it.
Third claim is that when you do a Schrodinger cat experiment, you start with one cat, you end with one cat, and the single cat is either alive or dead. That was Schrodinger's point. He thought that's obvious. You can't maintain all three. As a matter of logic, you can't maintain that the wave function is complete, that it always evolves by Schrodinger's equation, and that the cat ends up alive or dead.
What does complete mean once more? What does complete mean? Complete means if I give you the wave function of a system, you can derive from that somehow all the physical facts about it. All of its physical characteristics. Does that mean that it has that you could find out whether it's spin up and then spin? It means you could find out anything that's a physical fact about it. I'm not telling you what the physical facts are, but any physical fact is determined by the wave function.
So are there theories that say no, you can't? Sure, pilot wave theory. I mean the de Broglie-Bohm or Bohmian mechanics. In Bohmian mechanics, the ontology, the beables, there are two. There's the universal wave function, the universal quantum state, that's one. And then there are these particles, and the particles are like classical point particles. They always have definite positions, they move around, okay?
But if I just give you the quantum state, that doesn't tell you where the particles are. Where the particles are, the actual configuration of particles, is not derivable from that. It's an additional piece of physical data. And the way that theory works is you say, okay, how does the wave function evolve? Answer, always by Schrodinger's equation. Okay, how do the particles move? How does this particle configuration change? That's given by a second equation called the guidance equation, and the guidance equation has as an input to it the wave function.
So that's a theory where you say the wave function is not complete. There are additional physical facts. I could have two systems that have exactly the same quantum state but are different physically because the particles are in different locations. So you can't maintain that the wave function is complete and it always evolves by Schrodinger's equation and that the cat ends up either just plain alive or plain dead.
Okay. Got to give up one of those three. If you give up the completeness of the wave function, then you have what's called a hidden variables theory like Bohmian mechanics, pilot wave theory. If you give up the Schrodinger evolution, then you have an objective collapse theory. You say, okay, the wave function doesn't always fall by Schrodinger's equation.
and if you don't give up those two then you're stuck with many worlds and you say well the cat doesn't end up either alive or dead somehow one cat becomes two cats or many cats some of which are alive and some of which are dead okay those are only three options just as a matter of logic i mean i have a paper about this called three measurement problems which goes through in more detail this is a nice thing because if somebody says here's my understanding of quantum mechanics
you can immediately ask them these diagnostic questions okay according to quantum state complete according to you does it always evolve by schrodinger's equation according to you does a single cat end up either just alive or just dead and if they say oh i i you know i want to maintain
The kind of intuitive answer to all three, you say, I'm sorry, you can't, right? Then it's not a coherent theory. And once they answer which one they're going to give up of those three, then you sort of understand where you are and you understand how the questioning goes on from there. So yeah, this is extremely, extremely important. Yeah.
So, I mean, look, Penrose has gone for an objective reduction theory and that this follows, if you go and read von Neumann's book, Mathematical Foundations of Quantum Mechanics, von Neumann, and this was taken as kind of the Bible of a clean mathematical formulation of Copenhagen, von Neumann is very clear. He says states, physical systems can undergo two completely different sorts of evolution.
Process A and Process B, or Process 1 and Process 2, I can't remember what he uses. One of them is Schrödinger evolution and the other is a collapse. Right, right, right. And they're very different. And if you think that, then you have an objective reduction theory. Now that's the way Penrose wanted to go for a long time. And then the question, if you have a reduction theory, you say, okay, well, when does this collapse occur? What triggers the collapse?
Or does anything trigger the collapse right what governs the collapse if you will you better not say well it collapses when you make a measurement because then you what when do you make a measurement the measurement now you're back to against measurement. Penrose wanted to tie the collapses to gravity so. Any had a sort of.
Likely hand-wavy account of that, which actually a similar account was given by a guy named Karolyi Hazi long ago. He was also, when Karolyi Hazi was great, I heard him once when I was very young, and he would say, well, as it were, you have these two gravitational situations, which by Schrodinger evolution,
It's like, you know, they're separating and there's a gravitational difference between them. And then it's at some point that gravitational gets so big. And this is what Karolyi has, he said, he said, then Gabriel comes and blows his trumpet. And the blowing of the trumpet was the signal, the wave function shall now collapse and it'll either go this way or that way. Right. And so, you know, Penrose's idea was to tie somehow these collapses to gravity.
The Girardi-Romeini-Weber theory doesn't do that. It doesn't tie them to anything. It just gives you a fixed probability per unit time per particle that there'll be a collapse. So these are objective reduction theories. It's a way to go. The objective reduction theories make in principle slightly different predictions, empirical predictions, than the other theories. So you can kind of test in the lab. And the GRW kind of theory
I mean, people keep doing tests and ruling out certain parameters space, and there's not much parameter space left for it. In another 10 years, it may be empirically, empirically refuted. To reiterate its completeness. So is there more information necessary to specify what's physically going on? And in the case, just to be clear, again, bohmian mechanics and pilot wave are synonyms. Yeah. Okay. And so in pilot wave theory, it says there's a wave and a pilot.
No, the wave is the pilot. Okay, sorry. The wave, the quantum state, call it the quantum state. That's the thing that's described mathematically by the wave function. Okay. Its job physically is to pilot or guide the particles. Okay. To determine how the particles move. And it does so via a precise equation called the guidance equation.
i mean bell bell presented it that way but bone when he did it i mean the original idea was de broglie and then de broglie dropped it for bad reasons and then bone rediscovered it but bone presented the theory as a kind of newtonian theory with something he called the quantum potential that turns out to be a bad idea bell presented it in terms of the guidance equation which is the right way to do it
So the piloting is done by the wave as it were by the quantum state and what pilots or what it guides are the particle. A surfer on the wave is the particle something like that? You can think of it that way. Now the thing that you have to wrap your head around that's hard for this theory is if you have a single particle you can really think of it that way. You've got kind of a wave in physical space and it's telling the particle where to go. But as soon as you get two particles or three particles or four particles.
whether they interact or not, the weight function is now defined not on physical space, but on configuration space, which is a 3N dimension. So for four particles, it's a 12 dimensional space. And this is just standard quantum mechanics. This is not peculiar to this theory.
And essentially what that high dimensional object is doing is governing the entire configuration of particles. It's not as if, as it were, one part of the wave is telling this particle where to go and another part of the wave is telling this particle where to go. It's that the whole wave is telling the entire collection of particles how to evolve. That's why it's a kind of non-local theory. That's how the non-locality gets into it.
It's a kind of global holistic theory where everything is being done together at the same time, if you will. Are there separate pilot waves for each of the particles or is there one pilot wave? No, no, there is one. I mean, in this theory, if you take it seriously, ultimately, there is one quantum state and that one quantum state is the quantum state of the entire universe. And what it determines is how the entire configuration of everything in the universe
And then there's a very interesting question to which there are very interesting and precise answers. Well, if that's true, that's not what we ever use in our calculations. When I go into the lab, I don't use the universal wave function
of everything, I assign a little psi. So in the literature, there's big psi, which is the fundamental thing. It's the wave function of everything in the entire universe. And then there's little psi, which is something I would assign to a subsystem. And then there's a question of, well, what's the status of this little psi that I actually use? And there's a very nice answer to that. It's called the conditional wave function. The answer is mathematically clear. This is all covered in my book as well. And the conditional wave function
sometimes evolves by Schrodinger's equation and sometimes does something that looks like a collapse. Interesting. All of that is just a consequence of these two equations that are at universal scale. So there's a really beautiful explanation in this theory of why people use collapses to do calculations and to the extent to which that's justified.
even though in this theory there are no collapses on the single fundamental quantum state which is the quantum state of the entire universe. There's a wonderful answer to that and it's an answer that's not available in other theories. I mean there's a nice story to be told. And you use this word theories which is interesting because when one takes quantum mechanics in university they're taught or they're told you're going to learn quantum theory and you would say at least from my observation from your talks
Well, a theory needs to have an account of what is so it needs to have some ontology, or some way that the mathematics relates to what is, you can't just speak about the mathematics. So that's why you're not calling, I've been listening to you're not calling these interpretations of quantum mechanics, you kept saying these are different theories, because you don't see quantum theory as having multiple interpretations, you see each of them as separate quantum theories. That's right.
What you're taught when you take a course is not a theory. Why? Because a theory postulates the existence of certain things. It tells you, I am a theory about the physical world. Here's what I'm saying is in the physical world. And here's how it behaves. That's what a physical theory is.
What you're taught is a scheme for making predictions. Okay? It's a predictive recipe. It says, oh, in this circumstance, use this mathematical object and solve it in this way. And then you pull out of thin air born's rule to make probabilistic predictions. Okay? It's a prediction making apparatus, but it doesn't even attempt to be a theory. It doesn't pretend to be a theory.
Because if it were a theory, you could say, okay, how do I understand this wave function that you're using? Does that represent anything physically real? And it would give you an answer. The answer might be yes, the answer might be no, but it would be an answer. And what you're taught doesn't give you an answer. That's why people use this phrase, shut up and calculate. They'd say, look, I've given you this wave function, I've told you how to manipulate it to make your predictions.
Now shut up and calculate. Don't ask me whether it represents anything physically real. And so it's not a theory. And then what people call interpretations of quantum theory are really explicit physical theories that make precise postulates about what exists and how it behaves that are designed
What's wrong with the shut up and calculate mindset?
So why can't we simply have both? Hey, there are some people who are going to investigate subterranean, find the sharks and the giant squid. And then there are those on the top. They're like, Hey, I'm going to build the cities and make technology. And sure, you can say so, so exist. What difference does it make for me? At some points you provide insights and they come up to the surface and maybe we inform you, you do you will do us. Is there something wrong with that?
Division of labor. I mean, it's the same division of labor one has typically between engineering and fundamental physics. So let me just give you an analogy. Suppose I'm an engineer. I'm interested in building bridges. And I'm just fooling around at some point with different metal alloys. And I make an alloy that has some really nice properties for bridge building, right? It's light and it's very sturdy and so on. And I say, ah, good.
I'm going to use this to make my bridges. I can now make lighter bridges and stronger bridges. Good. As an engineer, I'm happy. And somebody comes along and says, but why that alloy? I mean, why is that alloy so much stronger than other alloys? Now, as an engineer, I don't have a complaint about the engineer saying, I don't care. I don't care why it works. I don't need to understand why it works for my job. I just have to know that it works.
I have to know that it is in fact light and strong, which I can test. And look, it is. If you, if you physicists want to try and figure out why this much bismuth and this much tin and this much, you know, iron happens to form a structure that's so light, fine, go to it. I as an engineer could not care less. My job is done when I built the bridge and I'm sure it's going to work. So that's a division of labor.
i'm not gonna i'm not gonna rag on the engineer you can say well at least they ended up building a bridge you know they made life better people can now travel across the river i'm not making a value judgment i am saying from my personal point of view i want to know why it works it's not to me i'm just a nosy person okay i'm not focused on on uh practical
I'm just curious. I would be interested to know why that particular mix of metals is so much better than other mixes of metals. And that's the kind of thing that the physicist would get into and try to describe in fine detail what's this underlying structure. Now, you know, some physicists have decided, from my point of view, when it comes to quantum mechanics, they're happy to be engineers.
Right. They say, okay, you've given me this wave function and you've told me how to calculate with it. And that's enough for me to design my chips or, you know, blah, blah, blah. I'm happy. I'm but personally, I'm not happy. I don't really care about the chips. I want to understand what's really going on. Right. I want to understand what the world is really made of. So I want to go deeper than that. But that's that's, you know, I can't say that that's intrinsically better thing to do. Certainly,
You would not have made all the material progress that has been made using quantum mechanics if people got stuck because they didn't understand it. So shut up and calculate was efficient. But it's really bad if people think there's something wrong with the small minority of us who are not satisfied with it. There's something illegitimate.
about people who say, but I really want to know what's going on, not for any practical purpose. And of course, most of physics is like that. I mean, when they built the large hadron collider to try and see if the Higgs particle existed, it wasn't because you would make better computers or build bigger bridges or have some practical consequence that came out of what the exact mass of the Higgs is.
Nobody could foresee they would have any interesting practical consequences. You're just curious, right? What's going on? To me, that's the fundamental impulse of being interested in physics. It's just, you know, thalmas. It's wonder at the universe. It's trying to understand things. Now, some people aren't driven by that and they're more interested in building things or they're more interested. And I don't want to say they're, you know, what they're doing is bad.
But, you know, leave at least a few of us who are motivated by pure intellectual curiosity to go about our business. Yeah. And I find that there's a bit of condescension from the hardcore physicist toward the philosophers. I don't know if you have an understatement. Yeah, I wish that wasn't there. It's getting better, though. It really is getting better. I mean, if you didn't notice, there was just within the last week, I think I saw it on Facebook, an editorial published in nature physics.
Okay. We don't know who wrote it. It's just an editorial from the editorial board, not signed, which is basically saying, you know, there's really a reason to look at foundations of physics. Right. It's not that everybody has to, you know, but it's really a worthwhile subject matter for physicists to think about foundational issues. And to me, this is just wonderful because that's, I mean,
I know people, I mean, I was lucky because I'm in the philosophy community and philosophers are happy with the questions I'm interested in. But physicists trying to do foundational work, you know, I mean, they have a very, very, very, very hard time just at a practical level, because it's not valued. It's just not valued in physics departments. It's considered somehow illegitimate. Do you see consciousness as a foundational question?
Consciousness I see is the hardest question there is. I think the mind-body problem is the hardest conundrum. You can't even imagine what a solution could look like. I don't know
I mean, if you sort of think, let me give an example. If you go back to ancient Greece and they said, I wonder what matter is made of and the atomists say, well, I think it's these little hard atoms and all they can do is move and some of them have hooks and some of them have eyes and sometimes they hook together and, you know, okay. You can at least understand
In a sketchy way, how such an apostolate could account for cables and chairs and things burning and, you know, okay, they become unhooked or they get hooked or blah, blah, blah. I mean, at least you see maybe it's wrong. Maybe it's completely off target. It turned out it was off target. But I mean, that's the story. But at least you could see from the beginning how if it were true, it could account for what you want accounted for, which is the behavior of objects at macroscopic scale.
In the case of consciousness, I don't even know what a solution to the mind-body problem could look like without putting any other constraints. I just don't know. Essentially, what you're asking for is some physics that when you apply it to a physical body in the structural form of a human brain or a dog's brain or, you know, I don't know, an ant's brain, I'm not sure.
Right? I don't know. Are ants conscious? I don't know. That somehow you see that, gee, that physical behavior would be accompanied by or create or whatever a subjective state like pain. And I don't understand how that could be. So I think that's the hardest problem there is. And because it's so hard, I don't work on it because I don't even understand what a solution would look like. I mean, I've pointed out
one of the first papers i published was a paper called computation and consciousness which is just pointing out that a certain kind of attempt to answer that question can't be right but i don't know what the right answer is i don't even know i i i'm you know so i it's not that i don't think there's a problem there i think there's a problem there that's so intractable yes that i don't want to
Bend my time banging my head against a brick wall. Now I firmly believe I'm not a dualist. I firmly believe that anybody who had a brain that was physically identical to mine and the neurons and everything were doing things physically identical to mine, they would be having conscious experiences that would be qualitatively identical to mine. So I'm not a dualist.
I think that consciousness in some sense supervenes on physical activity because I think that's all I am. I'm ultimately a physical thing. It therefore depends on quantum mechanics just because physics depends on quantum mechanics. But I don't see any clue of a comprehensible account of why
Subjective physical states should accompany any sort of physical activity. So, you know, you work where you think you can make progress and you just, you know, you acknowledge where you don't think you can and where there are problems.
You mentioned that you ruled out a certain class of consciousness theories that are computational. I don't know if that was the correct wording. Regardless, can you expound on that? Is it similar to the Penrose-Lucas argument? There's a whole class of theories that are called functionalist. And the basic idea is what makes a system conscious is kind of independent of the particular physical stuff it's made of. It's rather how that physical stuff behaves.
Functions in a certain way. Sorry, is this the same as substrate independence or no? Yeah, it's a sort of kind of substrate independence, right? That's a kind of general class. Then in that general class, there's a much more specific realization of it that's called computationism that says what's really important is to understand what computations the system is doing.
where I then understand a computation in terms of a Turing table, using Turing's way of understanding a computation. Such a theory would be committed to saying that any system, that there's some calculation or some computation on some input that can be done, such that any system doing that computation on that input will be conscious in a certain way.
All right, let's say, and again, by conscious, I mean, the paradigm case is pain, right? Have a toothache, right? Any system that satisfies a certain Turing table performing a certain computation on a certain input will experience subjectively a toothache. And what I have as an argument is that that can't be right. Because you can implement the Turing machine table
in peculiar ways and not ways that anybody would actually do in practical life but theoretically you can implement a Turing machine in ways that the Turing description depends on a bunch of structure that's not physically active through a certain period it's physically completely inert it's not doing anything and according to this
Whether that structure is there or not will determine whether there's a to think. But on the other hand, you can see that whether that stuff is there or not makes absolutely no physical difference to the makes no difference to the physical activity going on. So if you think that the conscious state really is determined by physical activity, it shows that this computational conceptual apparatus is the wrong one to be bringing to you.
but that you can read that paper i mean it's a journal philosophy will you be able to send me the title of it so i can include it in the description it's called computation and consciousness who's public in general philosophy i mean i could pull up my cv i think it's 1989 but you can find it i'll leave a link to it in the description so does that include
formulations like integrated information theory? Well, not exactly, because as I understand it, the IIT idea involves information theoretic concepts that go beyond the Turing description. But on the other hand, I know Scott Aronson's criticisms of the IIT and I find them completely convincing, so I don't think the IIT has any more hope.
of being an account of consciousness than the term machine stuff does. Have you heard of the global neuronal workspace theory? The words are a little familiar to me, but I don't know it in a way in any detail to be able to comment intelligently about it. Have you heard of Yosha Bach? Yosha Bach, who has his computational consciousness model. I don't know what it's called.
Okay, what are your opinions on Wolfram's physics theory? Because it's entirely computational and seems to at least he believes it produces consciousness because we're here and he believes his theory to be correct. Have you taken a look into his physics project? Again, not in any detail. I mean, anybody who says
The foundations are computation. Computation is just the wrong concept to be bringing to physics. Why? Because it's because computers have the physical structure they have that they're capable of doing the computations that they're capable of doing. That is, the physics determines the computational structure, not the other way around.
So, you know, somebody, I mean, it's a bit in the ballpark of, again, the idea that everything is mathematical, right? All physical reality is just mathematical, Max Tegmark's idea. It's so far off base to me that I don't really feel like I need to spend a lot of time on it. You know, I think the physical world, you know, I don't know, space-time, it's got particles, it's got fields, it's got some quantum state.
Trying to understand how all that stuff fits together, trying to figure out what the small-scale geometry of the physical space is and space-time. This all strikes me as recognizable speculative physics, but you go beyond a certain point and I don't even recognize it as speculative. Okay, so let's just get to a couple questions on time. There's entropic time and then there's thermal time. I don't know if those two are the same.
But what are your views on those explanations as to the arrow? I don't understand even what that means. There's time. There's a variable we normally designated with a T. When we assign numbers to it, we tend to assign numbers so that as time goes forward, the numbers get larger.
To say that there's entropic time, I don't even understand what that means. I mean, it's true that if a system is not at its entropic maximum, it's not thermal, that typically as time goes forward, it will become closer to thermal equilibrium.
That's a fact about, and we understand why that's so, I mean there's pretty good statistical understanding of why as time goes forward entropy tends to increase, but I don't think that means there's a new kind of time, there's just time.
I think what's being indicated is that entropy is the reason for the direction of time. If the universe eventually, which it could in some models, if it comes to thermal equilibrium,
And so there's no entropy gradient. The entropy just reaches its maximum and stays there at least for a very long time. It's still the case that the time is going from the past to the future. I mean, time will still go on. Why would anyone think that time would cease to pass?
The state would constantly be changing. I mean, if you're in thermal equilibrium, things are still changing, right? I mean, a gas in a box comes to equilibrium. Its micro state is still changing all the time. And its change is a normal change from past to future. And the causal structure is that these two particles bounced off each other. And therefore, all of that goes on, even though there's no entropy gradient. I mean, why would anyone
even postulate that the passage of time depends upon an entropy gradient. It's the other way around. The passage of time explains, together with certain statistical considerations, why typically and predictably entropy goes up if it has a place to go. But that doesn't make time depend on entropy, it makes entropy depend on time.
Someone that people keep referencing to me to look up is Henry Berkson. And then they say that Berkson and Bohm knew each other or had or they developed their theories together or where they were influential to one another.
So now that I know that you're an extreme Bohm fan, or at least you're a proponent of the pilot wave theory, what are your thoughts on Berkson? I read a book about Berkson and Einstein. I think, you know, Berkson was tremendously confused. I wouldn't recommend anybody spend time reading Berkson. And quite honestly, I wouldn't recommend anybody spend a lot of time reading the kinds of things Bohm wrote
In the 60s when he got involved with Krishnamurti and he becomes kind of semi-mystical and there's the Implicit Order. I mean he wrote a bunch of kind of weird stuff. You know, you want to look at his 1952 papers and then even better look at Bell's presentation of the theory, which as I say, Bohm presents it in terms of a thing called a
Quantum potential and using Newtonian dynamics and that's a bad idea. It's not really a Newtonian theory It's a great theory. Okay, and it was also again a theory that already de Broglie had seen it's the theory is kind of Almost inevitable way to understand quantum theory once you ask questions in a certain way But you know poor bone, I mean he got he got attacked
for political reasons in the United States, kind of exiled to, you know, he had a life that was, you know, one of being rejected, having been at Princeton and having talked to Einstein and having been influenced by Einstein, and then he ends up, you know, in Brazil and then eventually at the LSE. I have a lot of sympathy for the man as a human being. But I think it made him, his rejection by the physics community,
made him more open to align with Krishnamurti and some people who were of a more mystical bend. Personally, I don't see anything in that that's helpful for me to do physics. So I would say study the theory
I mean, he wrote this later book called The Undivided Universe, and there's some useful things in that together with Holland or not, anyway. But I wouldn't get too wrapped up in Bohm the individual because he was, I mean, he was quite, unless you're just interested in him as a personality, I think a lot of the peculiarities of him as a personality are not useful for understanding physics. Did you ever get to meet Bell?
