I see what they are getting at, and I understand it just fine, but the biological textbook definition of heritability [h = Var(genetic) / Var(phenotypic)] is almost worthless even in the scientific context in which it is used — because it doesn’t answer any important questions. It is explicitly designed to avoid those, in fact.
The whole area has been so corrupted by attempting to route around even thinking about possible racial differences that it trashed itself.

