Review:
I was reading "The Black Swan" on the flight home. It's got some fascinating stuff on the problem of induction, but overall, the author makes so many snarky hits at concepts and ideas he doesn't really bother to completely understand. I'm halfway through, and it seems to be getting a bit better, but I have to see this book so far as infuriating, decontextualized bile in the grand scheme of things that makes a few good points, and I think I will probably read the rest with such a viciously critical eye that I will probably miss any possibility of enlightenment.
You see, the book is about Black Swans, which are defined to be transformative, unexpected, rare events. The one-in-a-million events, like the invention of the wheel, the onset of a war, etc. Now, after a foreword asking us to imagine all such events in our life and in history, the author (Nassim Nicholas Taleb) claims that most of our lives absolutely hinge on these catastrophically powerful Black Swans, that these events are so transformative that they leave in their dust the gradual changes. This is the general assumption of the book, from which all the rest follows. Taleb's history as a trader gives him a wealth of examples to draw upon to illustrate Black Swans, and the consequent failures of predictions. Taleb also finds a number of historical examples: Wars, far from seeming inevitable, actually take almost everyone by surprise in the beginning. We have failures in predictions, and we are governed by unfathomable forces that are unfathomably rare.
Now, this narrative is innocuous enough, and the empiricism and humility and skepticism the narrative seems to engender should render the book fit for any readers looking for marvelous exposition. But there's a problem. Taleb is a shockingly unpleasant human being, who writes of his exploits with the empiricism of a sales pitch, the skepticism of a high-falutin' charlatan, and the feigned, egregiously false-hearted humility of a pick-up artist. Everyone Taleb decides is wrong is a "fool" or a selfish "fraud", or both. Deprecating humor without real self-deprecation. Human nature causes us to be stupid, okay, but simply by his special life story, he's able to be a lot less stupid, and tell us so, is his main point. He takes a broad, often thoughtful, sweep of the history of ideas, which is commendable. However, his reduction of this history of ideas into his self-serving (to the point of arrogance) and others-tearing-down (to the point of narcissism) is disgusting, and it's hard to take his work seriously. People and ideas he doesn't like are trapped by their own delusions, people he does like are polymaths with good senses of humor. The fox is better than the hedgehog, is what he gets from Isaish Berlin's famous dichotomy. Everything is about whether these people and ideas grasp "the point" (which is ostensibly the big idea of whatever Taleb is writing about at the time) and if they don't, then they are overrated and discarded, and if they are not, they are brilliant and underrated. In Taleb's hegemony of the Black Swan, Umberto Eco's library and the idea of a library containing vastly more unread books than read books, trumps all of the worthfulness of Plato ("Platonicity", or a tendency towards reductive categorization, is (often rightfully) mocked mercilessly), Kant, and all of economics. Everything that confirms the Black Swan is elevated above everything that doesn't yet understand the brilliance of the Black Swan.
I see all sorts of praise for books like this that posit "bold", "counter-intuitive", "deep" theories of everything, but so often (I guess, following Nietzsche) it seems like these books are not really about philosophy but about these "geniuses" who go around to interesting and exclusive clubs and have so many interesting and exclusive friends where they get to say all of these interesting and exclusive things in big meetings. They get to pontificate about how they made their fortune (You have no idea how cool it is to make a million dollars in an evening. It's the most placid feeling in the world, because that amount of money allows you to be comfortable without being arrogant. I slept for twelve hours straight, because I had the ultimate comfort of being right!), make funny stories and personal anecdotes that inevitably illustrate some aspect of their main point*, and generally, take 70 pages to make a simple point and make it into a gigantic revelation for the reader. 70 pages, a chunk of text replete with so many arbitrary neologisms (which are nearly always twinned: For every Extremistan is a Mediocristan in which things are the opposite) that it requires its own thesaurus of cute names. This is David Berri except on subjects that actually matter to our day-to-day lives, without an associated rise in gravitas, respect, and humility with the importance of his subject.
