A few weeks ago, Paul Krugman set off a debate by claiming that liberal economists could do a very good job at explaining what conservative economists think, but the conservatives just don’t understand the liberals. Regardless of the empirical truth of that statement, the idea is an important one: when there is a respectable disagreement (as opposed to one where the other side are just obvious crackpots), and important skill is to be able to put yourself in the mind of those with whom you disagree. Conservative economist Bryan Caplan formalized the notion by invoking the idea of a Turing Test: could a liberal/conservative do such a good job at stating conservative/liberal beliefs that an outsider couldn’t tell they were the real thing? Ilya Somin, a libertarian, actually took up the challenge, and made a good-faith effort to simulate a liberal defending their core beliefs. I actually thought he did okay, but as he himself admitted, his “liberal” sometimes seemed to be more concerned with disputing libertarianism than making a positive case. Playing someone else is hard!