Predicting the Future (1)

John Perry and Ken Taylor host Philosophy Talk, questioning the predictability of the future with Nassim Taleb. They delve into the challenge of accurate future predictions due to black swan events, emphasizing the uncertainty and randomness of the future. Taleb, a professor of uncertainty, discusses his journey from philosophy to trading and his skepticism towards statistics. The episode also explores long-term planning with the Long Now Foundation's 10,000-year clock project, promoting hope and responsible decision-making for a better future. Hello, I'm John Perry. And I'm Ken Taylor. Thank you for downloading this episode of Philosophy Talk. Your support helps keep Philosophy Talk on the air and online. We appreciate your support. And as always, thank you for listening. And thank you for thinking. Welcome to Philosophy Talk, the program that questions everything except your intelligence. I'm John Perry. And I'm Ken Taylor. We're coming to you from the studios of KALW in San Francisco. Continuing conversations that began at Philosopher's Corner on the Stanford campus. Today, Predicting the Future with Nassim Taleb, author of the best-selling and controversial book The Black Swan. School boards, stock market gurus, actuaries, bookies, climatologists, meteorologists, and demographers. They're all in the business of predicting the future. But is the future really as predictable as we think? Is our confidence in knowing what the future will bring a mere illusion? Are these guys really just charlatans and frauds? If the future is less predictable than we imagine, how do we plan for it? We'll tackle these questions in three parts. We'll begin by following the lead of the great philosopher, David Hume, and cast a skeptical eye on our ability to predict the future. As Hume said, believing that the future will be like the past isn't grounded in either experience or reason. Maybe Hume was right. In part two, we'll take up Nassim Taleb's idea that the future is shaped by a sequence of black swans. That is a sequence of highly improbable, but highly consequential events that forever alter the course of things. We'll ask him why it's important that we're now living in what he calls the land of extremistan, although we evolved in what he calls the land of mediocristan. Finally, we'll take up the practical question of how we're supposed to go about living in extremistan, a world governed more by chance and randomness than many of us are prepared to admit. But Ken, let's start by getting straight about your reasons for skepticism about our ability to predict the future. After all, we seem to be pretty good at it. I bet you're confident that the sun will rise tomorrow. I bet you have plans for after the show. Yeah, I'm going to go and coach my son's baseball game after the show. And I bet you think that your son will, in the future, as he has in the past, obey your every word. I bet you think that if he hits the ball with the bat, the ball will take off in an opposite and equal whatever. I'm sure you think the field is in the same place it was yesterday. So aren't you predicting the future all the time? Well, but here's my reason for being skeptical. Look, I want to predict the future. What do I have to go on? Well, I've got the past and the present. Okay, how can I use that to predict the future if I think the past and the present are guides to the future? But how can I be confident of that? I mean, it's not logical because if I think, well, the past and present are like that, so the future must be like that. It could be different, right? And it's not experience that teaches me that the past will be like the present. I haven't experienced the future. Past will be like the future. I haven't experienced the future yet. Your buddy Hume taught us that. John? Yes, that's right. What's in this dilemma? Does our belief that the future will be like the past come on the basis of experience or reason? Well, if it's on the basis of reason, it must be some kind of logical truth that the future will be like the past, but it's obviously not because we can conceive of its being quite different. Did we learn it from experience? Well, we have experienced many past pasts and many past futures, and the past futures have by and large been like the past pasts. So can't we, by induction, assume that future futures will be like future pasts? But the question of induction is what's at issue. So it seems pretty hopeless. Yes, and on a practical ground, think of how many times people are wrong about the future. I mean, think of all the acquaintances or friends who turned out utterly different from what you thought they were on the basis of your past with them. Or think of all the wise men, the so-called wise men, who failed to foresee the collapse of the Soviet Union, failed to anticipate the Iraqi insurgency, failed to anticipate the collapse of the stock market or the bursting of the dot-com bubble. We're often wrong about the future, John. My most confident thing about the future is that the same idiots that have been wrong, the same talking heads who have been wrong for year after year, will continue to be on the same shows for year after year, which leads to the question, why make future plans at all? But there are people that are experts at this, who make their living by trying to give us a picture of the very distant future. Our roving philosophical reporter, Zooey Corneli, went out and talked to someone involved in some very long-term planning. Think about something a really long time from now. No, longer than that. I'm talking your grandchildren's grandchildren's grandchildren are long dead. It's the kind of time frame Alexander Rose and the Long Now Foundation are trying to get people to think about. A 12-hour dial that you see in most clocks is not very relevant for our type of clock. They're building a millennium clock designed to last 10,000 years. That's as long as human civilization has existed. Our dials show a lot of celestial events, planets, stars, as well as the Gregorian year in five digits. In other words, right now it's 02007. The group plans to put the clock inside a mountain