You are right. My thinking was indeed imprecise here. If we assume that there exists a set of skills such that each skill, if practiced consistently, prevents one from having a specific set of irrational beliefs, then we can impose a partial order on sets of beliefs by observing which set of skills is implied to be absent by each set of beliefs. This partial order can be seen as a ranking of irrationality of different sets of beliefs, and the set of skills shared by a group of people places a lower bound with respect to the partial order, which can then be metaphorically called a "waterline."
Of course, the crucial assumption here is that it is possible for humans to acquire a set of reasoning skills so thoroughly and reliably that they will actually apply them to all issues, no matter what. I don't think this is possible, and with this in mind, I still don't think the "waterline" concept is useful. If anything, it's dangerous because people may fall into the trap of thinking that they are above a certain waterline, whereas in reality, there are issues where due to all kinds of biases even the very basic skills are failing them.
Of course, the crucial assumption here is that it is possible for humans to acquire a set of reasoning skills so thoroughly and reliably that they will actually apply them to all issues, no matter what. I don't think this is possible, and with this in mind, I still don't think the "waterline" concept is useful.
Fallacy of gray. Even if there are no actual magical superrationalists, clearly some people are better skilled than others, and a group of people would behave differently depending on this level.
This post grew out of a very long discussion with the New York Less Wrong meetup group. The question was, should a group dedicated to rationality be explicitly atheist? Or should it make an effort to be respectful to theists in order to make them feel welcome and spread rationality farther? We argued for a long time. The pro-atheism camp said that, given that religion is so overwhelmingly wrong on the merits, we shouldn't allow it any special pleading -- it's just as wrong as any other wrong belief, and we'd lose our value as a rationalist group if we began to put status above truth. The anti-atheism group said that, while that may be true, it's going to doom us to be a group exclusively for eccentric nerds, and we need to develop broad appeal, even if that's hard and requires us to leave our comfort zone.
Things got abstract very fast; my take was that we need to get back to practicalities. Different attitudes to religion have different effects on different types of people; we need to optimize for desired effects and accept what tradeoffs we must. We can't appeal equally to everyone. So I came up with a sort of typology.
The Four New Members
Annie