Sure. Yeah, I met him three, four times. What was that like? Oh, he was just the most wonderful human being. I mean, completely unpretentious, completely open, completely, I mean,
Bell, I'll tell you a story. Unfortunately, he was supposed to come to Rutgers when I was there, and he died about a week before he was supposed to come, which was a great tragedy for us, but I met him
at a philosophy of science association meeting, but not really talk to him there. I met him in a riche at a meeting and I met him at Rutgers. He came to Rutgers before and talked to him individually. But an interesting thing happened. I'll just tell you a story because it's so much to me, Bell. The first time I heard, I mentioned this Gerardi-Romini-Weber theory, which is an objective collapse theory, and Bell
has a paper called Are There Quantum Jumps, which is again in this book, which is his presentation of the GRW theory. Although he goes beyond in a certain ways what they did, but a very nice presentation. And the first time I heard about that theory was from Bell himself. It was in Ariche at this NATO summer school. And I was somebody who grew up
with the common idea that, okay, there are these collapses of the wave function and they're associated with measurements and measurements are associated with consciousness. This kind of idea you associate with Eugene Wigner. And therefore, oh, this is amazing. This is where consciousness connects with physics, right? That consciousness causes the wave function and so on. So you think the most important question there could be is what triggers these collapses?
and bell presents the grw theory and the thing is in that theory nothing triggers them they just happen at random with fixed probability per unit time per particle there's no environmental trigger there's no anything they just happen okay they just happen at random um and and so bell gives this talk and i went up to him afterwards and i said
I said, you're satisfied with this theory? I mean, here we've been for years and years worried about what causes these collapses, what causes these collapses and their answers, it just happens. And you're happy with that? And he just he was so mild. I mean, I can't express to you what a nice man he was. And he was so mild. And he just kind of smiled. And he just looked at me and he said, you don't appreciate what they've done. And he was absolutely right.
And there was no point arguing with me at that moment, but he said exactly the right thing. What they did was they took this idea of a collapse and gave it a precise mathematical formulation. And so they turned a notion into an actual physical feeling.
Whether you like it or not, I mean there are things about the theory one could dislike, even Girardi, who I got to know quite well, never thought it was a final theory, always thought it was a kind of transition point to a better theory, but they managed to just write down some equations, not mention measurement, not mention observers, and get a theory that gave you basically the right prediction.
So Bell was like that, and he was just always tremendously on point, on target. But he just seemed to be the nicest person in the world. I mean, after he died, there was a eulogy session for him at the Philosophy of Science Association. I remember Phil Pearl, who had been working also on objective reduction theories, continuous ones. He got up and he said, for him,
Bell, you know, he was somebody, Phil was somebody working on these foundational issues and he said for him it was like Bell was kind of a knight on a steed to protect them, right? I mean he was the guy, they're doing things that are very unpopular in the physics community and they could always count on Bell to be there to kind of protect them and to say you're doing, you know, what you're doing is important.
And yeah, and you know, you think the guy, I mean, he was a very down-to-earth guy, he was an engine, essentially an engineer, designing magnets at CERN. And he was doing all of this foundational work literally on his weekends, and when he had a sabbatical, you know, he wasn't paid to do it. But he had a tremendously sharp mind, and he had a tremendous physical intuition, right? What the physical world, and Einstein talked about having a kind of
empathetic understanding of the physical world. I mean, you felt like Bell had that. He wasn't satisfied with doing a calculation, right? He wanted to understand what was going on. You get that feeling out of Feynman and Reed Feynman too, right? That he just driving something mathematically, that wasn't enough. He wanted a picture. He wanted, you know, he wanted a feel of what the physics was really about. And he was just, you know, tremendously
Professor, thank you for spending so much time with me. Thank you. I mean, it's been fun. I hope it was more or less what you were looking for. It was more. It was more. Thank you. All right. Well, I do have a side question. How do you keep your mind sharp?
You may say, well, my mind is a sharp, it's dull, but how is it that I don't think that you have a dull mind? So I imagine you have some practice, maybe it's running, maybe it's your read, maybe you have a certain ritual. I mean, look, what I do that actually has an influence on that, I don't know for sure. I mean, I know from the little I read,
that some of the things I do like crosswords actually don't make any difference. So people say if you do this and this and this and then they think well no actually and of course people do say that exercise makes a difference and I do try to get a reasonable amount of exercise maybe that helps. I don't know. I don't have the empirical basis to tell you
There is a question about Judea Pearl because he has a book on causation. I wanted to ask you, apparently you have some objections to his, but if it's too long, then we'll save it for the next time if we speak again. Yeah, I mean, look, I mean, there's a funny thing about Pearl. If you're interested in Pearl, I should just mention it's relevant to understanding. Is it what Pearl's doing is is
is both similar to and was influenced by work that my dissertation director Clark Gleemore did together with Richard Shinus and Kevin Kelly on causal modeling and causal discovery. And I just kind of have a feeling that Pearl gets more attention for that relative to
what he did and what they did, then it's warranted. So I'm a little, I mean, I'm a little bit, I don't want to say defensive, but I do like, if you wanted to look into it, you should look into Gleamor's work, which was some of it earlier than Pearl's. And again, Pearl really learned some of some of the techniques from them. So it's not a criticism of his work. It's more about the attribution.
Yeah, there's a bit about attribution. There's some stuff that he does in his book. I have this review of this more popular book. You probably saw that in the Boston Review. And there is some stuff there that I just disagree with. It's not the technical stuff. It's this
you know, this kind of more hand-wavy stuff. The technical stuff is just writing down causal graphs and representing various interventions and then seeing what kind of statistics you should expect in different experimental situations, given different causal structures. And all of that I don't have any objection to, but it is the same thing, more or less, that Cloumo was doing.
The podcast is now finished. If you'd like to support conversations like this, then do consider going to theories of everything.org. It's support from the patrons and from the sponsors that allow me to do this full time. Every dollar helps tremendously. Thank you.
▶ View Full JSON Data (Word-Level Timestamps)
{
"source": "transcribe.metaboat.io",
"workspace_id": "AXs1igz",
"job_seq": 9464,
"audio_duration_seconds": 10617,
"completed_at": "2025-12-01T01:27:32Z",
"segments": [
{
"end_time": 20.896,
"index": 0,
"start_time": 0.009,
"text": " The Economist covers math, physics, philosophy, and AI in a manner that shows how different countries perceive developments and how they impact markets. They recently published a piece on China's new neutrino detector. They cover extending life via mitochondrial transplants, creating an entirely new field of medicine. But it's also not just science they analyze."
},
{
"end_time": 36.067,
"index": 1,
"start_time": 20.896,
"text": " Culture, they analyze finance, economics, business, international affairs across every region. I'm particularly liking their new insider feature. It was just launched this month. It gives you, it gives me, a front row access to The Economist's internal editorial debates."
},
{
"end_time": 64.514,
"index": 2,
"start_time": 36.34,
"text": " Where senior editors argue through the news with world leaders and policy makers in twice weekly long format shows. Basically an extremely high quality podcast. Whether it's scientific innovation or shifting global politics, The Economist provides comprehensive coverage beyond headlines. As a toe listener, you get a special discount. Head over to economist.com slash TOE to subscribe. That's economist.com slash TOE for your discount."
},
{
"end_time": 93.524,
"index": 3,
"start_time": 66.305,
"text": " Tim Modelin is a professor at NYU, a visiting professor at Harvard and Carnegie Mellon, as well as an influential philosopher of science specializing in the metaphysical foundations of physics and logic. Today we discuss various interpretations of quantum mechanics, including Bohmian mechanics, which is also known as pilot wave theory. We talk about super determinism, as well as how quantum mechanics can be formulated without observers. Also reformulations of the foundations of physics without numbers,"
},
{
"end_time": 118.968,
"index": 4,
"start_time": 93.524,
"text": " and also how to place time as primal rather than a derived notion or rather than spatializing time, which is what's generally done. Next time with Tim, we'll also get to the liar's paradox and computation and consciousness. Write down your questions below because Tim will most likely be coming on again for a part two. Today's sponsor is Brilliant. If you're familiar with Toe, you're familiar with Brilliant, but for those who don't know, Brilliant is a place where you go to learn math, science, and engineering"
},
{
"end_time": 140.691,
"index": 5,
"start_time": 118.968,
"text": " Through these bite-sized interactive learning experiences, for example, and I keep saying this, I would like to do a podcast on information theory, particularly Chiara Marletal, which is David Deutsch's student, has a theory of everything that she puts forward called constructor theory, which is heavily contingent on information theory. So I took their course on random variable distributions and knowledge and uncertainty."
},
{
"end_time": 157.466,
"index": 6,
"start_time": 141.015,
"text": " in order to learn a bit more about entropy. Now there's this formula for entropy, essentially hammered into you as an undergraduate, which seems to have fallen from the sky. However, when you take Brilliant's course, it was the first time that I could see that it's an extremely clear and intuitive formula. That is to say that"
},
{
"end_time": 181.015,
"index": 7,
"start_time": 157.466,
"text": " It would be unnatural to define it in any other manner. Visit brilliant.org slash toe, that is T-O-E, to get 20% off the annual subscription, and I recommend that you don't stop before four lessons. I think you'll be greatly surprised at the ease at which you can now comprehend subjects you previously had a difficult time grokking. At some point, I'll also go through the courses and give a recommendation in order. Thank you and enjoy."
},
{
"end_time": 205.009,
"index": 8,
"start_time": 181.015,
"text": " Professor, I should let you know that I'm extremely, extremely excited to speak with you. You're one of the brightest minds and most clear minds, and those don't go hand in hand usually. Someone else that has those two qualities is John Bias, who's down to earth but extremely bright at the same time. You learn a lot reading his stuff, yeah. Usually when I interview someone, they have one or two distinctive ideas, whereas for you, I feel like I flip through your papers and there's such a"
},
{
"end_time": 235.009,
"index": 9,
"start_time": 205.606,
"text": " A multitude is like Douglas Hofstadter. I'm going to have to use this as an introductory and hopefully you'll come back on to speak about all of them. What is it about your work that's most exciting? Parenthetically, you can talk about what that work is. Well, because I'm in a philosophy department, I have this amazing freedom to pretty much look into whatever I feel like looking into, which is why if you look through my work, it covers"
},
{
"end_time": 262.824,
"index": 10,
"start_time": 235.401,
"text": " a variety of different fields. Most of my work has been in foundations of physics, although I have some other stuff in semantics, a paper on consciousness and various other philosophy of science topics. At the moment, what's most exciting is some work I've been doing for the last, gosh, five years."
},
{
"end_time": 289.189,
"index": 11,
"start_time": 263.217,
"text": " which is going in a direction I never really anticipated, which is working on discrete geometry and ways to understand space and time as having a discrete rather than continuous structure. And just within the last six months or so, I got to the point of actually having a structure"
},
{
"end_time": 315.52,
"index": 12,
"start_time": 290.794,
"text": " and doing some calculations and suddenly finding relativity kind of falling out into my lap from foundations literature about as unrelativistic as you could possibly imagine. So that to me has been very exciting. Whether it will go anywhere at the end of the day or not, I don't know. But if nothing else, it's an illustration that relativistic structure may emerge"
},
{
"end_time": 341.544,
"index": 13,
"start_time": 316.254,
"text": " in a very natural way from a foundation that looks completely unrelativistic. So that to me has been quite exciting. So that's what I've been doing most recently. I mean, there's some other things trying to get numbers out of physics. I don't like having numbers in physics, so I've been trying to get rid of them and I made some progress on that. A variety of things. Okay, you're going to have to elaborate on that last point."
},
{
"end_time": 372.5,
"index": 14,
"start_time": 342.995,
"text": " Well, okay, so I can say some things very simply. If you go back to ancient Greece, which is always a nice place to go back to, and ancient Greek mathematics was divided into two sub-disciplines, arithmetic and geometry, and arithmetic was the theory of numbers, and of course for the Greeks the numbers are, well, what we would normally say, what we would call the counting numbers, right, the positive integers, although in a way,"
},
{
"end_time": 390.35,
"index": 15,
"start_time": 373.473,
"text": " The numbers for the Greeks started with two, not with one, which is, as I like to point out, also the way we talk, right? Why? Because if someone says, do you have any pets? And I say, yes, I have a number of pets."
},
{
"end_time": 409.582,
"index": 16,
"start_time": 390.657,
"text": " The number is one. That's like a joke, right? If I say I have a number of pets, you think I have at least two. If I say I have a number of pets and the number is zero, then it's really a bad joke, right? Then you say that just doesn't count as a number of anything. And we don't normally count one as a number of things."
},
{
"end_time": 439.189,
"index": 17,
"start_time": 409.582,
"text": " So, you know, if you just follow that language, you'd say, yeah, when you're when you're talking about the numbers, you're talking about beginning with two. That's why one was very special for the Greeks. One. We all have a number of PhDs. Yeah, exactly. You know, so it's not so strange anyway. So there was number theory and then there was there was geometry, which was the theory of of continuous magnitudes of space, basically. I mean, and between the two of them,"
},
{
"end_time": 464.582,
"index": 18,
"start_time": 439.753,
"text": " The balance was very much on the side of geometry. I mean, you had all of Euclid, 13 books of Euclid, with amazing proofs. Okay. You know, proofs of things you would never imagine would even be true, much less provable. And Euclid, everybody knows Euclid, not that many people have heard of Nicomachus of Rhodes, who was sort of the top guy in arithmetic."
},
{
"end_time": 488.78,
"index": 19,
"start_time": 464.855,
"text": " And that was the balance for a couple millennia was that if you look at math, it was the queen of math was geometry and number theory was kind of lower down. And if you look at Newton and Galileo and so on, all of their proofs are basically use the tools of geometry."
},
{
"end_time": 512.824,
"index": 20,
"start_time": 490.247,
"text": " And what you notice is that if I tell you the physical world has a geometrical structure, that seems like a very natural thing to say. I mean, geometry was thought to be about the structure of space itself, and space itself is part of the physical world. If I tell you the world has a numerical structure, it's not quite"
},
{
"end_time": 538.831,
"index": 21,
"start_time": 513.302,
"text": " Immediately so clear what I could mean I mean people it's reported that the Pythagoreans who love numbers said things like a horse is the number 10 and you just go like what in the world are you talking about? How can a horse be the number 10? The connection between numbers I think it's a very good therapy"
},
{
"end_time": 564.019,
"index": 22,
"start_time": 539.872,
"text": " Whenever you're looking at mathematical physics and you've got algebraic structure, numerical structure, to ask yourself where it came from, because usually it's introduced by means of some convention, like by means of, say, putting down a coordinate system and assigning numbers as coordinates. And because that's not intrinsic to the subject,"
},
{
"end_time": 594.514,
"index": 23,
"start_time": 564.804,
"text": " There are an infinite number of ways to do that, and then you have to be very careful about what in your representation is a mere artifact of the conventions and what corresponds to something that's really there. This is the kind of thing Einstein says. I mean, the reason it took him 10 years to get from special relativity to general relativity, and he says this, I think, in the autobiography, in the Shope volume, is that he kept thinking that the"
},
{
"end_time": 622.944,
"index": 24,
"start_time": 595.026,
"text": " coordinates had to have some physical significance. And it took them a long time to understand that they don't. And in order to have a mathematical structure that's capable of handling curved spacetime, it has to be, there are no special coordinate systems, right? In a flat space, Cartesian coordinates are special, and in a flat spacetime, Lorentz coordinates are special."
},
{
"end_time": 648.797,
"index": 25,
"start_time": 623.66,
"text": " And you tend to only use them and they're very convenient. In a curved space, in a variably curved space, there are no special coordinate systems. And you're really confronted with the fact that the coordinate values don't mean anything by themselves. And then you have to think hard about what does mean something, right? What are the quantities that really characterize the space itself? So, you know,"
},
{
"end_time": 675.503,
"index": 26,
"start_time": 649.65,
"text": " There's an interesting history here. It starts actually with Descartes among people in La Geométrie, where he introduces not what we would call Cartesian coordinates, but he's very clear because... I'm sorry to be going on so long, but you started... No, it's fine. Please. You know, it was known from the Middle Ages, and especially the Arabic scholars, that if you're doing number theory,"
},
{
"end_time": 703.114,
"index": 27,
"start_time": 676.118,
"text": " Then you can use algebraic methods to solve problems and that the algebraic methods were really powerful. And so what Descartes asked himself is, huh, I wonder if there's a way that I can somehow use algebra to solve geometry problems. And what he was interested in was doing that. He's very clear about what he's doing. And in fact,"
},
{
"end_time": 721.715,
"index": 28,
"start_time": 703.541,
"text": " One of the things he realizes is that in numbers, when you're dealing with numbers, one is a special number, right? It's the only number that squares to itself besides zero. It's the only number that when you multiply it by anything else, you get the same thing back. So one is just picked out."
},
{
"end_time": 749.377,
"index": 29,
"start_time": 723.063,
"text": " from the algebraic structure as really special. But when you're doing geometry, there's no unit that's special. There's no length that's special. So Descartes realized that in order to convert geometrical problems into algebraic ones, he had to arbitrarily choose a unit, which is really important. Is this overcome with affine spaces rather than just a vector space?"
},
{
"end_time": 777.602,
"index": 30,
"start_time": 749.838,
"text": " Well, look, he's dealing in a continuum. I mean, his interest in geometry are in a continuum, and no length is special in a continuum. I mean, you've got all these things. The other thing that is worth noting is that if you're dealing with numbers, part of what makes something a number field, why we even call a set of objects numbers at all, is that you can multiply them. And when I multiply two numbers, I get another number."
},
{
"end_time": 800.213,
"index": 31,
"start_time": 778.78,
"text": " but there's no there's no corresponding geometrical operation if i give you two magnitudes if i give you two line segments say there's no operation corresponding to multiplication that will hand me back another line segment the best you can do in a euclidean space you can make an area right you can make a rectangle"
},
{
"end_time": 816.766,
"index": 32,
"start_time": 800.913,
"text": " But the rectangle is a different kind of thing. I mean the units in which you characterize it are different than the lengths. So there's no natural object that corresponds to multiplication in geometry."
},
{
"end_time": 834.155,
"index": 33,
"start_time": 817.517,
"text": " But we use multiplication all the time in doing physics because we've converted everything into numerical representations, right? Just to be clear, let me just see if I'm breaking it down correctly, if you don't mind. So when you say, hey, if we have two lengths or two line segments, it's not obvious how to multiply them."
},
{
"end_time": 853.541,
"index": 34,
"start_time": 834.155,
"text": " right so someone may say well can you add them you can add the lengths okay you can add the lengths yeah you can wait and then they may think well no i can multiply well you think you have some way of multiplying let me tell you this first line segment was actually the number one so your way of multiplying actually doesn't change it so you think hey i did some complex"
},
{
"end_time": 881.357,
"index": 35,
"start_time": 853.541,
"text": " and then I can just say no this first line was number two so that's why you say that there needs to be a unit picked out in order to multiply when you're thinking about line segments well it's not it's not just which comes first I mean certainly multiplication should be an operation that commutes it shouldn't matter what order I give them to you the real problem is the number one is the multiplicative unit so when I multiply any other number by one I get the same number back again"
},
{
"end_time": 894.753,
"index": 36,
"start_time": 882.534,
"text": " and there's no geometrical length that's picked out as special. That's why Descartes says in order to convert"
},
{
"end_time": 922.125,
"index": 37,
"start_time": 895.128,
"text": " a geometrical problem into an algebraic one, I have to make an arbitrary choice of unit. I'm going to say, oh, this length counts as one. If I do that, then I can define something that you would call multiplication, right? Now, suppose I have this special length that I've just picked out of the air. Oh, that's length one. Now I give you two other lengths, a and b, and then you make a rectangle out of a and b."
},
{
"end_time": 947.671,
"index": 38,
"start_time": 922.79,
"text": " which has an area. And now you say, well, if I want another rectangle with the same area, but one of the sides is the unit, this arbitrary unit, how long does the other side have to be? And then that will give me a length. But it's only relative to the choice of that unit. If I choose a different unit, I'll get a different one. So how are you going to create all of this"
},
{
"end_time": 976.971,
"index": 39,
"start_time": 948.251,
"text": " richness of standard multiplication and addition using geometry? You don't try to replicate. The thing about multiplication is that again it's a function that takes two numbers as input and gives you a number back as output. So I can just iterate and multiply and re-multiply and cube and so on. The proper operation in geometry if I give you two lengths"
},
{
"end_time": 1006.817,
"index": 40,
"start_time": 977.483,
"text": " Is it, I can construct something out of it, but what I construct out of it is not a length, it's a surface. And if I give you a surface and another length, I can construct a volume. And if I only have three dimensions, I'm done, right? Because if you now give me another length, there's nothing I can do. If it's a four dimensional space, okay, I could make a hyper volume. And if it's five dimensional space and so on. But a lot of what happens in physics,"
},
{
"end_time": 1035.913,
"index": 41,
"start_time": 1007.534,
"text": " You don't go up above the dimensionality of the space. And so basically what I've been playing with, which is, I think, I don't know if it's an original idea, but it's certainly not one that's been much investigated, is something I call full discrete geometry. And the full means you have zero dimensional things points. You have one dimensional atomic things. You can think of the lines. You have two dimensional atomic things."
},
{
"end_time": 1059.753,
"index": 42,
"start_time": 1036.271,
"text": " Atomic surfaces, you have three dimensional atomic things, atomic volumes. And if it's a three dimensional space, then you've got those things to play with. And so you can't just keep, whereas if you have numbers, again, I can just take a number and multiply it by self and then again and then again and then again and then again as many times as I want and I'll still have a number. I never, you know, the multiplication operation"
},
{
"end_time": 1089.138,
"index": 43,
"start_time": 1060.572,
"text": " can be is a binary operation but it extended to any to multiplying any any number of numbers right any finite number there's no way that when you multiply outside of the dimension so it should jump out but instead it goes back to the beginning no no it's just not it's just not a defined operation because the operation again the the the operation is one that for example can take two one-dimensional elements and give you back a two-dimensional element"
},
{
"end_time": 1119.309,
"index": 44,
"start_time": 1089.599,
"text": " And it can take a one dimensional element and a two dimensional element and give you back a three dimensional element. But if you give it something else, it can't give you anything back because there are no four dimensional elements or five dimensional, right? So it just doesn't have, you know, it's not that it recycles back or anything. It's just not a defined operation. Now, when you say that there are no four dimensional nor five dimensional, what you mean is in the theory that you're constructing, what you would like would to be in this manifest image of three dimensional space, you want to get back to that rather than say,"
},
{
"end_time": 1145.981,
"index": 45,
"start_time": 1119.821,
"text": " something that's a bit more outrageous or unproven? The formal theory can handle four dimensional spaces, five dimensional spaces, a million dimensional spaces, 11 dimensional spaces if you want to do string theory. It's not that the formal apparatus is limited, but you make a decision, okay, I'm going to look at two dimensional"
},
{
"end_time": 1160.964,
"index": 46,
"start_time": 1146.869,
"text": " Think Verizon, the best 5G network is expensive? Think again. Bring in your AT&T or T-Mobile bill to a Verizon store today."
