Samir Varma is a physicist, investor, and author of the book: The Science of Free Will, How Determinism Affects Everything from the Future of AI to Traffic to God to Bees.' We discuss AI consciousness, the theoretical versus practical aspects of free will, and how determinism informs our understanding of human behaviour, moral philosophy, and even financial markets.
“Even though there is no free will in theory, because the laws of physics are deterministic… we don’t know what we’re going to do until we’ve done it.”
“The only way to find out how your life will end is to live it.”
Varma delves into the physics of free will, his insights on the applicability of chaos theory, and why certain financial and societal structures might need reevaluation.
The conversation touches on the evolution and mechanics of intellectual property, the intricacies of cricket, and the profound layers that make Pink Floyd's music a unique listening experience. Varma also shares his varied projects, including innovations in solar energy and traffic management.
The discussion culminates with practical life advice focused on pursuing meaningful and passionate work.
Summary contents, transcript and podcast links below. Listen on Apple, Spotify or wherever you listen to pods. Video above or on Youtube.
Contents
00:23 Will AI Gain Consciousness?
02:11 The Concept of Free Will in Theory and Practice
04:19 Quantum Theory and Determinism
18:58 Chaos Theory and Free Will
23:20 Death and the Reorganization of Atoms
25:37 Animal Consciousness and Communication
30:23 Implications for Society and Justice
37:48 Emotions and Rationality
41:08 Misunderstandings in Finance
44:39 Debunking Market Value Myths
47:43 The Momentum Factor Explained
57:14 The Unique Appeal of Pink Floyd
01:02:40 Understanding Cricket's Complexity
01:09:58 Overrated or Underrated?
01:19:28 Insights on Writing and Creativity
01:23:11 Final Thoughts and Life Advice
Transcript (This is AI assisted, so mistakes are possible)
Ben: Hey, everybody. I'm super excited to be speaking to Samir Varma. Samir is a physicist, investor, and inventor. He has written the book, The Science of Free Will, How Determinism Affects Everything from the Future of AI to Traffic to God to Bees. Samir, welcome.
Do you think AI will gain consciousness?
Samir: That's a difficult question, not because I don't think I know the answer, but because I don't know if I can convince anybody of the answer. So let me say this yes, definitely. I don't know when or how long it will take, and I also don't know how difficult it will be for people to accept that it is conscious.
Ben: And I suppose if humans can't distinguish the difference, so if it seems AI is as consciousness as something else and we can debate what this consciousness is, then I'm guessing in practice that would be consciousness just based on silicon rather than carbon, is that how we get there?
Samir: Yes, and yes, exactly.
And the point is that, as I take great pains to say in the book, the rules of physics that the AI operates upon, or operates with, if you wish, are the exact same rules of physics that we operate with. There's no difference. We're made of atoms, the AI is made of atoms. All those atoms are identical, all those atoms are following physical laws, the standard model of particle physics, because atoms are made of fundamental particles, and the AI follows those laws, same as we do.
So there's absolutely no reason to think it cannot be conscious.
Ben: And so there is some emergent property, some complexity property, of certain arrangement of atoms, whether they're silicon or carbon, which then we perceive as this consciousness. Is that what could happen then for AI?
Samir: Exactly. Couldn't say it any better.
Ben: So I think this hints at this idea of something in practice versus theory. And I'm really interested as to how you develop this idea of, okay, there might be free will in theory, determinism, laws of physics, but functionally or in practice, because we cannot compute what we're going to do or what certain of these systems are we going to do until we run them until we see them in practice, you've got a functional.
How did you come about it and would you like to explain the concepts as you see them now?
Samir: Yes, so I came about it because I've been thinking about this for a very long time. And in fact, I have to tell you a funny story about it. My professor, my advisor for my PhD thesis was a guy called George Sudarshan, who should have won two Nobel Prizes but didn't win either, never mind but he was extremely interested in philosophy.
And I remember telling him when I was a young 20 something year old, I used to call him boss. I said, boss, you will never catch me doing any philosophy. This is not hard science. And he's you wait. So I started thinking about this stuff and it just stuck in my head for years and years.
And then it in fact occurred to me that you can have functional free will, even if you don't have theoretical free will, because you cannot actually compute what you yourself are even going to do. Forget somebody else under those same laws, even if the laws are known. And I realized that, this was probably the right answer, but I didn't do anything about it until Tyler pushed me to actually write the book and so I did.
And in that, I had to coin two phrases to put the two concepts together, and one phrase was free will in theory, where I'm using the word theory as a physicist would, that is to say an explanation of a bunch of collective facts, and free will in practice, where I'm saying, look, Even though there is no free will in theory, because the laws of physics are deterministic, what do we really mean when we say we have free will?
Because we all feel like we have free will. And the answer is that we don't know what we're going to do until we've done it. And so the only way of knowing what we're going to do, the only way to find out how your life will end, is to live.
Ben: That makes sense to me. Guess when I was asking around, and I'm not a physicist by training, although I did some science, there were some ideas or people were saying from quantum theory, that quantum theory wasn't necessarily deterministic, but you have this whole set of arguments in your book that essentially it is.
And so people say in quantum theory, you've got this idea that the observer makes a difference, and I guess you have this collapsing of the wave function. But if I understood your argument correctly, that's still deterministic. So how do you fit quantum theory or quantum thinking into these laws of physics and into free wheeling theory?
Samir: Yes, so that, that actually is the most crucial of all questions, and the most important one. And the answer is the following. The first is to understand that no physicist that I'm aware of takes the Copenhagen interpretation, which I will explain in a second, of quantum mechanics particularly seriously.
It's an aid to calculation. It gives you the right answer, but it is effectively an approximation. What is the Copenhagen interpretation? The Copenhagen interpretation put forward by Niels Bohr, which is why it's called the Copenhagen interpretation, is that there is a classical system that is an observer, And there is a quantum system that is being observed.
And at the moment of observation, all of physics is suspended, don't ask any questions, don't look behind the curtain just, you will get a probabilistic answer. That's the Copenhagen Interpretation. No one that I'm aware of takes it seriously, in terms of it actually being real. And more and more experiments, and I go through this in the book, show that it probably isn't real.
In fact, it's not real. I use probably because I'm still a scientist, but really it's not real. So what is real? What is real are two things. The first is that the observer is made up of the same stuff as the observee. Therefore, all that can possibly be happening is that the wave functions of the observer are getting entangled with the observe, with the wave function of the observe V and that what is an, that's what makes an observation.
That's the first point. And that's all completely deterministic. The second point is that. When we then assign a single value to that court observation, to that interaction between the observer and the observee we have to assign a single value even though there's a range of possible outcomes. But, that range of possible outcomes is self deterministic.
No one can change them. In other words, if I do an experiment here in New York, and you do an experiment there in London let's say that we, and we set up the experiment in exactly the same way, So we set up an electron such that its spin is unknown, and we have two detectors at, specific angles, and we do a bunch of experiments, we will find the exact same probabilities.
In other words, the probabilities are themselves deterministic. They're not under anyone's control, and so therefore they cannot be the source of free will anyway, even if you decide to take the Copenhagen Interpretation seriously. That's the point. And so then you really are left with it's all deterministic, because there is nothing else in physics, nothing that is not deterministic.
Ben: And you mentioned that there have been a few experiments which are backing this kind of argument. Do you want to highlight your favorite one or any that you think that this means that this is the correct response and this is why the Copenhagen explanation of it isn't real or there is a kind of useful heuristic to, to grapple your mind with?
Samir: Yes. So my favorite is that the statement of the Copenhagen interpretation is simply that a extended object, a classical object, a large object, a big object, whatever. Big means made of many things, not just one thing. Is going to cause the wave function to, quote, collapse, unquote.
That means that large object itself cannot ever be in a superposition. But we've already been able to put a bacterium, I believe, in superposition. So if you can put a bacterium in superposition, then the entire argument collapses. There is no, which is fun. It's a pun. The entire argument about the wave function collapse, collapses.
It doesn't make any sense because a bacterium is by definition a classical object.
Ben: Okay, I think I get that. And I guess another line of evidence, so one that I came across a few years ago, which first started me down this route on the free will in theory idea, was from neuroscientists actually more than physicists.
where there's a set of experiments which show that these decisions seem to happen unconsciously as in before we seem to be consciously aware of them before our intent comes through and our body or however it is actually already reacted to it. Again, arguing for a form of determinism within that.
