[elefanten]

I'd add a corollary along the lines of:

"We also have extreme difficulty parsing what may be caused by (epi)genetics and what may be caused by differences in experiences or development"

Not only do we lack fine-grained data with which to make that comparison, we probably lack the epistemological capabilities and experimental infrastructure to even attempt gathering such data.

This doesn't have to be mutually exclusive with the argument(s) that population-scale difference in distribution of [whatever] have been experimentally observed. It's just that for most conceivable [whatever] with some kind of relevant social impact, we still can't assign causality with high confidence to the actual (epi)genetics themselves.