In his lunch with Philip Tetlock, Robert Armstrong concludes: “Tetlock’s demonstration that strong commitment to theory or ideology makes poor forecasters —and that even best forecasts come to resemble guesswork when they reach more than a few years into the future — raises an unsettling possibility. It suggests that human affairs are mostly random and intractable. Incremental gains in foresight are possible, but there is no deep order to life…” But it’s hard to look at our amazing self-organizing universe and conclude there is no deeper order–quite the contrary.
The better lesson is that the order is hopelessly complex, that this renders central planning typical of government highly inefficient, and that proponents of government intervention are recklessly over confident of its success. Evolution and free enterprise, which feel their way in the dark by running millions of failed experiments to stumble upon a small number of unexpected truths, is far more effective. The robustness of their discovered improvements should be respected. And criticism of them should be viewed skeptically–more likely, the result of incomplete and mistaken analysis rather than true insight.
A view of the world that appreciates how little we understand recognizes that randomness makes the risk of finding improvements enormous. As such, enormous rewards are needed to make risk-taking economical. I make these points in my upcoming book, The Upside of Inequality: How Good Intentions Undermine the Middle Class (September 13).