Ananyo Bhattacharya - John von Neumann, Jewish Genius, and Nuclear War

1. 0:00 – 0:30

   Intro

   1. ABAnanyo Bhattacharya0:00

      (instrumental music plays) I try to lay out the context of this. I mean, this was after the most destructive war that the world had ever known. Millions of people had died. And von Neumann had predicted this, um, the Holocaust very s- you know, successfully years in advance. And he now was convinced that within a decade, there would be a third world war with nuclear weapons.

2. 0:30 – 2:29

   John Von Neumann - The Man From The Future

   1. DPDwarkesh Patel0:30

      Okay. Today, I have the pleasure of speaking with Ananyo Bhattacharyah, who is a science writer, who has worked at the Economist and at, uh, Nature, and most recently, he's the author of The Man From the Future: The Visionary Life of John von Neumann, and it was a extremely enjoyable read, super interesting. And so, before we jump into the questions, Aananyo, I'm wondering if you can kind of give context to my audience and summarize the life of this giant.

   2. ABAnanyo Bhattacharya0:57

      (laughs) Uh, well, that's not an easy task, but I'll give it a go. So he was, um, born in Budapest, uh, in around 1903 to this wealthy Jewish family, and pretty early on, um, they realized that, um, there's something quite special about him. So he can do these long six-figure calculations in his head by six, and he's learnt calculus by eight, right? And he's teaching himself the finer points of set theory, um, by kind of 11, right? So he's going on long walks with, uh, Eugene Wigner, who was a childhood friend of his and a future Nobel Prize winner, and, uh, Wigner's a year older than him, and he's, he's teaching Wigner set theory at that age. So it's, it's kind of clear that even among geniuses, as he would be later on at, uh, Los Alamos, for example, or, um, at the, uh, Princeton and at the Institute for Advanced Study, uh, where he'd be recruited along with Einstein, that he was kind of, um, a cut above even all of these, uh, incredibly clever people. And, uh... And so yeah, so he grows up in this, uh, quite privileged Budapest surroundings. Um, their home was, um, often visited by,

3. 2:29 – 16:04

   The Forgotten Father of Game Theory

   1. ABAnanyo Bhattacharya2:29

      um, the, the greats of the time, um, it was an incredibly cultured city, and, um, his, um, father, Max, was a kind of successful banker, so they were quite wealthy. I mean, he was a self-made man. Um, but, um, he had, as a result, von Neumann, who was one of, uh, three brothers, actually, he was the eldest, he had the benefits of, um, kind of a top flight education as well.

   2. DPDwarkesh Patel3:00

      Yeah. So, you know, um, w- right before we did the interview, I was c- thinking about what... Uh, you know, I have a computer science degree, and I was thinking about, okay, what portion of my computer science degree can be traced back directly to von Neumann? I was go- just going through just like an initial glance at a few of the classes that I took where, like, a large part of the fraction of the content, uh, came from von Neumann, right? So you could like... Okay. Algorithms, linear programming, um, you know, merge sort, li- like, probably like a quarter of my curriculum. Um, quantum computing, uh, you know, density, and density matrix, uh, von Neumann entropy. Hardware, von Neumann, uh, the, uh, you know, the von Neumann architecture for my, uh, the computers. Um, you know, e- even like in my organizational ethics class, you know, that, that, that game theory, um, th- that comes up. Uh, you know, theory of computing, uh, you know, finite state machines, um, cellular automata. So, like, I, uh, it's astounding to me that this person is responsible for probably like a third of everything I learned in college. Um, and so it was, uh, it was, um, very interesting to then get to read the history of this person and the ideas that he came up with and interacted with. Um, a- and now, uh, o- one very interesting part about the context, um, uh, context surrounding von Neumann's work is, you know, he was part of this group, as you talk about, called the Martians. They were, uh, Hungarian and Central European Jews who migrated to the United States in the early 20th century, and w- um, as Scott Alexander has a fun blog post title about this, he says, uh, the, um, the, the (hand clap) nuclear, uh, the, the nuclear bomb was a high school science project for a bunch of, uh, Hungarians, uh, because the, uh, eh, a- a lot of the scientists who worked on the, uh, n- nuclear bomb were par- uh, like, went to the same high school. So what was the cultural or other factors that made this group of people so... I mean, uh, produce so many geniuses?

   3. ABAnanyo Bhattacharya4:45

      Right. So they were, um, all Jewish, um, and von Neumann attributed this kind of pressure to succeed, um, to growing up in kind of Central Europe between the two World Wars, um, being surrounded by sort of antisemitism. Now, Budapest was relatively tolerant, but it was in the air of Central Europe at the time. And he said that, um, he felt a pressure to, um, succeed or face extinction. I mean, they were constantly under this huge, relentless, uh, kind of psychological pressure, um, to kind of do the impossible. And, you know, von Neumann, in his letters from 1930, um, by which time he's safely in the, in the US, he's predicting disaster. He's pre- he predicts pretty accurately that there will be a second world war, and he predicts that, um, European Jews, um, will face extin- e- extinction. Um, so he, he is very equi- acutely aware of this. Of course, um, there are circumstances around, uh, Budapest at the, at the time which, um, was able... Which meant that geniuses of, um, this sort were, uh, nurtured. So there were...... private schools, and they were all inevitably private schools, and they were a- almost all boys' schools as well. And, um, von Neumann went to one of, uh, three, I think, elite schools in Budapest at the time. Uh, Teller, for example, and Wigner and, um, Szilard, um, are all part of this Martians- part of this group called the Martians later. They all went to these, um, kind of elite schools. And, um, von Neumann was spotted quite early on by his maths teacher who, um, told his father, you know, "Your, your boy is exceptional. Let me arrange special tutoring for him." So von Neumann gets picked out even from this group of exceptional people, and he's given, um, a special course at the University of Budapest, and, um, it's, uh, his teachers are all just amazed at his abilities. So the joke was, um, later on when all these guys met again at Los Alamos to work on the American bomb project that they had these funny Hungarian accents, and they had these almost supernatural, um, intellectual abilities. Um, so the joke was they must be from another planet. Now, when Wigner was asked about this, he said, "Ah, there is no Hungarian phenomenon. The only phenomenon that needs explaining is Johnny von Neumann." So you can, you can tell from, from those sorts of comments what kind of person he was.