Do you find any of those experiments compelling or any other evidence from neuroscience, which you weigh up on either side?
Samir: Yes. So I think I may have put this in a footnote about fMRI experiments showing that you are, you make a decision before you're conscious of making it. I have nothing.
Against those arguments that are all 100 percent correct. The only problem is that they don't actually seal the deal, as it were, because there's still loopholes that you can come up with that say that doesn't necessarily mean there isn't free will, because, what if the particles have free will or whatever it is, something along those lines.
It doesn't completely seal the deal, but it's the physics that actually seals the deal completely. That was why I did it that way.
Ben: That reminds me, as a joke, because I remember my chemistry teacher or professors always joking that chemistry was upstream from biology. So they said all biology can be explained by chemistry.
And even the chemistry that you can explain all neuroscience. And then the physicist coming in and going the thing about chemistry is all chemistry has to follow the laws of physics. So really physics is upstream of both chemistry and biology and everything human. And I guess this is one of the arguments you make, but I was always intrigued because the mathematicians.
would come in and go actually physics is downstream from mathematics. So everything you can describe in the language of mathematics. And again, I'm not a complete expert on this, but it seems that there are some things that maybe we can describe in mathematics, which we haven't as yet found or described within.
Physical laws, although we're not there yet. Do you see physics and maths on the same sort of bucket? They're just looking at things and obviously a lot of physics is described in the language of mathematics when you get there. So maybe there isn't a difference, but certainly you would argue everything is downstream from physics is maths downstream from physics as well.
Samir: That's not a simple question to answer. And there's a reason for that. It depends on whose viewpoint you want to take on this. One viewpoint is Stephen Wolfram's, who I quote in the book frequently, is that he creates an object he calls the Rouliad. And the Rouliad is the entangled limit of all possible computations.
And he says that's an object, such an object must exist, and because such an object must exist, Therefore, the universe exists. That's his argument, roughly speaking. Is that accurate? I don't know. But it is his argument. And if that argument is true, then yes, physics is downstream of mathematics. On the other hand, there is the traditional physics view, which is that the universe follows deterministic physical laws.
You can write down those physical laws in basically any way you wish. One way of writing them down is via physics equations and those equations are, I believe the term is isomorphic to reality. That is to say they're in a one to one correspondence with reality. That's a fair, more traditional physics view.
In that case, physics and mathematics would be on at the same level because there would be other ways of of expressing those laws, perhaps not as compactly, perhaps not as easily. So that's the second point. And then the third point is. That, again, if you take Stephen Wolfram's view, everything is downstream of computation, including mathematics.
So computation comes first, and then physics and mathematics follow. So it's, that's a little bit of an open question. I don't know is the right answer. Do
Ben: you lean anyone, any particular things? Do you have a, do you have a sense, even though we, I guess the problem with Wolfram's view is currently thinking about it.
We have no real way of. proving or disproving it, as such, we might get there. But do you have a sense I pick up that you lean slightly towards the physics view, but maybe you quote Wolfram a lot, so maybe you are slightly persuaded by him.
Samir: I'll tell you what I am persuaded by. I do lean towards the physics view.
I am persuaded by one of his arguments which is that everything can at bottom be thought of computationally. And the reason I think that is that I believe fundamentally okay, let's back up one second. Everything in physics except for space time is already discrete. The only question is space time discrete?
The way we do traditional physics, it's not. It's a continuous object. Thinking of that computationally is then an approximation rather than reality. However, again, I'm pretty sure that most physicists working in, say, quantum gravity or in grand unified theories or whatever, would more or less agree that the chances that spacetime is discrete too are pretty, pretty darn high.
If spacetime is also discrete, then Wolfram's view that everything is computational is a very good way of thinking about it, because if everything is discrete, you can represent everything computationally. And if you can represent everything computationally, then it's a great way of thinking about it.
Ben: Huh. I hadn't thought about that way. I'm going to have to dwell on it, but that's quite interesting. Everything is computational. That reminds me, I sat next to dinner to a physicist, a sort of non mainstream physicist, Julian Barber, who over dinner tried to convince me that time didn't really exist.
There are these other concepts of ways of thinking about it and holes and things but from your physicist view and I think he noted that was a lecture and they had a lot of theoretical physicists and in the room when you asked physicist to say, is time a discrete idea that not as many hands went up as you thought, although I guess people think about space time.
But I guess this might touch on this. So do you think time is actually a thing out of it of itself, or have we just made it up as humans?
Samir: Time? I believe is almost certainly an emergent property as opposed to a fundamental property. Does it exist? Sure, it exists in the same way that you and I exist.
As a collection of something. But, is it absolutely fundamental? I don't think so. And the reason I think that's the case is because of, we could get into this if you like to, but it's because of entropy. Entropy always increases in the universe. And because entropy always increases in the universe, you have an arrow of time.
And that leads to all kinds of other questions about, for example, how come, the universe used to be in a low entropy state and now it's in a high entropy state and so on and so forth, and that leads to a whole host of complicated questions. But the fundamental laws of physics are all what's called time reversible.
In other words, you can't tell whether a picture of what's going on is moving forwards or moving backwards. And, therefore the arrow of time is emergent as opposed
Ben: to fundamental. I haven't thought of time as an emergent property or collection or something. In which case, time has somewhat more unco in common with consciousness than it does with something else, if consciousness as well is potentially an emergent property.
Samir: I think actually they're the same thing. Think in some sense about what consciousness actually is. If you are a quantum system, you are, I am, then in effect there's all kinds of things going on in parallel inside us. That's the whole point of quantum computation is that you can do all kinds of parallel things at the same time.
We're made of 10 to the 27 atoms. The 10 to the 27 atoms are made up of countless subatomic particles. All of them are in all sorts of weird quantum superpositions all the time. The very existence of those quantum superpositions. It tells you that there's all kinds of stuff going on in somebody's brain, but what they're conscious of is only one thing at a time.
It's our way of taking parallel stuff, parallel computation, if you wish, and serializing it. So in effect, consciousness is the serialization of parallel physical events.
Ben: That does make it, in a way, our sense of time is consciousness. Yes.
Samir: And so here's another thought experiment. If you do the mathematics of general relativity, then as you approach a black hole, time for you, as seen by an outside observer, slows down.
You move, ever so slowly. More slowly. As far as the outside observer is concerned, when you reach the event horizon of the black hole, for you, time will have stopped. Him looking from the outside will never see you cross the event horizon, ever. But, what that means is that from his point of view, all the computations going on in your brain have now just stopped.
And if your brain is doing no computation, how can it be conscious?
Ben: I'm gonna have to think about that one. That's an interesting thought experiment. That reminds me, I was having a conversation with someone Julian Goff and a couple of other reading around with this idea that potentially a whole universe gets born in the middle of a black hole or a black hole could give birth to universe.
Have you come across those ideas and do you think they might possibly be true?
Samir: There's some evidence. In fact, just came out in the last couple of days. I haven't read the papers yet. That suggests that's actually a slightly stronger possibility than we thought.
Yeah, it's completely possible. You could have We could, for example, be living inside a black hole in our entire universe. And that black hole could be inside another universe. for example, or inside a multiverse or whatever you want to call it.
Ben: Yeah, that does seem to be, yeah, possible in the implication.
I wanted to bring through some of the ideas of chaos theory, I guess here, and talking about computation, this idea of computation irreducibility, I think is this Wolfram idea. Yes. Essentially, we take it back to this idea that Unpredictability, I sense we have to run the program, we have to do it before we actually know what we're going to do, and, this is coming through to free world in, in, in practice, and I guess chaos theory shows for some of these systems, think about the weather we've only got moderate ability to know what's happening with the weather.
We need to let it play through. So what do you think that chaos theory has to add to this idea of free world and how does it intersect with all of those ideas?
Samir: We actually have three entirely independent sources of free will in practice. And the reason I say they're independent is that if you want to say that we don't have free will in practice, you would have to reject all three.
Let's start with the first one. The first one is simply that, that we are made of 10 to the 27 atoms. Simulating 10 to the 27 atoms is hopeless. We're never going to be able to do it. Furthermore, If you want to, say, predict what you or I are going to do over the next one second, it's not just those 10 to the 27 atoms that we need to focus on.