in eastern Nevada. The sun's rays will hit it at the same time every day, ensuring that it keeps the correct time. Basically, we only have to engineer the clock to be accurate enough to last during a non-sunny time, which could be as long as a year or two, based on historical records from either meteor impacts or volcanic eruptions, potentially man-made events like nuclear winters as well. The visitors will provide energy by winding the clock. Rose says the goal of this project is to shift the way people think about the future. Environmental issues, educational issues, hunger around the world, these are all problems that are impossible in four years, but if you look at them as 50-, 100-year or even longer problems, then they are potentially solvable. And so we're trying to provide that kind of frame, at least if people see us doing a crazy 10,000-year project, maybe they would rein it back and do a more sensible 100-year project. With global warming, wars and the dizzying pace of technological advances, how can we imagine what things will be like in our own lifetime, let alone 10,000 years from now? But Rose says we shouldn't give up on the future. I think fundamentally our project is about hope. We will be here, it's likely we will be here in the next 10,000 years, so what are the choices that we make now that make that next 10,000 years a better one instead of a worse one? That becomes a pretty easily definable thing fairly quickly. Things like cutting down an old growth redwood forest very obviously cannot be replaced on the commercial scale that it's being harvested at. So therefore, that's a choice that's taking something away from the future. It starts to change your perspective in that way. While the Millennium Clock is ostensibly about the future, Rose says it doesn't really matter whether it actually lasts 10,000 years. It's a piece of theater in a sense, but at the same time it's an extremely real effort. We're not kidding around, we're still very much building it and what it does to the present generations I think is more valuable. Whether or not they see it, if they just know that it exists and it becomes a meme that they can work with, that's the real value. In a sense, Rose says, the clock will provide a way for us to communicate with people we'll never live long enough to meet. We don't have very many 10,000 year traditions or stories at this point, but if the memory of humanity continues, then it will certainly develop. For Philosophy Talk, I'm Zoe Corneli. I'm John Perry. With me is Ken Taylor. And our guest today is Nassim Taleb. He's Dean's Professor in the Sciences of Uncertainty at the University of Massachusetts Amherst. He's author of the best-selling book, The Black Swan, The Impact of the Highly Improbable. Nassim, welcome to Philosophy Talk. Hi, thank you very much. Thanks for inviting me. Hi, Nassim. Now, you're a professor of the Sciences of Uncertainty. Now, I bet your PhD doesn't say PhD in Sciences of Uncertainty or your MBA. So how did you get there? I mean, tell us just a little bit about your journey from wherever you started to being a professor of the Sciences of Uncertainty. It's a great honor to be among you guys because I wanted to be a philosopher first, okay, when I was a child. A worthy ambition. A worthy ambition. Yes. Yeah, that was my aim. And I eventually ended as close to that goal as possible. But in the meanwhile, I became a statistician, got an MBA, became a trader, and then later on got a PhD in a domain related to applied statistics. I worked as a trader for 18 years in the markets dealing with complicated mathematical products and spent some time thinking about my black swan idea. So when I wanted to teach something, I had choices. I can't teach statistics because I don't believe in most statistics. I think they're tools for us to fool ourselves into thinking that we know more than we actually do. So I couldn't really teach it. I decided to focus on the psychology of our error, our errors in statistics. So you know, like the anti-statistics. And there's a field called decision science, and that's the one I taught. Now I will no longer teach in it because of other obligations, but that's the one I taught, which is a combination of, I would say, one-third mathematics and statistics, one-third psychology of randomness, and one-third, I don't know if you can say, economics, business-applied entrepreneurship theory and stuff like that. So that's a great story. I mean, so from a would-be philosopher to the science of statistics to being a trader I should say, to psychology, to being a professor of uncertainty, I mean, it sounds like you are a black swan, but given that, why don't you tell us? What is a black swan? What's the significance of the title of your book? So I got to that profession, I only have one single interest, which is uncertainty and rare events. And it so happens that uncertainty straddles so many disciplines, which explains this diversity. But let's dig into your black swan idea a little bit now here. Tell us about the idea of a black swan, and then I'd like you to relate it after you tell us about the black swan to Hume's problem of induction, whether it's the same or different. The black swan, you see, before the discovery of Australia, we had no reasons to believe that swans could be of any other, no logical reason to believe, could be of any other color than white, because we had never spotted swans of any different color. And sure enough, one single observation, the sighting of an ugly black bird, destroyed centuries, perhaps millennia of confirmation. So my black swan isn't quite a bird, nor it's a logical problem. My black swan is an event. And it's an event that has three properties. The first one, it's highly improbable based on information we have or logical information, whatever we want to call it, based on reasoning or information we have at a certain period of time. Secondly, it is very immensely consequential. So I think that the Great War was a black swan, the Internet is a black swan, Google is a black swan, a lot of recent events, Harry Potter is a black swan. So it is very consequential, second point. So you