The Black Peacock
I finally got around to the much-discussed Black Swan book by Taleb. If you have not read it, I advise against doing so, but if you feel like it anyway you might want to stop reading now -- there will be spoilers.
The book is allegedly about low-probability high-impact events, which is what Mr. Taleb means by a "black swan". I disagree. I think the book is about Mr. Nassim Nicholas Taleb as he sees himself (or as he would like to). There are indeed a few ideas regarding probability and its perception by humans, none of them new, as far as I can tell, but these are positively drowned out by autobiographical anecdotes and narrative of self-admiration. The author is also deliberately insolent and provocative throughout the book, which makes it easy for a certain type of reader to identify with him and enjoy virtually emptying buckets of shit on physicists, economists, statisticians and general public. The author does spare a few names, such as Popper, Mandelbrot and Poincaré, and seems to declare respect for them, but even then it is a condescending "yeah, you are OK, dude". Personally, while reading all this I could not get rid of a nagging question: "Dear Mr. Taleb, who the hell are you to judge?"
It is clear that the author is not very much into most of the subject fields he mentions, but I would wager he is good at what might be called applied psychology. Essentially the book is just a four-hundred-page-long provocation. Its sole purpose is to attract attention, create a scandal, It seems to have succeeded, too, selling several million copies and making it to the NYT Bestseller list. In fact, in the book itself the author cheerfully recites which prominent scientists and executives he managed to offend personally and seems to wallow in the attention it created. He appears to be one of those people who realize there is no such thing as bad publicity and enjoys being criticized as much as being praised. This also implies that by writing this I am further promoting Mr. Taleb. My only excuse is that this post is unlikely to reach any sizable audience, so the impact should be negligible, compared to that of numerous critics and supporters writing in top journals.
I do find some of the ideas presented in the book useful. For instance, the fact that many events we encounter in real life have probability distributions significantly different from Gaussian, or that humans often tend to see causal connections between events that are in reality unrelated. What I object to is sticking labels on each one of these basic facts (the latter is referred to as the Narrative Fallacy, for instance) and then pretending that everyone else is trying to prove the opposite, just so you can feel like a knight in shining armor fighting against overwhelming odds and human stupidity. Besides, in most cases all the advice he has to give us is "be aware of all the cognitive biases and don't rely on statistical models because they are inaccurate". Overall, I think the guy is just trying to make some "fuck you" money, as he himself calls it.
Another review points out that there is a shorter and more informative description of the same ideas presented in The Fourth Quadrant
The book is allegedly about low-probability high-impact events, which is what Mr. Taleb means by a "black swan". I disagree. I think the book is about Mr. Nassim Nicholas Taleb as he sees himself (or as he would like to). There are indeed a few ideas regarding probability and its perception by humans, none of them new, as far as I can tell, but these are positively drowned out by autobiographical anecdotes and narrative of self-admiration. The author is also deliberately insolent and provocative throughout the book, which makes it easy for a certain type of reader to identify with him and enjoy virtually emptying buckets of shit on physicists, economists, statisticians and general public. The author does spare a few names, such as Popper, Mandelbrot and Poincaré, and seems to declare respect for them, but even then it is a condescending "yeah, you are OK, dude". Personally, while reading all this I could not get rid of a nagging question: "Dear Mr. Taleb, who the hell are you to judge?"
It is clear that the author is not very much into most of the subject fields he mentions, but I would wager he is good at what might be called applied psychology. Essentially the book is just a four-hundred-page-long provocation. Its sole purpose is to attract attention, create a scandal, It seems to have succeeded, too, selling several million copies and making it to the NYT Bestseller list. In fact, in the book itself the author cheerfully recites which prominent scientists and executives he managed to offend personally and seems to wallow in the attention it created. He appears to be one of those people who realize there is no such thing as bad publicity and enjoys being criticized as much as being praised. This also implies that by writing this I am further promoting Mr. Taleb. My only excuse is that this post is unlikely to reach any sizable audience, so the impact should be negligible, compared to that of numerous critics and supporters writing in top journals.
I do find some of the ideas presented in the book useful. For instance, the fact that many events we encounter in real life have probability distributions significantly different from Gaussian, or that humans often tend to see causal connections between events that are in reality unrelated. What I object to is sticking labels on each one of these basic facts (the latter is referred to as the Narrative Fallacy, for instance) and then pretending that everyone else is trying to prove the opposite, just so you can feel like a knight in shining armor fighting against overwhelming odds and human stupidity. Besides, in most cases all the advice he has to give us is "be aware of all the cognitive biases and don't rely on statistical models because they are inaccurate". Overall, I think the guy is just trying to make some "fuck you" money, as he himself calls it.
Another review points out that there is a shorter and more informative description of the same ideas presented in The Fourth Quadrant