},
{
"end_time": 1185.213,
"index": 47,
"start_time": 1165.64,
"text": " Jokes aside, Verizon has the most ways to save on phones and plans where everyone in the family can choose their own plan and save. So bring in your bill to your local Miami Verizon store today and we'll give you a better deal."
},
{
"end_time": 1215.691,
"index": 48,
"start_time": 1187.142,
"text": " Let me just lay out some thoughts. So firstly, you make a reference to a number there by saying three-dimensional or four-dimensional. And this is strange because you're trying to build a world without numbers, but then we need a number in order to specify this world. So how does that work? So that works because the Greek numbers, the counting numbers, are perfectly fine if you've got something to count, right? It's not that I hate all numbers. I can say there are two chairs in the room because I'm just counting one, two."
},
{
"end_time": 1245.503,
"index": 49,
"start_time": 1215.691,
"text": " But what happened historically, of course, if you want to convert geometry into arithmetic, you need more than the counting numbers. And if you look back at the history of it, it's very interesting. The Greeks really only recognized officially the positive integers as numbers. Then you can push a little and try and convince people there are these things that you call rational numbers. And that had a little bit of resistance to it, but not too bad."
},
{
"end_time": 1263.592,
"index": 50,
"start_time": 1245.981,
"text": " Then, well, there are two things you have to do. You eventually have to go to the irrational numbers if you want to do Euclidean geometry. That had a lot of resistance to it, right? There are people who said, no, those don't make any sense because if I try and express them in decimal form, it goes on forever. It gets lost in infinity."
},
{
"end_time": 1291.834,
"index": 51,
"start_time": 1263.592,
"text": " and and then you also need the negative numbers which you know there were people who were very upset about that and you can see why right i mean you say look okay i understand what you mean when you say you have two sheep i understand what you mean even when you say you have zero sheep but what would it be to have negative two sheep right right um there there can't be less than zero so negative numbers and it's just in the record in the historical record there was a lot of resistance to negative numbers"
},
{
"end_time": 1318.609,
"index": 52,
"start_time": 1292.449,
"text": " so i don't have any problem with integers if you have something to count and and the counting measure in this discrete geometry you use counting measure because you have a finite number of points or a finite number of lines or a finite number of surfaces in a certain situation and you can just count them and the counting is important so it's not that i hate all numbers"
},
{
"end_time": 1345.435,
"index": 53,
"start_time": 1318.968,
"text": " I just, when you use numbers, you should ask yourself, okay, where are the numbers coming from? If they're coming from counting something discrete, okay, that's a good answer. If they're not, then you need a fancier story about why you're using them at all. Let's get quickly philosophical, if you don't mind. And then I want to get to your views on quantum mechanics. There's different theories of truth. So there's correspondence, coherence, and pragmatism are the main three."
},
{
"end_time": 1372.739,
"index": 54,
"start_time": 1347.073,
"text": " For the correspondence, this just made me think, well, in the correspondence theory of truth, could there be minus two because minus two wouldn't correspond to something of the world? But then it seems strange because many people who say, I believe in the correspondence theory of truth are perfectly fine using numbers, real numbers. I never thought about that. What does the correspondence theory of truth have to say about real numbers or even complex numbers? Okay."
},
{
"end_time": 1400.265,
"index": 55,
"start_time": 1373.097,
"text": " I think the difficulty you're having is that you're touching on an extremely, extremely difficult subject. And one about which I don't have much to say because it's too deep for me. And that's the ontology of mathematics. It's mathematical reality. So if I'm in some intuitive sense of correspondence theorist, I say, look, at least for certain sentences,"
},
{
"end_time": 1428.063,
"index": 56,
"start_time": 1401.288,
"text": " What makes that sentence either true or false is whether it correctly or incorrectly describes some non-linguistic thing, the subject matter. And for physics, the subject matter is the physical world, so we kind of know at least what the target is. But mathematical items, as Plato pointed out, are not physical items. So if you want to be basically a correspondence theorist, as I do,"
},
{
"end_time": 1455.435,
"index": 57,
"start_time": 1428.319,
"text": " You're naturally going to be a Platonist in mathematics, which I am, right? That is, I think they're mathematical truths. And if you ask what makes them true, the answer is, well, mathematical reality. And if you ask me what mathematical reality is, I'm going to say, well, it isn't physical reality for sure. It has its own kind of existence, very much the way Plato would say. And there are facts about it. And some of the facts"
},
{
"end_time": 1483.029,
"index": 58,
"start_time": 1455.845,
"text": " We can discover, and some of them will never discover, as Gödel assures us. Okay, that's fine. There's a world that has more facts about it than we're able to. It's kind of amazing we can discover anything about it. I shouldn't be very surprised I can't discover everything about it. So what you're really asking, from my point of view, is not about the physical world at all. It's about the nature of purely mathematical reality."
},
{
"end_time": 1512.125,
"index": 59,
"start_time": 1483.592,
"text": " And again, I very firmly believe there is such a thing, and it's independent of physical reality, right? I mean, physics could have come out all kinds of different ways, but however physics came out, it wouldn't change the mathematical facts at all. So physics is dependent on the mathematical reality, but not vice versa? Well, physics is not dependent on mathematical reality in a way. See, mathematical physics is a discipline that uses mathematical items"
},
{
"end_time": 1541.442,
"index": 60,
"start_time": 1512.688,
"text": " to attempt to describe the physical world. And the mathematical items exist. They are what they are, and they are what they are independently of what we think. And we can find out about them by, you know, in a math department, you can investigate the structure of the mathematical items. The deep question, which is always there, is why should these mathematical items serve as good representations of physical reality?"
},
{
"end_time": 1571.578,
"index": 61,
"start_time": 1542.671,
"text": " Now, this is where I would say, again, for many numbers, I'm puzzled. Why am I using numbers anyway? For certain geometrical things, again, purely abstract mathematical geometrical things, it seems to me perfectly plausible to say, well, they're good because the physical world literally has a geometrical structure. You know, Newton thought that space was a three-dimensional Euclidean. How do you three-dimensional Euclidean space structure?"
},
{
"end_time": 1600.93,
"index": 62,
"start_time": 1572.108,
"text": " And it could have had. It turns out not to have, but it could have had. And if it did have, then you'd perfectly well understand why Euclidean geometry, which remains a perfectly good mathematical investigation into Euclidean space, you would understand why it has so much utility for physics. Three dimensional Euclidean geometry, right? I mean, 10 dimensional Euclidean geometry exists just as well abstractly"
},
{
"end_time": 1618.507,
"index": 63,
"start_time": 1601.578,
"text": " When I think about Platonism, I think of it as being extremely mystically religious, and I don't scorn it"
},
{
"end_time": 1643.456,
"index": 64,
"start_time": 1618.814,
"text": " For being so, and I think Plato thought of the forms of something mystical as well, something related to God. I can ask your thoughts on that. But what I wanted to say was more about it's strange because you were saying that the physical reality doesn't depend on the mathematical. However, the fact that we can use it, it gives some visual analogy. It's as if we're these physical beings and we can reach into the closet of mathematics and pull out something and then use it as a tool here."
},
{
"end_time": 1670.742,
"index": 65,
"start_time": 1644.565,
"text": " It doesn't seem like the mathematics reaches into us and then uses it there, although perhaps it does and we don't know. So it seems like that's a world that we pull from. Is there an interaction between these two worlds? There's a huge turn in Plato, the turn between the early and the middle dialogues, which is indicated in Mino. And in Mino is where you have a kind of standard Socratic"
},
{
"end_time": 1683.933,
"index": 66,
"start_time": 1671.749,
"text": " dialogue cross-examining Nino and showing that he doesn't really know what virtue is, what Arete is. And then he gets all messed up and says, I have nothing more to say. And Socrates says, well, why don't we keep talking?"
},
{
"end_time": 1712.176,
"index": 67,
"start_time": 1684.514,
"text": " And Meno quite reasonably says, wait a minute, you tell me you don't know what virtue is. And I now don't know what virtue is. Why should we keep discovering? We're two ignoramuses. Why should we keep discussing with each other? And Socrates says, well, you know, and then he kind of indicates the theory of reminiscence. But the important point is he then brings out this mathematical question. If I have a square. Say two units on a side, so it's four square, its area is four square units."
},
{
"end_time": 1730.094,
"index": 68,
"start_time": 1712.773,
"text": " I want to build a double square that is one whose area is twice as big, eight units, along what line do I build it? So that's a perfectly good question and he brings this slave over who's not learned any geometry and the slave starts out"
},
{
"end_time": 1759.394,
"index": 69,
"start_time": 1730.828,
"text": " saying oh you double the side and then he points out they do some counting and say no that's not right that gives me 16 and he says well let's try three well that doesn't work that gives me nine and then the slave says i don't know right and then socrates draws in you know the diagonals and and again they just count and easily show that the square on the diagonal has twice the area now i don't think the thing in the world mystical about is the opposite of mystical right mystical suggests some kind of"
},
{
"end_time": 1784.497,
"index": 70,
"start_time": 1759.872,
"text": " inarticulable presentation that you believe without reason, right? And this is just the opposite. The reason you believe that the square on the diagonal has twice the area is you've got a proof, right? That's what Euclid gives you. He's not a mystic, right? Euclid's not a mystic. He's a guy who can prove things."
},
{
"end_time": 1808.695,
"index": 71,
"start_time": 1785.606,
"text": " The thing that interested Plato is, look, I can actually literally know things about geometrical figures, know them, have proofs of them, have demonstrations, know what the truth is. And he says in Timaeus, which is the closest he comes to giving us the physics, Timaeus does"
},
{
"end_time": 1837.517,
"index": 72,
"start_time": 1809.019,
"text": " He says, look, don't expect that kind of knowledge of the physical world. The best you can expect of the physical world is what he calls an echos logos, which is a plausible story, right? A likely account of plausible story. And I'll give you a plausible story, but I can't prove that it's right. So Plato just recognized correctly that we have much better epistemic access, more secure epistemic access,"
},
{
"end_time": 1867.619,
"index": 73,
"start_time": 1838.404,
"text": " to mathematics than we do to physics. And that's not because he's a mystic. It's the opposite. He's kind of a hyper rationalist. Okay, so two thoughts. And I don't know if this is apocryphal or not. But Pythagoras, at least reputedly said that there was something ill about the square root of two. No, he didn't. I'm sorry. Okay, so at least infamously, infamously, or reputedly. This is a this is a fake story. interested in the truth here. There's a wonderful book called The Mathematics of Plato's Academy."
},
{
"end_time": 1897.261,
"index": 74,
"start_time": 1868.012,
"text": " When we say there was an issue about the square root of two,"
},
{
"end_time": 1924.531,
"index": 75,
"start_time": 1897.91,
"text": " We mean by the square root of 2 a number. Let's go back again. 1.414... The Greeks did not recognize the existence of any such number. They didn't believe there was any number which if you multiplied it by itself you got 2. Why? Because they thought the numbers were the counting numbers and obviously you can't multiply any counting number by itself and get 2."
},
{
"end_time": 1948.166,
"index": 76,
"start_time": 1925.435,
"text": " Now what they did recognize by a very interesting geometrical proof is that if you have a square and you have its side and you have its diagonal, that those are incommensurable. That is, that the length of one cannot be expressed as a ratio in terms of the length of the other."
},
{
"end_time": 1978.2,
"index": 77,
"start_time": 1948.916,
"text": " Now that doesn't pick out the diagonal as the bad guy or the side as the square bad guy. Incommenciability of that sort is a relation between two lengths. It doesn't say, oh, the side is good and the diagonal is bad or anything like that. The ratio of lengths between the side and the diagonal does not correspond to the ratio between any two integers. That's what they knew. But it didn't seem to bother them."
},
{
"end_time": 2008.49,
"index": 78,
"start_time": 1979.394,
"text": " In fact, they thought it was quite interesting. There was a whole terminology about links that are allogon, which would be like these that can't be expressed in terms of ratios, and they had further subdivisions among them, and nobody was upset about it. It's just an interesting fact. What led you down this route of trying to make geometry primal, like underive elements of reality?"
},
{
"end_time": 2033.285,
"index": 79,
"start_time": 2009.531,
"text": " Just mostly the thought that I could understand, I could see how a geometrical representation could in principle more or less directly correspond to the structure of the world, that the physical world could literally have a geometrical structure. And I don't understand how the physical world could literally have an algebraic structure."
},
{
"end_time": 2057.568,
"index": 80,
"start_time": 2035.077,
"text": " So I like to get along as much as I can with geometry. This word geometry is interesting. When I was in university, I always thought of geometry as I forgot the exact name, but I took some course where it's almost like Euclidean geometry or it is Euclidean geometry and there was inversions. I had to do some problem with inversions. This is all I remember. What is that called? Analytic geometry. Is there some word for it?"
},
{
"end_time": 2082.739,
"index": 81,
"start_time": 2057.875,
"text": " Like planar geometry, something like that. I mean, I'm not exactly sure. Well, anyway, it's with what you construct with lines and circles. And then there's differential geometry. And to me, I see that as extremely algebraic. Firstly, it's also contingent on the real numbers. Differential geometry is basically what you use when the geometrical structure"
},
{
"end_time": 2112.841,
"index": 82,
"start_time": 2083.183,
"text": " changes from place to place. So in Euclidean geometry, you have all these nice symmetries, right? The structure of Euclidean space is the same everywhere, so there's kind of translational symmetries, and it's the same in all directions, so they're rotational symmetries, right? It's actually the same even under expansions and contractions, which is really unusual. That's not true in even a space of uniform curvature."
},
{
"end_time": 2141.732,
"index": 83,
"start_time": 2113.097,
"text": " So it has all these nice symmetries but you could be living in a world where the geometry just doesn't have these nice symmetries and then the geometrical structure changes from place to place and then you have to use differential geometry which just means that you characterize the geometry local in some little region and then in the next one and then in the next one with all these little overlapping maps that cover the different regions and you don't have a constraint that the geometry looks the same"
},
{
"end_time": 2168.234,
"index": 84,
"start_time": 2142.227,
"text": " in all the regions. Now, none of that suggests using numbers, and you don't need to use numbers. I mean, numbers tend to get in again with you when you introduce coordinates, but they're coordinate-free characterizations of all of this geometry. So, as I said, if the space is curved, then no coordinate systems will be particularly great"
},
{
"end_time": 2180.247,
"index": 85,
"start_time": 2168.626,
"text": " The nice thing about spaces of constant curvature and especially zero curvature is that there are these nice coordinate systems like Cartesian coordinates."
},
{
"end_time": 2209.616,
"index": 86,
"start_time": 2180.794,
"text": " So forgive me, because it's been a while since I've been in university. The way that I recall it is that you define a smooth structure by saying that so-and-so is infinitely differentiable. To define differentiability, you have to say that it goes from the reals to the reals. I'm sure you have some chart in between, it's like Rn, but somehow you have a curve, and that's R to the M, so R to the manifold, and then from the manifold down to the chart. So you require the reals here, then the reals on the chart, and then you say it's differentiable based on"
},
{
"end_time": 2239.07,
"index": 87,
"start_time": 2209.616,
"text": " I don't know if that's correct. Please forgive me if I'm incorrect. That's the normal way it's presented. I mean, what you said is perfectly accurate is the way it's taught, that you throw in functions from the space to the reals or the R to the N as part of the story. Yeah. And then you talk about whether a function from Rn to Rn is continuous and so on."
},
{
"end_time": 2268.609,
"index": 88,
"start_time": 2239.411,
"text": " By appealing to the structure of the real one that is the usual way it's presented it doesn't have to be done that way and It should be obvious that there's something quite artificial about it because if i just have a space in front of me Where's this function into the real numbers coming from well it only comes by somehow laying down coordinates which could be done in infinitely many different ways and choosing some"
},
{
"end_time": 2295.35,
"index": 89,
"start_time": 2269.053,
"text": " arbitrarily and what you would prefer from as a pure as a calculational matter what they're doing is fine because it's easier to calculate with numbers as a conceptual matter you're better off avoiding this use of numbers where you can because they're only introduced in a kind of artificial way"
},
{
"end_time": 2322.125,
"index": 90,
"start_time": 2296.732,
"text": " So yeah, I mean, your memory of that is perfectly correct, but people teaching math are not. I mean, especially unless you really go deep into the foundations of mathematics, they're teaching you a tool. And when you learn calculus and vector calculus and so on, they're giving you a tool for doing calculations and grinding out answers."
},
{
"end_time": 2350.93,
"index": 91,
"start_time": 2322.517,
"text": " that's a very effective tool. What happens if you're a philosopher like me is that it's not so interesting learning how to use the tool. I mean, when I was doing my PhD in history and philosophy of science, because I was specializing in physics, I had to do graduate level work in physics. And I took one course, and I remember a perfectly fine physics course,"
},
{
"end_time": 2377.108,
"index": 92,
"start_time": 2351.698,
"text": " Interesting, for my purposes, interesting things were said on day one, the first day, because it was on the first day that the professor was trying to set up, what are we doing? And then the rest of the course, which was like, how do you use Green's functions to solve differential equations? Okay, it didn't matter to me. I mean, it just it made no difference for my interests."
},
{
"end_time": 2403.797,
"index": 93,
"start_time": 2377.432,
"text": " Now, if I needed to get numbers, if I needed to actually grind numbers out, I really needed to understand all the rest of that course. But that's not what I do, right? I'm more interested in, well, why am I using numbers anyway? You know, not how exactly do I use them or how do I solve these equations, right? Only rarely, once or twice in my life have I had to actually solve an equation to make progress."
},
{
"end_time": 2430.23,
"index": 94,
"start_time": 2404.787,
"text": " Can you give me an instance of them, of the one or two times where you had to use an equation? Oh yeah, I mean, I remember this very well. My very first book, which is Quantum Nonlocality and Relativity, there's a chapter which is about, part of it is about trying to mimic"
},
{
"end_time": 2460.555,
"index": 95,
"start_time": 2431.903,
"text": " Non locality violations bells inequality through a certain trick which involves having particles or objects that you're observing that just don't show up disappear that never show up in the data okay and if you're very clever. You can use this. Gap in the data. So the data that does actually get recorded will look like it violates balance inequality even though something nothing non local is happening."
},
{
"end_time": 2487.312,
"index": 96,
"start_time": 2461.544,
"text": " And then there was a question that arose, which is, well, how efficient can you be? That is, you want to say, I can rule out this kind of trick if my detectors have a certain efficiency. If my detectors are 100% efficient, if every time I throw a particle at a detector, it registers, then I can't use this trick because the trick really depends upon there being these missing data."
},
{
"end_time": 2516.135,
"index": 97,
"start_time": 2488.114,
"text": " But then you say, okay, well, how exactly how much efficiency do I need before I can rule out this loophole? And to answer that, I had to actually solve a differential equation, which I offhand, I didn't quite know how to solve, I could write it down. And I was actually, I was in Croatia off the peach, I had a big yellow legal pad, and I had a little hand calculator."
},
{
"end_time": 2538.012,
"index": 98,
"start_time": 2516.493,
"text": " and I swear to God I sat there and and numerically simulated an integral by hand with this calculator until I got it close enough to say hey that looks like pi over four and then once I saw that I said oh yeah it is pi over four then I could work out right how it goes. How long ago was that? Oh that was"
},
{
"end_time": 2568.575,
"index": 99,
"start_time": 2538.592,
"text": " 30 years ago. Okay, so this is before Wolfram Alpha. Oh, yeah. Oh, yeah. This is this is before you were born. Okay. Oh, yeah. Okay. So just to be clear, efficiency, because I haven't heard that term efficiency Houston. I'm not an experimentalist. Yeah, the worst course in physics for me was the experimental part. I'm much more of a mathematical person. Well, let me see. Let me give an example. When we see the double slit experiment, you see these dots, dots, dots, dots, dots, dots, dots. Well, sometimes"
},
{
"end_time": 2592.295,
"index": 100,
"start_time": 2569.07,
"text": " You throw out a photon and it just doesn't register. Right. A dot doesn't click. Yeah. And then you think, okay, well, well, if it's not going to click, it's not going to click in some uniform manner. So the pattern that I eventually get this oscillating pattern, it is characteristic of the photon. But you're saying there are some ways of designing an experiment such that the missing data gives you a pattern that's actually not there."
},
{
"end_time": 2621.852,
"index": 101,
"start_time": 2592.671,
"text": " There's a way, if you're an experimentalist, there's just a straightforward question, how efficient are my detectors? I have a beam and I know how many particles on average are coming through this beam and then I know how many particles on average are being detected by my detectors. And there's going to be a gap, right? I mean, stuff gets lost. Nothing is 100% efficient. In my book, I describe a certain trick"
},
{
"end_time": 2647.022,
"index": 102,
"start_time": 2622.858,
"text": " It's a very conspiratorial trick. It's not anything that would be physically natural. I mean, the whole thing is kind of crazy, but you're just doing logically possible. There's a trick you can use if there's high enough inefficiency in the detectors so that the registered data would violate the Bell inequality"
},
{
"end_time": 2673.797,
"index": 103,
"start_time": 2647.466,
"text": " without anything non-local and the trick involves a very judicious and special use of cases where data disappears right where no data shows up so it never gets into the recorded data and you can do that i mean it's not conceptually it's not really hard to see how it works at all it's just"
},
{
"end_time": 2697.841,
"index": 104,
"start_time": 2674.735,
"text": " There's a price to pay. You're simulating, by use of this trick, sending information where really no information is being sent. But you're simulating it by use of this trick. But the more information you have to simulate sending, the more inefficient the thing has to become. You have to more and more often"
},
{
"end_time": 2727.551,
"index": 105,
"start_time": 2698.319,
"text": " Use this trick of saying, well, one of the particles is just going to refuse to answer, right? I mean, normally you say I take a particle and I send it through a spin device and it's either going to come out up or down. But what if the particle has a third option, which is to just go mute and not give you any answer at all and then get that run gets deleted from the data. And so there's just a question of what those numbers look like. What is the most"
},
{
"end_time": 2757.79,
"index": 106,
"start_time": 2728.217,
"text": " efficient way you can run that trick, because if your detectors then have higher efficiency than that, then you can say, well, they can't be doing it, right? It's a loophole. There were these kind of loopholes that people talked about in tests of Bell's inequality, and the detector inefficiency loophole was a big one for a long time. But generation after generation of experimentalists just did it better and better and better and got higher and higher efficiencies, and all those loopholes have now been closed."