cannot, although it has a very small probabilities taking place, you cannot ignore it. And number three, which is a vicious property, which is where there is psychology in it. Although it is prospectively unpredictable, retrospectively, it becomes predictable in our minds as we teach ourselves disciplines, and we try to interpret them after the fact, thinking that, you know, we could, we did predict them. In fact, all we do is postdictum. And that is, the main part of the book is about how we fool ourselves into thinking that we understood the past, or what you say, what you call just before the past, past. So we understood that. So just a little footnote here, now, I mean, the discovery of black swans really wasn't a black swan capital, because except for a few logicians who had to go change their examples, it didn't have too many consequences, did it? That's not a criticism, I just, I wouldn't know. No, no, exactly. In the book I say it, I say my black swan, I had to tie it to that bird simply because it's a nice image and because we like images. We always swayed by stories. But secondly, it is, I had to fold it into that, it's part of that classical problem was one twist. My black swan is an event of massive consequence. Yeah. Okay. Well, the most second step, sorry, the most, we're going to take a break, but the most massive consequence of Australia, in my view, is a discovery that people hanging upside down still enjoy beer. On that note, you're listening to a philosophy talk. Today we're discussing the future with Nassim Taleb, author of The Black Swan. We've been talking about what Hume calls the problem of induction and Nassim calls the black swan problem, the problem of justifying predictions about the future based on observations about the past. In our next segment, we're going to see whether history really is a series of black swans. Why not try something unpredictably yourself? Join our conversation. You can call 1-800-525-9917. That's 1-800-525-9917. Give us a call and be surprised at what you say. Black swans in your telephone calls and emails when philosophy talk continues. Have you ever tried to predict the future? Did it turn out to be right or wrong? Luck if it was right, bad luck if it was wrong, or was it skill? Do you really think there's a way to know in advance how things will turn out? Have you ever encountered a black swan, something really surprising and baffling? How did you cope with it? This is Philosophy Talk. I'm John Perry. And I'm Ken Taylor. Tell us your thoughts. The toll-free number 1-800-525-9917. That's 1-800-525-9917. Or you can email us at comments at philosophytalk.org. Our guest is Nassim Taleb, author of The Black Swan, The Impact of the Highly Improbable. Nassim, I've got a little talisman for you. Look, I live in Silicon Valley, and I know something with utter confidence about the future. The next person I meet will not be twice as tall as I am, three times as tall as I am. Will not be. Now, I live in Silicon Valley. What I don't know, the next person I meet may well have 40 or 50 or 100 times the wealth that I have. And I can't really predict with any confidence how much wealth the next person I meet will have. So I can predict sometimes, and I can't predict other times. What's going on there? Bingo. So what I tried to do with my black swan, the black swan, and we'll see it later when I talk about Hume, is not so much a document about skepticism, so much as it is a map to where we can be gullible, you see. So what I did is I divided the world into two different provinces, or two different, you know, what we see in two different variables. The first one I call mediocristan. And mediocristan is, you know, a domain, and I call it the domain, in which the exception is not consequential. So if I have a thousand people on a scale, and I pick the heaviest possible person I can find, you know, on a planet, who can still be called a person, and add that person to the scale, the total will not, will barely change, will change by 0.3%, nothing, okay. Now that domain I call mediocristan, where an exception cannot be a big consequence, because as your sample becomes larger, you see, no single event can be consequential to the total. That domain you can be gullible, because your mistakes are not, a single mistake is not going to be very significant. Now a second domain I call extremistan. If you have the very same sample of a thousand people, and you pick the wealthiest person, that wealthiest, if you have Bill Gates in there, okay, Bill Gates will represent close to 100% of the total. So your total will depend on whether you have an exceptional person in it or not. So in other words, your total will not be stable, and that domain, extremistan, is the one that prevails for social, economic, modern variables. So let's stop, I want to stop you on that last point, that's what you said, that prevails for social, economic, modern variables, I want to, so you think, so height or weight or something, that's really kind of biological, physical, and physical, is it that you think physical quantities don't have this kind of extreme... Many physical quantities, not all physical quantities, but the one in which we trained our observations as pre-humans typically are of mediocrity. So in other words, an exception can happen, can be consequential, but it's never going to be massively devastating. So our intuition, we understand those variables rather well. Right, but what is it about social variables that make them different? Social variables, it's very simple, I mean, the best test is, after this I'm going to have a huge meal, and I hope you guys have a big brunch, okay, even if I, I'm in New York City where we eat bigger portions, but even if I try to kill myself eating, okay, it's not going to make a dent in my annual consumption. So I cannot become overweight in one single episode, probably I will never become thin on a single episode, that's bad news, but then you can become rich in a second or poor in a second, you see? So the social, the money, what has money in it, what I call scalable variables, money, things that are electronic, you know, these values that today