A well-known hypothesis in evolutionary theory is the sexy-son hypothesis (Weatherhead & Robertson 1979). It states that a female will sexually prefer males who are sexy, because this preference ensures sexier male offspring, which in turn will propagate the female’s genes more effectively.
An intriguing circularity
The hypothesis is clearly circular: It explains female sexual preference as a preference for sexy males (whose sons will be more likely to be sexy). But whatever males conform to female preferences are by definition sexy. So the hypothesis can be restated as follows:
Females prefer males that females prefer.
Perhaps this circularity should not be seen as a logical flaw, but as reflecting the causal loops inherent to these processes. In particular, females will prefer males that other females prefer.
Objective-trait preference versus female-judgment-copying
Any objective-trait preference (say a female preference for dark hair) is consistent with the sexy-son hypothesis, as long as the preference is widespread in the female population. (Dark hair is an objective physical trait. If most females find dark hair sexy, then finding dark hair sexy will be evolutionarily advantageous to a particular female.)
However, the sexy-son hypothesis is also consistent with a more generalized female preference mechanism: beyond any innate preferences for particular objective traits, the female may have a preference for any trait at all that she perceives to be preferred by other females. This is the female-judgment-copying variant of the sexy-son hypothesis.
At the level of group selection, such an abstract and flexible sexual preference mechanism may rapidly promote the genes of a group of males who suddenly achieved some cultural success (e.g. developed a tool or technique), because it would enable female preferences to self-organize and unify within a single generation in favor of the males in question.
It is imaginable that an initial nonsexual preference becomes sexualized in a self-organized manner: For example, a particular female may choose a male based on a nonsexual judgment (e.g. cultural achievement). Another female perceiving this may then become sexually attracted to the male or his most salient qualities.
Females may prefer natural or cultural male traits
The preferred traits could be natural or cultural, whether females have innate preferences for certain objective traits or a tendency to copy other females’ preferences. However, a tendency to copy the judgments of other females (or of conspecifics in general) may offer the more straightforward account for sudden strong sexual preferences among females for males with certain cultural distinctions (celebrities, rock stars, as well as members of respected groups).
Should the sexy-son hypothesis be replaced by a sexually-successful-son hypothesis?
What ultimately matters to the propagation of the female’s genes is not whether her offspring will be perceived as sexy, but whether her offspring will be sexually successful.
Consider a son who is not sexy in any conventional sense but seductive nevertheless: He is quite likely to successfully propagate the female’s genes. Next, consider a sexy son with no sex drive: He is not likely to help propagate the female’s genes. Next, consider an unsexy son with strong rapist tendencies: He is more likely to propagate the female’s genes.
Strong sex drive and even rapist tendencies may lead to a more effective propagation of the female’s genes, so a female sexual preference for these traits may be evolutionarily successful.
This suggests that the sexy-son hypothesis should be replaced by a sexually-successful-son hypothesis and may explain the widely noted female preferences for male promiscuity and male physical dominance during sex.