We need to focus on every subatomic particle that might impinge upon us in the next, say, one second, if one second is the period of time over which we're going to look. That means that we need to, since the speed of light is about 300, 000 kilometers per second, that means we need to know about every subatomic particle between us.
and a radius of 300, 000 kilometers. So that's a lot of space and a lot of particles. We're not going to do it. Okay. That's point one. You could object and you could say that's just an engineering challenge. Come on. Eventually in the fullness of time to use the old, yes, minister phrase, we'll be able to do it.
Fine. That's a reasonable argument. But then you have argument number two. And argument number two is chaos. And that is that in many of these complex systems, their outputs are exquisitely sensitive to their inputs. So if they're exquisitely sensitive to their inputs, it means that in many cases inputs are always real numbers, right?
Some number followed by a decimal followed by lots of other numbers. And essentially they're infinite because you can never have an exact real number. What happens if the system the outcome of the system depends upon the 70th decimal place of the number? And the 70th decimal place is a 6 instead of a 7.
Now, the outcome could be completely different. So that puts a significant limit on our possibility of being able to make a prediction, is that we can never exactly measure anything, and so we can never exactly know any real number that's an input to a predictive system. But you could again argue back to me and say, listen, wait a minute no, that's not good enough.
That's still an engineering challenge. Okay, so we'll measure to the 175th decimal place. Come on. Fine. To which I say, all right here's my trump card. I win. These are my aces. If you are willing to believe that the universe is fundamentally discrete, then the universe itself can be thought of as a computation or at least something that follows the rules of computation.
It doesn't need to be a system designer. It's just a system and it follows rules. Those rules are going to be, as of necessity, stuck under computational irreducibility, which is what Wolfram calls it. And computational irreducibility is the statement that even very simple rules, forget rules as complicated as the universe, just incredibly simple rules, I like to say rules that even a five year old can follow, will produce output that cannot ever, under any circumstances, be shortcutted.
That is to say, the only way to know what the output is to follow the number of steps required for the computation to get to that point. And that's actually a limitation of science from within science. So that's basically saying that, look, you can know the rules by which something operates. But you may not have any shortcut to calculate the result.
You may not be able to make a prediction, but it doesn't mean you've not done science. And the inability to make a prediction doesn't invalidate the science.
Ben: Yes, that makes sense to me with that. You can't reduce it. And actually, so putting That together with your determinism view Reconciles the two ideas of free will and no free will as being both valid but somewhat incomplete that you have to put them Yes together. So I was going to see what some of this might be in practice Let's take it as read as in terms of what it might have for society but before that there were a couple of Things taken to the extreme or more extreme, which I thought were really interesting Before trying to deal with moral and society implications and one which was on death Because you're writing on it essentially just argues death is a rearrangement of atoms, which I guess in theory Is just true in a deterministic world.
We, the atoms have just rearranged. Maybe these emergent properties have disappeared. So how should we think about death? And if death is just this reorganization of atoms, should that somewhat comfort us?
Samir: Yes. And we can say more than that. We can say several things. The first thing we can say is that death is a process, not an event.
So different parts of your body are irreversibly, unarranged, as it were, at different times. and could possibly be brought back to life if we have the right technology. Second, since you're only made of an arrangement of atoms anyway, if there was some way of recording what that arrangement is at any given point in time, then there's no law of physics that says you can't put that arrangement back together in the future.
So you could, for example, back yourself up today, and, reconstitute yourself when you die, as of this point. The same way you reconstitute a computer. There really is no difference. That's the second point and the third point is that the moment you understand that you are nothing more than an arrangement of atoms and the only thing special about you is your arrangement, then the theological point of a soul actually is answered.
What is a soul? Your soul, your essence, your being. Is the arrangement of your atoms. That's what you are. And does that arrangement exist always? It depends on whether you believe Plato or not. In some sense, an arrangement can always exist even if it's not instantiated. And in some other sense, that's not true, because you have to instantiate it for it to exist.
It's, it's up to you how, where you want to look at it, but that is in effect a proper scientific, grounded, non nonsense explanation of what a soul is.
Ben: Thinking about that definition, then animals, say an octopus or a dog, would have a form of soul as well. Absolutely. Not everyone that we humans understand.
Absolutely. No question. So I was reading your article on how dogs can't really communicate with us or we don't seem to communicate with them really well. As in we don't really understand them but they obviously can do things. That we can't do through smell and all sorts of other things. And it also struck me.
This was really true. I think about the octopus a little bit. I'm having watched them and read about them because I think an octopus might have its form of humor, its own form of society, its own form of culture, which essentially means I think there's an argument that. As if we'd met an alien, we might as well meet an octopus.
That it's a kind of system that we're very, we interact with really poorly. To the extent that octopus has a humor or has a society, it's so different to our own way of encountering that, that we don't understand octopus humor. You think that kind of might be true then? This is how he thinks, and seemingly you would think that animals have souls.
Samir: I would say 100 percent that's true. Everything you just said is true in my opinion. And I think demonstrably true actually, so it's not even really my opinion, but never mind. Ah and I would say two other things. The first is that related to what you just said, and this is why bees are in the title of the book.
Did you know that bees can get PTSD? This blew my mind. This absolutely blew my mind when I was doing the research. It's got a brain about the size of a pin. pinhead, sorry, right? And it can actually have emotions, it can dance, it seems to have feelings it has, mental breakdowns, PTSD. If something that small is exhibiting things that look like emotions, then really we, there's no question that all these animals have some sort of sentience, not like ours, and we have difficulty communicating with them.
And that's why I also gave the example in the book. From from another researcher, Katya Grace, I think her name is. And she said, why don't we trade with ants? Because if we could trade with ants, they could do all kinds of useful things for us. So we could do useful things for them. For example, they could detect cancer for us pretty well because of, how well they, because of the way they're structured and, we could build houses for them or give them food or whatever else they wanted.
The reason we can't communicate is the reason we don't trade with them is we can't communicate with them and there's the issue It's exactly the same as the issue with the octopus. Dolphins, for example, have been known to save drowning humans Yeah They see a human drowning and they come and they push the human up to give it air so that Smart enough to know that this person is in trouble.
There was an accident video not long ago of, was it in, it was a giant ape, I forget which one, that saw this man struggling to get out of the water in, I think, Bangladesh. And he walks up to the bank of the river, puts his hand out, pulls the man up, walks away.
Ben: That suggests to me that bees might get influenced by psychedelics and, or some form of, the equivalent of bee psychedelics maybe they only have to dance around a lot. But what do you think then is happening when humans take? psychedelics, these altered states of consciousness, I guess it's only just a rearrangement of atoms on the one level, but is it showing us anything else?
Samir: The current research suggests that's what that what is happening when you take hallucinogenics is that your brain is entering a more plastic state. Where it can be more, where the atoms can, if you wish, be more easily rearranged, which is why you find that they're starting to be very useful for things like mental health, PTSD.
Exactly. And so that we know for sure. The second thing that we can tell probably from the existence of anesthesia. Is that consciousness is the tying together of a large number of events. Because you're completely alive, you're still breathing, all the rest of it. But you're unconscious, you have anesthesia, and you don't know what's going on.
It's as if time stopped for you. Which, by the way, also ties back to what I said earlier, why I think it's taking parallel computation and making it this way, but that also tells you that experience informs this as well.
Ben: Ah, I hadn't really thought about that, but that does also seem to make sense. Okay. So moving from animals and some of the science and theory, although we put a bit of practice, I wonder how this should apply to. Society or if anything, so I think there are some philosophers who argue that free will is a useful or even necessary illusion for morality and law.
I wondered what you thought about this idea. I guess you wrote around. whether evil exists and essentially that is a human or social construct that we seem to have decided around that as well. But do you think therefore free will is a useful concept that whether it's in practice or however that we should do because of what philosophers argue and does this tie into your idea of what evil is in our society?
Samir: Partly. I hesitate to criticize philosophers too much largely because 99 percent of the time I don't understand what they're saying and, particularly when you start throwing it out, throwing out large Greek words like ontology. I start saying to myself, I'm not entirely sure I understand what you mean, and I'm not entirely sure that you understand what you mean, but anyway, never mind.
Ben: It's like talking to the octopus.
Samir: Exactly. You got it. That's exactly right. Exactly right. So the thing that they seem to always forget is that society is also following deterministic physical laws. Every atom in the universe is following deterministic physical laws. And so there's nothing wrong with one collection of atoms saying that we are going to treat this other collection of atoms as if it were A, B, C, D, E, F, G, H.
It's completely fine. That is a i, I go into this in the book about levels of explanation. That is a level of explanation that is above quantum fields. So to get from quantum fields to say law or morality or justice, you need to do all kinds of things. You have to first put a bunch of quantum fields together and make atoms.