* - PARODY WARNING AHEAD: But of course, Antoinette's error was that she was thinking with One Head. After reading this book, one would hope she would learn to think with Zero Heads next time and approach the problem of poverty with an open mind. It's too bad she learned her lesson so suddenly, without the possibility!
Appendix:
The above was just, like, my opinion, man, and I think it's kind of wrong to leave an opinion without any context. So let me give you an example of Taleb's dismissive arrogance: The book claims that use of the Bell Curve (i.e. the Gaussian normal curve) by sociologists and economists is a fraud. It gives predictions that use this curve the weighty title "Great Intellectual Fraud". It names this concept in the forward, and, after 200+ pages, is still motioning towards a magical "Chapter Sixteen" in which all of this vitriol will be explained as important, thoughtful, and necessary. Now, I'm fairly certain this Chapter Sixteen exists, given that I'm starting Chapter Fourteen after some 50000 prior words in which the order of words and chapters were preserved. But I don't know if it really matters of Chapter Sixteen exists, because (assuming I'm a reader only vaguely familiar with the term) I've been given dozens of psychological "anchors" attempting to endear me to this author and to make the idea of this "bell curve as fraud" idea incredibly palatable and even inevitable, without a slight bit of explanation or evidence other than the credibility of the author. I've been rhetorically set up for an explanation. For someone so keen to point out the narrative fallacy (apologizing for the central contradiction of the book's subject matter in the foreword) and the psychological ability to lead people on to bad conclusions, this is inexcusable leading on and narrative forming. If something is a Great Intellectual Fraud, then its own demerits by a clear-headed author should clearly expose this fraudulence.
Dozens of pejorative references to the bell curve, strangely indifferent references to Gauss (empiricism? Try Gauss and his friends literally climbed mountains to measure the curvature of the Earth.), and dozens of glowing references to people that dislike the Bell Curve. This is where I sit at the gates of Chapter Sixteen. Benoit Mandelbrot is given more play than Gauss. Mandelbrot, Pareto, that guy who wrote about the Pareto distribution, even fucking Zipf are given more play than Gauss in the history of ideas, just because Taleb (the skeptical empiricist who seemingly hadn't been exposed to Bayesian statistics before he wrote a book about probability and Platonic fallacies) doesn't like the way people use the normal distribution instead of his favorite distributions in which certain things are rare but important instead of banal and reducible.
You know what? I took stats: The normal curve does get too much play in introductory treatments, but it has nothing to do with rare events being too frequent (and therefore, for example, Pareto distributions therefore being preferable to normal distributions), any more than the traditional college emphasis on calculus has to do with "things, in general, actually being too discrete", which is essentially Taleb's analogous point with the normal curve.
The simplest reason the normal curve gets so much play is not because of a conspiracy of arrogant intellectuals. It's because of the Central Limit Theorem, which states that if you do the same thing a bunch of times (called trials) and you average the results you get a normal curve around the mean of what you did with a variance that relates to the variance of the individual trial is inversely proportional to the sample size. The variance heads to zero as you conduct infinitely many trials. If you don't know much statistics or much supplementary mathematics, but still have a lot to offer an institution, the CLT gives you an easy set of tools based on easily statable assumptions. Normality is just the mathematical form the tools happen to take. If researchers aren't using it well, I see no reason why they wouldn't butcher the use of Taleb's own favorite distributions, with as much consternation to the author. They still wouldn't think outside the "Platonic" box. They still wouldn't grasp the wonder of the void.
Now, if we don't know much statistics, and we're projecting complex things like the economy, should we probably say something about that? Should we say that we're using the normal distribution because it seems to make sense as a prior for such-and-such a reason? Should we talk about the error bars, and the limited descriptive power of what it is that we're predicting? Should we be a little more humble about the 400-page legislation's proposed impact on the 300 million person economy, if we're using not-very-incisive-nor-infallible tools? Yes, of course we should. Should we be a little humbler about the probability of the biopsy being cancer, and be more cautious with the intuitive probabilitistic bounds we place on statistics? Should we be more aware of our psychological limitations? Yes, of course we should.