},
{
"end_time": 2777.841,
"index": 107,
"start_time": 2758.524,
"text": " It's like the electrons are out to trick us. The physical story"
},
{
"end_time": 2803.916,
"index": 108,
"start_time": 2779.036,
"text": " It is not a natural one that anyone would ever come up with. You only come up with it if you're sitting here saying, I hate non-locality, I hate non-locality. I want some way to keep my physics local, yet consistent with the data that we have. And this is a trick, which as long as the data which you have, there were certain levels of detector inefficiencies,"
},
{
"end_time": 2833.422,
"index": 109,
"start_time": 2804.053,
"text": " It's a trick you could use. But there's never a reason to believe it's completely implausible, physically. There's no physical story behind it. It's just kind of looking, well, you know, this is possible. This is mathematically possible if I have enough inefficiency to play with. OK, now let's talk about quantum mechanics. What I noticed about you initially was I see you as bringing some sober reality to this whole realm of quantum confusion and mysticism."
},
{
"end_time": 2859.991,
"index": 110,
"start_time": 2833.814,
"text": " that generally sells in magazines, spookiness, strangeness, other abnormal qualities. And then Bell's inequality is generally sold as a trade-off between non-locality and statistical independence or realism. So firstly, is that the case? And can you tell the audience your view of what the heck is going on quantum mechanically? Okay, so yeah, let me"
},
{
"end_time": 2889.565,
"index": 111,
"start_time": 2861.152,
"text": " Let me start a little backwards here. This term realism has snuck into this discussion, and it really should just be expunged. Realism has nothing to do with anything here. In fact, from a philosophical point of view, scientific realism has to do with epistemology. It doesn't have to do with ontology, with what there exists. I mean, to say, gosh,"
},
{
"end_time": 2907.961,
"index": 112,
"start_time": 2891.015,
"text": " i can save i can save locality by just giving up realism the way people say it i often say oh yeah nothing actually exists but thank god it's local i mean it doesn't make any sense if you try and trace down people say oh bel made some assumption of realism he made no such assumption"
},
{
"end_time": 2935.964,
"index": 113,
"start_time": 2908.78,
"text": " There's no assumption made that you can just deny and then you're okay. The actual theorem of Bell, which I think is the most important and astonishing result, I mean there's the mathematical theorem and then the fact in the lab that these inequalities are violated, this is the most astonishing thing in the history of physics. And the theorem as the piece of mathematics, this is just a comment about it as mathematics,"
},
{
"end_time": 2963.046,
"index": 114,
"start_time": 2936.664,
"text": " Every theorem you're deriving something from some suppositions. There are two suppositions. There's one that's called Bell Locality. It's the way that he expresses mathematically what it would be for a theory to be local, relativistically local, meaning that there's no causation or effects that go faster than light. That all of the events"
},
{
"end_time": 2991.92,
"index": 115,
"start_time": 2963.643,
"text": " that can count as a cause of something happening lie in or on its past light cone, right? Nothing that would require going faster than light. And in addition, it will be relevant for where this goes. Nothing coming from the future either. I mean, Bell's locality condition also rules out reverse causation that the future could could somehow cause things in the present. So there's the locality condition"
},
{
"end_time": 3021.374,
"index": 116,
"start_time": 2993.114,
"text": " And then as a mathematical fact there's a second condition that he uses in the derivation which is called statistical independence. And from these two assumptions he derives an inequality and says okay if any theory satisfies locality and satisfies statistical independence then there are certain constraints on the sorts of correlations you could possibly see"
},
{
"end_time": 3051.032,
"index": 117,
"start_time": 3022.142,
"text": " in situations where Alice and Bob are in their labs doing experiments so far away from each other that light doesn't have time to get from one lab to the other. So there couldn't be any effects, if all effects had to go at most at the speed of light, there couldn't be any effects from what happens in Alice's lab on what happens in Bob's or vice versa. So that's the theorem."
},
{
"end_time": 3080.162,
"index": 118,
"start_time": 3051.8,
"text": " And then you go in and you see this inequality. I mean, first of all, you notice that quantum mechanics predicts violations of this inequality. And then you notice that you go in the lab and check and quantum mechanics, quantum mechanical predictions are correct. So you only had two suppositions from which this follows locality and statistical independence. So as a purely logical matter, you say, well, I have to at least deny one or the other. Now,"
},
{
"end_time": 3105.128,
"index": 119,
"start_time": 3081.084,
"text": " There are people who, for reasons I do not fathom, I just literally don't understand their reasons, are so deeply committed to locality that they're looking for any escape route. I mean, that's more or less what we were talking about with these loopholes, with these detection loopholes, right? They don't like the idea of non-locality and they're looking for any escape they can have."
},
{
"end_time": 3134.923,
"index": 120,
"start_time": 3106.067,
"text": " from accepting it. And these detection loopholes for many years were one until the experimentalists just closed those loopholes. They're gone. So the only other loopholes, they look at this and they say, well, I don't want to die locality. I guess I have to deny statistical independence. And the problem with denying statistical independence is that it's kind of crazy and it's crazy and conspiratorial. And on top of it, as I've said and other people have said, it undermines"
},
{
"end_time": 3157.637,
"index": 121,
"start_time": 3135.486,
"text": " All scientific method. I mean, if in order to get out of accepting non locality, you deny what's required assumptions that are required to do science. That's not a good deal. Right? That's a really bad deal. Speaking of that,"
},
{
"end_time": 3182.568,
"index": 122,
"start_time": 3158.217,
"text": " I sent you a video. I'm unsure if you had a chance to Yeah, yeah, I did. Okay, so as a preface for those who are watching the video is by Sabine Haassenfelder, and it's on super determinism. Either way, I'll give a brief description in the first 20 seconds or so. She says, some people think that if we do away with statistical independence, then all the science is undermined. And she says her attitude is no,"
},
{
"end_time": 3212.142,
"index": 123,
"start_time": 3183.422,
"text": " So, I'm curious to know your thoughts on that. Firstly, on super determinism and perhaps what her specific comments were. I mean, let me walk through this, although let me before, because I knew you were going to ask this, let me just point out to people, because it's much better than listening to me. If you're interested in this, be sure you take this book, John Bell's Speakable and Unspeakable in Quantum Mechanics, and very carefully read"
},
{
"end_time": 3236.186,
"index": 124,
"start_time": 3212.722,
"text": " Chapter 12, which is called Free Variables and Local Causality, because Bell nails it completely. And any questions you have, Bell is an extraordinarily precise writer."
},
{
"end_time": 3260.623,
"index": 125,
"start_time": 3237.568,
"text": " He's not so good, by the way, if you watch Sabina, she talks about some interview he gave, and he's not so good in interviews. He sometimes slips up, I would say, when he's talking off the top of his head, and says things which I think are just a bit sloppy and uncharacteristic, but when he writes, there are only a couple places in his entire book where I think he put a foot wrong in what he writes."
},
{
"end_time": 3276.374,
"index": 126,
"start_time": 3260.998,
"text": " That's everyone. If you're interested in this topic."
},
{
"end_time": 3303.968,
"index": 127,
"start_time": 3276.783,
"text": " Discussions with many people who want to talk about super determinism who've never read as far, you know, really have never read this particular thing, which is the thing that Bell wrote about. It's a free variables and local causality. And what I'm going to tell you is basically what Bell says. It has and let's just start off by saying this has nothing to do with free will. I mean, all this talk about free will has gotten into this discussion from the beginning. It was irrelevant."
},
{
"end_time": 3333.097,
"index": 128,
"start_time": 3304.633,
"text": " The term super determinism is bad because even as Sabina says you can't get more deterministic than deterministic. The issue is the mathematical assumption of statistical independence that's used in the theorem. Now what is that assumption? Let me just give you an example and why it's important and why I say if you deny it"
},
{
"end_time": 3359.531,
"index": 129,
"start_time": 3333.746,
"text": " empirical science kind of goes out the window. So suppose we're going to do a test and there are different experimental conditions. You might think of them as experimental in control or in our case, you're going to either set up a spin measuring device in the X direction or the Y direction or in some other direction. We have these different conditions."
},
{
"end_time": 3387.79,
"index": 130,
"start_time": 3360.043,
"text": " And I'm doing tests and I'm doing tests on a large number of items, a large number of rats or a large number of particles or whatever. And I'm looking at statistics. I'm looking at percentages of the ones I experiment on with these different conditions, whether they react this way or that way. And when I'm doing experiments on pairs on the correlations between the two sides. Okay. Now,"
},
{
"end_time": 3413.933,
"index": 131,
"start_time": 3388.404,
"text": " What i do if i want to do just to take the simple case everybody understands if i want to check whether smoking causes cancer in rats. I started with a big group of rats. And then i want to subdivide it into an experimental group in a control group. But i want to subdivide it in a way that there's no bias between the two groups that the two groups."
},
{
"end_time": 3443.814,
"index": 132,
"start_time": 3414.377,
"text": " before i run the experiment are statistically similar to each other okay now there's some ways i can check for that like if there are male rats and female rats i can just kind of count okay what percentage of male rats are in this group and what percentage are in this group and i can make sure by distributing them that they have the same percentages sure the thing is i want the groups to be similar in all respects even respects i'm completely unaware of i mean maybe there's some genes i don't know about"
},
{
"end_time": 3469.445,
"index": 133,
"start_time": 3444.718,
"text": " How do I do that? Well, I randomly assign the rats to the two groups. This word random is important here? The word random here is absolutely important. What the word random there means, it is the condition that Bell uses of statistical independence. It means that the sorting mechanism"
},
{
"end_time": 3490.947,
"index": 134,
"start_time": 3471.732,
"text": " Does not is not biased or does not respond or is not sensitive to the rats. So if I you can imagine flipping a coin, for example, I mean, that's a physical kind of what Bell would call a physical randomizer."
},
{
"end_time": 3513.814,
"index": 135,
"start_time": 3491.391,
"text": " There are other randomizers. I mean, he uses this example, which is very nice because it takes all issues about free will out of it. He says, look, you could use a pseudo random number generator, which works like this. Start at the millionth digit of pi. And then output a one or a zero, depending upon whether the next digit is even or odd. All right."
},
{
"end_time": 3538.66,
"index": 136,
"start_time": 3514.224,
"text": " So that'll give you a sequence of ones and zeros that is a pseudo random sequence. We call it random because it will pass every statistical test for randomness you can come up with. It'll be about the same number of ones and zeros, but there will be no observable pattern. And you could use that and say, okay, if a one comes up, then the rat goes this way. If a zero comes up, it goes that way."
},
{
"end_time": 3568.831,
"index": 137,
"start_time": 3540.265,
"text": " Now, why do I do that? Because if the sorting is independent, if whether the rat went this way or that way is independent of the nature of the rats, then with overwhelmingly high probability that's calculable, each subgroup will be statistically like the original group. If 60% of the rats had some unknown feature in the original group, then 60% in each of the subgroups."
},
{
"end_time": 3588.353,
"index": 138,
"start_time": 3569.787,
"text": " will have that feature. And therefore the subgroups will be like one another. The really important thing you want is you want the statistically the subgroups to be like one another. And that's why we use randomized procedures for making these groups."
},
{
"end_time": 3614.667,
"index": 139,
"start_time": 3588.78,
"text": " Now, if somebody says, now, suppose I do this experiment with the rats, I take this big group, I randomly divide them by one of these sorting mechanisms, right, by throwing dice or by this... Sure, sure. ...or whatever into these two groups, and then I expose one to smoke and I treat the other the same, except I don't expose it to smoke, and then I see a lot more cancer in this group, then we infer that the smoke caused cancer."
},
{
"end_time": 3634.872,
"index": 140,
"start_time": 3615.333,
"text": " Now if somebody says no no no really what's going on is this some of the rats were predisposed to get cancer no matter what. And other the rats weren't and it just turned out that the ones that were predisposed got sorted over into the ones into the smoke group and the ones that weren't got sorted over into the other group."
},
{
"end_time": 3664.735,
"index": 141,
"start_time": 3635.265,
"text": " And so the smoking had nothing to do with it. Now, this is the kind of argument that actual lawyers for the tobacco industry used to make when they were trying to argue that scientists had not that smoking causes cancer. And obviously, as a purely logical fact, yes, if if the sorting didn't work to make statistically similar subgroups, then that could be an explanation. But then you have to go on and say, I don't care how you sort sort any way you want."
},
{
"end_time": 3695.452,
"index": 142,
"start_time": 3665.486,
"text": " right? Sort based on on pi, sort based on the number of shares of stock sold on the shot stock exchange, sort based on anything you want, no matter how you do it, somehow, the cancer prone rats are going to go this way and the non cancer rats are going to go that way. Now, that will, as it were, as a purely logical matter, count as a as something that would make the prediction that this group will get more cancer than that group. But we would just say this is lunacy."
},
{
"end_time": 3722.756,
"index": 143,
"start_time": 3696.681,
"text": " Because you have no mechanism. You just don't like the result. You don't like the result, and you don't like what the result is telling you, so you're just grasping at straws. That's all that's going on here. That's what super determines. Somebody who just says, I'm just going to blankly deny statistical independence, no matter how you do this sorting, and in this case, you're doing these experiments"
},
{
"end_time": 3749.582,
"index": 144,
"start_time": 3723.08,
"text": " Alice is doing an experiment and Bob's doing an experiment. Each one is randomly choosing how to set their device. And it's there where the randomness comes in. How does the device get set? And they could be flipping coins or they could be running off pie or doing whatever the heck they want. And Bell's only assumption is that statistically, the group of particles that meet condition one"
},
{
"end_time": 3779.582,
"index": 145,
"start_time": 3750.435,
"text": " will be the same as the group of particles that meet condition two, where condition one and condition two are chosen by one of these pseudo random kind of processes. Now, Bell gives an even more extensive discussion of why this is true, but it's clear that when we say it would undermine science, experimental method depends upon, in all cases, using this. Now, I should say one more thing."
},
{
"end_time": 3806.869,
"index": 146,
"start_time": 3781.391,
"text": " There's another idea, which is retrocausation, which is a different idea. It's not just a blank denial of statistical independence. It says, oh, statistical independence fails because the nature of the particles when I made them is somehow affected by"
},
{
"end_time": 3835.503,
"index": 147,
"start_time": 3807.551,
"text": " At the time I made them, it's somehow affected by what experimental condition they would be exposed to in the future. So that's the future affecting the past. That's retrocausation. And all of this is to save locality? Well, that doesn't save locality. So the first point I want to make is that retrocausation violates Bell's locality condition."
},
{
"end_time": 3854.087,
"index": 148,
"start_time": 3835.828,
"text": " It would be kind of crazy to think look i have bob and alice and they're doing their experiments far apart and i absolutely will not believe that what alice does can have a causal effect on bob but i believe that stuff that happens to the future can have a causal effect on bob i mean that's that's even worse."
},
{
"end_time": 3880.64,
"index": 149,
"start_time": 3854.77,
"text": " And it's even worse because if you retro causation, then because you have causation from the, you certainly have causation from the past to the future. If you also have causation from the future to the past, then you're going to be involved in causal loops. And then it's very hard to even write down a coherent theory. I mean, Wheeler and Feynman tried to do it. It's not clear. They wrote down some equations. It's not clear if they have any solutions. If you just have non-locality,"
},
{
"end_time": 3903.49,
"index": 150,
"start_time": 3881.186,
"text": " The obvious ways to implement it will not involve any causal loops, you won't have that problem. But the fact is, to say I want to deny non-locality and therefore embrace retrocausation, well first of all, retrocausation violates Bell's locality condition, so you haven't gotten out of it anyway, and you've gotten yourself even to a worse problem."
},
{
"end_time": 3912.108,
"index": 151,
"start_time": 3904.326,
"text": " Forgive me because I'm uncertain here and I'll just fumble over my words and then you can perhaps cohere what I'm saying and say it better than I have."
},
{
"end_time": 3942.995,
"index": 152,
"start_time": 3912.995,
"text": " How far does this correlation have to be? So when I hear statistical independence, the opposite of statistical dependence, which I assume is a correlation. And it seems to me clear that even when we sort rats, if we try to do so randomly by picking the one millionth digit of pi and so on, going forward from there, that there is still some some dependence, it's extremely, extremely small. Now, how? Well, how do we choose the number one million? Well, okay, this person said, Okay, well, let's choose a random number generator to choose that one million or to choose some number of the digit of pi."
},
{
"end_time": 3962.09,
"index": 153,
"start_time": 3942.995,
"text": " Okay, well then when we press on the random number generator, why do we choose to press it five seconds from now versus 10 seconds from now because it would give you a different number. So essentially what I'm saying is that the world that we chose the rats from every single thing here is influencing our decision."
},
{
"end_time": 3987.278,
"index": 154,
"start_time": 3962.568,
"text": " So every single thing is in the sort of rats temperature is influencing our decision, the fact that the rats have a certain amount of hair is influences our decision and so on. So just just there's a there's a extremely tiny, extremely tiny amount 0.000000002. But I don't know, is that the fact that it's non zero? Does that have any implication? Or does it need to be non zero sufficiently large?"
},
{
"end_time": 4010.64,
"index": 155,
"start_time": 3987.892,
"text": " Okay, so let me try and address this. There are two parts of this, I think. Let me try and address both parts of it. Let me just take my simple case so everybody can follow. I have a big group of rats, and some percentage of them, say, have a certain gene. I don't even know about the gene, but you say 60% in the big group have it."
},
{
"end_time": 4029.292,
"index": 156,
"start_time": 4011.63,
"text": " And what i want to do is now partition this into two groups of rats and i want the two groups to be statistically like each other and therefore statistically like the original group which had sixty percent now if i just flip a coin. It just standard statistics."
},
{
"end_time": 4052.824,
"index": 157,
"start_time": 4029.838,
"text": " will tell you, all right, there'll be little fluctuation. It doesn't mean that you'll get exactly 60% in each of your subgroups, right? One subgroup might end up with 61% and the other subgroup with 59%. Those are just statistical fluctuations. Using standard statistical methods, you calculate the likelihood of these fluctuations. If you're worried about them,"
},
{
"end_time": 4069.684,
"index": 158,
"start_time": 4053.302,
"text": " Then you use a bigger group. The probability of the fluctuations is smaller and smaller without limit. You can't drive them to zero, but you can drive them below any epsilon."
},
{
"end_time": 4095.282,
"index": 159,
"start_time": 4070.418,
"text": " Just to be clear, what I'm wondering is, is the presence of any epsilon, does that undermine the... No, no, because the violations of Bell's inequality says a certain observed correlation in the lab cannot be greater than a given number. I see, I see. And the actual observed violation of that is so far, it's not close, right? It's not like"
},
{
"end_time": 4117.09,
"index": 160,
"start_time": 4095.623,
"text": " Gosh, it's just a tiny bit over. And if I just have a little epsilonic thing somewhere, no, no, no, the violation is radically far away from the limit. So if you say, oh, maybe the subgroups aren't exactly the same, of course, they're not exactly the same. But the chances of them being sufficiently different"
},
{
"end_time": 4144.019,
"index": 161,
"start_time": 4117.944,
"text": " to account for what's observed and furthermore of course what's observed is exactly what's predicted by quantum mechanics and if if what you were seeing was just the result of some random fluctuation then you wouldn't be able to predict it i mean that's the point about random fluctuations is you can't predict them at all but what you observe in the lab is precisely what you get out of a quantum mechanical calculation yeah there's one other thing i just want to if i can find it quickly from um"
},
{
"end_time": 4175.265,
"index": 162,
"start_time": 4146.493,
"text": " This thing of bells because it's relevant to what you said. Well, we could do this. We could do this. He says Actually, here's a relevant a relevant thing to what you just said He says He's talking about you've got some Things you treat as free variables in this case the choice of"
},
{
"end_time": 4189.65,
"index": 163,
"start_time": 4175.725,
"text": " of whether to do experiment a or experiment be the choice of allison bob nick he says of course there's an infinite infamous ambiguity here. About just what and where the free elements are."
},
{
"end_time": 4213.66,
"index": 164,
"start_time": 4190.247,
"text": " The fields of Stern-Gerlach magnets could be treated as external. That's then treating those as free variables. So I can just imagine those. I set them when I do the calculation however I want. Or fields and magnets could be included in the quantum mechanical system with external agents acting only on external knobs and switches. Or the external agents could be located in the brain of the experimenter."
},
{
"end_time": 4243.558,
"index": 165,
"start_time": 4213.66,
"text": " In the latter case, the setting of the instrument is not itself a free variable. It's only more or less closely correlated with one, depending on how accurately the experimenter affects his intention. As he puts out his hand to the knob, his hand may shake and may shake in a way influenced by the variables V. Remember, now this is now exactly to your point. This is Bell I'm reading, right? Remember, however, that the disagreement between locality and quantum mechanics is large, up to a factor of root two in certain sense."
},
{
"end_time": 4267.193,
"index": 166,
"start_time": 4244.053,
"text": " So some hand-trembling can be tolerated without much change in the conclusion. Quantification of this would require careful epsilonics, okay? And, I mean, there's another thing that he talks about, which I just... Alright, again, let me just jump to the end. He's talking about, this is the part that I wanted."
},
{
"end_time": 4286.578,
"index": 167,
"start_time": 4267.944,
"text": " Consider the extreme case. He's now talking about random generators, so we want something that's setting this equipment either this way or that way. For Alice, something that's setting it this way or that way. For Bob, it would be a bad idea to have it due to the, as it were, rim of the experimenter because the experimenter would have to be just sitting there"
},
{
"end_time": 4313.66,
"index": 168,
"start_time": 4287.108,
"text": " in a very boring way going oh now up now down and probably as we know people are not even good at creating random sequences right human beings are not good random number generators right right here's what he said he says consider the extreme case of a random generator which is in fact perfectly deterministic in nature and for simplicity perfectly isolated in such a device the complete final state right what it chooses"
},
{
"end_time": 4330.043,
"index": 169,
"start_time": 4315.094,
"text": " perfectly determines the complete initial state. Nothing is forgotten. And yet, for many purposes, such a device is precisely a forgetting machine. A particular output is the result of combining so many factors."