Then you gotta put the atoms together to make molecules. Then you gotta put the molecules together to make, amino acids. Then you gotta put those together to make, organs and then humans, and then. And then you have to create, philosophy, and then you have to, on top of philosophy, you have to create law, and justice, and morality, and so on.
By the time you've got all the way up to the top the aggregate doesn't have a heck of a lot to do with the substrate. Or as Philip Anderson, when the Nobel Prize said, more is different.
Ben: Okay, I guess that makes sense. Yeah, so it's just emerged. It's so far away from those fundamental pieces Does that say the same that it doesn't really have any commentary then on? prisons and rehabilitation because there's a line of argument which suggests that if some actions are people's actions are a little bit more beyond their control than We might then we might have thought then particularly prisons or justice reform should concentrate much more on her rehabilitation than punishment.
There are other economic reasons for maybe why that might be the case. Do you think that gives any weight to those sort of ideas or to your point? Is it just so far above that? We've had to make this agglomeration of things that it really can't say too much about. What people should think about punishment or rehabilitation.
Samir: So that's a great question. And I know that you addressed this in your last conversation with Rebecca Lowe as well. Physics actually has something to say about this, believe it or not. And the reason is that if we are thinking of the brain as a computational system, then you can ask the following question.
Is pathological behavior A problem with hardware or a problem with software? That is to say, is it a problem with the neural circuits you were born with or is it a problem with the programming that was encoded in those neural circuits as you grew up? My very strong suspicion is that it is easier to fix software than it is to fix hardware.
I'm not 100 percent sure of that, but I'm pretty sure that's right. So then you run into another issue of, okay, so X committed a murder, say. Now we have good technology, advanced from today, and we realize that the reason X committed a murder is because there was some screw up in their software and we can fix it.
Is there any point now in punishing them or sending them to, to prison? Probably not. Just fix it and move on. Otherwise you just ruined a productive member of your society. And as you just pointed out we now know that the reason their behavior was very bad is because of their software. On the other hand, now supposing it's hard to fix hardware, what do you do?
I don't know how to fix it, but I can see that there's a problem in a hard way, which is a problem we see now because I think 30 or 40 percent of criminals already have mental health problems, which kind of sucks that we put them in prison. But it's hard to fix, so what do you do? One argument is very straightforward, is if you can fix it.
If you can't fix it and you need to protect society from that person, then you protect society from that person, presumably by incarcerating them or whatever it is. So the only question I'm left with, is under what conditions is punishment justified and the only way that I can think of justifying it is as a deterrent to others.
For example if somebody decides to go around swindling people, say, then how do you stop that kind of behavior? You have to stop it presumably by saying there's a significant threat that you're going to go to jail for a long period of time. Now, of course, People will argue back that person's brain made them do it.
That's true. But then society made me do the opposite too. That's why that argument doesn't really fly, because society runs on the laws of physics too. But beyond that, physics can say some things. If you can fix it. If you have to protect society, protect it. And after that, you're left with a hard problem.
Ben: Yeah, that's quite a clear. So it does make some cases easier, actually, potentially quite a lot. And I guess we do that to some extent, if someone has brain damage, very clearly, then we treat them differently. And we can argue to what extent is a mental condition, brain damage or not. And I guess we are debating that.
But to your point, if it is because of that, and because of that you're in jail, and if it's that sort of hardware problem, then you're not going to, that's not going to be so much of a deterrent. If you've got this sort of mental condition, which causes it, causes you to do X or makes X much more likely, then that deterrent factor is is not going to be such a big thing.
But in the case of swindling, it's a lot less clear whether those deterrents might be. So those are hard cases, but there are definitely. Larger categories of easier cases and we can observe that as to your point that a lot of people in the system who go into the jail system have mental conditions or You know another way of looking at it if you test them on normalized iq and things you get very low levels which suggest these brain function hardware disabilities that we would have compared to the average and other populations I wonder does then physics or this have anything to say about A few other emotions.
So we mentioned Tyler previously. Tyler Cowen, I think, recently said, he basically never feels regret. And he thinks it's useless. And that's for a whole other set of things. And he tends to have quite a very well regulated emotional state. But I think you, you wrote about this from a slightly different point of view.
So I don't know, is, does physics have anything to say about regret? And I guess we can extend this. Does it say, have anything to say about other. Emotions, I guess this would be jealousy or I don't know about falling in love, which I've only just thought about, but certainly we can start with regret and maybe move to a couple of other emotions.
Samir: So regret is the one easy one to answer from physics, which is why I put it in the book is that because you don't have free will in theory and you only have it in practice. Fate exists. In fact, fate must exist. So therefore whatever was going to happen was always going to happen. So there's absolutely no point in having regret.
So you have free will in practice to make a choice. But you don't have free will in theory for the outcome of the choice. So therefore you might as well be like, the Buddhists have been saying for whatever 2000 years now is that, do your duty as it were, as you see it. And then the outcome is the outcome and it is whatever it is.
Or as modern management theorists like to say, it's the process that matters, not the result. Or as traders too, good traders will always prefer the bad outcome. With the good process to the good, with the bad process to the good outcome. Because one is repeatable and the other one is not.
I think it's the same idea. So that's the one emotion that you can more or less say is useless. except to the extent that it teaches you something.
Ben: Yes. Are there any other emotions that we could put either in the useless or the very useful bucket? None immediately spring to mind. I guess anger is not that helpful either, unless it is teaching you something, but I guess that's quite close to regret.
Samir: Yes. And the other thing is though that You may know this better than me. I read a paper some years ago that said that a human brain that has no emotions cannot be rational, which I thought was interesting, because for whatever reason, your emotions are necessary to your rationality.
Ben: I could see that.
Emotions widely defined are giving us all sorts of signals, which help us do learning at this kind of network level. And without those signals, then your learning goes way off. And these kind of learning mechanisms are useful for these frameworks of rationality which we seem to I like to pivot towards.
So I can see you can get it getting there, or at least would be very impaired. You can see it adjacently, like people without pain find it's really hard to live in the real world because pain is giving you all sorts of information and signals, even though you might not think Oh, it hurts.
Why could I not just get rid of that? Actually, if you say pain is actually a form of emotion. We could talk about emotional pain, but the, with that, and we put it into physical pain, but it is in that bucket of factors which seem to be doing something for us. Yes, so I think
Samir: that's the one thing we can say.
Is that from physics? I don't know. Everything's from physics. I don't know. Which might be useful. Beyond that, no, I don't know that there are any other emotions that are really generally entirely useless, but that is one of them. So we, because, envy and jealousy and so on can sometimes be productive and vice versa, non productive too, most of the time, but not always.
Ben: So what is most misunderstood about finance? We touched a little bit about actually process being more important than outcome, particularly in investing. But what do you think around your work and things and finance, some of that's complexity of markets and other things, but what do you think maybe the general person should maybe know about finance or markets?
I guess this is particularly financial markets but doesn't know some. observation that you have about what's misunderstood about finance?
Samir: So the most important thing to understand about finance is that every single theory put forward about why stock prices are what they are, why asset prices are what they are is wrong.
All of them. Some of them are catastrophically wrong and the rest of them are just merely wrong. And so you need to be incredibly careful in reasoning from statements like the market's overvalued. I'm going to sell or the PE ratio of the stock is so high. It can't possibly a good investment or as people like to say.
The market is currently at its highest of evaluation and history shows that over the next 10 years, it will have a lower than average rate of return because its current valuation is so high and so on. Those statements are all demonstrably false, but it doesn't seem to stop anybody from making them.
That's the first issue. The second issue is that financial markets, as you just said, are very complex systems. and we want simple explanations. And so what we do is we even try to mathematically model these simple explanations and we try to create factors. That's everybody's favorite word. So there's a momentum factor, and there's a value factor, and there's a growth factor, and there's a size factor, and there's a, all this other stuff.
The problem with all of these factors is that they're incredibly unstable. Can you calculate them at any given point? Sure. Are they going to remain that way in the future? Absolutely not. So what did you bother calculating them for? The reason you bothered calculating them is that you have, done a whole bunch of finance courses either in NDA school or, in a PhD in finance or whatever it is, and they've given you these complicated mathematical models that, allow you to do linear regressions.
