My definition of intelligence is that it’s the ability to reason and learn. And the learning refers to a specific type of learning, learning by reasoning, or in other words learning by figuring stuff out for yourself. There are at least two other significant types of learning that I have identified which are not intelligence: they are learning by memorization, or in other words rote learning, and learning by mimicking (which is how small children learn to speak and which is why they are better at learning foreign languages fluently than adults are).
In addition to reasoning, there are other mental abilities which are not intelligence. They include ability at both short and long term memorization, and computational abilities. There are probably a lot of other mental abilities which I haven’t mentioned, but what they all have in common is that they are not intelligence.
It’s likely that some of these non-intelligence abilities are correlated with intelligence for biological reasons. A larger size brain has been shown to be correlated with g, and it make sense to me that a bigger brain means more brain-power, which might result in better reasoning ability as well as being better at other mental abilities.
Mentioning g, it’s not clear to me if g is the same thing as what I define here as intelligence. Anyone have any opinions on that?
Idiot savants are autistic people who are not that intelligent yet have some amazing mental abilities such as super memorization skills or computational abilities.
The game of chess doesn’t require as much intelligence as you think it does because it’s primarily about memorizing openings and end game techniques, and being able to compute board positions many moves out into the future. The fact that the chess program Fritz can beat any human chess player demonstrates how chess is not about intelligence, because computers are not intelligent. Computers compute. Computers can compute chess positions many moves in advance better than humans can, and thus they can beat the best human chess players, even Gary Kasparov.
In contrast, the much more practical skill of teaching a computer how to understand language is something we have not been successful at doing. Language comprehension requires reasoning ability which computers lack.
Discovering calculus required a very high level of reasoning ability, but once smart people developed it it, it’s possible to train not-so-bright people to do it, as long as you can overcome the problem of keeping them motivated to learn a subject which they would normally find exceedingly boring and pointless.