@xwaldie (Katy Waldman) advises us to beware of evolutionary explanations that rest on what men find attractive. In Another Dingbat Sexual Selection Theory (Slate Feb 2013), she comments on a theory that explains the evolution of small breasts in East Asian women.
Women may well be insulted by the suggestion that the reproductive success of East Asian women depends on the mammary preferences of East Asian men. Waldman argues that women's choices are at least as important as men's - indeed, women may have more reason to be choosy in their sexual partners than men - and notes that in most species it is the male that invests in display (e.g. fancy plumage), while healthy females can get laid without needing to develop unwieldy protruberances.
Meanwhile, men may be insulted by two contradictory propositions - either that they will have sex with anything that moves, or that they will only have sex with women whose chest fits some predefined norm.
As I see it, one of the main problems of evolutionary biology is that for any plausible hypothesis, one can invent any number of equally plausible alternatives.
For example, let's start with the idea that a male seeks to produce as many viable offspring as possible. The survival prospects of his genetic line are surely improved if his offspring are as genetically varied as possible, which means that the male should try to maximize the variety of his sexual partners. Thus if the majority of the available women have one characteristic - say small breasts - then the pursuit of genetic variety might attract him to woman with the opposite characteristic - namely large breasts. But this pursuit of variety might equally be satisfied by variation in other physical characteristics, or even non-physical characteristics such as personality and intelligence.
Of course females also benefit from having genetically varied children. There is some evidence of what scientists call "negative frequency-dependent preferences" - in other words, fancying the unusual. A recent experiment suggests that the relative attractiveness of bearded and clean-shaven men goes in cycles: the more beards there are, the less attractive they become.
However, variety needs to be balanced against other factors. There are many reasons why a woman may prefer to have all her children with the same man. And there are also reasons why a man may prefer to have all his children with the same woman rather than casually impregnating many different women.
So even if sexual attractiveness is based on some genetic programme, we cannot infer the logic of the programme either from observing actual and attempted couplings, or from asking subjects to rate photographs of the opposite sex. Such evidence may lead us to dismiss some hypotheses as unlikely, but do not help us choose between equally likely alternatives.
It may be fun for scientists to speculate why a particular characteristic developed in a particular group of humans at a particular point in prehistory, and it may help the scientists get noticed by journalists, but theories based on sexual attractiveness are highly unreliable. After all, sexual desire is so polymorphous.
Updated 17 April 2014
Katy Waldman, Another Dingbat Sexual Selection Theory (Slate Feb 2013).
James Morgan, Beard trend is 'guided by evolution' (BBC News 16 April 2014)
Zinnia J. Janif, Robert C. Brooks and Barnaby J. Dixson, Negative frequency-dependent preferences and variation in male facial hair. Biol. Lett. April 2014 vol. 10 no. 4 20130958 16 April 2014 doi: 10.1098/rsbl.2013.0958