To extract what the factors are at any given point in time So what does that tell you that they're going to outperform the next n years? So then you appeal to history and you say look, you know from 1950 to 1965 the value factor did really So if I bought lots of cheap stocks out of outperform the market great Does that mean that from 1965 to 1980 the value factor will also work?
No. So then what did you bother with that for? Some of it is very good academic work in the sense of trying to create a narrative or an explanation of data, but that doesn't make it predictive. And that also doesn't mean that the explanation is in fact correct. And I think this sucks in a lot of people.
So for example, whenever you look at Wall Street research, they'll always give you the PE ratio and they'll tell you the price to sales ratio and they'll compare this ratio to that ratio and all the rest of it. I don't know. Your guess is as good as mine. I don't know if any of it is true. And then you ask what's the value of an asset?
What should it be? So then you'll get people saying we'll have a discounted cash flow model, or I have a dividend discount model, or I have a growth, whatever is model. Okay, great. How come it doesn't reduce to the current value then? Oh, cause the market is wrong. That's really dangerous. The market is whatever the market is.
That's what people are willing to pay for the asset. How can you say the market value is wrong? Now, then people will come back and say but it's a rule of thumb. If I can buy a dollar's worth of assets for 50 cents then I've done a good job. And, maybe the thing will get up to a dollar at some point in the future.
That was the original Benjamin Graham approach. Yes, for a short period of time, back in the 1930s, it was possible to buy 50 a dollar's worth of assets for 50 cents, it's not true anymore. So I don't know what the purpose of that discussion is as well. So what I'm really saying is either you find something that is statistically valid, and I don't mean statistically valid in the sense of I'm going to do this incredibly sophisticated econometrics and I'm going to tease out that effect, even if it kills me, that's not what I mean.
It should be so obvious it hits you with a two by four across the face. Either you do that or you do everything on a case by case basis. You really get into the business. You understand its drivers. You understand its risks. You basically become like a line level manager, if you wish, of that business to understand everything about that business.
And when you do that, then you may have some insight about what that business might do in the future and whether it's worth buying or shorting or whatever it is. But this general nonsense that is spread around all the time, it's just nonsense. And it drives me crazy.
Ben: Yeah, I think I broadly agree. So obviously I do the latter in terms of the day job of really this deep analysis and don't hold very much.
But I am intrigued by what I call my factor quantitative frenemies because obviously they're a big part of the market. And you're right, so people argue about the value factor, works, doesn't work, is it going to work in the future? No one knows, all the quants will just argue about it, and they'll have it in their scorecard or not.
But one which we touched upon in our conversation over email, which does seem to have worked for a while historically which has always intrigued me because I ask different people who look at this, and no one really seems to know why. So this is the other thing which really intrigues me about this thing, is this momentum factor which you actually can calculate relatively simply and still you know, if you calculated it five years ago, still seem to have some there, but actually, if you put gun to a head, people, because they don't quite know why it works and there's different views, says, are you like, for sure, even telling me 99 percent sure it's going to work in the next five years?
They were like no, we can't really say that because we've only got this historic evidence and we're quite not quite sure it works. But so are you saying that future looking for a lot of factors? And what do you think of my momentum? And I might roll that into that. So if we're saying that a lot of stuff on this macro level, we really can't predict outer sample, even though we might look at history can we say the same for the risk environment?
Is it possible to say actually with these 27 events going on in the world or these other things, we might say that the world is in a riskier place. And so that might be riskier for some sort of markets, even though we don't know what's going to happen. Can we say something about the risk environment, even if we can't say something about future returns?
Samir: Yeah, so those are two interrelated and extremely deep questions i'll try to answer them if I get off on a tangent you can stop me and re ask the question, because I might forget where I'm going, because there's a lot of dense stuff here to unpack, because this is, it's a very good question.
Okay. Let's start with the momentum factor first. You're right. It is the only demonstrably real factor in the entire market. And so what is the momentum factor? The momentum factor is simply the statement that if over the medium term something has outperformed the market, then over the next short term, it is likely to continue outperforming the market.
And the reason that everyone is pretty confident that this factor is true is that it shows up in essentially all markets, essentially all, all around the world, essentially. Over, over the last 300 years worth of markets, depending how far back you go. And it shows up in stock markets, shows up in commodities markets, shows up in, all kinds of things.
Why? There are, I believe, multiple interrelated reasons for that. The first reason is that people have a limited attention span and a limited quantity of information that they can process at any given time. And of course they're also in some sense doing Bayesian updating. So let's say that you have a stock that's, doing something.
It's doing, let's say it's doing reasonably well, but nothing great. And then a piece of good news comes out. Now as a good Bayesian updater, you would say, okay my priors were that this is an average stock. Oh, some good news came out. I'm going to update it to being it's a little bit better than usual.
But the problem is that piece of news might actually be really good. And it may mean that the stock is, in this particular case, severely undervalued. But you can't update your beliefs fast enough to incorporate all of that into the stock price. So that means that as your beliefs are updated, you will update the stock price and the stock price will go up.
That's one reason for momentum. The second reason for momentum
Ben: And the downside as well.
Samir: And that's true on the downside as well. Precisely. You got it. That's it. That's it. Completely true. The second reason for momentum has to do with the risk thing that you just brought up. Oddly enough, while it is pretty darned difficult to write models that will give you alpha, which is the fancy way of saying I can pick stocks that outperform it is actually not so difficult to write models down that will predict beta.
Or to put it another way, to predict the quote risk in the market. Now beta is a terrible definition of risk, but at least it's predictable to some extent. What that tells you is that in general, volatility in the market is clumped. That is to say, you have periods of quiescent volatility where basically nothing happens.
And then you have lots and lots and lots of volatility compressed into a very short timeframe. It's a, it's the same as the life of my father was a fighter pilot and his life consisted, as you always said, of sheer boredom followed by complete terror. It's no different in the markets because that's, it's sheer boredom followed by terror.
There's nothing in the middle. So you have that second point, you have punctuated equilibrium, if you wish, or however you want to think about it, of volatility. And to the extent that you want to think about volatility as risk. It's predictable. That's the second point. The third point is that you also don't know what the unknown unknowns are, to use Donald Rumsfeld's famous phrase.
And you can't model those unknown unknowns, so you always have to have some margin for error, if you wish, when you are trying to decide whether to buy something or not buy something. Which will also always cause you to be slightly delayed, or severely delayed for that matter, in your updating of the valuation of something.
object or asset. That's what typically leads to things like, for example, NVIDIA. No one realized when the first news came out about just how big AI was going to be and just how much it was going to need GPUs, except for people that were working in the artificial intelligence industry. Many of whom I know personally who were not professional investors who loaded up the truck on NVIDIA.
And they're now multi zillionaires because it's gone up, whatever, a hundred fold or something like that. That's specialized industry knowledge that is not incorporated into the stock price. And couldn't anyway, because no professional would say, wait a minute, this thing's going to go up a hundred folds.
I'm going to buy, 50 times higher than it is today. It's not going to do it. So it takes time valuation should be anyway. So you put all of that together. And you will actually get more momentum in stock prices, and you'll actually see the momentum being slow on the way up, relatively speaking, compared to how it goes on the way down.
Because on the way down, because most people are long, you have the situation that you'll get stock starts to drop, you get a margin call, so it drops some more, so you get a margin call, so it drops some more, so you get a margin call, it drops some more. That's why it drops so suddenly. Goes up more slowly because of updating of information, drops quickly because of things like margin calls and liquidity constraints, and risk metrics and concentration limits and all the other stuff that you know, very well from, your other job that you have to follow, even if you know that they're not a good idea.
Ben: That's the most comprehensive explanation of momentum that I've heard, and I hadn't appreciated that it's been seen across so many markets.
I know it has exhibited in other things, but I hadn't realized it was so well. Preserved and longstanding too, by the way. Yeah. And longstanding for this amount of time. But it has a, essentially a, call it a psychological, in part, psychological and behavioral route, which is probably why it's proving so.
So robust, because as long as humans think like this, it will likely continue. Yes.
Samir: And so then that leads to the question of what happens if there are more and more computers being run by AIs that are doing the trading, but that's a different question.
Ben: Yes. But I guess if AIs have also thought that momentum is a real thing, then they will likely mimic to some extent humans for at least quite a time.
Yes. Maybe there might be a different equilibrium. Yes. Yes. And the second point
Samir: is of course that AIs Don't own their own money, right? So the money is always some humans.
Ben: Yes.
Samir: And you know very well from, again, your professional career, that the time at which a human wants to pull the plug is the very time at which the strategy, if it's a good strategy, is going to start making again, making money again.
