A better case against g
I used to think Spearman’s Law of Diminishing Returns (SLODR) was a good argument against the existence, or relevance, of g. That is where g-loading drops off at the higher ranges of intelligence.
I no longer think that it is a good argument against g, because it might just be proportional. If a baby’s arm is 30cm, and the foot is 10cm, then the difference is 20cm. And if an adult’s arm is 75cm and foot 25cm, then the difference is 50cm. So the difference is larger, but only because everything is larger. It is proportional, the difference grew 2.5 times along with the foot and the arm. Maybe this is how SLODR works too, and if so, it is not a good case against g.
But it doesn’t matter, because there is a better case.
Powerlifters compete in 3 lifts, the bench press, the squat and the deadlift. I am sure you could find a general factor of strength if you checked the stats, but they measure all three lifts anyway. Why? Because each lift uses, and thus measures, different muscles.
Similarly with IQ tests. The best tests use several different types of batteries. Why? Because the brain has different ‘muscles’, so to speak.
Humans are genetically coded to have roughly equal abilities in different types of intelligence. Similarly, humans are genetically coded to have 2 legs that are roughly equal in length.
It is not absolutely certain their legs will be exactly the same, but most humans have legs that are very close to the same. Is there a general factor of leg length? Technically no, there is no universal law that states that all legs must have a general factor of length, and that all brains must have a general factor of intelligence.
g might be useful for some things, like big studies on normal populations where there is a need for short tests, but it is not real in the strict sense.