   4. DPDwarkesh Patel7:43

      I, I'm actually curious to boil it down what ac- actually, what, what exactly was going on that produced so many geniuses. I mean, uh, one, one, uh, one thing you proposed was maybe it was the private schools, but, uh, I mean, as you just said, you know, the, the, he, he had like, uh, he was, he had taught himself integral and differential calculus by the time he was 10 and knew, like, four languages. So, uh, may- maybe that aided his growth, but I'm curious. I- it, it seems like he was a- already on the path to becoming, like, a world star scientist.

   5. ABAnanyo Bhattacharya8:12

      Yeah, I mean, he was renowned as a mathematician really early on. I mean, he, he, as soon as he finished his PhD where he resolves this incredibly difficult, um, paradox in, in set theory or helps to resolve it by, um, sort of 22, and then he goes to, uh, Göttingen where quantum theory is being invented by another whiz kid actually, Werner Heisenberg, of course who's just a year older than von Neumann. And, uh, von Neumann gets really interested in quantum mechanics, and he, um, produces this first mathematically rigorous version of it, um, in a- a- a few years later. So, I mean, von Neumann clearly, I mean, he was just an exceptional... he had an exceptional brain. Now, his, um, grandfather was apparently, although he wasn't academically, um, particularly successful, he had started his own very successful business, but what was interesting was that he had these calculational abilities that were actually better than von Neumann's. So von Neumann remembers asking his granddad these incredibly long sums and his granddad would come back with, um, with answers pretty quickly. And von Neumann, despite all of his genius, he was never able to a- um, you know, kind of match these, um, these abilities himself. And of course there's a lot more to higher mathematics, as we know, uh, than, uh, being able to do really long sums, but it's k- kind of interesting that, uh, there's some, you know, genetic, um, predisposition there that we can, uh, we can see.

   6. DPDwarkesh Patel9:57

      Um, one interesting possibility that I've heard is you have Jewish em- em- emancipation in Europe in, like, um... was it the 18th or, uh, 19th century? And then afterwards you have this, a tremendous streak of Jewish achievement that's halted by the Holocaust, so, you know, you, you have this brief window where this group of extraordinary people are able to achieve great things before, you know, before they're forced to emigrate or, you know, other things happen. Um, and it... I mean, it makes what happened in Europe during that time even more tragic when you consider wha- wha- what was stopped. So, you know, one, one question I have, um, is y- you have this, uh, person who is incredibly prolific. Would he have been able to achieve as much as he did if he were born, say, uh, today given that w- a lot of the low-hanging fruit has been picked? Um, is it just that he got into science and mathematics at a time that there was just so many different ideas combining and left to explore, or, I mean, do, do you think that any... at any other time a person like, uh, von Neumann would have been able to be as prolific?

