The rest of 2021 is shaping up to be a sequel to a movie that never should have been made in the first place: from the people who brought you Pandemic 2020 comes the smash hit sequel Pandemic 2021: Delta Variant. New CDC masking “guidelines” make me wonder about further restrictions, all from “the experts” who are following “the science.” There are limits to what “the experts” can know and what “the science” can teach us: they can estimate probabilities arising from different courses of action, but they cannot tell us exactly what to do. An expert can say “This model predicts that with 90% of people wearing masks all the time and maintaining ‘social distancing,’ Covid transmission falls by 90%.” The expert cannot say “Therefore, the government should mandate masking and distancing” with smuggling in a bunch of auxiliary assumptions about what society’s goals should be, who should choose, whose preferences matter, and which tradeoffs are morally significant. I know myself, my family, and my friends pretty well, and I think we’re pretty well-equipped to make choices that balance our physical, mental, emotional, and spiritual health while accounting for our duties as good neighbors.

Here’s a nice illustration from Strange Planet that I’ve used on economics quizzes. One being says to another who is salting his food, “Why sprinkle minerals before ingesting?” The second says, “I enjoy sprinkled minerals.” The first says, “But this could decrease your final revolution count.” The first says, “Perhaps I prefer fewer revolutions and more minerals.” The principle: trade-offs are everywhere. The first being is free to disapprove, and his expertise might be informative. It is not, however, decisive.

It seems reasonable that before we cede important social decisions to experts, we should consider the very real possibility that the articulated and centralized wisdom of experts isn’t that much better than the unarticulated and decentralized wisdom of the masses reflected in prices and social conventions. Here are five recommendations for books you should have on your nightstand if you want to learn more.

1. Roger Koppl, Expert Failure. Koppl has done a lot of very important work on ordered hierarchies of expertise in general and the institutional limitations of forensic science in particular. Expert Failure is an explanation of the institutions and organizations in the “market” for expertise and shows how the financial, reputational, and epistemic incentives induce (wait for it) Expert Failure. If nothing else, we should remember that experts respond to incentives, too, and if you understand why a student puts a lot of time and effort into studying for an exam that’s worth 30% of the grade and little time and effort into a homework assignment that’s worth 0.1% of the grade, then you have a pretty good idea of how replacing the dispersed knowledge of the many with the articulated knowledge of the few can go sideways.

2. Philip Tetlock, Expert Political Judgment: How Good Is It? How Can We Know? This book created quite a splash when Tetlock explained that expert political judgment really isn’t that good. Princeton University Press’s website describes the book’s takeaway point succinctly: “Classifying thinking styles using Isaiah Berlin’s prototypes fo the fox and the hedgehog, Tetlock contends that the fox–the thinker who knows many little things, draws from an eclectic array of traditions, and is better able to improvise in response to changing events–is more successful in predicting the future than the hedgehog, who knows one big thing, toils devotedly within one tradition, and imposes formulaic solutions on ill-defined problems.” That last descriptor is really important when we are evaluating pandemic response. What are mandates, lockdowns, price controls, and barriers to innovation but “formulaic solutions?”

3. Philip Tetlock and Dan Gardner, Superforecasting: The Art and Science of Prediction. Bryan Caplan has described himself as a “fundamentalist Tetlockian” in light of Tetlock’s two books on this list. Tetlock and Gardner look at what it takes to predict well, and they buttress Tetlock’s conclusion from Expert Political Judgment: hedgehogs who gather a lot of different bits of evidence from a bunch of different places, use it to inform their estimates of different probabilities, and then assess how well they are doing. While Tetlock focuses on the best of the best, it’s another useful exploration of the importance of relying on decentralized knowledge.

4. F.A. Hayek, The Counter-Revolution of Science: Studies on the Abuse of Reason. Hayek’s analysis of “the abuse and decline of reason” was one of his important intellectual projects. Beginning with his 1937 paper “Economics and Knowledge” and going through his 1945 paper “The Use of Knowledge in Society” and his 1948 collection Individualism and Economic Order, Hayek explored how there is a lot more to “social knowledge” than what can properly be called “scientific.” The expert planner can know a lot of very important things, but as Hayek emphasizes over and over again, knowledge about “the particular circumstances of time and place” does not confront the planner as data. In the face of repeated appeals to expertise and science–both of which, again, inform even when they cannot decide–Hayek’s lessons are indispensable. 

5. Thomas Sowell, Knowledge and Decisions. I recently listened to the audiobook of Knowledge and Decisions twice. It’s a masterpiece that extends Hayek’s analysis and that shows, once again, how a lot of essential social knowledge is not “scientific” in the strictest sense. Throughout the book, Sowell emphasizes the importance of incentives and suggests that we should be wary of people who wish to control others but who don’t suffer serious losses when they are wrong. Sowell’s emphasis on tradeoffs and decentralized, non-”scientific” knowledge is a crucial contribution to how we should understand experts’ role in public policy.

By all means, we should listen to the experts and consider the science. They are undoubtedly informative, but once again, their opinions and conclusions are not decisive. The argument for decentralized rather than centralized pandemic responses is not simply a matter of rights and obligations, though these are important. When we turn experts into masters rather than advisers, we throw away a lot of important knowledge. Experts definitely have their place, but we have to remember that they respond to incentives (just like everybody else) and have their own epistemic limitations (just like everybody else). If you find yourself with a lot of extra time on your hands in this second year of “two weeks to flatten the infection curve,” it might be a good idea to read up on when experts get it right, when they get it wrong, and why “trust the experts” and “follow the science” aren’t as simple as they might appear at first.