occupy a lot of our time, but in fact were not so present in our past, or in our remote past, well, for these the sky's the limit, so an exception can easily represent a big share of the pie. So now, there's nothing more social, and I guess part of extremist stand-in, than what used to be your job, buying and trading stock, and every day I tune into Jim Cramer, he predicts the future, he's always right, what's going on? I mean, sometimes he's wrong, but there's always an explanation, right? No, it's always, when you're wrong, typically, I learned from trading, my training days is when you're wrong, it's because your mother-in-law was in town, or you have a specific reason, but when you're right, it's because you're skilled, you see, so when you're wrong, it doesn't count, there is what's called the attribution bias. So, the stock market seems like very much an extremist stand-in kind of thing, is that right? Anything, well, I mean, the most extremist stand-in of variables is academic fame, you see, you have a million academics laboring in the world, and a very small number get most of the citations and credits. Well, I agree with most of your book, but there I think, of course, you're totally wrong, it's academic merit is totally based on skill and talent. You're lifting a lot of exact, we're talking about the black swan, predicting the future, but I guess Nassim Taleb, who's the author of that really nice book. So, I mean, in extremist stand, a very small error can be extremely consequential, and typically, the domains that are part of extremist stand are extremely difficult to predict, whereas variables that are in mediocre stand are usually very easy to predict. So, just one final question before we go to some callers. You were a stock trader, you had clients, you, to say in the book, your clients, you lost a lot of clients because you wouldn't make any predictions. But what did you do for the remaining ones? Just tell them. Who knows? Yeah, who knows, exactly. You have to be honest with yourself that predictions are usually used more for therapy than any other reason. So, in this conversation, 1-800-525-9917, or send us an email, emails at comments.philosophytalk.org, and we have Mark in San Francisco on the line. Welcome to Philosophy Talk, Mark. Hi. Hi, Mark. So, you mentioned the stock market. In the popular press, the popular wisdom says that markets eventually will always go up even after, say, a 20-year bear market. I'm wondering if the author believes that this could eventually prove wrong, and if so, would that event not be a black swan because any trigger to make that happen would be of much greater consequence than, say, the complete breakdown of financial markets that would never recover? Okay. I'm not going to make a prediction now. No. Just the idea. Okay. It's a very nice question, but I will tell you that I don't buy anything in the press concerning hypothesis about finance for several reasons. Number one, these predictions or these estimations come from experts, and experts in finance, and now we have data, we have a lot of data, are no better than cab drivers. So, you should listen to the opinion of cab drivers because for two reasons. Number one, they're cheaper, and two, they can provide the same therapy without your reliance, but you're not going to make big decisions listening to cab drivers. So spare yourself a lot of risk and a lot of aggravation, and do not listen to experts in finance. Their track record is not good. But let me ask you a question. Thanks for the call, Mark. Let me ask you a question. If I have these extreme events that are highly consequential, one might think that if I have a long enough time horizon, then it's going to wash out because it's going to go down, and it's going to go back up, and it's just going to ... Is that right or wrong? Not quite. Not quite. The idea of long means very, very long, and I don't know if you plan on ... I know the other gentleman wants to live 10,000 years, so maybe for him it's different, but for you and I, we probably have much more modest plans, so long is not going to be long enough for these to average out. That's the first statement. The second one, that as time ... As we are advancing, as time is moving, the world is becoming less and less tractable, less and less predictable. I mean, the Google phenomena of a company going from basement C in Stanford, where you guys are, to the most successful company in history of the world, in no time, and probably back to basement C, it will take just as short a period of time, that's something that could not happen in the past. The world is becoming less predictable. Projects today that were ... I mean, in Roman times, we used to have a much better handle of our time to completion of big, grandiose projects. Today we can't put a building on time, you see? I want to bore in on this a little bit, because it goes back to your thing about social goods being more extremist. I mean, I want to figure out more about why that is, and I also want to connect to your thing about evolution. You say, we evolved in a context, and it was mediocristan, in which there weren't these extremes, or the extremes weren't as consequential, but what is it about social life that makes us live more in extremistan? Let me just give a little follow-up on that. Most examples you've given are ones in which you can have millions of people perceive an event, like the writing of Harry Potter book, or what somebody says about a stock, or something like that. Is it the printed word, or the possibility of language that makes the difference? You're right. I think, actually, I would go back to DNA. Anything that replicates itself effortlessly. Let me give you an idea. Say that you're a writer, and you're successful. If you, well, the lady from Harry Potter, her publisher does not call her up every time they want a new book. She can just stay home, actually, it's an even better idea to be dead, because you sell more books when you die. So you don't even have to be there to reproduce yourself. I call this scale, the scale is a limit. In other words, the contribution is not into a specific output, it's just