Always. They always pull money at the exact wrong time.
Ben: Yeah, it's that emotional thing.
Samir: And that's going to happen in AI too.
Ben: And also, we've seen these AIs are taking on, I don't know, for want of a better word, these human consciousness things, if you're, I'm more, I, in fact, I've always generally been quite polite to these non human things anyway.
But if you're polite to your AI agents, like your GPTs and stuff, you get better answers. And you ask the same thing and you just ask them to be better and you're nice to them. And in general, on average, that's humans too. So there's this kind of human consciousness. Trait. So I think we might have this symbiotic thing, and wherever, however they've learnt it, and I guess we don't quite understand how, and in the same way that we might don't quite understand some of these emergent properties then that's how it then that's how it would be.
Thinking about emergent properties, actually this has come across as well. Do you think they, They fit on other levels, so when we have consciousness, we maybe have time, we have actually maybe emergent properties from ants and bees and things like that. I was thinking, is the emergence of someone like a Donald Trump or Elon Musk or on the other side, someone like a Greta Thunberg, almost an emergent property of society.
If we have enough thinking around climate, then someone like Greta appears. Because if you were to have written a story as a fiction story, people would say I don't believe that a schoolgirl sits outside parliament in Sweden and is suddenly a figurehead for a whole movement. Similarly for Trump, maybe slightly less spectacularly, but if I wrote Trump's story as a fiction story, I think most people would say I don't believe that would happen, but also both things have happened.
And I was debating with someone the other day who thought that maybe it's just some sort of emergent property of societies and complexity that we get these figures which emerge from movements within society. Do you think emergent properties can work on that level?
Samir: Yes, there's some evidence to suggest that's true.
I don't know how well applicable it is. The evidence I'm thinking of is from Thomas Schelling. Yeah. The economist who won a Nobel Prize. Isn't he the guy that put forth the theory about, what is it, focus points or something like that?
Ben: Yeah.
Samir: Where he was basically saying, imagine that somebody tells you, I'm going to meet you in New York City at noon on September 25th.
How do you know where to meet? And, you'll be drawn to the clock at Grand Central Station because that's more or less the most obvious place to meet, right? I, my guess is that it's something similar. So each of these people is the focal point for a large group and they almost become the sort of anointed leader because everybody needs a point at which to focus.
Coordination point. Isn't that the name? Coordination point. Yeah.
Ben: That makes a lot of sense actually. I hadn't quite thought of it like that, but. It seems to make sense of what we see. Okay, so thinking about aliens and all of this and bring it back, what do you think, if an alien came to Earth, they would need to understand about Pink Floyd?
About how great Pink Floyd is, or maybe music in general, but explain to an alien, or maybe explain to someone who hasn't come across Pink Floyd, why they should listen and what it means.
Samir: Ah, gosh. So the first problem, of course, is you need to have the right hearing receptors, don't you?
Yeah. Because if your anatomy is such that you don't have, I don't know, hearing from 20 hertz to 6, 000 hertz, let's say. It's not clear that you're going to understand it. That's the first thing. And I guess the second thing is that, are you going to be the type of person that's philosophically inclined to, a long, slow buildup of lots and lots of, tension, which is then released.
So I don't know, but if I was trying to, let's say I was trying to explain why Pink Floyd is so great to other people, I would say several things. The first is that it is, in my opinion, the first genuinely English rock band. Every other band, including the Beatles before that, who I of course adore was American.
They sang like Americans, they composed like Americans. The music was very American, etc. The first actually English rock band to talk about things like meadows and green grass and, put a cow on the cover of an album, that kind of thing tongue in cheek and also sing with an English accent, not an American one, was Pink Floyd.
That's the first thing. The second thing is that they were also the first rock band, I believe, where all the members were upper middle class or higher. Because in those days, anyway, it was a way to get out of poverty. You didn't have for example, David Gilmore being a professor's son, or, Nick Mason being the son of a very successful business man or whatever, forming a rock band.
So their sensibilities were different. That's the second thing. And then the third thing is, they were not to put too fine a point on it, particularly talented as musicians. like with their instruments. So what that meant is they had to be incredibly creative to actually do something that was different and interesting and would be something that their audience would like.
And so they hit upon this, the idea of concept albums where the entire album revolves around the concept. The lyrics are serious and the music is incredibly detailed. textured, and builds up and really makes you think and has lots and lots of layers so you can listen to it over and over and over again.
And I think that's really the difference between them and everybody else. Who else would put a 23 minute song on the back end of an album, Echoes I'm thinking of, that, included backwards guitars and seagull cries in the middle. And, put it out seriously, it's one of my favorite songs of all time, actually, is Echoes.
I think it's amazing. I, it's just, they're so different than everybody else, that if you haven't listened to Pink Floyd, you haven't listened to an entire genre of
Ben: music. I hadn't appreciated that Britishness or that Englishness and I think you might be right. I'd have to, I have to think about that.
Not such an expert in music, but that also layering because that was adjacent to another question I was asking, is that, what makes it, What makes music in general, the ability to listen over and particularly Pink Floyd? Because people like Pink Floyd can listen and you can get as much pleasure from it Throughout and in fact, some people would say even increasing pleasure as you discover those layers that complexity …
Samir: So music is I think Probably our most abstract language and the more abstract the language is, the more meaning you can discover in it because each time you Past the language again because of the abstractness it can mean something different That's the first thing and the second thing is that music is very hard to remember unless you're a real expert musician Because it's very hard to remember you can remember you remember the theme you remember the lead guitar you remember Most of the lyrics, but you don't necessarily remember all the little stuff that they do during the song and that's really the difference between say, for example, Pink Floyd and somebody who's popular today as a pop star is they don't have those little things that have been interspersed in all the way that you pick up on the 27th listen.
For example, it's not there. Yeah. So it's very sweet up front. But there's no substance to it in the backend. And it's forgotten, six months later. But when you've got something where there's been, overdubs and overdubs to overdubs and a little instrument inserted over here for, a few seconds and another instrument put over there for a few seconds and a chord held for two minutes over with somebody does a solo, kind of thing.
You're like, wow. 'cause it's different. It's just not something you hear.
Ben: Yeah. And I guess that explains also why we'd get. The same or if not more enjoyment from live performances because it adds even more layers Yes within that seeing it. Yes
Samir: Yes, I saw david gilmore live in your home city in in october and he was unbelievable Absolutely, just unbelievably good at the Royal Albert Hall.
It's just, your jaw goes like this.
Ben: Yeah, excellent. Moving on to a another non other topic cricket. So what do you think is most misunderstood? about cricket. What actually, I guess the average American, doesn't understand cricket at all. I once tried to liken particularly a test match to a five act Shakespeare play that actually the sporting form of cricket was the closest you came to some of these dramatic narratives.
But they still didn't understand. But in any event, what do you so much like about cricket? And what do you think is misunderstood perhaps by the average American?
Samir: Let me tell Americans a few things about cricket. The first is that the oldest international sporting fixture is, or was, the 1845, I think, test match between the United States and Canada.
Cricket has a long history in the U. S. George Washington played cricket and wrote about it in his memoirs. Cricket was a popular sport in the U. S. until 1905 or 1910 or something, until the ICC, at that time, the Imperial Cricket Conference. became all sniffy with Americans playing cricket. And so Americans switched to baseball.
Until then, it was a big sport. Swing bowling was invented in America by an American from Philadelphia who died only a couple of decades ago, actually. So that's funny. Cricket has a long history here. And yet, now Americans know nothing about it, which is funny. The second thing is that Americans think that a five day test match, oh my god, it's so long, it must be boring.
But they're all watching in droves four day golf tournaments. And nothing happens on, I love golf, but nothing happens in a golf tournament, honestly, until about the last nine holes anyway. So how come everyone's watching? So that's the next thing. It's a way of not bothering to understand.
The third thing that they don't understand is that unlike, say, the NFL, which I also adore. But I will criticize like mad for other reasons. Cricket has been incredibly careful in trying to update itself. And I really appreciate that. I really appreciate the fact that now, leg before wicket, it is properly adjudicated and, batters can't just stick their pad out and stop the ball and say I was outside the off stump.
I'm not out. They fixed that. The replay system is excellent. I love the fact that in the replay system, you see the same replay that the umpire is seeing and the umpire is then with a microphone explaining to you what he's seeing and why he's making the decision he's making. So there's no arguments.
