Rationality by Other Means

by DataPacRat 2 min read26th Dec 20126 comments


Political violence is a terrible thing - but, sometimes, not quite as terrible as the alternative. I recently commented that a post focusing on such things might be worthwhile, and since the voting has been positive, here we are.

From a Bayesian/rationalist/winningest perspective, if there is a more-than-minuscule threat of political violence in your area, how should you go about figuring out the best course of action? What criteria should you apply? How do you figure out which group(s), if any, to try to support? How do you determine what the risk of political violence actually is? When the law says rebellion is illegal, that preparing to rebel is illegal, that discussing rebellion even in theory is illegal, when should you obey the law, and when shouldn't you? Which lessons from HPMoR might apply? What reference books on war, game-theory, and history are good to have read beforehand? In the extreme case... where do you draw the line between choosing to pull a trigger, or not?

I've cobbled together /a/ set of answers to such questions, based on what I've learned so far of economics, politics, human nature, and various bits of evidence. However, I peg my confidence-levels of at least some of those answers as being low enough that I could be easily persuaded to change my mind, especially by the well-argued points that tend to crop up around here.

As just one starting point, 'freedom', 'liberty', and 'justice' (often with a halo effect list of other virtues) are often considered some of the highest values to rally around, and to fight for. And it's certainly quite attractive to be able to say you're fighting for them - but I have a suspicion, and a rather strong one, that that such reasoning may be closer to post-hoc rationalizations than is generally considered.

An old saying goes, 'Amateurs study weapons; professionals study logistics'. For a number of reasons, economies tend to do best when as much competition as possible is done within them. Since a natural tendency of individuals and groups who achieve economic success is to use their leverage to push for even greater success relative to others, even if doing so causes others to pay externalized costs, it requires special efforts to promote the efforts of the 'little guy' against entrenched interests to allow new entrants into a market to have any chance of competing successfully in it. Thus, those economies which tend to perform best tend to be those with the most focus on individual rights, of allowing small-scale enterprises to successfully use the legal system, of allowing as many individuals as possible to put their hand to taking advantage of whatever opportunities they find. As those states with better economies tend to win wars against those with worse ones, it is easy to observe that those states with a greater focus on liberty tend to beat those with a lesser one... and thus quite natural to conclude that those values are, in and of themselves, the values worth fighting for.

A similar line of argument could be made about the important virtues of the democratic process not being what's commonly said about it.


Even such a rudimentary analysis of the principles many people think are worth fighting for puts the whole prospect of political violence in new light. Even if the analysis itself is wrong, it seems to open up uncommon avenues of thought, which could lead to a more accurate estimation of large-scale conflict, of which side is more likely to win, what reasons will be proclaimed for the victory, and what the most valuable efforts might be to nudge the odds one way or the other.

Then again, I could be wrong. In which case, I'd prefer to know as soon as possible.