But the CLT is a brilliant theorem which ties together the continuous and the discrete, makes the interpretation of certain ubiquitous datasets trivially easy, and in its most general form is a staggeringly gigantic mathematical statement at which I can only smile shake my humble head in ignorance. It makes unfathomable dimensionality comprehensible to people that don't have much mathematical sophistication, and gives people with such sophistication a much richer toolset. That humanity imperfectly grasps this gift is not evidence that we need to drop it. That the CLT is the one thing keeping humanity from grasping the Void of Buddhism (so to speak) is laughable. That it's one of many things keeping humanity from the Void is equally laughable, but also easy to disguise behind the talents of a good author. It is a gift, from God, mathematics, empiricism, or a Black Swan. The CLT is the shit that works, and we have to recognize it.
It's just given a little bit too much play. And it's cool to just say that, I think. I don't know that academics would be so offended by just saying that.
God, it feels like I'm writing an obituary for Taleb here, like the dead are humming around me as I speak ill of them, so little respect do I have for how he has spent his vital mind in banal self-serving narratives. God, it's hard for a reviewer not to mirror the form of the work under consideration, and Taleb's text is a dismal collection of dismal opinions brightened up by clarity and darkened by hateful rhetoric, to a picture of a man, and I guess my sense now was that, all together, this was a picture of spiritual death.
I mean, it just turns into a tally at some point: Taleb actually calls the "last thinking mathematician" Poincare. I'll just list Grothendieck (and, what the hell, John Baez, too) and his unimaginably vast creativity and conscience, in response to this disgusting ignorance and bitterness, leaving a casual search of mathematical philosophy to the reader.* It's glaring ignorance and arrogant foolishness. For someone that calls so many people "frauds" and "foolish" it's amazing how little domain knowledge is required to refute so many of his snarky asides.
* - Wikipedia. In the meantime, some of Taleb's other crimes against art: When introduced to the intractability of the three-body problem, he chastises mathematicians who build more complex models to deal with the complexity encountered instead of giving up and dealing with the problems of ineffable complexity, claiming they "missed the point" of the thought experiment. He rediscovers Bayesian interpretation of probability in a clumsy section claiming that imperfect knowledge about a variable is functionally equivalent to that variable being random (a philosophy which has already been well-formalized into a useful branch of statistics called Bayesian statistics). He uses a long thought experiment about a turkey's lifespan of 1000 days of feeding followed by Thanksgiving to illustrate the failure of linear models (so naive of that turkey), and in an unrelated section chastises a hypothetical "statistician who works for a bank". The statistician's crime? Being a "nerd" who lives a 9-to-5 life and declares that the 100th head has a 50% chance of coming up after 99 heads. He is dismissed as foolish, compared to a street-smart Manhattan version of Rush's version of Tom Sawyer.
Enough of that:
But enough of that. I want to talk about something less unpleasant. Joe Posnanski is the Tim Duncan of sportswriters.
It should be hard to write a blog when your reader's vocalization of your column is - like Duncan's voice - a syrupy lilt that resembles the serenity of a dribble in its syllabic consistency and its airy ambience.
It should be hard to write a blog when you are free - like Duncan is - from the egotistical tearing down of sports figures (including competitors), preferring instead the rare censure and the common praise.
It should be hard to write a blog when you do the same thing every time - like Duncan does - but this is not actually true of either Duncan or Posnanski. Within the apparent simplicity of a column or a move springs the accumulated thoughts and passions of a life well-spent, distilled into the end result with surreal levels of craft and intelligence.
Linking to one of his columns is pointless, just like linking to a film of Duncan highlights: You'll understand his greatness, but you won't understand why. Start at the most recent and work your way back, or google whatever subject you can think of in sports. That's how the rest of us found him, and none of us have any regrets about it.