   7. ABAnanyo Bhattacharya11:01

      Um, no, I think you've really hit the nail on the head there. I think there was definitely a historical moment. I mean, in terms of people with brains like von Neumann, they're pretty hard to think of, but, you know, in terms of raw mathematical ability, you, you look at somebody like Terence Tao, uh, today or, um, you know, you, you consider, um, uh... the- there's, there's a few other pure mathematicians who could maybe, um, match, uh, von Neumann's sort of brain. I mean, it, it was extraordinarily unusual, but maybe not, you know, once in a century unusual, but, um, extremely unusual. But I think, um, there are a few things that, um, kind of mark him out. One is, yeah, the historical moment. So he arrives on the scene in, um, kind of, you know, 1910, 1920s, and is, he's immersed in, um, kind of a maths that's going through this logical crisis and it's gonna spur people like Alan Turing and Kurt Gödel to think really hard about these step-by-step...... proofs. Um, how do we, how do we prove stuff properly without, uh, getting into these awful paradoxes? And that would lead later on, that step-by-step thinking would be extremely influential when people came to think about programming, uh, you know, and algorithms and things like that. So there's, um, so there's that side of things. And then, of course, science just explodes. You know, you've got, um, masses of funding, of course. Quantum mechanics becomes the atom bomb basically within a space of 25 years. You have huge amounts of money suddenly being thrown at, um, at science and, and then you get big science. And, you know, economics, uh, you know, thanks to Von Neumann, um, in, in large part becomes suddenly more mathematical. But now, um, with that massive funding and the continued, uh, funding of science, I think there's been, um, a great degree of, of specialism. Um, I think the time when one genius of Von Neumann's stature could contribute so productively to kind of, you know, everything from pure mathematics right the way through quantum mechanics, to various fields of physics, to, you know, non-linear equations, and, uh, to, uh, distill out the modern form of, of the computer, the programmable computer, to automata theory, um, you know, uh, come up with a proof that, um, machines could reproduce themselves, I think, uh, sadly, that that was really, uh, uh, a, a brief moment of the 20th ce- century that made it possible. But the second, um, thing that's, that's incredibly rare about Von Neumann that I noticed, he actually embraced this idea of applying maths to real world problems. Whereas many mathematicians, many academics of all sorts actually would, uh, rather eschew, uh, you know, the, the real world. They don't want very much to do with it. They (laughs) , um, when it comes to mathematicians, they'd rather be left alone in their ivory tower to prove theorems. And Von Neumann did a lot of that. He, he left behind a, you know, a massive, um, amount of pure mathematics. But really my, my book focuses on, um, the stuff that he left behind that came about from engaging with, with the real world, and there's a huge amount of that. And, um, I think that's, that's also what made him really quite exceptional. The only other person that I can think of that was, that is now as gifted, uh, mathematically a- as he was and has shown, um, some interest in these sort of practical affairs is Stephen Wolfram. Um, so, um, uh, but, you know, Wolfram was born (laughs) in the wrong time, I think. Um, perhaps if he'd have been born in 1903, uh, you know, he might have been a, a Von Neumann-esque figure. But, um, so there's definitely a combination there of good luck, um, a historical moment, and just, you know, a particular attitude. Maybe because he was brought up i- in a, you know, by a bank of father who was not afraid to get his hands dirty. I mean, this was... and he was an investment banker happily investing in, um, firms, in technology fir- the technology firms of the time. People, uh, uh, you know, he invested in a Jacquard loom company, um, which used, used punch cards to program looms. You know, that, that made a, an impact on Von Neumann, obviously, at the, at the time. So I think, yeah, there's a, a, um, a combination of reasons that, uh, Von Neumann

4. 16:04 – 19:45

   The last representative of the great mathematicians

   1. ABAnanyo Bhattacharya16:04

      was so influential.

   2. DPDwarkesh Patel16:05

      W- W- Wolfram could have been a great scientist in another time. I guess he just ended up (laughs) writing, uh, writing as a mathematical software in our time. Um, uh, uh, not, not to say he hasn't tried other things. Um, so y- uh, you suggest that maybe it was his, uh... it was the time he spent working on practical problems that helped him achieve so much, and I wonder if the opposite may not be true that is it possible that because he got i- recruited into all these different projects that the government had going on at the time, especially because of World War II, you know, ballis- ballistics research and nuclear implosion devices and then advising with, like, uh, Cold War strategy. Um, wa- was this in some sense a distraction from the, uh, uh, you know, uh, the basic research that he might have otherwise done and have been more productive at?

   3. ABAnanyo Bhattacharya16:51

      Well, I mean, (sighs) uh, you know, Bronowski thought, you know, that Von Neumann had kind of wasted this, the, his incredible talent. But to me, the more I looked at his work, the more I realized that for him, um, this engagement with the real world was actually vitally important. And, um, you know, it need not have been the work for the military, but that is where, at the time, in the, in the... unfortunately, in the early to mid 20th century, a lot of the challenging problems were. I mean, designing the, the atom bomb, which is where he made some key contributions. And then later on, of course, the emergence of the computer is deeply, deeply linked to the mathematics of the atom, atom bomb. And, um, arguably, you know, it was, it was his engagement with these areas that led him to, to think and be in a position to kind of spur computing. And as I argue, he was kind of a godfather of the open source movement. Um, you know, his, uh, proof, um, of, um, that automata could, could reproduce themselves and evolve, all of this thinking, um, came about because he was, I think, deeply engaged with the real world, and that, that makes him unusual. And he argues as much, um, quite openly in an essay that he did call The Mathematician.... and, um, where he says that if mathematicians retreat too far, kind of into their ivory cha- towers, if, if the maths becomes just maths for the sake of maths, with no, um, input from, um, kind of the real world, then, um, the, he, he said, "It, it became baroque, um, uh, and, and, uh, not interesting." So I find it really difficult to believe that if von Neumann had sheltered himself away, and somehow been left alone or didn't engage with the sorts of problems that he did, whether it was the computer or to his military work, that he would've left behind the kind of interesting aura that, that he did. Uh, he, he wouldn't have been von Neumann, right? I mean, you, you can see it's so deeply ingrained in his personality to be, um, to be out there thinking all the time, and to be thinking about, um, you know, key problems that, um, that yeah, it's difficult to imagine a von Neumann that, that wasn't like that, that was tucked away. Uh, and I think that, that as a kind of intellectual biographer, that that makes him kind of incredibly interesting (laughs) but also incredibly challenging,

5. 19:45 – 26:31

   Did John Von Neumann have a Miracle year?

   1. ABAnanyo Bhattacharya19:45

      um, to tackle.

   2. DPDwarkesh Patel19:46

      Yeah, that's what makes your book so interesting, um, is, is that you are a biographer of ideas. So, you know, I mean, uh, a lot of other biographies about scientists really frustrate me because you get to hear all these details about their life, uh, you know, uh, which, which, which is also interesting, but you never get to engage with their ideas, which is probably a big part of what reading about a scientist should be about. S- and you do that really well, so, you know, that, that, that was super fun. Uh, did John von Neumann have a miracle year?