the information that replicates itself. Whereas a baker has to bake bread, or a massage professional, or a dentist, these people have to perform, they're paid by the hour. So in the domain one, Mediocre Stan, your income will not grow geometrically, it will have some boundaries, the number of hours you can work in a day. In the second one, you have no boundaries. So these domains that have no boundaries will be completely dominated by exceptions, just like the lady from Harry Potter. You know that in 1995, I think, five novels out of 16,000 represented more than half the sales. Five novels in the English language, okay? And things are getting worse, because the ecology is getting less and less diversified, so we have fewer winners now, and these winners take everything. So let me connect this to the ability or inability to predict the future. I take it, the point is, that if you took a writer, and say, predicted on the basis of her efforts, or his efforts, what his or her income might be, so you look at past writers, past writers, past writers, past writers, and you say, oh, here's a writer, and you predict on the basis of the information, how good the writing is, what the quality of it is, this extremist Stan structure makes it impossible, given a writer, to predict anything about their income, right? The next writer... Exactly, because this environment, the link between action and consequence is becoming very, very fuzzy, okay? Because you have a lot of writers who have equal skills, say, one of them is going to win, and is going to take everything, so you have to predict a lot of randomness, because you don't have a mapping, one-to-one, between skills and success, whereas for dentists, it's much easier, because a good dentist will typically have a successful career. So a good cook, if a cook is good, he's going to be able to deliver food, and people will be happy. In domain two, which is the domain that is dominated by extremes, you have a smaller number of winners, and these people win everything, so it's harder to predict. I see, I see, I see. Let me ask you about another part of the black swan idea, the retrodiction, the, you know, we can't, we narrate, when it happens, when a black swan happens, we say, oh, I knew that was going to happen, but you think of that as kind of confabulation, and we kind of design to do that, or something like that. Tell me more about that. Yeah, well, I mean, there are two elements in that. The first one is the way our brain works. History runs forward, seen backwards, you know, as Kierkegaard said, but, so I explore that mechanism about how we play events in our minds, and the second one has to do with information itself. History kills information, kills causes, like, let's take the first war, the Great War, okay? The first war took place, and then we looked at what preceded it, and effectively, there was tension between Great Britain on one hand, Austria and Germany on the other, okay? So we say, okay, tension causes war, because you have tension, and then you have, you see war, and then it preceded by tension, and you teach every single history, child in history class that tension, you know, caused the Great War. But effectively, history kills information. If you look at the record, you'd notice that there was a lot of tensions in history that did not cause war, and these are not recorded anywhere. Right. You see? So this is sort of like the bias we have in our representation of the world. I understand, I understand. You're listening to Philosophy Talk. We're talking about predicting the future with Nassim Taleb, author of The Black Swan. We'd like you to join our conversation, 1-800-525-9917. That's 1-800-525-9917. Or send us an email at comments at philosophytalk.org. And Paul and Santa Clara's on the line. Welcome to Philosophy Talk, Paul. Hi. This is a great show. It's wonderful about the black swan phenomenon. I could interject something here, perhaps. I believe people might actually have a sense of the future, which I would call the premonitive. And based on what is really existence in a future state, sort of an aspect of inevitability, what would you think about that? I'm not quite sure what you mean by that. Nassim, do you understand what he means by that? Yeah, I can link it a little bit to something that was mentioned earlier when you talked about Hume and Hume's induction. I mean, the problem of induction is being able to generalize outside of what you see to what you did not see, you see. So it's the same thing with predicting the future based on a past or generalizing based on a small sample. It's exactly the same problem. And it looks like we humans are naturally good at generalizing in some areas, not others. So children are equipped with this ability, and effectively Hume was wrong. We have a tool for induction wired into us that makes us generalize in some domain and not others. So to give you an example, if you show children, and discuss it in a black swan, if you show children pictures of people, a member of a tribe of a certain skin color, and ask them and tell them how to describe people who are not in the picture, they'll describe people that are very similar, same skin color. So they generalize based on skin color. But if you show pictures of members of a tribe, they're all overweight, and ask them to describe someone who's not part of the picture, they will not describe an overweight person. So it looks like we have some intuitions about dealing with the unknown or unseen based on the scene that are wired into us, and that do work rather well in some domain, and did work rather well in some domain. I mean, people could detect famines and stuff like that. However, the problem that we have today is that most of the randomness we encounter that matters, we do not have any innate sense for because it is not part of our heritage, and because it is wild randomness. I understand your point. I mean, it's a deep point. It's that we're evolved for a certain kind of environment in which information about the past related to information about the future in a certain way, it was structured in a certain way. And now we live in an environment