21 Ideas from The Black Swan
I keep coming back to The Black Swan: The Impact of the Highly Improbable by Nassim Nicholas Taleb (Random House, 2007; the second edition from 2010 adds a long essay called “On Robustness and Fragility”). It is not a tidy book and it does not try to be. Taleb wanders, argues with economists, tells stories about his childhood in Lebanon, and then drops a line that makes you put the book down for a minute.
So I wrote down 21 of the ideas that stayed with me. No deep analysis here, just the ideas as I remember them. If one of them grabs you, the book says it much better than I can.
1. Where the name comes from Link to heading
Europeans had only ever seen white swans, so for centuries “a black swan” was how you said something was impossible. Then Dutch explorers landed in Western Australia in 1697 and found black ones. A single bird cancelled a belief that millions of sightings had built up.
That is the whole metaphor. A Black Swan, in Taleb’s sense, is an event of that kind: the market crash of 1987, the rise of the personal computer, the First World War, the attacks of September 11, the book nobody wanted to publish that sells twenty million copies. He gives it three parts.
- It is an outlier. Nothing in the past pointed at it, so no reasonable forecast contained it.
- The impact is extreme. It does not nudge the numbers, it decides them.
- It looks predictable afterwards. Once it has happened we produce the reasons, and we walk away believing it was foreseeable and that next time we will see it coming.
All three have to be there. A rare event with no consequences is a curiosity, and a large event that everyone saw coming is just news. What makes a Black Swan dangerous is the combination, plus the last part, because the explanation we invent afterwards is what stops us preparing for the next one.
2. The turkey Link to heading
A turkey is fed every day for a thousand days. Every meal makes it more confident about the future. Then comes the Wednesday before Thanksgiving. More data made the turkey both surer and more wrong.
3. Mediocristan and Extremistan Link to heading
In Mediocristan one sample cannot move the total much. Put the heaviest person alive in a room of a thousand people and the average weight barely changes. In Extremistan one sample can be almost the whole total: book sales, wealth, deaths in a war.
4. The problem of induction Link to heading
The bird from the first section is an old problem in logic, and Taleb takes it from Hume and Popper. Counting confirmations never proves a general rule, while one counterexample kills it outright. So the useful move is not to hunt for support but to look for the thing that would break your belief.
5. Absence of evidence is not evidence of absence Link to heading
Not finding something is not the same as it not being there. Taleb calls the mix-up the round-trip fallacy, and once you notice it you see it everywhere in reporting about risk.
6. Silent evidence Link to heading
Cicero tells of Diagoras, who was shown tablets painted with worshippers who had prayed and survived a shipwreck. He asked where the pictures of the drowned ones were. We study the survivors because they are the ones still around to be studied.
7. The narrative fallacy Link to heading
We cannot keep raw facts in our heads, so we build a story around them. The story is easier to remember, and it quietly throws away every fact that does not fit.
8. Confirmation bias is the default setting Link to heading
We look for what agrees with us. Taleb’s point is not that this is a moral failing but that it is our normal mode of operation, and it needs active work to fight.
9. The ludic fallacy Link to heading
Casino odds are tame. The rules are written down, the dice have six sides, the probabilities are computable. Taleb’s example: the big losses at one casino came from a tiger mauling its performer, an employee who did not file tax forms for years, and a kidnapping. Nothing at the tables.
10. The bell curve is the great intellectual fraud Link to heading
His words, not mine. The Gaussian works fine for height and shoe size. Applied to markets, wealth and casualties it hides exactly the events that matter.
11. Fractals are a better kind of wildness Link to heading
Taleb follows Mandelbrot here. Power laws and scale invariance will not predict the next big event, but at least they do not pretend it cannot happen.
12. Scalable and non-scalable work Link to heading
A dentist gets paid per patient and there is a ceiling. A novelist, a singer or a founder can sell the same work a million times. Scalable work brings big winners, long tails of nothing, and Black Swans.
13. The triplet of opacity Link to heading
Three things make history look tidier than it was: we think we understand what is going on, we rewrite the past after the fact, and we trust facts and experts more than they deserve.
14. Epistemic arrogance Link to heading
We are mediocre at knowing things and excellent at feeling certain. Ask people for a range they are 98% sure about and reality lands outside it far more often than twice in a hundred.
15. The scandal of prediction Link to heading
Forecasts are published without their error rates. Taleb’s complaint is that the size of the error is usually the more interesting number.
16. Some experts are experts and some are empty suits Link to heading
Chess players, physicists and livestock judges really do get better with experience. Economists, political forecasters and stock analysts, on his reading of the evidence, do not. The difference is whether the field is Mediocristan or Extremistan.
17. The further out you look, the less you know Link to heading
Long projections do not just get a bit fuzzier. They cross into territory where the model itself stops applying, and the honest answer is to widen the range until it is useless, or to stop.
18. Black Swans come in two signs Link to heading
Some fields have a good tail: publishing, venture capital, research, anything where one hit pays for everything. Others have a bad tail: banking, insurance, bridges. Know which one you are standing in.
19. The barbell Link to heading
Put most of your money somewhere boring and safe, put a small part in things with a huge upside, and skip the medium-risk middle where you cannot measure the risk anyway. Losses are capped, gains are not.
20. Tinker, and be a fool in the right places Link to heading
Penicillin, the laser and the internet were not planned outcomes. Apelles the painter threw a sponge at a canvas in frustration and got the horse’s foam right. Taleb’s advice is to collect as many cheap chances at good luck as you can.
21. The antilibrary Link to heading
Umberto Eco’s library was full of books he had not read. Taleb likes this: the shelf is not a trophy case of what you know, it is a map of what you do not. Read the book and you learn something. Look at the unread ones and you remember how much is missing.
Should you read it? Link to heading
Yes, and slowly. The Black Swan is repetitive, opinionated and rude about people Taleb disagrees with, and it is still one of the few books that changed how I read the news. Start with Part One. If the turkey gets you, the rest will follow.