},
{
"end_time": 4356.937,
"index": 170,
"start_time": 4330.35,
"text": " of such a lengthy and complicated dynamical chain and this is going what you were saying well what if it they do it a little later if i have you know suppose i have like when they do lotteries they have these ping pong balls that are bouncing up and down in this cage okay that could be a deterministic system so that some set of causes account for why this ball came up but that set of causes is huge it involves all the other collisions with all the other balls and all this other stuff going on so he says"
},
{
"end_time": 4381.049,
"index": 171,
"start_time": 4357.483,
"text": " Yet for many purposes such a device is precisely a forgetting machine. A particular output is the result of combining so many factors of such lengthy and complicated dynamical chain that it is quite extraordinarily sensitive to minute variations of any one of many initial conditions. It is the familiar paradox of classical statistical mechanics"
},
{
"end_time": 4404.343,
"index": 172,
"start_time": 4381.493,
"text": " that such exquisite sensitivity to initial conditions is practically equivalent to complete forgetfulness of them. To illustrate this point, suppose that the choice between two possible outputs corresponding to A and A' depended on the oddness or evenness of the digit in the millionth decimal place of some input variable. Then fixing A or A' indeed fixes something about the input."
},
{
"end_time": 4430.503,
"index": 173,
"start_time": 4404.787,
"text": " i.e. whether the millionth digit is even or odd, but this peculiar piece of information is unlikely to be the vital piece for any other distinctively different purpose, i.e. it is otherwise rather useless. With a physical shuffling machine, we're unable to perform the analysis to the point of saying just what particular feature of the input is remembered in the output, but we can quite reasonably assume that it is not relevant for other purposes."
},
{
"end_time": 4460.23,
"index": 174,
"start_time": 4430.998,
"text": " In this sense, the output of such a device is indeed a sufficiently free variable for the purposes at hand. For this purpose, the assumption one, which is the statistical independence assumption in the proof, is then true enough and the theorem follows. Now, if you just read that paragraph carefully and understand it, it puts this whole thing about, I'm going to deny statistical independence to bed."
},
{
"end_time": 4487.363,
"index": 175,
"start_time": 4461.391,
"text": " I mean, this is just Bell. Okay, he understood this objection. He understood that logically, yes, he has two premises. So logically, you can deny either one. But then he says, look, denying statistical independence, that's just not physically the kind of thing a physicist could ever do and remain a physicist. Here's what's interesting. I respect you. I respect Sabin. I think you're both extremely bright people."
},
{
"end_time": 4516.681,
"index": 176,
"start_time": 4488.183,
"text": " Why is it then that extremely bright, respectable people believe in super determinism if it's so obviously incorrect? So here's one answer. Nima Arkani Hamed said that there's this mistake that the public has about physics, which is that almost every moment in time, we need some radically new paradigm that we have to be revolutionary. He said, no, what is required is conservative revolutionaryism or revolutionary conservatism, whichever. And what he meant by that is"
},
{
"end_time": 4544.138,
"index": 177,
"start_time": 4517.432,
"text": " Well, let's take a look at Einstein. People say that with special relativity, what he did was extremely revolutionary. He said, no, he was revolutionary about one aspect and was extremely conservative about others. He said, these are certain aspects that I want to hold on to and here's what I'm willing to give up. And then he came up with this theory. Do you see it as some people are extremely conservative about locality and they're willing to be extremely revolutionary in other respects, like let's dismiss with"
},
{
"end_time": 4567.671,
"index": 178,
"start_time": 4544.565,
"text": " Statistical independence. Do you see it as that sort of issue or is it something else? I think it's something else. Look, you have to remember physicists, philosophers, everybody are human beings and then human beings have their foibles. I'm going to illustrate this."
},
{
"end_time": 4596.049,
"index": 179,
"start_time": 4568.148,
"text": " One of the physicists who early on recognized and advertised Bell's result, because you have to bear in mind that if not for some kind of very lucky things happening, Bell's result might have gone entirely buried and nobody would have even known about it. I mean, it was published in the first volume of a brand new journal that went out of business within a couple years."
},
{
"end_time": 4625.828,
"index": 180,
"start_time": 4596.408,
"text": " So people wouldn't even know about the journal. It happened to fall in the hands of Abner Shimony, and Shimony read it and understood how important it was, and Shimony happened to know John Clauser, who was looking for an experimental project. Okay, there's a whole history here, but it's a very iffy thing, right? That Bell's result could have just completely disappeared. But one of the physicists who early on"
},
{
"end_time": 4649.633,
"index": 181,
"start_time": 4626.937,
"text": " understood the importance of Bell's result was David Merman, and Merman was a good popularizer and he wrote some articles. He also wrote articles that were published even in philosophy journals explaining what Bell did. And the original Bell result, the one that Bell proves,"
},
{
"end_time": 4675.23,
"index": 182,
"start_time": 4650.213,
"text": " is a statistical result it says if you there's a certain kind of experiment if you do it over and over and over and over again you then accumulate statistics about correlations between outcomes on these two sides and this bell inequality puts mathematical constraints on how strong those correlations can be and that's what's violated and merman said he always had a suspicion"
},
{
"end_time": 4697.688,
"index": 183,
"start_time": 4676.305,
"text": " that the probabilistic nature of these predictions was really essential. The fact that you were dealing not with strict 100% predictions but only percentage predictions was somehow very deep to this result. He said he could never quite articulate why."
},
{
"end_time": 4717.841,
"index": 184,
"start_time": 4699.002,
"text": " And then some years later, Greenberger, Horn, and Zeilinger came up with this very nice example. It's the one I use all the time now instead of Bell's original example, which instead of involving pairs of particles involves triples of particles. This is in my book if anybody wants to read it. It's a beautiful thing to be explained in five minutes."
},
{
"end_time": 4746.34,
"index": 185,
"start_time": 4718.302,
"text": " But the nice thing about it is that the predictions there are not statistical. They're 100% predictions. They say, according to quantum mechanics, you should get this kind of result every time. So if it even fails once, quantum mechanics has failed, unless they're experimental error. And when Merman saw that,"
},
{
"end_time": 4774.275,
"index": 186,
"start_time": 4747.09,
"text": " he appreciated he wrote a very nice article about it and he said at the end i used to think that the probabilities here were really important i was wrong okay i think you know this just shows i was wrong because here you have a locality result that doesn't involve probabilities they're all one in zero and you know you can't ask for more than that from somebody okay he had his suspicions he stated them publicly when he got"
},
{
"end_time": 4803.285,
"index": 187,
"start_time": 4774.582,
"text": " Reason to see that those were wrong. He retracted them publicly and he wasn't upset about it. I mean, why would you be upset about it? I mean, I often say the nicest feeling you can have when you fall asleep at night is that you understand something you didn't understand in the morning or you've corrected an error in your thinking. I mean, people don't like to admit errors to different degrees. Right. And that's just human beings. Right. I mean, that's true of all human beings. You have to"
},
{
"end_time": 4831.869,
"index": 188,
"start_time": 4804.138,
"text": " Try to get familiar enough with at least some of the content that you can really make an independent judgment of your own about who's right and who's wrong. And then start more trusting the people who are right and being more leery about the people who are wrong. There's nothing else you can do. It's unfortunate. It's hard work, right? And anybody trying to pick up"
},
{
"end_time": 4854.155,
"index": 189,
"start_time": 4832.329,
"text": " physics from the kind of public presentation is not going to be in a position very well to do that. As I say, if you're interested in super determinism, get Bell's book, read chapter, it's only four pages, read chapter 12, read it carefully and think, think for yourself."
},
{
"end_time": 4877.125,
"index": 190,
"start_time": 4855.162,
"text": " And ask yourself whether you think denying the statistical independence assumption that he uses is something that has any physical plausibility. And if you're wondering, that's the best thing to do. I'm not asking you to take it on my word."
},
{
"end_time": 4904.292,
"index": 191,
"start_time": 4877.637,
"text": " Sure, sure, sure. So let's talk about the PVR theorem. Do you see it as one of the landmark theorems on par with Bell or second or even perhaps greater than Bell's? I would say I think it's in a certain way as important as Bell's result, but not as surprising. Not as surprising given Bell that we already know about? No, no, no, not as surprising even before Bell."
},
{
"end_time": 4921.459,
"index": 192,
"start_time": 4904.974,
"text": " Let me say why. The PBR result says we use this thing, the wave function, this mathematical object at the center of the mathematical apparatus of quantum mechanics. Everybody uses it to calculate."
},
{
"end_time": 4951.937,
"index": 193,
"start_time": 4922.995,
"text": " And again, one question you should ask yourself is, why is that? Why are we using this thing? One possible answer is, well, because it's a pretty good representation of some actual physical reality in an individual system. It represents some physical characteristic of the individual system in front of me and a relevant physical characteristic."
},
{
"end_time": 4963.916,
"index": 194,
"start_time": 4952.654,
"text": " Now, even apart from Bell, if you ask, well, should I take this wave function seriously in that sense, or should I rather think"
},
{
"end_time": 4992.193,
"index": 195,
"start_time": 4964.292,
"text": " it's something merely epistemic that so the wave function is kind of for a single particle let's just talk about a single particle because it's easier although it's a little misleading single particle the wave function is defined over all of space so you can sort of think like a field and it's spread out and and the schrodinger equation tells you how it evolves and it evolves by a wave equation good okay is there anything in it if i send a single particle through a two slit experiment"
},
{
"end_time": 5016.527,
"index": 196,
"start_time": 4994.548,
"text": " I represent that single particle in part using a wave function and the wave function is spread out and the wave equation carries that wave function through both slits and from there each slit produces its own, as it were, little wave and then they interfere like water waves and you get this interference pattern."
},
{
"end_time": 5044.462,
"index": 197,
"start_time": 5017.159,
"text": " Physically what we see if I do this two-sit experiment over and over again with individual single particles is what shows up on the screen is this interference pattern with these bands through it, light and dark bands. Now as soon as you see that, you say, gosh, okay, well that wave function must really represent something physical that does interact with both slits. That's how interference works."
},
{
"end_time": 5067.108,
"index": 198,
"start_time": 5044.923,
"text": " Right? If each individual particle merely went through one slit and nothing interacted with the other slit, then you couldn't possibly get this kind of interference. So the idea that you should take the wave function seriously as representing a real physical feature of individual systems"
},
{
"end_time": 5096.988,
"index": 199,
"start_time": 5067.91,
"text": " So nothing to do with anybody's knowledge, nothing to do with epistemology, nothing to do with statistical characteristics of large groups of systems. It represents some physical reality that pertains to an individual system. That's just from two-slit interference. Everybody should believe that. And everybody I know does believe it. This was not something the people that I know that work in foundations, nobody ever questioned that."
},
{
"end_time": 5117.295,
"index": 200,
"start_time": 5097.449,
"text": " Because you just say just the two, now two slit interference does not violate Bell's inequality. So this is not non locality. This is a different thing. And this is again another place where, you know, Feynman makes a mistake when he says, well, he wrote this before Bell, but he says all the mysteries of quantum mechanics are somehow packaged into two slit. No, they're not because Bell is"
},
{
"end_time": 5143.012,
"index": 201,
"start_time": 5118.677,
"text": " But if you just think about those kinds of familiar interference experiments, done with single particles and where you accumulate the data over time, anybody looking at that is going to say, gosh, that wave function must represent a real physical characteristic of individual systems. Now, what the PBR theorem does is it gives you something even sharper."
},
{
"end_time": 5170.776,
"index": 202,
"start_time": 5143.985,
"text": " It says if despite all that, you want to think that there's something epistemic about the wave function, that it doesn't really represent physical reality of the individual system, it somehow reflects your information or something like that, then the PBR... This is also called cubism or no? Well, that's sort of what the cubists say. Cubism is a very difficult doctrine to even understand what it's claiming. Let's leave it aside for the moment."
},
{
"end_time": 5191.476,
"index": 203,
"start_time": 5171.118,
"text": " But someone, you know, a common idea was that no, the wave function is kind of spread out, not because anything physical was spread out, but because we're ignorant about where something is, right? And the spread outness reflects us not knowing where it is, not that there is something in these different regions."
},
{
"end_time": 5220.145,
"index": 204,
"start_time": 5192.073,
"text": " So I think anybody who just thought about two slit would say, no, that can't be right. It can't be merely epistemic. And again, David Merman makes this point in one article, he made the point very nice with a dialogue, a kind of dialogue with somebody who's trying to be epistemic. And they say, oh, you know, the wave function spreading out is just a matter of our ignorance. And then in response, somebody says, so it's our ignorance that goes through the two slits."
},
{
"end_time": 5250.947,
"index": 205,
"start_time": 5221.101,
"text": " Right. And that makes no sense, right? I mean, something physical is interacting with both slits. It has to be. So what PBR did was they took the observation that I made, which is not a proof, and they gave a sharp definition of what it is to regard the wave function as representing something real about the individual system as opposed to"
},
{
"end_time": 5274.309,
"index": 206,
"start_time": 5251.305,
"text": " representing something about merely our knowledge of the system. And then they said, well, if you believe it's epistemic, by our definition, here's an experiment that you'll make the wrong predictions about. I mean, you'll make predictions, you'll have to make predictions that violate quantum mechanical predictions."
},
{
"end_time": 5296.459,
"index": 207,
"start_time": 5274.735,
"text": " And by the way, they also use statistical independence assumption in their proof, but that's fine because the statistical independence function is fine. They use one as well. And it's a nice proof. I mean, and people who reacted at the time who said this is the most important proof since Bell, I think David Wallace said that. I think he's right. I think if you're doing foundations,"
},
{
"end_time": 5322.807,
"index": 208,
"start_time": 5297.073,
"text": " it just i mean you could say it was the last nail in the coffin and the coffin was already shut that you shouldn't have been an epistemicist in the first place for you know straightforward physical reasons right how do you explain these interference bands um but if somebody again wanted to be very recalcitrant and and and stubborn and say no no no i just really think there's something epistemic about"
},
{
"end_time": 5339.343,
"index": 209,
"start_time": 5323.148,
"text": " This wave function, the PVR theorem kind of puts that to rest and so you can move on. So yeah, I think it's a very important. I think it's an extremely important there. I think as I say, it's not as astonishing in what it's claiming as Bell."
},
{
"end_time": 5367.005,
"index": 210,
"start_time": 5339.804,
"text": " There is a word, a phrase that you use called foundations of quantum mechanics or foundations, but that's the standard for foundations of quantum mechanics. Infamously, when you go to a physics course, if you get a physics degree and you ask about, well, what does this so and so mean, especially when it comes to quantum field theory, quantum mechanics, they say, hey, go to a philosophy department. Essentially, what I'm asking is what textbooks should one read? Let's say they're at the graduate level in order to become familiar with quantum foundations."
},
{
"end_time": 5381.493,
"index": 211,
"start_time": 5367.005,
"text": " Is there a set of five books that you feel like, hey, read these in this order that gives you a great introduction? Well, I mean, the thing I was once did a thing at the end, they ask you name five books in your field that you would recommend."
},
{
"end_time": 5410.913,
"index": 212,
"start_time": 5382.295,
"text": " And I said, and I was only half joking, I would say, I said, okay, get this book and read it five times. I mean, you know, this is a collection of articles. There are different levels, but it's not a single presentation. But Bell is really on target. Really, really beautifully on target. You have to be careful when you read it because he's so precise in the way he writes that he'll say something once and he won't beat you over the head with it a hundred times and you have to pay attention."
},
{
"end_time": 5439.991,
"index": 213,
"start_time": 5411.237,
"text": " I beat people over the head. My technique, as I say, is to use a two by four and whack you over the head a hundred times and say the same thing. Bell's using a scalpel and sort of just like that. It's like Dirac. Yeah. I mean, it's just, I mean, it's, it's beautiful. And he's really, again, if you're worried about super determinism, just read this four page paper of Bell's. It's all there. So you should read that. If you're, if you're used to a physics, if you're a physics major and you're used to a physics presentation,"
},
{
"end_time": 5469.531,
"index": 214,
"start_time": 5440.435,
"text": " There's this nice book by Travis Norsen that came out, I don't know, five years ago. I think it's called Foundations of Quantum Mechanics, but it's really written for physics students as an introduction to quantum theory, but with attention to the historical background and conceptual detail that's not in a normal physics book."
},
{
"end_time": 5496.596,
"index": 215,
"start_time": 5470.333,
"text": " And he goes over interpretations as well? Yeah. Then, you know, David Albert, who people may not know, got his degree in theoretical physics from Rockefeller University, but was interested in foundations and therefore switched to a philosophy department. So he's in the philosophy department at Columbia. You know, he wrote a book, which is more philosophical and a little more out there in certain ways called quantum mechanics and experience."
},
{
"end_time": 5521.886,
"index": 216,
"start_time": 5497.227,
"text": " And it's not like Norsen's book, which is aimed at physics students. David's trying to take as much math out of it as possible and kind of look at the conceptual issues. I mean, you have my book, which is Philosophy of Physics, Quantum Theory, which is more math than David has and less math than Travis has."
},
{
"end_time": 5545.418,
"index": 217,
"start_time": 5522.21,
"text": " Alright, it's in between the two. And every one of the books you mentioned, for those who are watching, that'll be in the description. I think, you know, between those four books, you certainly have everything you need. And the book by you, what was it called? Philosophy of Physics, Colon, Quantum Theory."
},
{
"end_time": 5574.189,
"index": 218,
"start_time": 5545.981,
"text": " There's a little set of books. It's a series of books that are supposed to be introductory books for philosophers to different issues. And I did the philosophy of physics one, but there are two volumes. One is space and time, which is an introduction to space time theory and relativity. And the other is quantum theory, which is about quantum theory. And someday I'm going to do a third part or put them all together and do something on statistical explanation and thermodynamics and entropy and stuff. But that's somewhere in the future."
},
{
"end_time": 5604.002,
"index": 219,
"start_time": 5574.889,
"text": " Hi, I'm here to pick up my son Milo. There's no Milo here. Who picked up my son from his school? Streaming only on Peacock. I'm gonna need the name of everyone that could have a connection. You don't understand. It was just the five of us. So this was all planned? What are you gonna do? I will do whatever it takes to get my son back. I honestly didn't see this coming. These nice people killing each other. All Her Fault, a new series streaming now only on Peacock."
},
{
"end_time": 5633.2,
"index": 220,
"start_time": 5606.032,
"text": " What the heck is a be-able? A be-able. Be-able. Yeah, so what the heck is a be-able? And why did Bell invent this terminology? Because to me, as an outsider, it just sounds like something that exists, some ontological object. That's all it is. But that's, you see, the word ontology, I can testify from personal experience, it scares physicists, right? You say, what's the ontology of their theory? And they're like,"
},
{
"end_time": 5652.176,
"index": 221,
"start_time": 5633.677,
"text": " Ontology, because they didn't learn Greek, right? And ontology just means in Greek, ta onta are the things that exist, and logos is an account. So the ontology of your theory is, tell me what your theory postulates to exist. And"
},
{
"end_time": 5678.985,
"index": 222,
"start_time": 5652.944,
"text": " Bell wanted a word for the same thing. Now, he could have used ontology, but he didn't. I don't know why. I mean, he doesn't have a philosophy background. Maybe it didn't occur to him. And he even says in a paper, he says in the theory of local beables, he said, well, yeah, he wanted a word whose root was to be or to exist as opposed to observable. I mean, what he was doing was setting a contrast with the term observable."
},
{
"end_time": 5705.265,
"index": 223,
"start_time": 5679.77,
"text": " Because so many physicists would say, oh, physics is about observables, observables, observables. And Bell said, no, no, no, it should be about be-ables, things that exist. I mean, he said he considered be-er, B-E hyphen E-R, but then that looks like beer, right? And then on the model of observable, he put be-able,"
},
{
"end_time": 5728.592,
"index": 224,
"start_time": 5706.169,
"text": " I would just call it the ontology of the theory, but philosophers would just call it the ontology of the theory and physicists would just get used to it. But you can call it the beables, it's fine because Bell used that. It's just what is your physical theory postulate to physically exist? Now this is a question which should be the first question, it's not a philosophical question."
},
{
"end_time": 5754.77,
"index": 225,
"start_time": 5729.241,
"text": " It's not, oh, philosophers are interested in that. No, physicists, right? That's what your physics is actually postulating to physically exist, right? It's the first question you should ask of any physical theory if you want to understand it. What does it postulate to actually exist? And unfortunately, because foundations of physics is not part of a standard physics education, which it isn't,"
},
{
"end_time": 5774.957,
"index": 226,
"start_time": 5755.299,
"text": " This is get very nervous around questions like this and sometimes they they they suggest answers that don't kind of make any sense so there's a physicist who i will remain unnamed who was asked this question at one of our conferences. What's that what you know what are the vehicles of your theory what is your theory possibly and came up with the answer matrices."
},
{
"end_time": 5794.65,
"index": 227,
"start_time": 5775.452,
"text": " No, that's not even in the right ballpark. A matrix is a mathematical object. A matrix is a set of numbers in an array. You're not going to tell me the physical world is made up of sets of numbers in an array."
},
{
"end_time": 5812.483,
"index": 228,
"start_time": 5795.094,
"text": " The world just has a mathematical reality and"
},
{
"end_time": 5831.305,
"index": 229,
"start_time": 5813.183,
"text": " The physical world has no reality that's any different than any other mathematical object. So every mathematical object is physically, I mean, again, there's a certain point at which quite honestly, you just say, I don't really want to spend my time with this."
},
{
"end_time": 5862.295,
"index": 230,
"start_time": 5832.346,
"text": " It's just, you know, no, there's a physical world, right? It's true that we can make really good predictions about it using some mathematical representations and that's a good thing to think about and why those mathematical representations are so effective. That's a really good thing to think. But what you're thinking about when you think about that is, okay, what might the physical world be like such that this mathematics makes such good predictions?"
},
{
"end_time": 5884.462,
"index": 231,
"start_time": 5863.012,
"text": " And that's why I said that's why I like geometry because one in the case of geometry. At least a possible answer is this abstract geometry is such a good representation of the physical world because the world literally has that geometrical structure, right?"
},
{
"end_time": 5909.462,
"index": 232,
"start_time": 5885.128,
"text": " then of course it's going to be pretty good at making predictions because you've got it right you know you've got a good representation of the way it really is and in the case of geometry i can understand that as a possibility in the case of you know pure number theory i don't understand what it would be to say look the world really is is has this numerical structure i don't know i just don't understand it i can't get i can't wrap my head around what would even mean"
},
{
"end_time": 5933.063,
"index": 233,
"start_time": 5910.418,
"text": " Going back to that geometrical structure that you were referencing in the beginning, is that something that you're currently working on or is there a paper that you can point people to to learn more about? I wrote a book called New Foundations for Physical Geometry, The Theory of Linear Structures. That came out a while ago and that was a way of"
},
{
"end_time": 5959.77,
"index": 234,
"start_time": 5933.677,
"text": " trying to cover the same content or conceptual territory that's covered by standard topology, point set topology, but in a different way, using a different foundation. And there was always supposed to be a second volume to that book, so the first volume was just pure math and the second volume was supposed to be application to physics. And I intended to write that and I thought it was going to be very straightforward."
