Antifragile

1. Some things gain from disorder. Taleb names a third category beyond fragile and robust: antifragile. Fragile systems break under stress, robust ones endure it, and antifragile ones improve from it. Muscles, immune systems, and evolutionary processes are antifragile by design. 2. Volatility and stressors are information. Suppressing small shocks — bailouts, over-medication, helicopter parenting, censored markets — hides risk that then concentrates and erupts later. Frequent small failures are healthier than rare catastrophic ones. 3. The barbell strategy combines extremes while avoiding the middle. Put most of your resources into the very safe and a small slice into the very risky with bounded downside. Avoid the moderate-risk middle, where you face most of the downside without the upside. This applies to portfolios, careers, and projects. 4. Optionality beats prediction. Having many small options with capped losses and uncapped gains is more powerful than forecasting the future correctly. Tinkerers, traders, and evolution all exploit asymmetric payoffs without needing to know what will work. 5. Via negativa — improvement by subtraction. Often the best move is to remove harmful things rather than add supposedly helpful ones. Skip the bad food, the toxic person, the fragile dependency. Subtraction is more reliable than addition because it is harder to be confidently wrong about it. 6. Skin in the game is the foundation of honest systems. Decision-makers who do not personally bear the consequences of their decisions transfer fragility to others. Bankers with bonuses but no clawbacks, pundits with no track record, and bureaucrats with immunity all illustrate the pattern. 7. Iatrogenics — harm done by the would-be healer — is everywhere. Doctors, central banks, regulators, and managers regularly cause net harm by intervening in systems they do not understand. The intervention bias overweights visible action and underweights invisible costs. 8. Lindy effect: the longer something has lasted, the longer it will likely last. For non-perishable things — books, technologies, institutions — expected remaining life rises with age. This is a heuristic for choosing what to trust, what to read, and what to copy in a noisy world.