I was reading "The Black Swan" on the flight home. It's got some fascinating stuff on the problem of induction, but overall, the author makes so many snarky hits at concepts and ideas he doesn't really bother to completely understand. I'm halfway through, and it seems to be getting a bit better, but I have to see this book so far as infuriating, decontextualized bile in the grand scheme of things that makes a few good points, and I think I will probably read the rest with such a viciously critical eye that I will probably miss any possibility of enlightenment.
You see, the book is about Black Swans, which are defined to be transformative, unexpected, rare events. The one-in-a-million events, like the invention of the wheel, the onset of a war, etc. Now, after a foreword asking us to imagine all such events in our life and in history, the author (Nassim Nicholas Taleb) claims that most of our lives absolutely hinge on these catastrophically powerful Black Swans, that these events are so transformative that they leave in their dust the gradual changes. This is the general assumption of the book, from which all the rest follows. Taleb's history as a trader gives him a wealth of examples to draw upon to illustrate Black Swans, and the consequent failures of predictions. Taleb also finds a number of historical examples: Wars, far from seeming inevitable, actually take almost everyone by surprise in the beginning. We have failures in predictions, and we are governed by unfathomable forces that are unfathomably rare.
Now, this narrative is innocuous enough, and the empiricism and humility and skepticism the narrative seems to engender should render the book fit for any readers looking for marvelous exposition. But there's a problem. Taleb is a shockingly unpleasant human being, who writes of his exploits with the empiricism of a sales pitch, the skepticism of a high-falutin' charlatan, and the feigned, egregiously false-hearted humility of a pick-up artist. Everyone Taleb decides is wrong is a "fool" or a selfish "fraud", or both. Deprecating humor without real self-deprecation. Human nature causes us to be stupid, okay, but simply by his special life story, he's able to be a lot less stupid, and tell us so, is his main point. He takes a broad, often thoughtful, sweep of the history of ideas, which is commendable. However, his reduction of this history of ideas into his self-serving (to the point of arrogance) and others-tearing-down (to the point of narcissism) is disgusting, and it's hard to take his work seriously. People and ideas he doesn't like are trapped by their own delusions, people he does like are polymaths with good senses of humor. The fox is better than the hedgehog, is what he gets from Isaish Berlin's famous dichotomy. Everything is about whether these people and ideas grasp "the point" (which is ostensibly the big idea of whatever Taleb is writing about at the time) and if they don't, then they are overrated and discarded, and if they are not, they are brilliant and underrated. In Taleb's hegemony of the Black Swan, Umberto Eco's library and the idea of a library containing vastly more unread books than read books, trumps all of the worthfulness of Plato ("Platonicity", or a tendency towards reductive categorization, is (often rightfully) mocked mercilessly), Kant, and all of economics. Everything that confirms the Black Swan is elevated above everything that doesn't yet understand the brilliance of the Black Swan.
I see all sorts of praise for books like this that posit "bold", "counter-intuitive", "deep" theories of everything, but so often (I guess, following Nietzsche) it seems like these books are not really about philosophy but about these "geniuses" who go around to interesting and exclusive clubs and have so many interesting and exclusive friends where they get to say all of these interesting and exclusive things in big meetings. They get to pontificate about how they made their fortune (You have no idea how cool it is to make a million dollars in an evening. It's the most placid feeling in the world, because that amount of money allows you to be comfortable without being arrogant. I slept for twelve hours straight, because I had the ultimate comfort of being right!), make funny stories and personal anecdotes that inevitably illustrate some aspect of their main point*, and generally, take 70 pages to make a simple point and make it into a gigantic revelation for the reader. 70 pages, a chunk of text replete with so many arbitrary neologisms (which are nearly always twinned: For every Extremistan is a Mediocristan in which things are the opposite) that it requires its own thesaurus of cute names. This is David Berri except on subjects that actually matter to our day-to-day lives, without an associated rise in gravitas, respect, and humility with the importance of his subject.
