Society is more akin to an iterated prisoner's dilemma: the same game, but played lots of times. In that situation, the ideal strategy is actually to cooperate.
Cooperation only happens when you iterate every time in a one-on-one fashion with the same players over and over again. For example: the relationship with your local grocer, who has a reputation to build. But it is the wrong metaphor for most real issues in social policy. Pollution is a tragedy of the commons problem.
This is a more important point that I think most people realize. The classic "flaw" in Libertarian thinking (both among self-described Libertarians and those who criticize it) is neglecting higher-order effects. The prisoner's dilemma is a perfect example of this. The game played once is very different from the game played repeatedly. You'll find the same sort of differences in numerous other game-theoretic situations.
Unfortunately, most supporters of Libertarian policies only get to the first-order implementation and then stop and are rightly criticized for the limitations and flaws that result. All of the complexities of higher-order effects are neglected, but those are the parts that actually make the theory workable in the real world.
I think you mistook my comment as an argument against Libertarianism - it was not intended to be such. My point was that people oversimplify both the implementation and the analysis of Libertarian principles. A system with many, many free actors can be amazingly complex, and what I'm suggesting is that for people to see the value of the system they need to recognize and embrace that complexity.
For example, the basic principles of a free market (e.g., supply and demand) are easy to state but result in complex transactions. When things don't "work out" for someone in the real work and they raise an argument against "free markets" (as in the current housing bubble), we need to be ready to point out all the complex ways that the free market worked correctly but where the non-free parts distorted it (government policies encouraging loans to unqualified applicants).
In complex systems with non-linear behavior, we need to be very careful that we understand the full context of an event or we won’t be able to extrapolate the results. That’s why first-order approximations fail so often in mathematics and in real life.
Excellent point. You're right, a basic humility about the human ability to understand and control complicated systems is at the root of libertarian principles.
I think it's somewhat absurd that Obama and Bush even pretend they can do anything about the financial crisis.