through social change, largely, that's not at all, we're like out of time. That's not at all like... We don't have the right instinct. Yeah. We're not... We don't have the right instinct. And let me give you another example to link it to how the world is changing. Okay? We are very good, and I explained it in The Black Swan, our track record is excellent at, I'm sorry to say it, at pillaging or at doing raids. Okay? You can forecast the outcome of a raid. Both parties, the one who's going to attack knows, and the one who's been, you know, the victim knows, you know they're attacking us and we're going to lose. We have a very good sense, and primates actually do that all the time. They go, they attack another group and take their territory. So, we are very good at doing raids, and effectively in small police raids on neighborhoods, you know, attacking drug dealers and stuff like that, we're very good at predicting. Now, ha, how about wars? Wars, we're not good at predicting. Yeah. We have not been good at predicting since Napoleon, for example. Yes. The world, and we don't know it. Look at, I mean, what happened the last war. I mean, I was writing The Black Swan before the last war, so it was quite depressing to see that I was right. But, I mean, look at the Vietnam War. Look at the first war. Look at the second war. Nassim, I'm going to ask you to hold that thought for a second. You're listening to Philosophy Talk. We're predicting the future, or failing to, with Nassim Taleb, a professor in the sciences of uncertainty. We've been examining the concept of the Black Swan and looking at historically significant examples. In our next segment, we'll ask our guests for some practical advice on how to live and plan for the future in the face of uncertainty, us residents of extremistan. Should we be more or less open to risk when we can't know what the future holds? Prediction, Uncertainty, and Risk, when Philosophy Talk continues. Okay, this program, I predict, is going to last another few minutes. I'm John Perry. This is Philosophy Talk, the program that questions everything. Accept your intelligence. I'm Ken Taylor. Our guest is Nassim Taleb, author of The Black Swan, The Impact of the Highly Probable, and we'd like you to join this conversation, 1-800-525-9917. That's 1-800-525-9917. We may have shaken your face in the future, but you still know that by dialing that number you can get a hold of us. Nassim, I want to ask you about 9-11. This might be considered a black swan. It was certainly an exceptional and unexpected event, and it was certainly consequential. But that third thing, I'm not so sure. In the first place, some people did predict it, like, for example, you in your previous book. And secondly, when Condi Rice, the national security officer, the one you would have thought would immediately come up with some kind of story, she said, oh, who could have thought of this? So how about 9-11? My feeling about 9-11 is that it was predictable. It's just that people didn't have enough meetings where they carefully went over our procedures and stuff, or they might have come up with the idea of locking the doors. I don't know about 9-11. And I'll tell you that after the fact, after an event, you always find people who predicted it. You need to know it's better than random. Like, every stock market crash has, like, 20, 30 people who produce an I-told-you-so right after. Is it better than statistical? I'm not that sure. I did not do a thorough analysis on that, so I can't really issue a pronouncement. But there is this thing. After something consequential happens, there are lots of us who blame the people in control for not knowing because we have the feeling that it ought to have been predictable. And you think that feeling is an illusion or something, right? Let me tell you that after an event, you automatically, because that's a retrospective distortion, you'll overestimate what people should have done or should have known before. But nevertheless, I don't know about September 11, but one thing for sure, Katrina was not a black swan. It was a gray swan. Katrina is incompetence. 9-11, I'm not that sure. Right. Okay. We've got some color in the line. David's in San Francisco. You're an earthquake predictor, I understand. I'm not an earthquake predictor. I'm a structural engineer, and we work with earthquake engineering. Earthquake prediction, interestingly, is something we don't put a lot of faith in. But the conversation is very interesting. I'm looking forward to picking up the book. The last comments about Katrina and September 11th are interesting because, yeah, you try to blame the experts after the event, but we are the experts, and we're busy trying to get the word out every single day and really not finding traction there. I think earthquake engineering is interesting in the Bay Area. I'm curious to hear what the authors' thoughts are on that. Is it a probabilistic prediction? Is it something we know will inevitably happen, but we can't say today, we can't say tomorrow, and we're not exactly sure what its effects will be? And I'll hang up and listen, but I'm also interested in the comment that was made earlier in the show that statistics and the mathematics of probability he does not find particularly helpful. So that's an interesting idea, and I'd like to hear more about that in the context of extreme events too. Thanks very much. I think this is a great question. I classify extreme events in two kinds. There's the gray swan, and we can use a certain class of probability distribution that I'm sure you use, like power laws. These seem to work, although not with a lot of precision, of course, but they can give us guidance. And the second one, the black swan, in which statistics can be either useless or worse than useless in the sense that you may rely on them. They're unreliable. So earthquakes tend to belong to the gray swan category. We know we cannot predict an earthquake, when it's going to happen, but we have a vague idea of the possible gravity of an earthquake and its consequence. Thanks for the question. But let me now stress one point in how to deal with the future and what