Merry Andrew wrote:What, exactly. is intelligence? I mean, going beyond the standard dictionary definition, how would you define it? I think your thesis -- which is certainly arguable -- depends on that definition.
Here is a short discussion, and definition, of intelligence.
If we are to regard intelligence as an ability then my definition would be this:
intelligence is the ability to recognise a standard set of recognisable objects. The term ability is not helpful however, because it introduces an unknown ('ability' - see bottom).
Rather, we could define intelligence in this way:
Intelligence is the numerical difference between a set of standard recognisable objects and a set of objects recognised as belonging to that set. What is an object in this context? The objects are taken from the world that the creature that is being tested inhabits. For example, visual, linguistic etc, for the human. The test that is constructed for IQ also includes objects that affect all the objects in the test, such as time constraints. The standard set of recognisable objects needs to be constructed carefully, and is different for each species. There are virtually no cross-species objects.*
There is no need to construe intelligence as an 'ability'. If there is confusion over what IQ is testing, then the confusion is really over the nature of the objects chosen for the test and we should not search for a mythical 'ability' to put the confusion to rights. The test for IQ should be clear enough and on that count need not introduce vague notions such as 'ability'. This is why, given the method for testing, intelligence is adequately, if not fully described as a number. Any other associations made with the 'size' of that number are socially driven by practical considerations or vanity.
*The introduction of cross-species objects to the intelligence test makes it unrecognisable as a test.