On Sat, Nov 29, 2003 at 02:15:54PM -0500, Robin Hanson wrote:
> That's pretty weak evidence, compared to the vast experience each of us has
> in using cleverness to better get along in the world.

Using ourselves as samples suffers from a selection bias. We don't live in
a time and place where smart people are being severely persecuted, but if
we did, we wouldn't have the opportunity to have this discussion.

I'm not saying that intelligence is not useful, just that its
social costs can help explain why we're not smarter. Of course
intelligence seems extremely useful, which is what makes our dumbness
puzzling. Is your position that the other known costs of intelligence
(more energy use, more difficult births, what else?) are sufficient to
explain this?

> If you can fake being dumb in particular areas, why can't you fake being
> dumb in all areas?

If you have to consistently fake being dumb in all areas, that greatly
lessens the benefits of being intelligent, since you can't apply it to
solve problems any time somebody can see either your efforts or the
results.

> Perhaps sufficiently general intelligences can fake anything, but if so
> humans do not seem to be sufficiently general.  Humans give off all sorts
> of subconscious clues about whether they are faking, and they are adept at
> detecting those clues in others.  Humans cannot lie with impunity.

I was saying that *if* we had sufficient general intelligence, we would be
able to fake being dumb in particular areas, therefore it is not an
evolutionary option to have high general intelligence and still be
convincingly dumb in areas where being seen as dumb is beneficial. The
fact that we don't actually have sufficient general intelligence to
flawlessly fake being dumb doesn't affect the logic of my argument.

To make sure we're not misunderstanding each other, you're says that
having higher intelligence is not so bad, because one can either (a) still
be dumb in certain areas, or (b) fake being dumb in all areas, right? I'm
arguing that (a) is not an equilibrium, and (b) reduces the benefits of
intelligence too much.

Reply via email to