> The larger the group, the more complex the system, the longer the system has been in place, the worst the results are. It seems to me that these are not mono

> The larger the group, the more complex the system, the longer the system has been in place, the worst the results are. It seems to me that these are not monotonically true. If I have a handful of competent allies in the wilderness or in hostile territory, I'm better off. Language is a complex system and an infrastructure; more complex language has costs vs. an exceedingly simple language, but increased complexity is very frequently a big net positive. You might argue that size, age, and complexity of systems are correlated with poor results, but you can not use that to exclude the possibility that maximum results can be obtained by some large, complex (undiscovered) system. It's reasonable to argue that the direction some country should move in today is towards libertarianism; maybe that is the direction to a useful local maximum. But you're arguing for a very strong and simple heuristic that would allow you to dismiss without evaluation any large, complex system; that seems to me to be a critical error.