* - PARODY WARNING AHEAD: But of course, Antoinette's error was that she was thinking with One Head. After reading this book, one would hope she would learn to think with Zero Heads next time and approach the problem of poverty with an open mind. It's too bad she learned her lesson so suddenly, without the possibility!
Appendix:
The above was just, like, my opinion, man, and I think it's kind of wrong to leave an opinion without any context. So let me give you an example of Taleb's dismissive arrogance: The book claims that use of the Bell Curve (i.e. the Gaussian normal curve) by sociologists and economists is a fraud. It gives predictions that use this curve the weighty title "Great Intellectual Fraud". It names this concept in the forward, and, after 200+ pages, is still motioning towards a magical "Chapter Sixteen" in which all of this vitriol will be explained as important, thoughtful, and necessary. Now, I'm fairly certain this Chapter Sixteen exists, given that I'm starting Chapter Fourteen after some 50000 prior words in which the order of words and chapters were preserved. But I don't know if it really matters of Chapter Sixteen exists, because (assuming I'm a reader only vaguely familiar with the term) I've been given dozens of psychological "anchors" attempting to endear me to this author and to make the idea of this "bell curve as fraud" idea incredibly palatable and even inevitable, without a slight bit of explanation or evidence other than the credibility of the author. I've been rhetorically set up for an explanation. For someone so keen to point out the narrative fallacy (apologizing for the central contradiction of the book's subject matter in the foreword) and the psychological ability to lead people on to bad conclusions, this is inexcusable leading on and narrative forming. If something is a Great Intellectual Fraud, then its own demerits by a clear-headed author should clearly expose this fraudulence.
Dozens of pejorative references to the bell curve, strangely indifferent references to Gauss (empiricism? Try Gauss and his friends literally climbed mountains to measure the curvature of the Earth.), and dozens of glowing references to people that dislike the Bell Curve. This is where I sit at the gates of Chapter Sixteen. Benoit Mandelbrot is given more play than Gauss. Mandelbrot, Pareto, that guy who wrote about the Pareto distribution, even fucking Zipf are given more play than Gauss in the history of ideas, just because Taleb (the skeptical empiricist who seemingly hadn't been exposed to Bayesian statistics before he wrote a book about probability and Platonic fallacies) doesn't like the way people use the normal distribution instead of his favorite distributions in which certain things are rare but important instead of banal and reducible.
You know what? I took stats: The normal curve does get too much play in introductory treatments, but it has nothing to do with rare events being too frequent (and therefore, for example, Pareto distributions therefore being preferable to normal distributions), any more than the traditional college emphasis on calculus has to do with "things, in general, actually being too discrete", which is essentially Taleb's analogous point with the normal curve.
The simplest reason the normal curve gets so much play is not because of a conspiracy of arrogant intellectuals. It's because of the Central Limit Theorem, which states that if you do the same thing a bunch of times (called trials) and you average the results you get a normal curve around the mean of what you did with a variance that relates to the variance of the individual trial is inversely proportional to the sample size. The variance heads to zero as you conduct infinitely many trials. If you don't know much statistics or much supplementary mathematics, but still have a lot to offer an institution, the CLT gives you an easy set of tools based on easily statable assumptions. Normality is just the mathematical form the tools happen to take. If researchers aren't using it well, I see no reason why they wouldn't butcher the use of Taleb's own favorite distributions, with as much consternation to the author. They still wouldn't think outside the "Platonic" box. They still wouldn't grasp the wonder of the void.
Now, if we don't know much statistics, and we're projecting complex things like the economy, should we probably say something about that? Should we say that we're using the normal distribution because it seems to make sense as a prior for such-and-such a reason? Should we talk about the error bars, and the limited descriptive power of what it is that we're predicting? Should we be a little more humble about the 400-page legislation's proposed impact on the 300 million person economy, if we're using not-very-incisive-nor-infallible tools? Yes, of course we should. Should we be a little humbler about the probability of the biopsy being cancer, and be more cautious with the intuitive probabilitistic bounds we place on statistics? Should we be more aware of our psychological limitations? Yes, of course we should.