   3. ABAnanyo Bhattacharya20:14

      You know what? Um, I don't know, and, uh, maybe you've looked at his publication record more closely than I have (laughs) and counted up the papers. But, you know, whilst, um, Einstein, for example, and Kurt Godel, when they were placed into this perfect environment that was the Institute for Advanced Study, right? They didn't have, you know, to teach anybody anything. They had massive holidays. They could do what they wanted. Well, Einstein's time there was, you know, really not very productive. He, he had, you know, his miracle year, right? Uh, in the kind of early 20th century, which was incredible. But then his time at, at, um, the IAS was not particularly productive, and he was trying to find his theory of everything. And Godel, after this incredible work in Europe on, you know, his incompleteness theorems, again, he, he spent a lot of time at the IAS going for nice walks with Einstein, and, uh, you know, and v- talking, chatting to von Neumann. Um, but of course, you know, there, there wasn't much coming out there in comparison. Now, when you, when you, you look at von Neumann's productivity at the IAS, I mean, he was invent- inventing whole new fields of mathematics. He was, um, bringing about the birth of the modern computer. You know, he had this project at the IAS, um, to, to, uh, bring a, a computer, uh, to them, against, you know, it has to be said, against the, uh, wishes of many of the IAS staff. But, um, you know, he was, he was... He'd written three volumes worth of operator, (laughs) operator theory, and he always joked, right? That, um, um, you know, a, a mathematician's productive years are over, um... uh, y- you know, at, at 30 or at, uh, 28. It was always 10 years away from however old he was at the time (laughs) . So, um, you know, he, he, he clearly felt that he had a lot more to do, and I think that's what made his kind of un- untimely death all the more tragic, uh, for everybody. But, you know, it, it was, uh, incre- in- incredibly painful for him. Uh, you know, nobody enjoys staring death in the face, but for m- for von Neumann, it, it was, um... it was extraordinarily, um, painful, yeah.

   4. DPDwarkesh Patel22:48

      And I think you mentioned the theory that it might've had to do with his s- spending time around nuclear tests, the, the bone cancer he got, which is, uh, you know, ironic but still tragic. Um, so, uh, w- we, uh, we know him very well for his work on computers. I'm curious why his, uh, research on cellular automata and constructors hasn't, uh, taken off, and why that isn't considered... well, why that hasn't been researched as, I guess, e- as fundamental as computers are. You know, there are... like David Deutsch has recently published about, uh, constructor theory. His, his claim is that a universal constructor, uh, is like as fundamental a tool as a universal computer is, something that can construct anything else. W- why did this, uh, train of thought kinda languish?

   5. ABAnanyo Bhattacharya23:29

      Well, I mean, that's fascinating, isn't it? Because, I mean, the book's called The Man from the Future, right? And I, I loved, um, von Neumann's, uh, proof of, um, his automata theory, uh, you know, his proof, um, uh, that automata could, could reproduce. And, uh, you know, he combines Turing's universal, uh, computer with, um, uh, with, with this idea of a, of a construction unit, and so he produces the universal constructor, right? And, um, I think in a sense, this is an idea that's still kind of ahead of its time. And just after I published, um, The Man from the Future in the UK, this group in the States, this, uh... published, um, their paper on xenobots.And these are kinda stem cells, and they, they sort of whirl around and collect other stem cells together in little groups. And then these stem cells themselves start to whirl around and, and collect more together. So they suddenly realized, "Wow, you know, this is a, the embodiment of von Neumann's self-reproducing automata, and it's only taken, what, uh, you know, 70 years for them (laughs) to make an appearance." And these stem cells were designed by, um, kind of a, uh, a neural net, so artificial intelligence. And here we are, you know, all of von Neumann's little influences coming together in this neat, neat package. I think maybe in another ten years' time, we'll be asking the same question again, "Why didn't anybody realize this stuff was important?" I mean, when, um, von Neumann's first biographer, Norman McRae, wrote about automata theory, he was extremely dismissive, barely, you know, gave it a few pages, as if it's, like, something quirky. Um, and, and now we're beginning to see kind of its, the influence of this extraordinarily powerful idea, if nothing else. We know that it inspired, um, those early pioneers in nanotechnology to think about universal constructors at the molecular level. We know that, um, RepRap, this idea of a 3D printer where you could print most of its parts... You know, I, I, I talked to, um, the inventor of that, and he said he was inspired by this, uh, this idea, von Neumann's idea. And, um, you know, in the '50s and '60s and '70s, you had people thinking about, "Well, how do we explore the universe? Well, why don't we make a probe that can make more copies of itself, you know, out in space by foraging on the planets it finds?" It's this incredibly fertile idea, and I think we're still just at the, uh, the beginning of really working out where this goes, and, uh, it's kinda (laughs) dangerous (laughs) , and it's kind of exciting, and who

6. 26:31 – 29:34

   The fundamental theorem of John von Neumann’s game theory

   1. ABAnanyo Bhattacharya26:31

      knows where it's gonna, gonna end?

