One important factor that drove the evolution of psychological sexual dimorphism is the difference in the minimum obligatory parental investment between the sexes. (This is the great insight of Trivers, 1972). Men, as the rather coarse saying has it, ‘can leave a bed unmade’. That is, a man need only invest a few minutes of effort and some sperm to produce a child. Women, on the contrary, must invest as much as men plus 9 months of pregnancy and, given the absence of baby formula on the African savannah, several months or years of breast-feeding. Moreover, before the advent of modern medicine, childbirth was very dangerous so a woman quite literally risked her life to have children. The minimum obligatory investment for men and women, then, is radically different, so we should expect the evolution of a dimorphic sexual psychology reflecting, as Trivers put it, 'female choice and male competition'. (This is, obviously, a crude simplification). And, not surprisingly, we have a mountain of empirical evidence that confirms this expectation.
There is a lot one can say about this theory, and the above sketch certainly does not do it justice or acknowledge the complexities and uncertainties of the empirical data. But a story I saw in a newspaper recently made me think of one of its features, namely, that a man could always be doing better. From the perspective of a man’s genes, women are an extremely valuable and limiting resource. This may seem a bit weird, so let me explain. There are (of course) a finite number of fertile women alive at any given time, and, since a man has such a low minimum parental investment, he could, in principle, impregnate tens of thousands of them. Women, on the other hand, have to carry and give birth to all their offspring, so the total number of children each woman could have in a lifetime is severely limited by comparison. Men have the potential to sire several orders of magnitude more offspring than women, and as a result there is an oversupply of willing males. (One interesting consequence is that there is a much greater variance in male reproductive success, which produces much greater variance in males in a whole range of traits. The variance in male mathematics grades, for example, is substantially higher than that of women).
In any case, the story that got me thinking about this again concerns one Desmond Hatchett (pictured above). Hatchett, an American man from Tennessee, is only 29 years old but, amazingly, has fathered 20 children. Not quite Ismial the Bloodthirsty (who reportedly sired at least 888 children) or Genghis Khan (who is the likely ancestor [pdf] of ~8% of Central Asian men, and ~0.5% of all men worldwide), but evolutionary speaking, not bad at all.
Trivers, R. (1972) "Parental investment and sexual selection" in Campbell, B. (ed.), Sexual Selection and the Descent of Man.