More about the politics of expertise.

Think of three points on an ideological triangle:

1. Point L, where you believe that markets are effective at processing information and solving problems. This position is to take a radically pro-market view, and to let markets fix their own failures. 2. Point C, where you believe that tradition incorporates the evolved use of information to solve problems. This position is to be very cautious about overthrowing existing institutional arrangements.

3. Point P, where you believe that expert technocrats should be in charge. You are comfortable with throwing out tradition and markets in order to cede power to experts.

I'll apply this framework to some people and to some issues.

Tyler Cowen is a mix of L and C. He is too C to go along completely with the brute logic of Robin Hanson or the bumptious enthusiasm of Bryan Caplan. But Barack Obama has managed to offend Tyler on both the L and C points, as you can see in the latter part of this interview with Matt Welch. I also find myself somewhere between L and C.

Expert technocrats have always been in charge to some extent. A C, then, would acknowledge this as an evolved use of information and be cautious about overthrowing existing institutional arrangements. Cs end up being moderates sandwiched by the L and P ideological extremes.

The ideological blunder of progressives is lack of grounding in reality. They arrogantly assume that there is sufficient expertise and that experts exist who could be competent technocrats - people like themselves. This is false, and it takes great energy - a lot of humming with their fingers in their ears - to maintain the fantasy.

The ideological blunder of libertarians is too much grounding in reality. They are mired in it and so some problems can't be worked. They rule such problems out of bounds, not the proper concern of society. In hill climbing search terms they readily find local optima and settle there, never seeking higher points.

But in both cases there is a further dynamic when there are multiple concurrent versions running. It's not technocracy, it's technocracies. Each technocracy may be an inept bungler, but they each bungle in different ways and so can teach one another and punch above their weight. Not all of course. Some are too stupid to improve.

Similarly, multiple libertarian societies will find different optima and be able to evaluate them against one another. It's like having multiple agents working the hill climbing problem, each starting from a different point and having varied capabilities. When they communicate the findings of each can be of use to the others.

If I was an alien having some drinks with my buddies and betting on the puny human contest I'd put most of my money on the Ls since they have a better long term strategy, but there may be some short term winnings for one of the lucky Ps. They don't know what they are doing but luck doesn't care about that. Just don't expect them to do it repeatedly or consistently.

All bets are off if the dynamic changes and some society comes to dominate everywhere. Sell short if that happens.