You rarely see dissent from batsmen anymore because there's nothing to dissent about. The other thing I really like is that cricket has realized that to make the long form interesting, it also has to have a short form. So now there's, a one day international and then there's a three hour T20 and there's a one and a half hour T10 too if you're playing in the UAE.
I think all those forms of cricket are valid. And all those forms of cricket now require different kinds of athletic skills. The long form is a different game than the short form, and now you're getting specialists, which I think is fantastic. And the other thing that's happened with cricket, and Ben Stokes is the perfect example of this, is that you have people that are tremendous athletes in their own right.
Not just people with just good hand eye coordination, but just amazing athletes that are playing cricket, that are not doing absurd things on a cricket field. Which you look at and you say, this isn't possible and they do it routinely. And lastly the three sports that I enjoy the most are cricket, squash, and the NFL.
And the reason I enjoy the three the most are those are the three, in my opinion, most strategic sports. That people have invented. Those are the ones that require, I think, the most thinking, really thinking hard at all times about what you're doing and why you're doing it. In other sports, you can get by with being bank crash wallop.
In many cases, I'm bigger than you, I'm faster than you, I'm stronger than you. In these three, you can't. You have to know what you're doing and why you're doing it.
Ben: That's really fascinating. I take the point particularly on cricket and technology. So a couple of podcasts ago, I did one. With Daisy Christodoulou, who wrote a whole book on VAR but particularly why it hasn't worked in football.
And she likes cricket as well, and she was explaining why it's worked in cricket to exactly how you said, but they haven't done any of these things within football. And also this the one you talk about in terms of leg before wicket, is really interesting how technology has actually given a sense of fairness to both the athletes and the audience.
No one really thought it was that fair. that batsman could do this kind of technical trick and get away and therefore all of the skill within the spinner was lost to what was a trick but within the letter of the law and even the batsmen themselves would go yeah we're doing this oh my god the spin is really good we're just going to try and not play them and now the technology this is a good pit of determinism within that can tell you is you can't really get away with that.
You have to play the skill of the spinner with your own skill and everyone can see that. And so everyone is acknowledged while the game has just got better, even when the bats batsmen has had to. Yes. Just think about
Samir: around the wicket in the old days, you couldn't get leg before when you were around the wicket, you couldn't get leg before if the batsman ran down the pitch, you couldn't get leg before the batsman hid his back behind his pad and pretended to play.
And so on, which just ruined the game. And now they fixed all of that. And then the other thing they've done, and this is why I want to criticize the NFL, is that the cricket has hired ex cricketers. Who know the game and trained them to within an inch of their life and made them full time umpires There's an umpire.
Is it goff? I'll get his name wrong who almost never gets an LBW wrong. You can appeal it as much as you like and you almost always lose the review why because It's clear the man spent an enormous amount of time studying video of balls to understand which would be in, which are hitting the wicket, which are not hitting the wicket.
I respect that tremendously. On the other hand, the NFL, can you believe it? The richest and most expensive sports league in the world doesn't have full time umpires. They're still part timers with other jobs. Come on! It's infuriating. And then the other thing they do, which drives me insane again, is that there'll be some crazy call on the field because the refs don't really know the rules properly.
So then instead of having somebody from, the TV booth, or whatever it is, overrule or call them and say, listen, you need to change this, whatever, they don't do that. And then the other problem is, which is even worse, they don't have a centralized review system in the sense of, okay, I don't know what call to make on the field.
I'm the referee, you up there, which is what you do in cricket. Go make the call. They don't do that. They make the referee truck to the sidelines to look at some tiny little monitor inside the television to make a call. Don't do that. NFL. Put the people in a central review system in New York or wherever it is, on screen, show the reviews, have them talk through the decision, and then relay the decision back to the field the way cricket does it.
But they don't do that.
Ben: Yeah, it's terrible. Exactly. What's going through your mind? How are you assessing it? Yeah, that's exactly it.
Samir: Yeah, I forget which hedge fund manager it is that said this, but it's absolutely true. He says the only reason now for the NFL to exist is to be a random number generator for sports fantasy and betting.
Yeah.
Ben: Great. Let's play a little underrated, overrated, and then we'll ask a couple of final sort of questions. So this is honor of Tyler Cowen. He used to do this in his podcast a lot, but less so recently. So yeah, overrated, underrated. What do you think of efficient market hypothesis?
Samir: Severely overrated.
Ben: Yeah, so this is the idea that they've got a more fancy way of saying it, but the markets are broadly efficient all of the time, or most of the time, with information. But I've always thought it's really weird, because, if that's true, it's extremely schizophrenic, even second to second, let alone day to day.
Samir: Yes, and there's two related issues there. The first is the Grossman Stieglitz paradox, which is that if there was no return to doing market analysis, nobody would do it. And if nobody did any market analysis, then stock prices would be completely insane, at which point there would be a return to do so there should always be some small return or some return to doing market analysis, otherwise there'd be no point in doing it. That's the first point. The second point is that they confuse the difficulty of outperforming an index. Like the S& P 500, with the efficiency of the market, and that's not the same thing at all.
You can have an inefficient market where it's difficult to outperform an index. And the way efficient market, exactly, for a lot of reasons. Yeah, that's the whole podcast. But there's no reason to think that the two are the same thing. And yet, Eugene Farmer won a Nobel Prize for that.
Ben: Yeah,
Samir: just bizarre. But anyway,
Ben: Universal basic income UBI overrated or underrated good idea or bad idea?
Samir: Both. I think it depends very much on how it's implemented. The studies suggest that in some kinds of implementations, it hurts the interest in work. But in other kinds of implementations, it really helps people, that are poor, no longer be poor and move up the ladder.
And I think we just don't know how we would do it so that we don't, ruin the capitalist goose, as it were, that sort of lays the eggs, as it, the golden eggs. It's tricky. So it's both overrated and underrated. In principle, it'd be fantastic. I just don't know if it's doable.
Ben: And then I was going to ask about Squash, but I guess we basically said that it's definitely going to be underrated.
But would you like to say why it's underrated? Is it just the strategic or is there more? And is there more about squash that you think people don't understand?
Samir: Squash to the uninitiated just looks two guys banging a ball against a wall. Over and over again. And it's not at all obvious why they're banging the ball against the wall each, shot after shot.
And it takes a little bit of while. Or talking to somebody in NoSquash to understand why they're hitting the shots that they're hitting. Because there's the pace of the shot, there's the height of the shot, there's the depth of the shot, and there's the angle of the shot, which are different depending on who's playing whom and why.
And then the shot combinations that they're going to play are going to be different. So to just give you one example, for the longest time, there was a this player, Paul Cole from New Zealand, who became world number one, and now world number four, who was playing against Ali Farag from Egypt, who was world number one, and a legend, and Ali Farag, people generally consider Ali Farag's forehand overhead shot to be a very good shot.
But Paul Cole discovered that by hitting at a certain angle and a certain speed, he could actually make it break down if he just kept hitting shots that Ali Farag had to keep hitting on his high forehand. So for quite some number of matches, Paul Cole won by making Ali Farag hit forehand overhead after forehand overhead after forehand overhead.
Wow. And eventually Ali Farag learned how to counter that, but it took a while. That's what I mean. That's the cleverness of squash.
Ben: Okay, but if you don't know
Samir: that you just say this guy just keeps hitting this ball
Ben: What's going on? It's interesting. I find this is a lot with essentially human cultural aspects like I think there's a lot of maybe high end food or not even high end food So a lot of chinese food For instance, you get really into it because of texture, but if you've never had textual food or you've never understood what textual food means or in fact, you think why am I eating this cold rubbery thing?
There's, that's a lot of reactions to non Chinese food eaters when they first eat cold jellyfish. They don't, my mom literally says they don't know, understand how to eat this. And I think it's true of some of that, but actually for something like squash and things, it doesn't take that much to understand a little bit to then go, wow.
That's that's amazing.
Samir: Yes, and the other thing that Squash could do, which they did for a short while, is I think they should put heart rate monitors on all of their players in real time.
Ben: Okay.
Samir: Because the physical demands in Squash Outweigh any other sport that I know of.
Ben: Okay.
Samir: These guys, for the short time they did it, showed that people's heart rates were in the 190s.
Sustained for an hour at a time. And unlike tennis, there's no 30 second clock that says in between points you can go sit in a chair.
