generalization is politically incorrect: a kindergarten teacher might generalise from "all the boys in my class are unable to sit still" to "all boys these days are unable to sit still". result: boys are going to be treated differently to girls.

i think there's a distinction to be made between generalization, which is a variety of inference from a limited sample to a wide population (with all the murky quagmire action that entails), and abstraction, which is informed ignorance.

abstraction is the act of, having assembled a certain set of observations of/information about the world, going on to selectively put information aside, assembling a simple model of the world to be used for future decision making.

between us and the world, we have a wide range of filters protecting us from the flood of sense perceptions we'd otherwise be subjected to. we can be semi-consciously aware of a funny smell, for instance, for some time before it registers consciously.

abstraction is a similar technique; by keeping our internal model of the universe pared-back, focused on the essential elements, we improve our ability to reason in a timely fashion about the world; we're enabled to reach sensible conclusions about whether going to lunch would be a good idea before the universe ends in heat death and lunch is no longer an option.

so, ignoring the ethical aspects of adopting the principle that "all boys these days can't sit still", it's possible to see such a position as both a generalisation of an observation of a limited sample set to a prediction about a larger population AND as an abstraction that facilitates timely reasoning about the chaotic real world.

that's by no means a justification if undesirable consequences flow from acts following from that mental attitude, but it's important to understand the mechanisms by which people (me included obviously) reach dodgy points of view, and why our brains might be arranged in such a way as to take us there.

fear dodginess in all its forms!