The "It must be “Let’s all beat up Evo Psych” Day!" article seemed very convincing when I read it, but now that I had some time to think about it, it seems much less convincing. Please tell me whether I am wrong...

The article essentially says that there are no "male genes" and "female genes", because everyone in every generation gets their genes from their father and their mother. So even when there is an evolutionary pressure only on one sex to evolve some skill, the other sex gets the skill automatically. When we find an evolutionary explanation why men have genes for some skill or trait, at the same time we found an explanation why their daughters have the same genes and the same skill or trait, too. And vice versa, when we find an explanation why women have genes for some skill or trait, we also have an explanation why their sons have the same genes and the same skill or trait, too.

A trait which would be different between sexes, would not only need genes benefiting one sex, but also some special genes to actively turn it off for the other sex. Otherwise, both sexes would have it. Even under assumption that a trait is helpful for one sex and irrelevant for the other sex, our default expectation should be to find the same genes and the same trait in both sexes. To develop otherwise, the trait would have to be actively harmful to the other sex.

For example it is useful for a man to have a penis, and it would be harmful for a woman; this is why both sexes remain different in this aspect. But we cannot use the same logic for things like color perception. Even under unrealistically generous assumptions that historically women always gathered plants, and men never did, and the color perception is useful for discerning plants, and completely useless for anything else... still, we should expect the color perception to be the same for both men and women. To expect a different result we would have to claim that color perception is actively harmful for men, which obviously is not the case.

Did I understand it correctly?

If yes, then...

Reality check: According to Wikipedia, 5-10% of men, but less than 1% of women have some form of color blindness. What?! We have just proved that this should not happen, unless men get some huge evolutionary penalty for not being colorblind, which is obviously not the case. But this information about color blindness is not some evo-psych story; it is measured data. So how is that possible?

Oh yes, there is this pesky little detail that men and women have the same chromosomes, except for the sex chromosome. Women have XX, men have XY. And the X chromosome happens to contain some data not directly related to reproduction, for example the genes for the color perception. Women get two version, men get one. It means that when something goes wrong with the color-perception genes, women have a backup copy, and men don't. Therefore the difference.

As far as I know (but I don't really know much about this), the Y chromosome does not contain much useful information besides specifying the male sex. So this mechanism alone could explain why men can be on average worse than women in some tasks (if the genes necessary for successfully doing the task happen to be located on the chromosome X), but not the other way around! -- By the way, which specific gene happens to be on which chromosome, that is partially a historical coincidence. There is no logical reason why the color-perception genes must be on the chromosome X. It just happened. It could have been on some other chromosome; and perhaps in some other species, it is.

But there is a more general problem with our assumption that men and women, as members of the same species sharing the same genes, must be the same in everything except reproduction (with some disadvantage for men if the genes are on the sex chromosome) -- the genes do not operate independently on each other. The results of a gene can be influenced by the internal environment, which in turn can be influenced by other genes. Different level of sexual hormones can make the same genes produce somewhat different results. Men and women do have different levels of sexual hormones, even during prenatal development. So even the same genes can work a bit differently for men and for women.

Please note that even a small difference can be noticed by people, because we observe each other a lot, for various reasons; we are a social species. You don't need to have all men go completely blind, just to notice that there is perhaps some difference between visions of men and women. A few more percent of partially color-blind men is enough for people to eventually notice. (A possible alternative explanation, that Patriarchy spread the myth of inferior male color perception for its evil purposes, and it just randomly happened to be true, does not make a lot of sense to me.) Also, we compare humans with humans, not with other species. Even if men and women both have verbal skills tremendously superior to other species, people still notice that women have these skills somewhat better than men.

So my alternative explanation is that the same gene can produce slightly different (but still observable) results in the male and in the female body, because of a presence of sex hormones. So if there is a greater evolutionary pressure on one sex to develop some trait, as a result the whole humanity may get a gene for this trait optimized for the given sex. Both men and women will have it, and it will do the same thing; it will just do the thing a little bit better in presence of hormones of one sex than of the other sex. It could be a difference between 100% and 99% efficiency. Just a tiny difference in verbal skills, or math skills, or navigation skills, or whatever. But in the context where these skills are critical, people will notice.

Of course this does not make every evo-psych hypothesis automatically true. But neither does it make a hypothesis automatically false just because it explained a difference between sexes by greater evolutionary pressure on one of them with regard to the given trait... as the linked article seems to suggest.