to do. My whole book, in my book, I wanted to do the opposite of what Karl Marx proposed. Marx said, okay, let's try to transform knowledge into action. My book is about the opposite, how to transform lack of knowledge into action. We live in a world we don't quite understand, and there are things we don't understand. So let's try to produce an action plan under this environment. And, okay, of course, once you accept that, then the answer becomes very easy. Number one, what are the areas in which the risk expert is not an expert? Stock market. The stock market is largely unpredictable. Therefore, I have a very simple advice. Don't take advice from someone wearing a suit if you don't have to, because that's a very simple heuristic. Wait a minute. That's a simple heuristic, but I don't quite see how it gets me through the day, how it lets me make investment decisions. Okay. An investment decision is very simple. I have my portfolio, or I suggest someone has a portfolio, say, 90% and zero risk securities, because when people tell you it's medium risk, they don't know anything about risks. Okay. We don't have a good ability. A risk is not something you measure like the temperature. We don't quite understand it, because it's extremely thin. So 90% and zero risk securities, and 10% and extremely risky securities. Hopefully, you know, the risky securities will, you know, something may come out of them. Okay. If nothing happens, then you're bounded. You can't lose more than 90%, more than 10%. Let me ask you a more basic question about this. How do I decide whether I'm an extremist stand or a mediocre stand? I mean, on the base, because it looks like I have to decide on the basis of observation and information. That's a good question. Aren't I trapped in Hume's thing? Because I don't know. Yeah, you are a little bit, a little bit, but I'll mitigate the problem. I don't know. You may, it's not too absurd to say that even, I know you guys in California and Stanford can produce a lot of strange things, but I don't know if you're going to be able to produce a two miles tall human being in the next 24 hours, right? Right. But the equivalent in socioeconomic phenomena is not as remote, you see. So it's a matter of degree. Okay. I think this is extremist stand. I think this is mediocre stand. That's number one. And the second distinction you make is don't try to guess the probabilities of events. Try to, because these are very hard to compute. What you try, you should work on is seeing how an event can affect you, you see. The reason I don't like the stock market is because I cannot produce a probability of a crash. I don't know what's the probability. But I know how a crash would affect me. I may have to go back to work, and I don't like working. But I have to decide how much, I don't know, I have to decide how much effort and energy to expend to protect myself against the thing, right? And then if I think it's not very likely, maybe I shouldn't, if it's not going to happen for a thousand years, you know, a thousand year flood. You don't know. Yeah, but the problem is a thousand year flood, you need to have 5,000, 6,000 years data to figure out a thousand year flood. And in markets, we barely have any data, you see. So we can't, it's not about, the point is not that we have a probability that's difficult. It's impossible to know the probability of stock market crash. That's my point. If I try to summarize your advice about living in an extremist stand this way, tell me if this would be more or less right. Okay. So there's going to be some black swans. You don't know what they're going to be. Don't wrap yourself around a particular black swan like the guy in your book that spends his entire life waiting for the Cossacks to invade unexpectedly. And then he misses it because he's having a drink at the bar. Right. That's a mistake. On the other hand, don't just stick your head in the sand and assume there's not going to be any black swans, right? Exactly. So the right way is somewhere in between. And I guess the metaphor is 90% in government bonds, not because the government's so great, but because it's the best thing we have going. And then 10%, I guess if I, I've taught at Stanford for 33 years, and if I had invested $100 in every symbolic systems major, that's one of our majors that's produced a lot of very successful entrepreneurs, most of those $100 would have gone down the toilet. But, boy, I would have made a lot of money on a couple of them. Yeah, there we go. That's exactly my point. What I do in a book is I make a distinction between two classes of black swans. There's a positive black swan, a good black swan, and then there's negative black swan. And I say that, you know, if you invest in a venture capital firm or invest $100 in one of these projects, you know, all you can lose is $100. So you're safe, you see? So the best thing is to take maximum risk that entails very small amounts of money, okay? When you have huge leverage, and it's only positive leverage. So you have what I call hunting for positive black swan, and you can spend some money. And over the next 33 years, you may hit on a couple of good ones. And they will pay more than, that's what book publishers do. Book publishers, you know, produce 16,000 novels, okay? And they make money on just very few of them. That's what thinkers do. They have billions of ideas that they make, and, you know, and one of them hits, and it's boom. There we go. Or, no, no, that's what science does. Science has a lot of scientists working, but only very few of them are going to win. It's sad because it's not the same scientists who can have 10,000 ideas, but science has thousands of scientists. Why not? So this is on the positive black swan. On the negative black swan, do not get involved in things, okay, like investment in the stock market in which the expert is patently not an expert. So I have a very simple tableau in my book, and I say, okay, these are the people who are not experts. Don't depend on them. Very simple. Okay, I just want to say, you know, your ambition was to be a philosopher. It's a very philosophical book, and I'm sure your ambition