Books I'm Reading: The Black Swan

Author Nassim Nicholas Taleb frequently uses the publishing industry as an example of one prone to what he calls "blacks swans"—a black swan is a highly unpredictable event that has far-reaching consequences and one that we rationalize after the fact, making up explanations to reassure ourselves that what happened was not as random as it seemed. He mentions best-selling books that seem to come out of nowhere, like the Harry Potter series—books that become extraordinarily popular for reasons that are easy to come up with after they take off, but that few could have suggested beforehand. He mentions events such as the 9-11 attacks. Taleb's book, The Black Swan (Random House, 2007) is something of a black swan itself. It appears to have been a major bestseller when it was new, around 2007. I bought my copy at about that time on a business trip to Tokyo but have read it only now.

Detailed, far-ranging, and not a little bit arrogant, The Black Swan is entertaining, if hard to follow in places, and somewhat repetitive. It's the kind of book you immediately want to read again after you finish, wondering if you really absorbed all there was to be absorbed—although the repetition and the author's self-congratulatory asides can become a bit tedious. To be fair, though, Taleb considerately suggests that about half the book can be skipped by those already familiar with some of his ideas.

In a nutshell, Taleb argues that we are very bad at prediction but like to pretend that we are quite good at it—that we are far too often groping blindly in the dark, hoping to find our way safely through the minefield of life, victims of confirmation bias and narrative bias, in particular. He has especially pointed barbs to hurl at policy makers, economists, and investment advisors, marking their activities as those of people habitually making bad decisions because of these biases and because of a misguided belief that they are working on meaningful evidence from the past, which Taleb suggests is, in the vast majority of cases, useless information.