},
{
"end_time": 5989.565,
"index": 235,
"start_time": 5960.35,
"text": " and i was writing away at it and then i was teaching and then i got interested because one of the things that happens in the first book is i develop a way of thinking about topology which applies both for continual and for discrete spaces because the standard point set topology doesn't really work well in a discrete setting it just gives you kind of trivial answers anyway i got interested in this idea of discrete geometry and then i started developing this thing i call full discrete geometry and so this will eventually come out"
},
{
"end_time": 6014.991,
"index": 236,
"start_time": 5989.906,
"text": " as the second volume of New Foundations of Physical Geometry. I can't say exactly when, I mean I've got six chapters that are all about spatial structure and a few chapters now about adding time in for space-time and I have some really exciting recent results which I was talking about which are not published but if you're interested I've given the talk"
},
{
"end_time": 6044.241,
"index": 237,
"start_time": 6015.811,
"text": " Lots of times, in fact, I just gave the talk for the International Space Science Institute and they've put it up on the web within the last couple of years. So yeah, I've got more work to do to finish that book. I kind of see how it's going to go. There's some calculations and things and some examples that I need to think through."
},
{
"end_time": 6072.5,
"index": 238,
"start_time": 6045.179,
"text": " Eventually that will come out. There's a phrase that I heard you say with Einstein. The general view is that he spatialized time, but instead what we should do is temporalize space. Right. Okay, so please expand on that because that sounds poetic. I don't know what it means. Okay, right. So one thing that people say that is just flat incorrect is that in special relativity somehow"
},
{
"end_time": 6101.92,
"index": 239,
"start_time": 6073.541,
"text": " Time is just another spatial dimension. Or that they're placed on equal footing. They're placed on, quote, as you say, equal footing. That's just a phrase also that you never hear anywhere else except in special relativity. So then that makes me think that people are just copying that phrase because you never use the word equal footing in math or physics except here when talking about space and time. Everyone repeats it like the powerhouse of the cells, mitochondria. Good, good, good. You've put your finger on something which is really smart and really perceptive."
},
{
"end_time": 6131.374,
"index": 240,
"start_time": 6102.483,
"text": " And you should use it all the time. When a whole bunch of different people are using identically the same language, it's an indication of a problem. Because if you really understand something, when you speak about it, you won't always use the same phrases and different people will use slightly different phrases or produce different analogies and so on, because they understand it. And then they're filtering their understanding through language to try and"
},
{
"end_time": 6154.104,
"index": 241,
"start_time": 6131.834,
"text": " Convey that understanding. When everybody's repeating verbatim the same phrase, they've just memorized that phrase. Right. It's an indication of lack of really understand. I mean, think about it. I agree wholeheartedly about that. Think about the term equal footing. If by equal footing you mean equal, then why don't they say as they say, why don't they say, oh, he made everything into time?"
},
{
"end_time": 6178.865,
"index": 242,
"start_time": 6155.794,
"text": " But they say, no, somehow he made everything into space. Right. Now we know what it would be. We know what it is to add an extra spatial dimension. So we start out thinking space has three dimensions. You think it might have four? Fine. We know the tool to represent that. It's four dimensional Riemannian geometry. Well defined mathematical subject. That's not relative."
},
{
"end_time": 6202.21,
"index": 243,
"start_time": 6180.401,
"text": " The geometry of special and general relativity is not Riemannian geometry. It's Lorentzian, and because it's Lorentzian, it contains a light cone structure that is pseudo. A purely spatial structure has nothing that corresponds to a light cone."
},
{
"end_time": 6232.858,
"index": 244,
"start_time": 6203.319,
"text": " To say they're on equal footing, I mean, again, it's kind of also a little vague what that means, but it certainly doesn't mean that you've just turned time into space. And the real problem of people saying that you've turned time into space is that space doesn't have a directionality. That is, if I were to just go to somebody in the street and say, you know, the United States runs from north to south and not south to north, they'd look at me like, what in the world are you even saying?"
},
{
"end_time": 6259.428,
"index": 245,
"start_time": 6233.558,
"text": " What do you mean? There's a north-south dimension, if you will, but it doesn't go one way rather than another. Whereas, if I tell you time goes from past to future, and not from the future to the past, everybody says, yeah, of course. I mean, that's a triviality. Everybody knows time has a direction. We're constantly getting older. We wake up when we're older."
},
{
"end_time": 6289.241,
"index": 246,
"start_time": 6260.009,
"text": " There's some that we'll get to that say time is an illusion, so I'll get your thoughts on that. Okay, but anyway, time, intuitively time has a direction and space doesn't. And if by equal footing, people took the lesson, which many of them do, is they say, oh, first of all, by equal footing, you mean you've spatialized time, somehow time has now just become more space, but space doesn't have a direction. So gee, I guess time doesn't have a direction. Now you're in a complete mess."
},
{
"end_time": 6315.674,
"index": 247,
"start_time": 6290.811,
"text": " Now when I say you're temporalizing space, what I mean is something very, when I said that, is something very precise. Take a relativistic space time and blind yourself to everything except just the temporal structure, the proper time structure."
},
{
"end_time": 6345.503,
"index": 248,
"start_time": 6316.527,
"text": " from that you can recover the whole thing okay sorry repeat that one more time from the time structure from the proper time structure you can recover the entire structure right information about the the whole space time geometry is contained within if you just give me all the causal curves all the time like or light like curves and their proper lengths in right in proper time"
},
{
"end_time": 6374.206,
"index": 249,
"start_time": 6346.374,
"text": " That nails down the entire space-time geometry. So to that extent, you could say time settles everything. Time has a direction. We're going to save that. Yep. But we agree space doesn't appear to have a preferred direction. Right. Because it doesn't. Somehow we can generate space from a preferred time direction. You can generate the spatial structure. Yeah. Yeah."
},
{
"end_time": 6395.811,
"index": 250,
"start_time": 6375.35,
"text": " There's only one spatial structure that's consistent. There's only one full relativistic metric that's going to be consistent with just the temporal part. Now is this written somewhere in the paper or is this an idea that you're toying with? Here's some stuff that's well known. This is not original to me."
},
{
"end_time": 6420.947,
"index": 251,
"start_time": 6396.459,
"text": " Just forget about, just give me the light cone structure. So we all know, I hope, that in a relativistic setting, space-time has a light cone structure. So any event, there's a double cone, a future light cone and a past light. And let me just comment on that. It's important that it's a double cone. That is, if I have an arrow pointing into the upper part of this light,"
},
{
"end_time": 6451.049,
"index": 252,
"start_time": 6421.561,
"text": " I can't continuously rotate it down into the bottom so it's pointing the other direction and always have it point timelike, that is, have it point within the cone. The timelike directions are divided into two disconnected groups, the future pointing ones and the past pointing ones. But the space-like directions, which are the directions that as it were point outside the light cone, they're continuously connected. I can get from any space-like direction to any other one just always remaining space-like."
},
{
"end_time": 6480.486,
"index": 253,
"start_time": 6451.51,
"text": " Right. So that's the sense in which time has a direction. There's a fundamental distinction between the future directed things and the past directed things. And there's no corresponding distinction in space. Now, what's well known is if I just give you the light cone structure, just the light cone structure, to specify a relativistic metric, I need 10 numbers. If I give you just the light cone structure, that nails down nine of them. Okay?"
},
{
"end_time": 6506.647,
"index": 254,
"start_time": 6481.783,
"text": " That's well known. This was proven a long time ago. So there's just one other number that I need. And if you give me not just the light cone structure, but the length, so these timelike curves have a length that's called the proper time. If you tell me that, that'll nail down the tenth number. Interesting. Done. Okay. There's no other wiggle room. You've got everything. This is well known. This is not anything special to me."
},
{
"end_time": 6533.473,
"index": 255,
"start_time": 6508.131,
"text": " And so just to be clear when we were talking about that if you're in the upper part of a light cone, then you can't go to the bottom part. We're in an orientable manifold or does this still apply? And it's always the assumption when you do relativity, always, always, always the assumption that the manifold is temporally orientable. So it's not like a Mobius strip where things get twisted up. And that's because time has two directions."
},
{
"end_time": 6563.131,
"index": 256,
"start_time": 6534.309,
"text": " I would like to know your thoughts on Occam's Razor, whether it's a good principle, what are its pros and cons, and this relates to the metaphors, but we'll get to that. So what are your thoughts, what are the pros and cons of Occam's Razor? Well, again, part of the problem is when people use the term Occam's Razor, they often have different principles in mind. So what Occam said was something like,"
},
{
"end_time": 6590.196,
"index": 257,
"start_time": 6563.473,
"text": " Don't postulate entities without necessity. The problem with that is, what do you mean by necessity? If you take that too strongly, then it's kind of a crazy view because you're never really necessitated to postulate things. So then you weaken it and you say, look, don't postulate entities without good reason. That sounds good."
},
{
"end_time": 6619.07,
"index": 258,
"start_time": 6590.623,
"text": " But that's kind of trivial. I mean, you should also not delete entities without good reason. Okay, you shouldn't do anything. You just use good reason. Okay, it becomes good advice, but kind of trivial. Yeah, sometimes people say, well, Occam's razor says, if I have two rival theories, and one postulates fewer things than the other. So again, we go to the idea of ontology or beables, then you should prefer the one with fewer."
},
{
"end_time": 6644.462,
"index": 259,
"start_time": 6620.299,
"text": " Many times people say, oh, that's Occam's race. Take the one with fewer. That's certainly not advice that physicists follow, and I think for good reason. Why? For example, take a classical electromagnetic field of the sort that Maxwell did. So it's got charged particles, that's part of the ontology, and it's got fields, it's got electromagnetic fields."
},
{
"end_time": 6668.336,
"index": 260,
"start_time": 6645.316,
"text": " Now you can imagine a theory that gets rid of the fields, a kind of action and a distance theory. This is what Wheeler and Feynman were doing. You just throw away the fields and you say, no, no, it's not that this charged particle affects that one by the effect of an intermediate field. It's that this one just directly affects that one, maybe with a time lag with nothing between them."
},
{
"end_time": 6689.565,
"index": 261,
"start_time": 6668.729,
"text": " Well clearly that second theory has less ontology than the first one because it's you know they both have the particles and one has the fields and the other doesn't doesn't mean it's a better theory in fact it's it seems like a much worse theory because you give up certain principles of continuous action and so on and in favor of this"
},
{
"end_time": 6715.384,
"index": 262,
"start_time": 6690.213,
"text": " I mean, no physicist would take that seriously. They'd say, look, wait, but I have a whole set of equations for the electric and magnetic fields. I understand how they behave. I understand how jiggling this charged particle creates an electromagnetic field that goes through space and time and then hits this charged particle in an antenna and makes it go up and down. And then I turn on my radio."
},
{
"end_time": 6743.592,
"index": 263,
"start_time": 6715.384,
"text": " Okay. And if somebody wants to explain that taking away the intermediary fields, that's just a very strange thing to do. I certainly shouldn't think that, you know, less is more in ontology. So again, if it's just, if the advice is, look, just think about the ontology when you're comparing two theories, compare their ontology and ask, what is the justification?"
},
{
"end_time": 6764.514,
"index": 264,
"start_time": 6744.684,
"text": " where one has more if one strictly has more what's the justification think about that justification that's good advice but it's not good advice to say no don't even think about that just just compare them and then and announce that the one with less ontology is better because you don't you have to think about what price you're paying"
},
{
"end_time": 6788.404,
"index": 265,
"start_time": 6765.077,
"text": " Yeah, get rid of it. With the Feynman case with Wheeler, was their mathematical treatment more complex than the original electromagnetic from Maxwell? Because the Wheeler-Feynman theory tried to get rid of the electromagnetic fields in favor of, and now this comes back to the stuff about retrocausation, in favor of having direct action at a distance"
},
{
"end_time": 6811.766,
"index": 266,
"start_time": 6788.78,
"text": " both to the past and to the future. So the way a charged particle behaves right now was determined by looking both on its past light cone and seeing where there are other charged particles on the past light cone and what they're doing, and on the future light cone and where there are other particles on the future light cone and what they're doing."
},
{
"end_time": 6838.831,
"index": 267,
"start_time": 6812.671,
"text": " and so you had instead of what we normally have which is a boundary condition a single boundary condition from which you generate the future in the past you had a boundary condition to the past and a boundary condition to the future and to the future they use this thing called an absorber a perfect absorber condition blah blah blah but just mathematically that's a much more difficult problem much more difficult problem to know you can write down some equations to know whether they even have any solutions"
},
{
"end_time": 6853.933,
"index": 268,
"start_time": 6840.503,
"text": " It's just mathematically very intractable. Okay, so then it's intractable, which means that they weren't able to produce or reproduce the results? Yeah, as far as I know. I mean, I knew some people who were interested in this theory."
},
{
"end_time": 6878.899,
"index": 269,
"start_time": 6854.292,
"text": " uh in munich and you know recent i mean within the last 20 years and we're working on it and to my knowledge they never exactly got a precise solution either but i haven't looked at it in much detail it's okay it's just clear on mathematical grounds when you have something like newton's equations if you give me an initial condition and tell me the positions and moment of all the particles"
},
{
"end_time": 6901.032,
"index": 270,
"start_time": 6880.179,
"text": " Then you can prove in certain circumstances there's a unique solution to the future, there's a unique solution to the past, and you can actually get your hands on it mathematically. But when you say, no, now I have a situation where I have one boundary condition to the past and one boundary condition to the future and I'm trying to solve in between,"
},
{
"end_time": 6923.985,
"index": 271,
"start_time": 6902.278,
"text": " it just mathematically i mean you can imagine how do you go about it i mean in the case of gluten literally you imagine on a computer you set some data here and then the computer computes okay then a little later this then a little later this then it just builds it up or a little earlier this a little earlier builds it down but if i if i give you data here and here and i have to somehow"
},
{
"end_time": 6947.807,
"index": 272,
"start_time": 6924.735,
"text": " fill in the middle in a way that's consistent with it, it's not even quite clear how you proceed. If Wheeler and Feynman were able to reproduce it, you would still have the same objection. It's complex. I'm not sure. I mean, you would really have to ask yourself, look, if the theory that postulates electromagnetic fields has this simple Newtonian presentation, which it does,"
},
{
"end_time": 6974.753,
"index": 273,
"start_time": 6948.746,
"text": " Right. I just give you an initial condition. Simplicity needs to factor in too. Sorry, I don't mean to interrupt. I apologize. So simplicity has to factor in as well. It's not just less assumptions. Is there another quality like aesthetics? What are the factors that need to be in place? Because Occam is the way it's ordinarily stated. You have two competing theories, whichever one has less assumptions but produces the same results. You choose that one. Yeah. I mean, again,"
},
{
"end_time": 7003.592,
"index": 274,
"start_time": 6975.623,
"text": " Even if you're a logician, if you're like a philosopher and you're used to these logical tricks, you say, well, what do you mean fewer assumptions? If I have a theory with 10 assumptions, then I just put a conjunction between them and I have a single sentence. That's where I was getting at. Well, that's my objection with Occam's razor. To me, it's unclear what counts as an assumption. So some people say the simplest is God did it. But I don't know if God is one assumption or if it's 10. I just pause at space time. Is that one assumption?"
},
{
"end_time": 7033.49,
"index": 275,
"start_time": 7003.592,
"text": " Space-time has a huge structure. So it's like a manifold is like second countable Hausdorff and then there's atlases and then there's orientations So is that one assumption? Yeah, and I think quite honestly It's a fool's errand to think that you can come up with some abstract statement of a methodological principle That'll be the right one to apply in all cases. I mean start with the problem in front of you The problem in front of you is going to be here's one theory. Here's another theory"
},
{
"end_time": 7061.596,
"index": 276,
"start_time": 7034.258,
"text": " And let's assume that they make the same empirical predictions or where they differ you can't check or something. So you can't say, I'm just going to do an experiment to try and decide between them. Then people will start producing justifications for thinking one is more plausible than the other. And your job is to think through those justifications. Now, I don't think any single sentence in Latin"
},
{
"end_time": 7090.299,
"index": 277,
"start_time": 7061.988,
"text": " is going to give you very helpful advice, universal advice about how to do that. It might be that in some sense one of the theories is simpler, is clearly simpler than the other with no penalty. Then you'd say, okay, the simpler one just seems more plausible. But usually there are trade-offs. How important are they? How much of this will you give up for that? There's not going to be a general calculus of this. You should just"
},
{
"end_time": 7103.746,
"index": 278,
"start_time": 7090.759,
"text": " Try to be as explicit as you can in the accounting."
},
{
"end_time": 7133.643,
"index": 279,
"start_time": 7104.462,
"text": " If on grounds other than empirical grounds I prefer A to B, let me say why and then just listen. The kind of justification you might get may be very specific to the particular mathematics being used. I mean, God knows what. I mean, don't expect some guy in the Middle Ages wrote down something that is going to give you the royal key to deciding between these things."
},
{
"end_time": 7162.278,
"index": 280,
"start_time": 7133.899,
"text": " And maybe there's no good way to decide between them. I mean, we should be prepared to have alternative physical accounts of the world that we don't actually have any particularly good reason to decide between. And if that's the situation, that's the situation. Speaking of what's tricky to decide between the illusion of time. So there are certain theories and certain interpretations of time."
},
{
"end_time": 7187.295,
"index": 281,
"start_time": 7162.79,
"text": " Yeah, so you did give me some warning you were going to ask that question. And I'm afraid I just have to say I don't understand relational quantum mechanics. I looked at it a bit. I talked to Carlo about it a bit. I just don't understand it."
},
{
"end_time": 7216.578,
"index": 282,
"start_time": 7187.961,
"text": " I know some people who looked at it more carefully and they were unable to find what they considered a coherent formulation of it. If you feel like you understand it, then you can explain it to me and I can try to respond. If you feel like you don't really understand it and can just repeat what other people say, we're a bit stuck. I just don't understand what the theory claims, even apart from time. My issue here isn't about time."
},
{
"end_time": 7245.418,
"index": 283,
"start_time": 7217.295,
"text": " I don't understand what it means to say that all that's real are relations. I mean, sometimes there's this tag phrase that people say, oh, relations without Rolada. And this is, you know, you're just contradicting it. How can you have a relation without Rolada? Or things only exist when they interact and then you want to say, but wait, if they weren't already there, how could they interact? I mean, how can interaction occur between two things that don't exist?"
},
{
"end_time": 7274.77,
"index": 284,
"start_time": 7246.408,
"text": " I literally don't understand what's being claimed. And again, if you feel like you want to explain it to me, I'm all ears. Again, one little data point was once I was puzzled about this, I was talking to Carla and we were talking about non-locality, right? So in the non-locality case, Alice does an experiment over here and Bob does an experiment very far away. And the issue is"
},
{
"end_time": 7299.599,
"index": 285,
"start_time": 7275.213,
"text": " the correlations between the outcomes of the experiments they do, right? That's what Bell's inequality puts constraints on in these correlations. And as far as I could understand, Carlos seemed to be saying that there is no outcome for Bob"
},
{
"end_time": 7328.933,
"index": 286,
"start_time": 7300.742,
"text": " just, you can't just talk about what happened in Bob's lab, right? You normally think you could. You say, okay, Bob did an experiment. Did he get a dot here or a dot there? You know, was it an up or a down outcome? And it's like, no, no, no, he only had an outcome relative to Alice. And furthermore, somehow that outcome only occurs when Alice and Bob interact and Alice and Bob only interact when after the experiment, they come back together and share data or something like that."
},
{
"end_time": 7357.449,
"index": 287,
"start_time": 7329.701,
"text": " And I said, but I don't understand, Carlos, so if I'm Bob, and I do my experiment, right, and I have my data, and then I'm waiting for Alice, Alice is to come meet me to share her data. And Alice comes and she says, you know, 10 minutes ago, I just got this result. Do I not believe her? I mean, do I literally just not believe? Yes, you got that result 15 minutes ago, long before I ever found out about it and long before we ever interacted."
},
{
"end_time": 7383.899,
"index": 288,
"start_time": 7358.404,
"text": " And Carlo just looked at me like I'd asked a question that never occurred to him. I mean, he didn't have a ready answer at the time. I don't know if he does now. And at that point I said, I just don't understand. I mean, I think Alice is there doing her experiments and getting her outcomes, and Bob is there doing his experiments and getting his outcomes, even though they're not interacting with each other in a normal sense of interaction."
},
{
"end_time": 7409.07,
"index": 289,
"start_time": 7385.606,
"text": " And those outcomes are what they are and they do or they do not violate Bell's inequality. I don't understand what the relational theory says about that case. I have some puzzlement because I asked Carlo, well, what's being related? And then I believe he said certain relations. Then I said, well, what are those relating? And is it not just some infinite regress of relations? And then he said, yes."
},
{
"end_time": 7438.677,
"index": 290,
"start_time": 7409.36,
"text": " This is a real good story about Bronx and his dad Ryan, real United Airlines customers. We were returning home and one of the flight attendants asked Bronx if he wanted to see the flight deck and meet Kath and Andrew. I got to sit in the driver's seat. I grew up in an aviation family and seeing Bronx kind of reminded me of myself when I was that age. That's Andrew, a real United pilot. These small interactions can shape a kid's future. It felt like I was the captain. Allowing my son to see the flight deck will stick with us forever. That's how good leads the way."
},
{
"end_time": 7458.831,
"index": 291,
"start_time": 7440.572,
"text": " There's a Dr. Seuss story about the lazy bees in Hotch Hotch."
},
{
"end_time": 7481.442,
"index": 292,
"start_time": 7459.053,
"text": " So, you know, they find that when the bee watcher watches the bees, they're busier, so they hire a bee watcher. But the bee watcher was lazy, so they had to hire a bee watcher watcher, and then a bee watcher watcher to watch the bee watcher. And, you know, it's the same thing people say, well, you know, reality doesn't exist until an observer observes something. And you say, but wait, did the observer already exist?"
},
{
"end_time": 7510.981,
"index": 293,
"start_time": 7482.022,
"text": " Oh, no, no, the observer only exists because some super observer was observing the observer. But did the super observer already exist? Oh, no, there had to be a super, super observer. And then you said, this is just silly. I mean, you know, why are you doing this? And if someone said, well, is it turtles all the way down? Yes, there's an infinite hierarchy of observers observing other observers, bringing them into existence. You just say, look, you can say this stuff, but why are you doing it? I mean, we have perfectly good accounts of the physics here that don't involve this kind of ridiculous gymnastics."
