This book is about things that gain from uncertainty or volatility. For instance, a candle is extinguished by a random wind blow, but a fire might just thrive from it. The candle light is fragile. Taleb says that a fire is not just robust against the volatile wind - it does not stay the same, but becomes better. So he proposes the term "antifragile" for such things.

Another example of his: being an employee at a big corporation for decades is fragile, being a taxi driver is antifragile. The employee may have stable income, but he gets complacent and an unforseeable "Black Swan" event (the book that made Taleb famous) could get him fired. The taxi driver constantly needs to reevaluate himself and adapt. He is constantly antifragile, the (average) employee is constantly fragile.

One more: Scientists or corporate executives are fragile with respect to reputation. A majority of people being dismissive could be fatal for their career. Artists and writers on the other hand thrive on criticism, all they need is a few followers as well.

In essence: Big successes in the world can, according to Taleb, be explained by the antifragility in which the actors were (be it that they were by accident or not). A lot of it comes down to not having complex models and strategies in this complex world. We are better at doing than thinking, if doing is done right: A mix of small-error-tolerating tinkering and experience-based heuristics, instead of complex theories which need to be correct about what happens when. The choice to be a free, independent actor instead of joining large complex structures. Keeping options open to yourself that will do less harm than good in the long run, whatever happens.

This is an especially strong message in a civilisation that is building more and more complex structures. Taleb is coming from the financial world (he was a quantitative trader), so he's seen all the needless and hurtful complexities that make up our economies.

Two telling quotes:

"In a complex domain, only time - a long time - is evidence." (p. 337)

"The best way to verify that you are alive is by checking if you like variations." (p. 423)

In the course of the book, Taleb explores the idea in various ways, among them:

1. A more mathematical (but readable for everyone) look at the issue. He introduces convex and concave payoffs to events. In essence: We might not have a good idea of the event x's distribution, but to be prepared for the payoff f(x), we can at least look at the function f.

2. Less is more. Substracting things from models and strategies is more often than not enriching. He uses medicine as an extended example.

3. Ethics of being antifragile. Many people can be antifragile to the cost of others who have to be fragile. Most currently, and in Talebs, experience, bankers are still anti-fragile as they're being bailed out by fragile tax payers. He argues that only of you have put value into y (time or work of building it, or money betting on it) you can be taken serious in your opinion about y - this is what he calls "having skin in the game".

Now, how is this book written? It is a wild ride of opinions, personal and biographic stories (all in favour of the author), many little tidbits and example from classic literature, repitions and also dialogues between more or less made-up characters. Honestly, I was close to giving up on it because the author thinks he is insightful and helpful but really isn't. It is mostly the overly creative chapter names and little chapter summaries that throw the reader out of the flow and make him forget where he is.

That's what you get if you refuse to accept an editor - a luxury Taleb can afford after his two best-selling books beforehand. However, this previous success (and experiences in the media world after that) also allowed him to be refreshingly harsh. Several professions, such as my current one (research), are not treated well and even specific persons are put in the spotlight. I was delighted that Taleb shares my refusal to like Thomas ("the world is flat") Friedman and Ray ("the singularity is near") Kurzweil for the simplistic visions they have about our complex world. He also picked Nobel price winner Joseph Stiglitz as his preferred target of critique, for being seriously wrong about the crisis, during the crisis, and having cherry-picked his past statements afterwards (this cherry-picking is currently too easy for all experts who have no skin in the game, I wonder if there is a future technology to make cherry-picking harder, given that lots of statements are actually published, as this is how economic commentators like Stiglitz or, more general, scientists, operate).

Several reviews have focused on Taleb's self-centeredness, favourably or not. I agree to a lot of what they have to say. However, the metaphor he proposes seems useful and, sometimes, the messenger is important:
1. Taleb rode around in the complex world (of finance) for decades so if he argues in favour of substracting complexities from the world, he is more believable than someone who comes from the outside.
2. He actually bet on the system failing (for years losing money) and in the end won big. It's debatable if that is ethically what you want your messenger to have done, but he can say that he not only said something was going to happen (many people did actually) but he had "skin in the game". Again, this is helping believability in most circles, especially inside the complex modern ones.