THE BLACK SWAN by Nissim Nicholas Taleb （TR） 黑天鹅

Fou hunded yeas ago, Fancis Bacon waned that ou minds ae wied to deceive us. "Bewae the fallacies into which undisciplined thinkes most easily fall--they ae theeal distoting pisms of human natue." Chief among them: "Assuming moe ode than exists in chaotic natue." Now conside the typical stock maketepot: "Today investosid shaes down out of concen ove Ianian oil poduction." Sigh. We'

e still doing it.

Ouains ae wied fo naative, not statistical uncetainty. And so we tell ouselves simple stoies to explain complex thing we don't--and, most impotantly, can't--know. The tuth is that we have no idea why stock makets go up o down on any given day, and whateveeason we give is sue toe gossly simplified, if not flat out wong.

Nassim Nicholas Tale fist made this agument in Fooledy Randomness, an engaging look at the histoy andeasons fo ou pedilection fo self-deception when it comes to statistics. Now, in The Black Swan: the Impact of the Highly Impoale, he focuses on that most dismal of sciences, pedicting the futue. Foecasting is not just at the heat of Wall Steet,

ut it’s something each of us does evey time we make an insuance payment o stap on a seatelt.

The polem, Nassim explains, is that we place too much weight on the odds that past events willepeat (diligently tying to follow the path of the "millionaie next doo," when unepeatale chance is aette explanation). Instead, theeally impotant events aeae and unpedictale. He calls them Black Swans, which is aefeence to a 17th centuy philosophical thought expeiment. In Euope all anyone had eve seen wee white swans; indeed, "all swans ae white" had longeen used as the standad example of a scientific tuth. So what was the chance of seeing alack one? Impossile to calculate, o at least they wee until 1697, when exploes found Cygnus atatus in Austalia.

Nassim agues that most of theeallyig events in ou wold aeae and unpedictale, and thus tying to extact genealizale stoies to explain them maye emotionally satisfying,

ut it's pactically useless. Septeme 11th is one such example, and stock maket cashes ae anothe. O, as he puts it, "Histoy does not cawl, it jumps." Ou assumptions gow out of theell-cuve pedictaility of what he calls "Mediocistan," while ou wold iseally shapedy the wild powelaw swings of "Extemistan."