},
{
"end_time": 7539.565,
"index": 294,
"start_time": 7511.544,
"text": " What's motivating you to say things like that? Yeah. So the way that I view it, and again, just so you were clear, Carlo, another huge admiration for Carlo, huge intellectual giants. So I'm Switzerland when it comes to this. I think that many people, they believe that and I'm not categorizing Carlo as this, I'm referring to those people about the observers and infinite regress of super observers. So not regress, but superset. I think that people"
},
{
"end_time": 7560.486,
"index": 295,
"start_time": 7540.265,
"text": " Including myself, including everyone. I think that people reason backward and we believe that we're reasoning from founded. We say first principles, almost no one is a first principles thinker. We have certain values that we're trying to preserve and we reason backward and we say, well, you know how a implies B is the exact same as not B implies not a, let me give you a story."
},
{
"end_time": 7580.384,
"index": 296,
"start_time": 7561.391,
"text": " I was driving with someone who is of extremely liberal temperament, and there's nothing wrong with that. And I said, oh, that dog breed is known for being extremely intelligent. And she's like, no. I'm like, what do you mean? No, it's just known for being intelligent. It has a higher IQ. She's like, what are you talking about higher IQ in dogs?"
},
{
"end_time": 7609.206,
"index": 297,
"start_time": 7580.691,
"text": " And then I realized what's happening is that she believed that what could be done is I could then infer onto humans and say, oh, well, there are different species here. So because that is not she wants to preserve that she reasons backward. OK, so I believe that that's what's occurring in many of these cases. Well, look, there's a there's a tradition. There's a tradition going back to boar and is kind of associated with the Copenhagen school or the Copenhagen spirit or whatever."
},
{
"end_time": 7624.07,
"index": 298,
"start_time": 7610.128,
"text": " that thinks that the great revolution in quantum theory is because the observer unlike in classical physics, unlike in Newtonian physics, the observer somehow plays a central role in quantum theory."
},
{
"end_time": 7644.821,
"index": 299,
"start_time": 7625.179,
"text": " And the problem is that on no actual existing precise understanding of quantum theory that we have, is that true? It's just that, I mean, for many years, for decades, there's been a intermittent series of meetings and the title of the meetings is quantum theory without the observer. And it's just"
},
{
"end_time": 7671.817,
"index": 300,
"start_time": 7646.015,
"text": " formulations, theories that would would account for the quantum phenomena, where in their formulation, they never mentioned observers. And there's a wonderful phrase that Bell has. At one point, he says, this is a theory that is neither requires nor is embarrassed by the existence of an observer, right? I mean, it's not that the physics is going to have trouble modeling observers, their physical systems, but"
},
{
"end_time": 7701.323,
"index": 301,
"start_time": 7672.193,
"text": " Gee, you know, I mean, we all think, don't we, that shortly after the Big Bang, a whole bunch of physical stuff happened and there was nobody there observing it? I mean, it's kind of, again, it kind of blows your mind when you try and see what are you even thinking if you think that physics needs observers to go on? It's just the opposite. Observers are physical systems of a certain kind, and you want the physics"
},
{
"end_time": 7726.425,
"index": 302,
"start_time": 7702.637,
"text": " to describe how they behave as by pure analysis of them as physical systems. And it's the same thing for measurement, right? A measurement, whatever you call a measurement is just a physical interaction between two things that's going to be described by some physical equations that don't mention the word measurement. And that's again, go read Bell's against measurement against quote measurement."
},
{
"end_time": 7751.237,
"index": 303,
"start_time": 7726.903,
"text": " He says the term measurement should not appear in the formulation of any theory that pretends to be foundational. Because measurement isn't the right kind of concept to appear in the foundations of physics. Measurement is a derivative concept. Observer is a derivative concept. They shouldn't appear in the foundations of physics."
},
{
"end_time": 7767.568,
"index": 304,
"start_time": 7752.09,
"text": " Lots of people convince themselves that, again, here is where I think there's something semi-mystical. There are people who like the idea that observers should become central to the universe."
},
{
"end_time": 7792.056,
"index": 305,
"start_time": 7768.626,
"text": " It annoys them that observers seem to be this tiny little special class of humans in particular, some tiny little specks on something not playing a very central role in physics. They have an emotional attachment of some sort, but it seems to be contrary to physics as we understand it."
},
{
"end_time": 7815.35,
"index": 306,
"start_time": 7792.534,
"text": " And certainly nobody's ever made progress doing that. Nobody's ever made progress saying, oh, here, I'm going to write the observer into the foundations of physics. And people have made progress doing the opposite. Okay, so two notes here. Number one, there exists perfectly coherent formulations, theories of quantum mechanics that don't involve observers. Sure. Okay."
},
{
"end_time": 7826.732,
"index": 307,
"start_time": 7815.811,
"text": " So that's interesting. The pilot wave theory, the GRW collapse theory, I mean the many worlds theory in so far as you can make sense of it certainly does not make any reference to observers."
},
{
"end_time": 7851.271,
"index": 308,
"start_time": 7827.381,
"text": " I mean, these are theories where they're written in the equations. I mean, again, Bell says somewhere, I want a theory where the physics is in the equations and not in the surrounding talk, right? These are all theories where you postulate a quantum state, maybe you postulate some local variables in addition to that, and you write down some clean equations about how they behave. And that's the theory."
},
{
"end_time": 7880.196,
"index": 309,
"start_time": 7851.698,
"text": " It doesn't mention observers, it doesn't mention measurements, it doesn't mention anything like that. There's some physical stuff that you postulate, it behaves in a certain way that you postulate, and then you see what happens. So pilot wave, many worlds, and then what were the other? The collapse theory, Jardhi-Ramini-Weber objective collapse theories. I mean, there are a class of so-called objective reduction theories. I mean, Penrose has a kind of sketch of a theory for many years. He's had a sketch of a theory"
},
{
"end_time": 7909.889,
"index": 310,
"start_time": 7880.196,
"text": " Which ties the collapses to gravity, the GRW theory doesn't. What do you make of Penrose's theory? I mean, I understand he takes the foundational questions seriously. And foundationally, you only have three options. Okay, that's why you can kind of have a nice taxonomy of solutions to the measurement problem."
},
{
"end_time": 7938.626,
"index": 311,
"start_time": 7910.947,
"text": " You can say that what you do is just logically you get into trouble if you try to maintain three principles. Principle one, that the wave function that's used in all these quantum theories provides a complete description of a physical system. And this was the EPR paper in 1935, the Einstein-Podolsky-Rosen paper, the title of it, is the quantum mechanical description of reality complete."
},
{
"end_time": 7966.954,
"index": 312,
"start_time": 7939.104,
"text": " Meaning, if I give you the wave function, can you somehow derive from it all the physical facts about the system? They were arguing you can't. They were arguing it's incomplete. Okay. Many, Bohr and Heisenberg wanted to say it was complete. So you have one principle. Is the wave function complete? Second question. Does it always evolve by Schrodinger's equation? Schrodinger's equation is linear. That's a very important feature of it."
},
{
"end_time": 7991.8,
"index": 313,
"start_time": 7968.063,
"text": " Third claim is that when you do a Schrodinger cat experiment, you start with one cat, you end with one cat, and the single cat is either alive or dead. That was Schrodinger's point. He thought that's obvious. You can't maintain all three. As a matter of logic, you can't maintain that the wave function is complete, that it always evolves by Schrodinger's equation, and that the cat ends up alive or dead."
},
{
"end_time": 8022.244,
"index": 314,
"start_time": 7992.551,
"text": " What does complete mean once more? What does complete mean? Complete means if I give you the wave function of a system, you can derive from that somehow all the physical facts about it. All of its physical characteristics. Does that mean that it has that you could find out whether it's spin up and then spin? It means you could find out anything that's a physical fact about it. I'm not telling you what the physical facts are, but any physical fact is determined by the wave function."
},
{
"end_time": 8049.377,
"index": 315,
"start_time": 8023.302,
"text": " So are there theories that say no, you can't? Sure, pilot wave theory. I mean the de Broglie-Bohm or Bohmian mechanics. In Bohmian mechanics, the ontology, the beables, there are two. There's the universal wave function, the universal quantum state, that's one. And then there are these particles, and the particles are like classical point particles. They always have definite positions, they move around, okay?"
},
{
"end_time": 8079.804,
"index": 316,
"start_time": 8050.128,
"text": " But if I just give you the quantum state, that doesn't tell you where the particles are. Where the particles are, the actual configuration of particles, is not derivable from that. It's an additional piece of physical data. And the way that theory works is you say, okay, how does the wave function evolve? Answer, always by Schrodinger's equation. Okay, how do the particles move? How does this particle configuration change? That's given by a second equation called the guidance equation, and the guidance equation has as an input to it the wave function."
},
{
"end_time": 8108.063,
"index": 317,
"start_time": 8080.794,
"text": " So that's a theory where you say the wave function is not complete. There are additional physical facts. I could have two systems that have exactly the same quantum state but are different physically because the particles are in different locations. So you can't maintain that the wave function is complete and it always evolves by Schrodinger's equation and that the cat ends up either just plain alive or plain dead."
},
{
"end_time": 8127.312,
"index": 318,
"start_time": 8108.473,
"text": " Okay. Got to give up one of those three. If you give up the completeness of the wave function, then you have what's called a hidden variables theory like Bohmian mechanics, pilot wave theory. If you give up the Schrodinger evolution, then you have an objective collapse theory. You say, okay, the wave function doesn't always fall by Schrodinger's equation."
},
{
"end_time": 8154.377,
"index": 319,
"start_time": 8128.114,
"text": " and if you don't give up those two then you're stuck with many worlds and you say well the cat doesn't end up either alive or dead somehow one cat becomes two cats or many cats some of which are alive and some of which are dead okay those are only three options just as a matter of logic i mean i have a paper about this called three measurement problems which goes through in more detail this is a nice thing because if somebody says here's my understanding of quantum mechanics"
},
{
"end_time": 8173.063,
"index": 320,
"start_time": 8154.872,
"text": " you can immediately ask them these diagnostic questions okay according to quantum state complete according to you does it always evolve by schrodinger's equation according to you does a single cat end up either just alive or just dead and if they say oh i i you know i want to maintain"
},
{
"end_time": 8193.609,
"index": 321,
"start_time": 8174.275,
"text": " The kind of intuitive answer to all three, you say, I'm sorry, you can't, right? Then it's not a coherent theory. And once they answer which one they're going to give up of those three, then you sort of understand where you are and you understand how the questioning goes on from there. So yeah, this is extremely, extremely important. Yeah."
},
{
"end_time": 8221.834,
"index": 322,
"start_time": 8194.394,
"text": " So, I mean, look, Penrose has gone for an objective reduction theory and that this follows, if you go and read von Neumann's book, Mathematical Foundations of Quantum Mechanics, von Neumann, and this was taken as kind of the Bible of a clean mathematical formulation of Copenhagen, von Neumann is very clear. He says states, physical systems can undergo two completely different sorts of evolution."
},
{
"end_time": 8246.527,
"index": 323,
"start_time": 8222.654,
"text": " Process A and Process B, or Process 1 and Process 2, I can't remember what he uses. One of them is Schrödinger evolution and the other is a collapse. Right, right, right. And they're very different. And if you think that, then you have an objective reduction theory. Now that's the way Penrose wanted to go for a long time. And then the question, if you have a reduction theory, you say, okay, well, when does this collapse occur? What triggers the collapse?"
},
{
"end_time": 8266.561,
"index": 324,
"start_time": 8247.227,
"text": " Or does anything trigger the collapse right what governs the collapse if you will you better not say well it collapses when you make a measurement because then you what when do you make a measurement the measurement now you're back to against measurement. Penrose wanted to tie the collapses to gravity so. Any had a sort of."
},
{
"end_time": 8285.947,
"index": 325,
"start_time": 8268.166,
"text": " Likely hand-wavy account of that, which actually a similar account was given by a guy named Karolyi Hazi long ago. He was also, when Karolyi Hazi was great, I heard him once when I was very young, and he would say, well, as it were, you have these two gravitational situations, which by Schrodinger evolution,"
},
{
"end_time": 8315.145,
"index": 326,
"start_time": 8286.408,
"text": " It's like, you know, they're separating and there's a gravitational difference between them. And then it's at some point that gravitational gets so big. And this is what Karolyi has, he said, he said, then Gabriel comes and blows his trumpet. And the blowing of the trumpet was the signal, the wave function shall now collapse and it'll either go this way or that way. Right. And so, you know, Penrose's idea was to tie somehow these collapses to gravity."
},
{
"end_time": 8340.333,
"index": 327,
"start_time": 8315.691,
"text": " The Girardi-Romeini-Weber theory doesn't do that. It doesn't tie them to anything. It just gives you a fixed probability per unit time per particle that there'll be a collapse. So these are objective reduction theories. It's a way to go. The objective reduction theories make in principle slightly different predictions, empirical predictions, than the other theories. So you can kind of test in the lab. And the GRW kind of theory"
},
{
"end_time": 8369.155,
"index": 328,
"start_time": 8340.947,
"text": " I mean, people keep doing tests and ruling out certain parameters space, and there's not much parameter space left for it. In another 10 years, it may be empirically, empirically refuted. To reiterate its completeness. So is there more information necessary to specify what's physically going on? And in the case, just to be clear, again, bohmian mechanics and pilot wave are synonyms. Yeah. Okay. And so in pilot wave theory, it says there's a wave and a pilot."
},
{
"end_time": 8397.295,
"index": 329,
"start_time": 8369.855,
"text": " No, the wave is the pilot. Okay, sorry. The wave, the quantum state, call it the quantum state. That's the thing that's described mathematically by the wave function. Okay. Its job physically is to pilot or guide the particles. Okay. To determine how the particles move. And it does so via a precise equation called the guidance equation."
},
{
"end_time": 8418.797,
"index": 330,
"start_time": 8398.695,
"text": " i mean bell bell presented it that way but bone when he did it i mean the original idea was de broglie and then de broglie dropped it for bad reasons and then bone rediscovered it but bone presented the theory as a kind of newtonian theory with something he called the quantum potential that turns out to be a bad idea bell presented it in terms of the guidance equation which is the right way to do it"
},
{
"end_time": 8449.087,
"index": 331,
"start_time": 8419.258,
"text": " So the piloting is done by the wave as it were by the quantum state and what pilots or what it guides are the particle. A surfer on the wave is the particle something like that? You can think of it that way. Now the thing that you have to wrap your head around that's hard for this theory is if you have a single particle you can really think of it that way. You've got kind of a wave in physical space and it's telling the particle where to go. But as soon as you get two particles or three particles or four particles."
},
{
"end_time": 8465.708,
"index": 332,
"start_time": 8450.247,
"text": " whether they interact or not, the weight function is now defined not on physical space, but on configuration space, which is a 3N dimension. So for four particles, it's a 12 dimensional space. And this is just standard quantum mechanics. This is not peculiar to this theory."
},
{
"end_time": 8493.029,
"index": 333,
"start_time": 8466.988,
"text": " And essentially what that high dimensional object is doing is governing the entire configuration of particles. It's not as if, as it were, one part of the wave is telling this particle where to go and another part of the wave is telling this particle where to go. It's that the whole wave is telling the entire collection of particles how to evolve. That's why it's a kind of non-local theory. That's how the non-locality gets into it."
},
{
"end_time": 8522.637,
"index": 334,
"start_time": 8493.387,
"text": " It's a kind of global holistic theory where everything is being done together at the same time, if you will. Are there separate pilot waves for each of the particles or is there one pilot wave? No, no, there is one. I mean, in this theory, if you take it seriously, ultimately, there is one quantum state and that one quantum state is the quantum state of the entire universe. And what it determines is how the entire configuration of everything in the universe"
},
{
"end_time": 8540.043,
"index": 335,
"start_time": 8522.944,
"text": " And then there's a very interesting question to which there are very interesting and precise answers. Well, if that's true, that's not what we ever use in our calculations. When I go into the lab, I don't use the universal wave function"
},
{
"end_time": 8569.821,
"index": 336,
"start_time": 8540.452,
"text": " of everything, I assign a little psi. So in the literature, there's big psi, which is the fundamental thing. It's the wave function of everything in the entire universe. And then there's little psi, which is something I would assign to a subsystem. And then there's a question of, well, what's the status of this little psi that I actually use? And there's a very nice answer to that. It's called the conditional wave function. The answer is mathematically clear. This is all covered in my book as well. And the conditional wave function"
},
{
"end_time": 8595.674,
"index": 337,
"start_time": 8570.401,
"text": " sometimes evolves by Schrodinger's equation and sometimes does something that looks like a collapse. Interesting. All of that is just a consequence of these two equations that are at universal scale. So there's a really beautiful explanation in this theory of why people use collapses to do calculations and to the extent to which that's justified."
},
{
"end_time": 8623.729,
"index": 338,
"start_time": 8596.203,
"text": " even though in this theory there are no collapses on the single fundamental quantum state which is the quantum state of the entire universe. There's a wonderful answer to that and it's an answer that's not available in other theories. I mean there's a nice story to be told. And you use this word theories which is interesting because when one takes quantum mechanics in university they're taught or they're told you're going to learn quantum theory and you would say at least from my observation from your talks"
},
{
"end_time": 8649.394,
"index": 339,
"start_time": 8624.053,
"text": " Well, a theory needs to have an account of what is so it needs to have some ontology, or some way that the mathematics relates to what is, you can't just speak about the mathematics. So that's why you're not calling, I've been listening to you're not calling these interpretations of quantum mechanics, you kept saying these are different theories, because you don't see quantum theory as having multiple interpretations, you see each of them as separate quantum theories. That's right."
},
{
"end_time": 8674.428,
"index": 340,
"start_time": 8649.855,
"text": " What you're taught when you take a course is not a theory. Why? Because a theory postulates the existence of certain things. It tells you, I am a theory about the physical world. Here's what I'm saying is in the physical world. And here's how it behaves. That's what a physical theory is."
},
{
"end_time": 8701.937,
"index": 341,
"start_time": 8675.043,
"text": " What you're taught is a scheme for making predictions. Okay? It's a predictive recipe. It says, oh, in this circumstance, use this mathematical object and solve it in this way. And then you pull out of thin air born's rule to make probabilistic predictions. Okay? It's a prediction making apparatus, but it doesn't even attempt to be a theory. It doesn't pretend to be a theory."
},
{
"end_time": 8728.063,
"index": 342,
"start_time": 8702.534,
"text": " Because if it were a theory, you could say, okay, how do I understand this wave function that you're using? Does that represent anything physically real? And it would give you an answer. The answer might be yes, the answer might be no, but it would be an answer. And what you're taught doesn't give you an answer. That's why people use this phrase, shut up and calculate. They'd say, look, I've given you this wave function, I've told you how to manipulate it to make your predictions."
},
{
"end_time": 8747.159,
"index": 343,
"start_time": 8728.575,
"text": " Now shut up and calculate. Don't ask me whether it represents anything physically real. And so it's not a theory. And then what people call interpretations of quantum theory are really explicit physical theories that make precise postulates about what exists and how it behaves that are designed"
},
{
"end_time": 8759.65,
"index": 344,
"start_time": 8747.91,
"text": " What's wrong with the shut up and calculate mindset?"
},
{
"end_time": 8785.247,
"index": 345,
"start_time": 8760.23,
"text": " So why can't we simply have both? Hey, there are some people who are going to investigate subterranean, find the sharks and the giant squid. And then there are those on the top. They're like, Hey, I'm going to build the cities and make technology. And sure, you can say so, so exist. What difference does it make for me? At some points you provide insights and they come up to the surface and maybe we inform you, you do you will do us. Is there something wrong with that?"
},
{
"end_time": 8812.415,
"index": 346,
"start_time": 8786.766,
"text": " Division of labor. I mean, it's the same division of labor one has typically between engineering and fundamental physics. So let me just give you an analogy. Suppose I'm an engineer. I'm interested in building bridges. And I'm just fooling around at some point with different metal alloys. And I make an alloy that has some really nice properties for bridge building, right? It's light and it's very sturdy and so on. And I say, ah, good."
},
{
"end_time": 8840.742,
"index": 347,
"start_time": 8812.995,
"text": " I'm going to use this to make my bridges. I can now make lighter bridges and stronger bridges. Good. As an engineer, I'm happy. And somebody comes along and says, but why that alloy? I mean, why is that alloy so much stronger than other alloys? Now, as an engineer, I don't have a complaint about the engineer saying, I don't care. I don't care why it works. I don't need to understand why it works for my job. I just have to know that it works."
},
{
"end_time": 8868.046,
"index": 348,
"start_time": 8841.578,
"text": " I have to know that it is in fact light and strong, which I can test. And look, it is. If you, if you physicists want to try and figure out why this much bismuth and this much tin and this much, you know, iron happens to form a structure that's so light, fine, go to it. I as an engineer could not care less. My job is done when I built the bridge and I'm sure it's going to work. So that's a division of labor."
},
{
"end_time": 8895.418,
"index": 349,
"start_time": 8868.729,
"text": " i'm not gonna i'm not gonna rag on the engineer you can say well at least they ended up building a bridge you know they made life better people can now travel across the river i'm not making a value judgment i am saying from my personal point of view i want to know why it works it's not to me i'm just a nosy person okay i'm not focused on on uh practical"
},
{
"end_time": 8925.486,
"index": 350,
"start_time": 8895.606,
"text": " I'm just curious. I would be interested to know why that particular mix of metals is so much better than other mixes of metals. And that's the kind of thing that the physicist would get into and try to describe in fine detail what's this underlying structure. Now, you know, some physicists have decided, from my point of view, when it comes to quantum mechanics, they're happy to be engineers."
},
{
"end_time": 8952.381,
"index": 351,
"start_time": 8926.101,
"text": " Right. They say, okay, you've given me this wave function and you've told me how to calculate with it. And that's enough for me to design my chips or, you know, blah, blah, blah. I'm happy. I'm but personally, I'm not happy. I don't really care about the chips. I want to understand what's really going on. Right. I want to understand what the world is really made of. So I want to go deeper than that. But that's that's, you know, I can't say that that's intrinsically better thing to do. Certainly,"
},
{
"end_time": 8980.794,
"index": 352,
"start_time": 8952.875,
"text": " You would not have made all the material progress that has been made using quantum mechanics if people got stuck because they didn't understand it. So shut up and calculate was efficient. But it's really bad if people think there's something wrong with the small minority of us who are not satisfied with it. There's something illegitimate."