   2. DPDwarkesh Patel26:33

      I think, um, uh, for me at least, his, h- his work here and, uh, the suggestion, the implications of it are even more scary than, like, the counterintuitive implications from his game theory work, because, um, like I... Uh, Robert Hanson has this paper, I forget the title, but the idea is, um, w- whatever force or, like, civilization or whatever is expanding fastest will be the one that controls most of, uh, the universe, uh, at least un- unless impeded by another one. And so if it's the case that this sort of, uh, von Neumann, uh, probes almost spread like a virus around the universe, uh, and turning everything into goop, may- maybe, like, the expected outcome of colonization is just that that's what the universe ends up looking like, where the low-hanging fruit, so to speak, has, uh, been burned away by, uh, uh, burned away by, uh, such probes. And, uh, it's an interesting, like, futuristic hypothesis and one I don't really hear much talked about, which I think is interesting.

   3. ABAnanyo Bhattacharya27:28

      (laughs) Well, you know, (laughs) that's one way it could go. Let's hope it doesn't go (laughs) that way. Um, you know, um, uh, may- maybe they'll, you know, build us a, a new home after we've trashed this one. Uh, who knows? Um, the, uh... Yeah, I, I think, of course, you know, these sorts of science fiction-y elements, maybe, maybe, uh, part of it is that, uh, no- you know, nobody wants to talk about automata theory because it's got these unsavory science fiction (laughs) elements attached to it. Um, you know, people would rather, uh, stick to, um, the von Neumann architecture and all that, all- all that sort of stuff. Um, but yeah, I, I mean, um, it's the fecundity really of the idea more than the mathematics, isn't it? It just, you know, that somebody can take this question that philosophers have been kicking around for sort of centuries, you know, can machines make more machines, can machines have babies, can machines reproduce, and he just says, "Yeah, well, let's, let's look at this mathematically, shall we?" And then he solves it, and, you know, we have the answer. And that's what I find gripping about von Neumann's work, and it's, it's kind of what I found overall as I was approaching this book that I wanted to show, that people... When you look at kind of popular science books or popular mathematics books, uh, the majority of them are really about kind of celebrating the maths or the science in and of itself, right? They rarely actually talk about maths as this kind of existential thing that humans have invented that underpins our technological world. We don't really think of it like that often, and with von Neumann, as I was writing about von Neumann, it became impossible not to, right? So take game theory. Um, what was he trying to do there? Well, this was rooted, again, in this very early 20th century idea amongst

7. 29:34 – 50:51

   The strong supporter of "preventive war”

   1. ABAnanyo Bhattacharya29:34

      mathematicians that maths was extraordinarily successful, so we can apply it to kind of anything. And, you know, why should we leave, um, the human mind and human behavior to psychologists when they've been so terribly unsuccessful in actually getting anywhere with understanding it? Let's, uh, let- let's try to do the maths on this. And so kind of that... I think it was that impetus that really drove a lot of mathematicians, including von Neumann, to tackle, um, the theory of games, which is really about conflict and cooperation, and I think that was kind of his motivation there. And, um, and again, you've got, you know, the, the, the very thing that kind of s- some pure mathematicians would say, "Oh, yeah, you know, von Neumann was wasting his time by being so involved with military work or, or, you know, this practical stuff. He was whizzing about looking for computational power." Well, you know, without that part of his personality...... would he have been so interested in, in game theory? Would he have done ... Would he have achieved what he did, um, e- you know, in, in those terms? Which is recasting economics in a, you know, in a completely different light, really.

   2. DPDwarkesh Patel30:51

      Yeah, yeah, it's almost like he foresaw the replications, uh, crisis in psychology or something. Um, you know, spe- speaking of his w- work on game theory, I think that part was e- e- especially, um, es- especially relevant today. Um, I'm curious how y- you know, th- th- his, you know, min/max theorem and th- theory of like zero sum games that makes it really har- easy to model, um, a mo- model two player games, um, the two player zero sum games, uh, y- like the one we had against the Soviet Union. I'm curious how he would've thought about a multipolar world where more than two parties have nuclear weapons and are possibly roughly equal in power, um, eh, oh, how would game theory generalize to that kind of problem?

