Aptitude, genius, intelligence, etc: none of these things can be
defined explicitly and unambiguously enough for measurement. What
can be measured is skill at a particular set of tasks, and it seems
to be a pretty good hack to assume that skill at manipulating
language and mathematics is a pretty good predictor of success
within our current crop of academic institutions.
But brains have many, many other skills besides the abilities to
manipulate language and mathematics, and IMHO these other skills
(collectively: every human talent and skill *other* than language
and mathematics) have made heavy contributions to the achievements
of the people whom *I* admire as geniuses. These other skills are not
measured on the SAT.
The dream of measuring intelligence is the dream of measuring
generalizably-extrinsic Skill. I don't believe such a thing exists.
We are each a bundle of different skills and aptitudes working together,
including some freakish individual skills each of us has that are
the result of a unique configuration of neurocomputational processes,
and the more closely I study such things, the more laughable the
notion of numerically measuring "general intelligence" as a single
scalar quantity seems. It will be even *more* laughable in the era
of Jupiter brains. Measuring general intelligence is like evaluating
an entire economy, except that this economy is all barter, so we
can't even do a summation of money prices of things. (And even if we
could, the analogy to that in our brains might bear no relationship
to what we think we mean by "general intelligence".)
-- Eric Watt Forste ++ arkuat@pobox.com ++ http://www.pobox.com/~arkuat/