But the CLT is a brilliant theorem which ties together the continuous and the discrete, makes the interpretation of certain ubiquitous datasets trivially easy, and in its most general form is a staggeringly gigantic mathematical statement at which I can only smile shake my humble head in ignorance. It makes unfathomable dimensionality comprehensible to people that don't have much mathematical sophistication, and gives people with such sophistication a much richer toolset. That humanity imperfectly grasps this gift is not evidence that we need to drop it. That the CLT is the one thing keeping humanity from grasping the Void of Buddhism (so to speak) is laughable. That it's one of many things keeping humanity from the Void is equally laughable, but also easy to disguise behind the talents of a good author. It is a gift, from God, mathematics, empiricism, or a Black Swan. The CLT is the shit that works, and we have to recognize it.
It's just given a little bit too much play. And it's cool to just say that, I think. I don't know that academics would be so offended by just saying that.
God, it feels like I'm writing an obituary for Taleb here, like the dead are humming around me as I speak ill of them, so little respect do I have for how he has spent his vital mind in banal self-serving narratives. God, it's hard for a reviewer not to mirror the form of the work under consideration, and Taleb's text is a dismal collection of dismal opinions brightened up by clarity and darkened by hateful rhetoric, to a picture of a man, and I guess my sense now was that, all together, this was a picture of spiritual death.
I mean, it just turns into a tally at some point: Taleb actually calls the "last thinking mathematician" Poincare. I'll just list Grothendieck (and, what the hell, John Baez, too) and his unimaginably vast creativity and conscience, in response to this disgusting ignorance and bitterness, leaving a casual search of mathematical philosophy to the reader.* It's glaring ignorance and arrogant foolishness. For someone that calls so many people "frauds" and "foolish" it's amazing how little domain knowledge is required to refute so many of his snarky asides.
* - Wikipedia. In the meantime, some of Taleb's other crimes against art: When introduced to the intractability of the three-body problem, he chastises mathematicians who build more complex models to deal with the complexity encountered instead of giving up and dealing with the problems of ineffable complexity, claiming they "missed the point" of the thought experiment. He rediscovers Bayesian interpretation of probability in a clumsy section claiming that imperfect knowledge about a variable is functionally equivalent to that variable being random (a philosophy which has already been well-formalized into a useful branch of statistics called Bayesian statistics). He uses a long thought experiment about a turkey's lifespan of 1000 days of feeding followed by Thanksgiving to illustrate the failure of linear models (so naive of that turkey), and in an unrelated section chastises a hypothetical "statistician who works for a bank". The statistician's crime? Being a "nerd" who lives a 9-to-5 life and declares that the 100th head has a 50% chance of coming up after 99 heads. He is dismissed as foolish, compared to a street-smart Manhattan version of Rush's version of Tom Sawyer.
Enough of that:
But enough of that. I want to talk about something less unpleasant. Joe Posnanski is the Tim Duncan of sportswriters.
It should be hard to write a blog when your reader's vocalization of your column is - like Duncan's voice - a syrupy lilt that resembles the serenity of a dribble in its syllabic consistency and its airy ambience.
It should be hard to write a blog when you are free - like Duncan is - from the egotistical tearing down of sports figures (including competitors), preferring instead the rare censure and the common praise.
It should be hard to write a blog when you do the same thing every time - like Duncan does - but this is not actually true of either Duncan or Posnanski. Within the apparent simplicity of a column or a move springs the accumulated thoughts and passions of a life well-spent, distilled into the end result with surreal levels of craft and intelligence.
Linking to one of his columns is pointless, just like linking to a film of Duncan highlights: You'll understand his greatness, but you won't understand why. Start at the most recent and work your way back, or google whatever subject you can think of in sports. That's how the rest of us found him, and none of us have any regrets about it.
No comments:
Post a Comment