Résumé de Adapt

Tim Harford’s *Adapt: Why Success Always Starts with Failure* argues that trial-and-error, not top-down planning, is the engine of progress in business, science, and government. Drawing on evolutionary theory, the British economist and Financial Times columnist outlines three principles for successful adaptation: try new things in small, low-cost experiments; make failures survivable rather than catastrophic; and learn from mistakes honestly instead of denying them. Harford builds the book around vivid cases. He describes how the U.S. military’s counterinsurgency manual, championed by General David Petraeus, emerged from decentralized experimentation in Iraq rather than central command. He recounts the 2008 financial crisis as a failure of institutions that hid risk instead of adapting to it, and contrasts the doomed Soviet planned economy with the messy resilience of market systems. Other chapters examine Google’s chaotic “20 percent time,” the race to develop a malaria vaccine, and the design of the Toyota Production System, whose famous andon cord lets any worker halt the line to expose problems. A key recurring idea is the “Palchinsky principle,” named after a Soviet engineer who was executed for criticizing grand industrial schemes. Harford uses it to argue that reformers must seek variation, survive failure, and select what works, rather than impose elegant blueprints. He also explores why individuals resist adapting, citing cognitive biases and organizational incentives that punish admitting error. The result is a defense of pragmatic, evolutionary problem-solving over heroic central planning, published in 2011 and widely cited in debates about economics, management, and public policy.