   3. ABAnanyo Bhattacharya31:37

      Yeah, I mean, so it's not at all clear, right, that von Neumann thought about nuclear strategy in kind of mini/max terms as a zero sum game. Eh, in fact there's quite a lot of evidence that he didn't. I mean, his, um ... He, for example, he took very little interest in the prisoner's dilemma. That wasn't cooked up by him, it was cooked up by people at Rand who were kind of ins- inspired and influenced by him. And, of course, prisoner's dilemma isn't a zero sum game, it's a, it's a non-zero sum game. But it became this, um, template with which many people thought about, um, nuclear strategy in the Cold War. Now, um, if you look at what von Neumann wrote, um, in Theory of Games and Economic Behavior with Morgenstern, what he was concerned with, his kind of solutions were based around cooperation. So, he was like, "Were there stable solutions to games, um, if a number of the players cooperated? And, you know, was this an optimal solution to the game?" Um, so you could, um, you could imagine, right, say if you play, I don't know, Monopoly and there's three of you, um, often what you'll notice is one player will start winning and then the, the two other players, even without talking to each other, they'll sort of gang up on them, right? They'll form a kind of alliance. And, you know, kind of von Neumann's the- early, um, um, look at game theory was based around increasing numbers of these kind of, um, alliances. So, if you wanted to know about a 10 player game, von Neumann tried to kind of think about how, you know, within this 10 player group you could get different alliances that were kind of stable and would lead to a winning solution. Wasn't entirely successful, and it took, um, John Forbes Nash later on to kind of, um, develop this idea of non-cooperative game theory, which was, um, hugely successful. Um, uh, but, um, that kind of doesn't chime well really with this idea of von Neumann viewing the world in these zero sum terms, right? He came from this rather central European background where they were used to discussing ideas in kind of bars and cafes over a drink and talking about, um, stuff quite freely and sharing and, um, giving credit, um, to others when, when it's due. And so, I mean, he, he was obviously proud of his own contributions and he was quite defensive about them, but he was also reasonably honest. If he had culled an idea from somebody else, he would totally be, um, be honest about that and give them credit. And so this kind of thread of thinking I think, uh, was, um, was quite important and it's been weirdly overlooked w- when it came to kind of this caricature of von Neumann that developed as a result of Kubrick using him as one inspiration for Dr. Strangelove later on. Um, now von Neumann's actual thoughts on nuclear strategy, he penned a paper in the '50s before he died, um, and, um, in that, um, he makes it clear that he doesn't ... He's not really talking about this preemp- the idea of a preemptive strike on the Soviet Union anymore. It's a lot more complicated. It, it's more like what evolved at Rand later. So, uh, you know, he was deeply uncomfortable with this idea that, you know, we had two or more sides with enough nuclear weapons to wipe out the world many times over. So he thought that if nuclear weapons ever were used, um, you know, you would have to be insane to just go all out. So, you know, he, he talked about kind of holding, holding back and, you know, you toss a ... If one person tosses a nuclear weapon over (laughs) and blows, blows up a city, then the other person does and it proceeds a little bit more, um, slowly. It doesn't escalate all at once into this massive catastrophic, um, nuclear war. But, um, the thing that people picked up, uh, most about his thinking was, of course, in this brief period after the Second World War where he famously said, "If you say, 'Bomb them tomorrow.' I say, 'Why not today?' If you say, '4:00,' why not 2:00?" And, you know, it's not entirely clear that he meant that a-In all seriousness, I mean, his daughter certainly thinks he was advocating for a preemptive strike, or at least he was asking people to think quite rationally about whether a preemptive strike on the Soviet Union might be worthwhile, given that he felt that it was almost inevitable Stalin, as soon as he'd developed nuclear weapons, would launch a, a kind of, um, strike on, on the United States. Um, he was, he was sort of arguing, "Well, you know, if we're in this situation where we're thinking about it, why sh- shouldn't we do it sooner, um, rather than later? And shouldn't we do it before the Soviet Union has enough weapons that, um, you know, they can, uh, fight back? And shouldn't we do something to ensure that nuclear power doesn't get into the wrong hands?" And, you know, whether that's a world government or whether the United States functions as a de facto guardian of nuclear technology, you know, it wasn't, uh, it wasn't clear. Um, I think the other thing that I sort of say in my book is, I try to lay out the context of this. I mean, this was after the most destructive war that the world had ever known. Millions of people had died. And von Neumann had predicted this and the Holocaust very sp- you know, successfully years in advance, and he now was convinced that within a decade, there would be a third world war with nuclear weapons. Now, if you imagine that, and if you think that, and if your past predictions have come true, then it allows you incredible scope to think in this kind of rather, kind of ruthless manner about, "Well, maybe we, maybe bombing, you know, the Soviet Union and wiping out, you know, 100,000, um, people's lives at the push of a button, may- maybe that's not, not as bad as it, um, as it could be when you consider that millions of people are gonna be dead in a decade, and, you know, p- potentially bringing all of human civilization to an abrupt, um, end. Um, well, maybe we can, we can stop that from happening." And, um, it's, uh, it turns out that it's a surprisingly common idea at the time in America, um, and elsewhere. Um, I mean Bertrand Russell, for example, the famous pacifist, um, also argued for a preemptive strike on the Soviet Union if they didn't give up their a- you know, their nuclear ambitions. And, you know, you dig around in the post, kind of in the late '40s, um, in this brief window after the Second World War when the US seemed to have a virtual monopoly on, on nuclear weapons, and you find suddenly that (laughs) you know, a lot more people, um, supported this idea, including a, you know, a large proportion, by the way, of the, uh, American public, um, than, than you think is, is possible.

   4. DPDwarkesh Patel39:36

      Um, y- you know, as you talk about in the book, there's like a ve- very interesting but extremely scary precarious scenario where two sides think, um, t- two sides have a nuclear weapon or think that both sides have a nuclear weapon, but neither one has developed the ability yet to defend their nuclear silos, um, against, uh, initial attack. So then, you know, both of them, uh, think that the other one, if they launch a first strike, there would be no deterrent, so then both of them are incentivized to launch that first strike, which is kind of like the opposite of mad. And, you know, the, that, that's, that's one worry, if like, I don't know, if, if, um, if nuclear technology gets better, in some ways that could have- make nuclear war much more likely because the, p- people could start thinking, "Okay, well, we can just take out all their, all, all, all, all, their entire arsenal but, but so they have no way to retaliate." Um, um, I, I'm curious what Neumann... Uh, you, you mentioned, you know, he had a good way of thinking about escalation. I'm curious how he would've thought about, um, you know, the one problem we have today is like, the, you can have cyber warfare, which is immensely destructive, uh, in an economic sense, but doesn't warrant or seem to warrant a sort of land war, and then you can have a land war like, I don't know, China takes over Taiwan or, you know, you have, uh, what's going on in Ukraine, and, but it seems like way too harsh to react with nuclear war. And I'm curious how von Neumann would have been able to think about these kinds of problems.