},
{
"end_time": 9005.469,
"index": 353,
"start_time": 8981.254,
"text": " about people who say, but I really want to know what's going on, not for any practical purpose. And of course, most of physics is like that. I mean, when they built the large hadron collider to try and see if the Higgs particle existed, it wasn't because you would make better computers or build bigger bridges or have some practical consequence that came out of what the exact mass of the Higgs is."
},
{
"end_time": 9034.582,
"index": 354,
"start_time": 9006.493,
"text": " Nobody could foresee they would have any interesting practical consequences. You're just curious, right? What's going on? To me, that's the fundamental impulse of being interested in physics. It's just, you know, thalmas. It's wonder at the universe. It's trying to understand things. Now, some people aren't driven by that and they're more interested in building things or they're more interested. And I don't want to say they're, you know, what they're doing is bad."
},
{
"end_time": 9064.821,
"index": 355,
"start_time": 9035.759,
"text": " But, you know, leave at least a few of us who are motivated by pure intellectual curiosity to go about our business. Yeah. And I find that there's a bit of condescension from the hardcore physicist toward the philosophers. I don't know if you have an understatement. Yeah, I wish that wasn't there. It's getting better, though. It really is getting better. I mean, if you didn't notice, there was just within the last week, I think I saw it on Facebook, an editorial published in nature physics."
},
{
"end_time": 9091.374,
"index": 356,
"start_time": 9065.879,
"text": " Okay. We don't know who wrote it. It's just an editorial from the editorial board, not signed, which is basically saying, you know, there's really a reason to look at foundations of physics. Right. It's not that everybody has to, you know, but it's really a worthwhile subject matter for physicists to think about foundational issues. And to me, this is just wonderful because that's, I mean,"
},
{
"end_time": 9119.309,
"index": 357,
"start_time": 9091.817,
"text": " I know people, I mean, I was lucky because I'm in the philosophy community and philosophers are happy with the questions I'm interested in. But physicists trying to do foundational work, you know, I mean, they have a very, very, very, very hard time just at a practical level, because it's not valued. It's just not valued in physics departments. It's considered somehow illegitimate. Do you see consciousness as a foundational question?"
},
{
"end_time": 9140.23,
"index": 358,
"start_time": 9120.179,
"text": " Consciousness I see is the hardest question there is. I think the mind-body problem is the hardest conundrum. You can't even imagine what a solution could look like. I don't know"
},
{
"end_time": 9164.224,
"index": 359,
"start_time": 9140.879,
"text": " I mean, if you sort of think, let me give an example. If you go back to ancient Greece and they said, I wonder what matter is made of and the atomists say, well, I think it's these little hard atoms and all they can do is move and some of them have hooks and some of them have eyes and sometimes they hook together and, you know, okay. You can at least understand"
},
{
"end_time": 9194.923,
"index": 360,
"start_time": 9165.367,
"text": " In a sketchy way, how such an apostolate could account for cables and chairs and things burning and, you know, okay, they become unhooked or they get hooked or blah, blah, blah. I mean, at least you see maybe it's wrong. Maybe it's completely off target. It turned out it was off target. But I mean, that's the story. But at least you could see from the beginning how if it were true, it could account for what you want accounted for, which is the behavior of objects at macroscopic scale."
},
{
"end_time": 9226.613,
"index": 361,
"start_time": 9196.783,
"text": " In the case of consciousness, I don't even know what a solution to the mind-body problem could look like without putting any other constraints. I just don't know. Essentially, what you're asking for is some physics that when you apply it to a physical body in the structural form of a human brain or a dog's brain or, you know, I don't know, an ant's brain, I'm not sure."
},
{
"end_time": 9256.049,
"index": 362,
"start_time": 9227.108,
"text": " Right? I don't know. Are ants conscious? I don't know. That somehow you see that, gee, that physical behavior would be accompanied by or create or whatever a subjective state like pain. And I don't understand how that could be. So I think that's the hardest problem there is. And because it's so hard, I don't work on it because I don't even understand what a solution would look like. I mean, I've pointed out"
},
{
"end_time": 9280.725,
"index": 363,
"start_time": 9256.578,
"text": " one of the first papers i published was a paper called computation and consciousness which is just pointing out that a certain kind of attempt to answer that question can't be right but i don't know what the right answer is i don't even know i i i'm you know so i it's not that i don't think there's a problem there i think there's a problem there that's so intractable yes that i don't want to"
},
{
"end_time": 9305.794,
"index": 364,
"start_time": 9281.101,
"text": " Bend my time banging my head against a brick wall. Now I firmly believe I'm not a dualist. I firmly believe that anybody who had a brain that was physically identical to mine and the neurons and everything were doing things physically identical to mine, they would be having conscious experiences that would be qualitatively identical to mine. So I'm not a dualist."
},
{
"end_time": 9332.278,
"index": 365,
"start_time": 9306.323,
"text": " I think that consciousness in some sense supervenes on physical activity because I think that's all I am. I'm ultimately a physical thing. It therefore depends on quantum mechanics just because physics depends on quantum mechanics. But I don't see any clue of a comprehensible account of why"
},
{
"end_time": 9351.357,
"index": 366,
"start_time": 9334.616,
"text": " Subjective physical states should accompany any sort of physical activity. So, you know, you work where you think you can make progress and you just, you know, you acknowledge where you don't think you can and where there are problems."
},
{
"end_time": 9380.998,
"index": 367,
"start_time": 9351.954,
"text": " You mentioned that you ruled out a certain class of consciousness theories that are computational. I don't know if that was the correct wording. Regardless, can you expound on that? Is it similar to the Penrose-Lucas argument? There's a whole class of theories that are called functionalist. And the basic idea is what makes a system conscious is kind of independent of the particular physical stuff it's made of. It's rather how that physical stuff behaves."
},
{
"end_time": 9409.155,
"index": 368,
"start_time": 9381.544,
"text": " Functions in a certain way. Sorry, is this the same as substrate independence or no? Yeah, it's a sort of kind of substrate independence, right? That's a kind of general class. Then in that general class, there's a much more specific realization of it that's called computationism that says what's really important is to understand what computations the system is doing."
},
{
"end_time": 9438.302,
"index": 369,
"start_time": 9409.667,
"text": " where I then understand a computation in terms of a Turing table, using Turing's way of understanding a computation. Such a theory would be committed to saying that any system, that there's some calculation or some computation on some input that can be done, such that any system doing that computation on that input will be conscious in a certain way."
},
{
"end_time": 9467.466,
"index": 370,
"start_time": 9439.445,
"text": " All right, let's say, and again, by conscious, I mean, the paradigm case is pain, right? Have a toothache, right? Any system that satisfies a certain Turing table performing a certain computation on a certain input will experience subjectively a toothache. And what I have as an argument is that that can't be right. Because you can implement the Turing machine table"
},
{
"end_time": 9496.937,
"index": 371,
"start_time": 9467.961,
"text": " in peculiar ways and not ways that anybody would actually do in practical life but theoretically you can implement a Turing machine in ways that the Turing description depends on a bunch of structure that's not physically active through a certain period it's physically completely inert it's not doing anything and according to this"
},
{
"end_time": 9523.507,
"index": 372,
"start_time": 9497.159,
"text": " Whether that structure is there or not will determine whether there's a to think. But on the other hand, you can see that whether that stuff is there or not makes absolutely no physical difference to the makes no difference to the physical activity going on. So if you think that the conscious state really is determined by physical activity, it shows that this computational conceptual apparatus is the wrong one to be bringing to you."
},
{
"end_time": 9544.462,
"index": 373,
"start_time": 9524.701,
"text": " but that you can read that paper i mean it's a journal philosophy will you be able to send me the title of it so i can include it in the description it's called computation and consciousness who's public in general philosophy i mean i could pull up my cv i think it's 1989 but you can find it i'll leave a link to it in the description so does that include"
},
{
"end_time": 9571.425,
"index": 374,
"start_time": 9545.35,
"text": " formulations like integrated information theory? Well, not exactly, because as I understand it, the IIT idea involves information theoretic concepts that go beyond the Turing description. But on the other hand, I know Scott Aronson's criticisms of the IIT and I find them completely convincing, so I don't think the IIT has any more hope."
},
{
"end_time": 9597.022,
"index": 375,
"start_time": 9572.022,
"text": " of being an account of consciousness than the term machine stuff does. Have you heard of the global neuronal workspace theory? The words are a little familiar to me, but I don't know it in a way in any detail to be able to comment intelligently about it. Have you heard of Yosha Bach? Yosha Bach, who has his computational consciousness model. I don't know what it's called."
},
{
"end_time": 9624.974,
"index": 376,
"start_time": 9597.739,
"text": " Okay, what are your opinions on Wolfram's physics theory? Because it's entirely computational and seems to at least he believes it produces consciousness because we're here and he believes his theory to be correct. Have you taken a look into his physics project? Again, not in any detail. I mean, anybody who says"
},
{
"end_time": 9651.203,
"index": 377,
"start_time": 9626.647,
"text": " The foundations are computation. Computation is just the wrong concept to be bringing to physics. Why? Because it's because computers have the physical structure they have that they're capable of doing the computations that they're capable of doing. That is, the physics determines the computational structure, not the other way around."
},
{
"end_time": 9682.056,
"index": 378,
"start_time": 9653.575,
"text": " So, you know, somebody, I mean, it's a bit in the ballpark of, again, the idea that everything is mathematical, right? All physical reality is just mathematical, Max Tegmark's idea. It's so far off base to me that I don't really feel like I need to spend a lot of time on it. You know, I think the physical world, you know, I don't know, space-time, it's got particles, it's got fields, it's got some quantum state."
},
{
"end_time": 9710.947,
"index": 379,
"start_time": 9682.432,
"text": " Trying to understand how all that stuff fits together, trying to figure out what the small-scale geometry of the physical space is and space-time. This all strikes me as recognizable speculative physics, but you go beyond a certain point and I don't even recognize it as speculative. Okay, so let's just get to a couple questions on time. There's entropic time and then there's thermal time. I don't know if those two are the same."
},
{
"end_time": 9736.186,
"index": 380,
"start_time": 9711.374,
"text": " But what are your views on those explanations as to the arrow? I don't understand even what that means. There's time. There's a variable we normally designated with a T. When we assign numbers to it, we tend to assign numbers so that as time goes forward, the numbers get larger."
},
{
"end_time": 9760.708,
"index": 381,
"start_time": 9736.834,
"text": " To say that there's entropic time, I don't even understand what that means. I mean, it's true that if a system is not at its entropic maximum, it's not thermal, that typically as time goes forward, it will become closer to thermal equilibrium."
},
{
"end_time": 9777.79,
"index": 382,
"start_time": 9761.954,
"text": " That's a fact about, and we understand why that's so, I mean there's pretty good statistical understanding of why as time goes forward entropy tends to increase, but I don't think that means there's a new kind of time, there's just time."
},
{
"end_time": 9797.176,
"index": 383,
"start_time": 9778.831,
"text": " I think what's being indicated is that entropy is the reason for the direction of time. If the universe eventually, which it could in some models, if it comes to thermal equilibrium,"
},
{
"end_time": 9819.155,
"index": 384,
"start_time": 9799.053,
"text": " And so there's no entropy gradient. The entropy just reaches its maximum and stays there at least for a very long time. It's still the case that the time is going from the past to the future. I mean, time will still go on. Why would anyone think that time would cease to pass?"
},
{
"end_time": 9849.309,
"index": 385,
"start_time": 9819.77,
"text": " The state would constantly be changing. I mean, if you're in thermal equilibrium, things are still changing, right? I mean, a gas in a box comes to equilibrium. Its micro state is still changing all the time. And its change is a normal change from past to future. And the causal structure is that these two particles bounced off each other. And therefore, all of that goes on, even though there's no entropy gradient. I mean, why would anyone"
},
{
"end_time": 9878.097,
"index": 386,
"start_time": 9849.77,
"text": " even postulate that the passage of time depends upon an entropy gradient. It's the other way around. The passage of time explains, together with certain statistical considerations, why typically and predictably entropy goes up if it has a place to go. But that doesn't make time depend on entropy, it makes entropy depend on time."
},
{
"end_time": 9892.398,
"index": 387,
"start_time": 9878.49,
"text": " Someone that people keep referencing to me to look up is Henry Berkson. And then they say that Berkson and Bohm knew each other or had or they developed their theories together or where they were influential to one another."
},
{
"end_time": 9917.517,
"index": 388,
"start_time": 9893.456,
"text": " So now that I know that you're an extreme Bohm fan, or at least you're a proponent of the pilot wave theory, what are your thoughts on Berkson? I read a book about Berkson and Einstein. I think, you know, Berkson was tremendously confused. I wouldn't recommend anybody spend time reading Berkson. And quite honestly, I wouldn't recommend anybody spend a lot of time reading the kinds of things Bohm wrote"
},
{
"end_time": 9942.654,
"index": 389,
"start_time": 9918.524,
"text": " In the 60s when he got involved with Krishnamurti and he becomes kind of semi-mystical and there's the Implicit Order. I mean he wrote a bunch of kind of weird stuff. You know, you want to look at his 1952 papers and then even better look at Bell's presentation of the theory, which as I say, Bohm presents it in terms of a thing called a"
},
{
"end_time": 9972.176,
"index": 390,
"start_time": 9943.08,
"text": " Quantum potential and using Newtonian dynamics and that's a bad idea. It's not really a Newtonian theory It's a great theory. Okay, and it was also again a theory that already de Broglie had seen it's the theory is kind of Almost inevitable way to understand quantum theory once you ask questions in a certain way But you know poor bone, I mean he got he got attacked"
},
{
"end_time": 10001.817,
"index": 391,
"start_time": 9972.995,
"text": " for political reasons in the United States, kind of exiled to, you know, he had a life that was, you know, one of being rejected, having been at Princeton and having talked to Einstein and having been influenced by Einstein, and then he ends up, you know, in Brazil and then eventually at the LSE. I have a lot of sympathy for the man as a human being. But I think it made him, his rejection by the physics community,"
},
{
"end_time": 10028.183,
"index": 392,
"start_time": 10002.551,
"text": " made him more open to align with Krishnamurti and some people who were of a more mystical bend. Personally, I don't see anything in that that's helpful for me to do physics. So I would say study the theory"
},
{
"end_time": 10057.79,
"index": 393,
"start_time": 10028.712,
"text": " I mean, he wrote this later book called The Undivided Universe, and there's some useful things in that together with Holland or not, anyway. But I wouldn't get too wrapped up in Bohm the individual because he was, I mean, he was quite, unless you're just interested in him as a personality, I think a lot of the peculiarities of him as a personality are not useful for understanding physics. Did you ever get to meet Bell?"
},
{
"end_time": 10074.923,
"index": 394,
"start_time": 10058.37,
"text": " Sure. Yeah, I met him three, four times. What was that like? Oh, he was just the most wonderful human being. I mean, completely unpretentious, completely open, completely, I mean,"
},
{
"end_time": 10090.742,
"index": 395,
"start_time": 10075.691,
"text": " Bell, I'll tell you a story. Unfortunately, he was supposed to come to Rutgers when I was there, and he died about a week before he was supposed to come, which was a great tragedy for us, but I met him"
},
{
"end_time": 10117.21,
"index": 396,
"start_time": 10092.039,
"text": " at a philosophy of science association meeting, but not really talk to him there. I met him in a riche at a meeting and I met him at Rutgers. He came to Rutgers before and talked to him individually. But an interesting thing happened. I'll just tell you a story because it's so much to me, Bell. The first time I heard, I mentioned this Gerardi-Romini-Weber theory, which is an objective collapse theory, and Bell"
},
{
"end_time": 10144.787,
"index": 397,
"start_time": 10117.875,
"text": " has a paper called Are There Quantum Jumps, which is again in this book, which is his presentation of the GRW theory. Although he goes beyond in a certain ways what they did, but a very nice presentation. And the first time I heard about that theory was from Bell himself. It was in Ariche at this NATO summer school. And I was somebody who grew up"
},
{
"end_time": 10172.278,
"index": 398,
"start_time": 10145.299,
"text": " with the common idea that, okay, there are these collapses of the wave function and they're associated with measurements and measurements are associated with consciousness. This kind of idea you associate with Eugene Wigner. And therefore, oh, this is amazing. This is where consciousness connects with physics, right? That consciousness causes the wave function and so on. So you think the most important question there could be is what triggers these collapses?"
},
{
"end_time": 10194.718,
"index": 399,
"start_time": 10173.882,
"text": " and bell presents the grw theory and the thing is in that theory nothing triggers them they just happen at random with fixed probability per unit time per particle there's no environmental trigger there's no anything they just happen okay they just happen at random um and and so bell gives this talk and i went up to him afterwards and i said"
},
{
"end_time": 10222.415,
"index": 400,
"start_time": 10195.213,
"text": " I said, you're satisfied with this theory? I mean, here we've been for years and years worried about what causes these collapses, what causes these collapses and their answers, it just happens. And you're happy with that? And he just he was so mild. I mean, I can't express to you what a nice man he was. And he was so mild. And he just kind of smiled. And he just looked at me and he said, you don't appreciate what they've done. And he was absolutely right."
},
{
"end_time": 10244.582,
"index": 401,
"start_time": 10222.858,
"text": " And there was no point arguing with me at that moment, but he said exactly the right thing. What they did was they took this idea of a collapse and gave it a precise mathematical formulation. And so they turned a notion into an actual physical feeling."
},
{
"end_time": 10272.619,
"index": 402,
"start_time": 10245.725,
"text": " Whether you like it or not, I mean there are things about the theory one could dislike, even Girardi, who I got to know quite well, never thought it was a final theory, always thought it was a kind of transition point to a better theory, but they managed to just write down some equations, not mention measurement, not mention observers, and get a theory that gave you basically the right prediction."
},
{
"end_time": 10303.66,
"index": 403,
"start_time": 10273.848,
"text": " So Bell was like that, and he was just always tremendously on point, on target. But he just seemed to be the nicest person in the world. I mean, after he died, there was a eulogy session for him at the Philosophy of Science Association. I remember Phil Pearl, who had been working also on objective reduction theories, continuous ones. He got up and he said, for him,"
},
{
"end_time": 10331.937,
"index": 404,
"start_time": 10304.718,
"text": " Bell, you know, he was somebody, Phil was somebody working on these foundational issues and he said for him it was like Bell was kind of a knight on a steed to protect them, right? I mean he was the guy, they're doing things that are very unpopular in the physics community and they could always count on Bell to be there to kind of protect them and to say you're doing, you know, what you're doing is important."
},
{
"end_time": 10361.408,
"index": 405,
"start_time": 10332.244,
"text": " And yeah, and you know, you think the guy, I mean, he was a very down-to-earth guy, he was an engine, essentially an engineer, designing magnets at CERN. And he was doing all of this foundational work literally on his weekends, and when he had a sabbatical, you know, he wasn't paid to do it. But he had a tremendously sharp mind, and he had a tremendous physical intuition, right? What the physical world, and Einstein talked about having a kind of"
},
{
"end_time": 10391.647,
"index": 406,
"start_time": 10362.159,
"text": " empathetic understanding of the physical world. I mean, you felt like Bell had that. He wasn't satisfied with doing a calculation, right? He wanted to understand what was going on. You get that feeling out of Feynman and Reed Feynman too, right? That he just driving something mathematically, that wasn't enough. He wanted a picture. He wanted, you know, he wanted a feel of what the physics was really about. And he was just, you know, tremendously"
},
{
"end_time": 10420.486,
"index": 407,
"start_time": 10392.005,
"text": " Professor, thank you for spending so much time with me. Thank you. I mean, it's been fun. I hope it was more or less what you were looking for. It was more. It was more. Thank you. All right. Well, I do have a side question. How do you keep your mind sharp?"
},
{
"end_time": 10441.118,
"index": 408,
"start_time": 10422.534,
"text": " You may say, well, my mind is a sharp, it's dull, but how is it that I don't think that you have a dull mind? So I imagine you have some practice, maybe it's running, maybe it's your read, maybe you have a certain ritual. I mean, look, what I do that actually has an influence on that, I don't know for sure. I mean, I know from the little I read,"
},
{
"end_time": 10462.927,
"index": 409,
"start_time": 10441.596,
"text": " that some of the things I do like crosswords actually don't make any difference. So people say if you do this and this and this and then they think well no actually and of course people do say that exercise makes a difference and I do try to get a reasonable amount of exercise maybe that helps. I don't know. I don't have the empirical basis to tell you"
},
{
"end_time": 10488.166,
"index": 410,
"start_time": 10463.951,
"text": " There is a question about Judea Pearl because he has a book on causation. I wanted to ask you, apparently you have some objections to his, but if it's too long, then we'll save it for the next time if we speak again. Yeah, I mean, look, I mean, there's a funny thing about Pearl. If you're interested in Pearl, I should just mention it's relevant to understanding. Is it what Pearl's doing is is"
},
{
"end_time": 10518.029,
"index": 411,
"start_time": 10489.684,
"text": " is both similar to and was influenced by work that my dissertation director Clark Gleemore did together with Richard Shinus and Kevin Kelly on causal modeling and causal discovery. And I just kind of have a feeling that Pearl gets more attention for that relative to"
},
{
"end_time": 10546.459,
"index": 412,
"start_time": 10518.592,
"text": " what he did and what they did, then it's warranted. So I'm a little, I mean, I'm a little bit, I don't want to say defensive, but I do like, if you wanted to look into it, you should look into Gleamor's work, which was some of it earlier than Pearl's. And again, Pearl really learned some of some of the techniques from them. So it's not a criticism of his work. It's more about the attribution."
},
{
"end_time": 10565.759,
"index": 413,
"start_time": 10546.852,
"text": " Yeah, there's a bit about attribution. There's some stuff that he does in his book. I have this review of this more popular book. You probably saw that in the Boston Review. And there is some stuff there that I just disagree with. It's not the technical stuff. It's this"
},
{
"end_time": 10594.309,
"index": 414,
"start_time": 10566.152,
"text": " you know, this kind of more hand-wavy stuff. The technical stuff is just writing down causal graphs and representing various interventions and then seeing what kind of statistics you should expect in different experimental situations, given different causal structures. And all of that I don't have any objection to, but it is the same thing, more or less, that Cloumo was doing."
},
{
"end_time": 10616.971,
"index": 415,
"start_time": 10596.493,
"text": " The podcast is now finished. If you'd like to support conversations like this, then do consider going to theories of everything.org. It's support from the patrons and from the sponsors that allow me to do this full time. Every dollar helps tremendously. Thank you."
}
]
}No transcript available.