was to get involved in a black swan and as a bestseller. So congratulations on both counts. And on that note, Nassim, thank you so much for joining us. Thank you very much. Thanks for inviting me. Our guest has been Nassim Taleb, Dean's Professor in the Sciences of Uncertainty at the University of Massachusetts Amherst, author of the bestselling and controversial book, The Black Swan, the Impact of the Highly Improbable. And I might also add a fun read. So, John, what did you learn today? Well, I learned quite a bit reading Nassim's book. I do want to put in a caveat here. Neither Ken or I were really in a position to grill him on his critique of statistics. So if you're interested in that, go to Google and Google The Black Swan and find some of these debates and reviews where statisticians have put up a little defense of their science. It's an extremely interesting set of topics coming under the broad rubric and philosophy of what we call epistemology and endlessly fascinating. And so I really enjoyed reading his book. It was kind of long and anecdotal, but I like books like that. Yeah, there's actually a special issue of I think it's the American Statistics Journal or something like that devoted to there's a sort of symposium on this book back and forth between Nassim Taleb and a bunch of professional statisticians. I've got to say, I do think this is a really fascinating idea. And I think it's probably right that we live in this world with these extreme distributions where one single event can have these astounding consequences. But here's the thing I wonder. He does distinguish between mediocrity and extremist. How can you tell? He kind of hesitated over that. How can you tell by observation when you're in such a thing? He seems to have an a priori argument that social factors are bound to create extreme distributions. I'm not so sure about that. I mean, social factors are probably a necessary condition for being an extremist, but maybe not a sufficient condition. By the way, in the book he gives a lot of credit to Karl Popper. And we did a show on Karl Popper. So you might want to, if you find this fascinating, you might want to hunt down that show and live stream it or get a podcast of it and listen to what we said about him. For the final word on future predictions, we turn to the crystal ball of Ian Scholes, the 62nd philosopher. Ian Scholes, when it comes to predicting the future, the go-to guy is Nostradamus. Before he started calling himself Nostradamus, he was just plain Michel de Nostredame, a traveling 16th century apothecary. As he grew older, he settled down and turned his attention to astrology and the occult. He wrote an almanac, which became a bestseller. He wrote more. It was in these almanacs that his first predictions were made, thousands of them. This almanac's popularity led him to become a kind of astrologer to the stars, doing horoscopes and offering advice to nobles and the wealthy. One of his greatest admirers was Catherine de' Medici, queen consort of King Henry II of France. He then turned to what has become his most famous work, the prophecies, which contains over 900 rhymed quatrains grouped into sets of 100, which he called centuries. It received a mixed reaction at the time, but since then, over 200 editions of the prophecies have been published. The prophecies themselves are what you might call cryptic or even woefully obscure. Take this sample. A man will be charged with the destruction of temples and sects, altered by fantasy. He will harm the rocks rather than the living, ears filled with ornate speeches. Well, what man, what temple, what rocks? Is it the rocks or the living with the filled ears? Now, this is the time of the Black Plague, giving rise to a lot of gloom and doom end-time prophecies, which were very popular. Apparently, Nostradamus wanted to get a piece of that, but didn't want to get in trouble with zealots by actually, you know, saying anything. Hence his vagueness, and hence why most of the quatrains are about generic disasters, battles, fire, flood, drought, that could happen anywhere, anytime. Still, over the years, Nostradamus has been credited with predicting the French Revolution, the atomic bomb, the death of Princess Diana, and even 9-11. Just after 9-11, this quatrain started making the rounds of the web. In the city of God, there will be great thunder. Two brothers torn apart by chaos. While the fortress endures, the great leader will succumb. The third big war will begin when the big city is burning. And it was signed Nostradamus 1654. This symbolism was supposed to uncannily reflect the events of 9-11. Well, Nostradamus died in 1566, so it's doubtful he could have made this prediction in 1654. The quatrain turned out to have been written in the 1990s by a Canadian student, Neal Marshall, to show how easy it is to be Nostradamus-like. Throw in some vague yet pretentious imagery, presto, you have a prophecy for any occasion, one size fits all. Back in 1566, meanwhile, Nostradamus' increasing ill health led him to draw up his will, after which he told his secretary, You will not find me alive at sunrise. Next morning, he was found dead. So, he got one right at least. I gotta go. Ian Shoes, the only man who can solve a philosophical problem in 60 seconds. Philosophy Talk is a presentation of Ben Manilla Productions and the trustees of Leland Stanford Junior University, copyright 2007. Our executive producer is David Demarest. Special thanks to Devin Strolovich, Daniel Elstein, Zoe Corneli, Merle Kessler, and Mark Stone. Philosophy Talk is sponsored in part by Powell's City of Books, on the web at powells.com. Support also comes from the Templeton Foundation and from various groups at Stanford University, the Friends of Philosophy Talk, and the members of KAOW San Francisco, where our program originates. The views expressed or misexpressed in this program do not necessarily represent the opinions of Stanford University or of our other funders. The conversation continues on our website, philosophytalk.org. I'm John Perry. And I'm Ken Taylor. Thank you for listening, and thank you for thinking. Philosophy Talk www.philosophytalk.org