   5. ABAnanyo Bhattacharya40:58

      You know, um, von Neumann, I mean, he was recruited by Rand, but the work that he did... And, and Rand became this kind of hot house for nuclear strategic thinking, right, in, in the Cold War, and it, um, influenced, um, American policy. But von Neumann, apart from this paper on, on, uh, nuclear strategy, he seems to have taken remarkable little interest (laughs) in, in the whole thing. I mean, when he was at Rand, he was, um, uh, computing various solutions, um, to kind of jewels. So, you know, he'd worked out the Minimax theorem and, um, so he was busy, well, you know, if you have a, a plane and a, um, I don't know, a tank or, you know, whatever, a submarine and a ship, you know, and they, they can see each other coming, at what point should they fire? At what, you know, at what point should they do this? And so he got kind of involved in that and computing, and he kind of lost interest in game theory again as soon as, as soon as computing came to the fore. So he helped... So whilst he was doing this, he ended up helping Rand kind of realize their own ambitions of, of having a computer. Um, so it's, it's, uh, not at all clear to me how much he'd still carry on being involved in the strategy, you know, in the, in, in the, um, nuclear strategy side. But of course, I mean, this idea of kind of if you are coming up with your best strategy, then you have to think what, um...... you know, your opponent will make of that, and you have to imagine that they're also, you know, an intelligent opponent, um, who's gonna be out for themselves. And that thinking is very deeply embedded into Minimax. And, um, and, uh, you know, and that, that was, that was clearly very influential, um, later on.

   6. DPDwarkesh Patel43:05

      One thing I find very interesting about von Neumann's work with the government, um, and in aiding these kinds of str- strategic conversations is, um, uh, at least from my understanding, it seems that a lot of the scientists during that time were, um, somewhat radical and sympathetic to socialism, you know, like Bertrand Russell or Oppenheimer. Um, and Von Neumann seems to be a very practical, non-radical, uh, person. I mean, uh, uh, you can think that's a good thing or a bad thing, but it, it seems like he broke from the conventional, uh, I guess, elite scientific culture at the time. I'm curious, what about his personality or background do you think made him that way? Or am I even characterizing the situation in the correct way?

   7. ABAnanyo Bhattacharya43:44

      Yeah. Um, no. I, I, I think that's fair. In fact, if anything, he was considered, um, kind of right-wing or at least a, a Cold War hawk, um, in certain circles. Um, I think if you look at him quite closely, I mean, you could argue in many ways he was, you know, something of a, a, a liberal. But, you know, um, at the time, some, you know, a lot of people felt that he was quite hawkish. Um, now, the reason for that is that, um, there was a, uh... Shortly after the First World War in Hungary, there were two things that happened. One was there was a very short-lived communist uprising, and that government lasted for six months, and it was pretty brutal. Um, you know, they, they reclaimed private property from wealthy fam- wealthy families and, and there was just general chaos and, um, beatings on the street and stuff. And, um, but then something happened afterwards, and a, a military, uh, essentially a, you know, a, a military government, uh, just marched in, uh, led by General Horthy, and they took control, and that turned out to be even worse. I mean, they, they, there was public hangings and, and rapes and, um, you know, thousands of people ended up dead. And, um, many Jewish people, um, at that time were seen to have been collaborating with the earlier communist, um, government, so, you know, many Jews were basically shot on the streets as well. Now, the von Neumanns were, uh, you know, uh, by dint of their wealth, um, they were kind of protected from this, but von Neumann saw all of this as he was growing up. And then, of course, later with the rise of the Nazis in Germany, he, um, you know, he had left Germany by then, but a lot of his formative years as a, as a scientist or as a mathematician w- were spent in Germany. And he adored, um, kind of interwar Germany in the, at least in the, uh, late '20s, and it, for him, it was this perfect intellectual climate. I mean, you have to remember that Germany was, you know, scientifically and mathematically definitely kind of the center of the world then. I mean, America just was nothing at the time. Um, it was only, you know, uh, kind of during the Second World War and post the Second World War that, um, from the '30s, um, late '30s onwards that America became this scientific and technological kind of powerhouse really, and, uh, you know, it benefited from many of these European scientists who, who left as a result of the Nazis. Now, he'd seen this and he, his lesson was that authoritarianism, you know, is something that we shouldn't tolerate. And so when he came to the States, his priority was to put his expertise into the hands of the democratic government there. And whilst he definitely was, um, advising them, he, um, you know, I, I, I got the feeling that, you know, he, he wa- he wasn't interested in making decisions on their behalf 'cause, you know, he, he... this was a democratically elected go- government. I think deep down he was a democrat. He felt he should work as hard as possible to give, uh, the US government the tools that it needed to overcome the Nazis and to, you know, and to, um, you know, main- maintain, uh, their lead as kind of the preeminent, um, democracy in the world. But, um, so he was kind of, um, I don't know, I think m- more, more allergic to authoritarianism, whereas I think, um, you know, uh, before the Second World War happened, before we knew what was happening under Stalin, there were many intellectuals who were willing to give s- you know, the communism, uh, you know, deep left, um, thinking m- more of a chance. Whereas von Neumann had kind of seen what that turned into, um, in Hungary, and he'd seen the, uh... Essentially, it became a kind of authoritarian regime. Um, he was deeply suspicious of Stalin, um, from day one for the very same reason, and he'd had these experiences of, you know, Eu- Europe being turned upside down by the Nazis, and I think that really shaped him very profoundly. Um, he became quite cynical about human nature as well at the same time. I think, you know, uh, deep down, he was... You know, superficially, he was, um-... kind of a good man and, um, uh, he, you know, he, he was nice to people and I think that's really where he started. Uh, he, uh, in his day to day interactions with people, he was, he was nice. He would do these incredible things very quietly behind people's backs that many other scientists wouldn't dream of like, you know, this builder, Hungarian builder who contacted him in the middle of the second World War and said, "I wanna learn more about maths, but I'm in America basically building stuff. Where do I find out more about maths?" So he writes to his friend (laughs) in wartime Hungary and gets them to send over a bunch of Hungarian maths textbooks. I mean, and, and later on, you've got people like, uh, uh, Mandelbrot who came over thanks to his reference and, you know, he was at Princeton in the IAS. And years later when Mandelbrot ran into problems with his boss, um, he goes looking for work elsewhere and he finds that, like, whatever a decade earlier, long after, you know, and this is long after von Neumann was dead, you know, von Neumann had sent out letters and talked to people saying that Mandelbrot is doing really important work, but, you know, he may struggle because what he's doing is so cutting edge. So if he does and he go, comes looking for a job, please (laughs) you know, give him a job 'cause this guy is brilliant. (laughs) And, you know, he does these little things and he, he of course helps, um, scientists leave, um, kind of Europe before the Nazis make that impossible. He gets, he helps to get Godel out, um, of Germany, for example. So, you know, he's this very conflicted personality. Um, so I think, I think, you know, he's, as you would expect, quite a complex, um, and thoughtful human being and he's not easily characterized as, you know, Dr. Strangelove or, uh, you know, a, um,