Ben: Yeah. Or
Samir: that you can bounce the ball 25 times. The ref will actually get on you and say, Continuous play please. So you get maybe 5 seconds between points. So these people are tremendous athletes, and I don't think people realize how absurd the athletic skill is.
Ben: Okay, great. Artificial intelligence or AI, the whole thing, do you think it's underrated or overrated?
Samir: I think it is severely overrated in the short run and severely underrated in the long run.
Ben: Yeah, I think that's true. I think we often, we're really bad at knowing how things play out, even over 10 years.
For some reason, once we go Over a certain amount of time. I look back, like I don't think anyone makes 10 year predictions, which don't get there by essentially luck or something like that. Whereas we think we're in really good and it's six months to a year, but actually the vast majority of things don't grow quite as quickly as we think they do on that timeframe.
But on a decade's timeframe we completely miss it every single time, which is fascinating.
Samir: Yes. I think part of it has to do with the difficulty of creating new workflows. For example, in the old days, you had people would set up mills next to streams, right? And then the streams powered the things that ground the wheat.
And then when, steam machines came in, they just basically took that water wheel away and put in a steam wheel. But that really isn't the most efficient way of using it. You have to figure out how to use steam power to grind wheat and other such stuff in smarter ways. And I think that's really the issue is we, it will take us time to figure out how to use AI.
Ben: Yeah.
Samir: So it's impact who is going to be. underappreciated in the long run in my opinion, but overappreciated now because everybody's hyping it.
Ben: Last one on this then intellectual property patents. Do you think patents are overrated or underrated?
Samir: Those that's a long and sad story for me. But but I think they're actually fairly underrated at this point.
And they're, they are underrated because people get hung up on this whole idea of a software patent.
Ben: Yeah.
Samir: And so then, the Supreme Court. Came up with this very silly decision called the Bilski decision, which basically eviscerated software patents. But essentially what the Supreme court doesn't seem to understand is that at the end of the day, all patents are a set of instructions.
Computer patents are no different than another set of instructions. And those computer patents move physical objects, i. e. electrons, in a computer. Calling them abstract is very silly. The result of this and I think I went into this in the book as well, is that by eviscerating software patents, they've cut down the amount of software innovation done by small companies.
And they made it much easier for large companies to protect their profits and protect themselves from competition, which I guess, I don't know if that was the intention, but that is in fact what they've done. So if we want more innovation from small companies, it would be a good idea to have intellectual property, and it would be a good idea to have intellectual property for all kinds of things.
And just be more severe and more strict with what is innovative, what is new, and what is well described in a patent. That's really the problem with patents is when you allow substandard patents through, not that it is quite an abstract idea, which is the latest, bugbear of the Supreme Court, which has screwed things up.
Ben: Yeah. I think the difficulty is we've got to such a state that you need really a real expert patent examiner. Yeah. Yeah. To know that. And also, cause you can see in the marketplace, we take experts and people know okay, this is an amazing innovation and you know what, this is an innovation, which probably isn't that amazing on whether, and if we could, you'd want to give actually more value to that, which is going to be really amazing.
I've always thought that actually in general patents are a little bit weak. They should probably be. Be stronger across the board and maybe particularly stronger for where you have to spend a lot of time in our R& D But actually our copyright is a little bit too long because copyright goes yes to your estate After you go, so that's yes a really long time Yes, and writing definitely builds on other writing in this way in a way that you know it doesn't for pharmaceuticals and software or in a different or in a different manner, but
Samir: There's a balance to be struck there, I agree with you, but we seem to have struck the exact wrong balance.
Copyright is too long, and too strong, and intellectual property is too weak and too short.
Ben: Yeah, exactly. Great, okay last couple of questions. One was about what your writing day is like, so when you try and do, when you're writing your book. I don't think there's a right answer to this, but are you a morning writer, evening writer?
Do you write in bursts? Do you like little notebooks and then do it big? Did you did you speak your book in? How does writing or creativity in general happen for you?
Samir: I did all of those. I wrote in bursts. I wrote in long periods of time. I spoke parts of the book in. I emailed myself notes. I made notes on my notes app.
Essentially, anytime a thought struck me that should be in the book, I immediately made a note of it somewhere. And then when I had some, free time, I'd just write it down. But that, to me, that was the one trick that I learned that was useful, is that anytime inspiration strikes, just write it down.
That's all. Even if it's just a short note to yourself, that's sufficient. Because then you'll be able to recover that thought. But once the thought's gone, it's gone. So don't lose it.
Ben: Yeah, I have that. I always keep a notebook by my bed because sometimes I might wake or half stir at 4 a. m And I'll just scroll something in that half, obviously my atoms have been slightly jiggled So I'm dreaming so I can write something because sometimes it won't make sense But sometimes oh, yeah, and then it will actually re trigger what that whole train of thought is Yeah, you need that thing to re trigger.
Otherwise it then it disappears. I guess the trace fades.
Samir: And you can't get it back.
Ben: And what current projects are you working on? I saw you was a book about what you which had known about politics as well as finance and something else. Is that currently what you're writing on or any other projects?
You have a substack as well. And obviously this book.
Samir: Yeah, I have, so the first is I've written most of that book, haven't finished it, but I've written most of it, but I realized that unless I sell a reasonable number of copies of this book. It's going to be hard to publish that one, so I've left it to finish if this one sells.
So please, it's called the science of free will, please go buy the book so I can write another one. The second project that I am working on is oh, by the way, I should share something about that project. So the reason I started working on that project is that more than one friend of mine said to me, I'm tired of reading the newspaper because I never know what's true and what's false.
Is there a way of distilling current knowledge about different topics, politics, economics, finance, into short pieces, a chapter at a time that I can read so that I have a framework for when I read the newspaper, I can try to figure out if what this person is telling me is an opinion or it's actually fact.
So that's what the book is about. It's to try to give you the framework so you can think about this stuff when you read the news yourself. So anyway, so that, that's Mostly done but not but you're not finished yet. So that's one project The second project is that I am co editing a book on string theory I'm doing some physics again, so that's fun.
And I've also been starting to write some physics papers again, so I have a chapter out for the book. That'll be in the book about the physics of quantum mechanics near black holes. And then I'm also working on some particle physics again, which is fun. So I'm submitting some papers.
They haven't been published yet, but maybe they'll be accepted. Let's see. I'm doing that. I have the substack, as you said. And then lastly, I do have some inventions that I'm trying to commercialize. One in solar energy which is Palm Energy Systems and another one in traffic management, which is trafficbid.
me. So the idea there is that London has had a great deal of success with congestion pricing. New York is having some success with congestion pricing as well. So why not bring congestion pricing to parking? So that's an idea that I have, which I'm working with a friend and co inventor on.
Various projects and of course my baby is my fun. So that's what I run. That's my day job.
Ben: Excellent. So maybe we finish on the last question is do you have any overall advice or life advice or thoughts for people? I guess we had one in terms of regret. It's not a very useful emotion, we've touched on it.
Trained as a physicist, you know a lot about investments. You're also an inventor. and a writer. So I don't know across all of that whether you have some thoughts you, you would share about what people should think about to do with their lives.
Samir: So I think the most important thing really is you have to work on stuff that you want to work on.
And if you're working on stuff that you don't really want to work on, to me that's a little bit sad. And I think you should try to figure out how you can change things so that you can work on stuff that you want to work on. I think that's really important. A lot of people say, for example, that they would like to wait until they retire before they do Project X or Project Y or Project Z.
And my feeling is, I guess sometimes there's no choice. But sometimes it's possible to work on that project sooner than when you retire. Maybe as a hobby, or maybe even as a sideline, or a side business or something or the other, is there a way that you can work on something that you really want to work on, and make it a way of making a living or something along those lines.
So that, that would be my biggest advice. Work on stuff that you want to work on, because if you work on stuff that you want to work on, you don't actually ever work. That's the great beauty of it.
Ben: That's great advice is, yeah, work on stuff work on stuff you want to work on and find time to do it.
And I would maybe add just reflecting on my own life. It's probably more doable than you think to find some time to do that thing on the thing that you agree with. I agree with that.
Samir: And also with current tools, particularly now with AI coming up, I think it's becoming easier than ever. Yes.
Because you can hire yourself the top assistant in the field from, chat GPT for 200 a month. Why wouldn't you?
Ben: Excellent. So on that note we'll once again say the book, The Science of Freewill do look out for it available on Amazon and in bookstores everywhere. And on that thank you very much.
Samir: Thanks, Ben. This was a great chat. Really appreciate you taking the time.