Milton Friedman was a magnificent economist, and I’ll defer to other economists to sing his praises. But it’s worth noting that, besides being an evangelical for free-markets, he was also a proponent of the rational-agent model. Those two facts aren’t a coincidence. Friedman firmly believed that, when left alone, people will intelligently act in their own best interest, and that the market will coordinate their actions to produce outcomes beneficial for all. The wisdom of crowds depends upon the rational wisdom of the individual.

Of course, prospect theory, behavioral economics, neuroscience, and neuroeconomics have dismantled many of the naive psychological assumptions that Friedman subscribed to. The more we know about the brain the less rational we seem. Herbert Simon was prophetic.

But Friedman wasn’t interested in our irrationality. He repeatedly argued that, because economists can make reasonably accurate predictions about the value of the dollar, or the rate of unemployment, the erroneous assumption of rationality is irrelevant. The powerful calculus of economics cancels out the bewildering psychology of its subjects. (Friedman summarized this philosophy in his famous “F-twist” argument for “positive economics,” which consisted of two separate claims: 1. Theories should be judged by the accuracy of their predictions; 2. Theories should not be judged by the accuracy of their assumptions.)

Tyler Cowen defends Friedman’s positivism by referencing Quine, since Quine also didn’t think much of theoretical assumptions. (See “The Two Dogmas of Empiricism“.) But Quine’s model of knowledge depended upon the interconnectivity of our facts. When Quine famously wrote that “the unit of empirical significance is the whole of science” he was pointing out that without the web of related knowledge, all empirical statements are pretty meaningless.

This was Friedman’s mistake: he wasn’t enough of a Quinean. If he was, then he would have wanted to draw connections between economics and psychology. He would have wanted to see how the facts of economics could be entangled with our neural anatomy. Quine isn’t an excuse for not testing your assumptions, or turning your assumptions into an unfalsifiable dogma; Quine simply reminds us that whenever possible we should try to connect our empirical observations together. This means that economists should always be trying to ground their mathematical predictions in an accurate biological model of human decision-making. Of course, when they do that, the assumption of rationality that Friedman believed in is no longer valid.