8. 50:51 – 54:44

   We can't all be superhuman

   1. ABAnanyo Bhattacharya50:51

      a bleeding heart liberal. (laughs)

   2. DPDwarkesh Patel50:54

      Uh, I, I, I understand what you meant, but out of context, um, "He was superficially a good man," has got to be-

   3. ABAnanyo Bhattacharya51:00

      (laughs)

   4. DPDwarkesh Patel51:01

      ... the, the best backhanded compliment ever.

   5. ABAnanyo Bhattacharya51:03

      (laughs)

   6. DPDwarkesh Patel51:04

      Um... (laughs)

   7. ABAnanyo Bhattacharya51:05

      (laughs)

   8. DPDwarkesh Patel51:05

      Uh, so the final question, I don't know if we have time. You know, yo- you're a researcher yourself, you know, you have a PhD in protein cli- crystallography, you're a medical researcher, and now you've analyzed, uh, John von Neumann's life, you know, prob- one of the greatest, probably the greatest, um, uh, geniuses of all time. What are y- Do you, have you like extrapolated some lessons about how to be prolific or how to come up with new insights in different fields?

   9. ABAnanyo Bhattacharya51:30

      (laughs) Uh, not at all. But I, I, I would thoroughly recommend if you're gonna write a book that you try not to give up your day job a year before the worst pandemic (laughs) descends, that we've known about, uh, you know, decades, descends on, on, and engulfs the planet. Um, thus ensuring that instead of working on your book about the, the cleverest, uh, person of the 20th century, uh, who works on abstract set theory (laughs) , uh, you end up having to home school a recalcitrant 10-year-old. (laughs) Um, so that's, that's one, you know, if you wanna be productive, don't do that. Okay. Um, but in other terms, I think, you know, it's, it's dangerous trying to, you know, come out with a kind of self-help book based on von Neumann's lifestyle, right? I mean, his first wife left him because he was too busy thinking and, um, you know, she took up with her son (laughs) . She had a, a graduate student, uh, Horner Cooper, who was, um, you know, a physics graduate student and, you know, and she was, you know, quite, um, the thinker herself. She ended up becoming this mover and shaker in, um, science admin. And, uh, you know, his second wife was, uh, very clever herself, um, Klara Dan. But, y- you know, I, he, he thought incessantly, um, from morning to night and, you know, even at the cocktail parties that he threw, um, he would sometimes just find noise conducive to work and he would just rush off, cocktail in hand to write down some, some theorem. I mean, what, what do you draw (laughs) , what, what kind of lessons do you draw from that? You know, the only lesson I draw is that, um, is that just don't do that. (laughs) You know, try and, try and forge some sort of, um, work schedule that, that kind of works for you. We can't all be superhuman and, uh, you know, his, you know, as we see, his relationships, his human relationships suffered. And he was, you know, deeply troubled as he, as he went out, at the close of his life as, as, you know, cancer was eroding his mental capabilities. I mean, he, he kind of rediscovered Catholicism. He'd converted when he was younger, but he had this, he was overtaken by this fear of mortality. And I think, you know, when we think about a productive life, I think, you know, we probably all want to go out on, on something of a high and not go out in abject terror. So yeah, you know, read about this incredible human being, but don't try to draw too many life lessons from it I think.

   10. DPDwarkesh Patel54:22

       Yeah, yeah. No, that's, that, that, that's definitely very fair. Uh, you're not John von Neumann, almost certainly. Um, so, um, Onanio, thank you so much for your time. I really appreciate you coming on the podcast.

   11. ABAnanyo Bhattacharya54:32

       Thanks very much. It was a, it was a pleasure. (music)

Episode duration: 54:44