Sexually antagonistic selection

This article clarifies part of a previous discussion addressing discrimination against unattractive women.

We consider the problem of transforming the average English man and the average English woman via natural reproduction over multiple generations into individuals similar to those shown in the following picture, respectively.  We disregard the desirability of this transformation.

Dorian Yates (left) and Elkie from Simonscans.

Let us specifically address the goal of producing masculine male offspring.  For a given male input, which of the following two women appears more likely to produce masculine male offspring?

Alexa B. from Met Art (left) and Amiee Rickards from Mystique magazine.

Some more pictures of Amiee Rickards.

Similarly, for a given male input, which of the following two women appears more likely to produce masculine male offspring?

Adri (left) and Natasha (variously identified as Corinna from Met Art, Natalie from Domai, etc.).

The face of Adri.

Again, for a given male input, which of the following two women appears more likely to produce a tall, masculine son?

Alana from bestyoungbabes.com.

Leona from ATK galleria.

The intuitive answer to the three questions above is the more masculine-looking woman.  However, since women will give birth to either male or female offspring, for a given male input, more masculine-looking women are less likely to produce feminine-looking female offspring.  The three problems above can also be posed in terms of which of two men, one more masculine than the other, appears to be better suited to producing a more masculine son or more feminine daughter for a given female input.  Once again, other things being equal, whereas more masculine men would tend to give birth to more masculine sons, they would also tend to give birth to more masculine daughters.  Therefore, how can one naturally achieve the goal considered here?  We consider numerous possible mating scenarios under the following assumptions:

The fertility of men increases, within limits, from below average masculinity to above average masculinity.

The fertility of women increases, within limits, from below average femininity to above average femininity.

In the table below, viability refers to the likelihood of a population thriving indefinitely.

It should be obvious that the most viable mating preference schemes considered above are #3 and #4.  Therefore, in a sexually reproducing species that has been thriving for a long time, one would expect to see few individuals with #1 and #2 mating preferences.  Consider also that sexually reproducing species descend from asexually reproducing species.  The problem of transforming a unisexual species to a species with two or more sexes requires a divergence that is similar to the divergence that is part of our goal, namely exaggerating the physical differences between the sexes, and it is clear that mating preference scheme #4 is best suited to the task.  Therefore, mating preference scheme #4 will be a major tendency in humans and similar sexually reproducing species, a legacy of the origin of sexual dimorphism and sexual reproduction.

Note that the practical implementation of mating preference scheme #4 would be similar to that of mating preference scheme #3 on several counts.  In mating preference scheme #4, masculine men and feminine women would be considered the most desirable on the part of most.  Hence, masculine men will usually mate with feminine women, which translates to a) mostly leaving average men the option of selecting average to masculine women, and average men will prefer average to masculine women; b) mostly leaving average women the option of selecting average to effeminate men, and average women will prefer average to effeminate men; and c) leaving effeminate men and masculine women mostly the choice of each other.  However, the difference from scheme #3 would be that in a number of cases, masculinized women will pursue and succeed in mating with masculine men even if they are not able to obtain masculine men as long-term partners, thereby increasing the likelihood that some of the resulting offspring are across-the-board very masculine men, but the resulting female offspring will tend to be masculine, too.  Similarly, some not-so-masculine men will succeed in obtaining feminine women as a result of compensatory desirable characteristics (such as riches), thereby increasing the likelihood that some of the resulting offspring are across-the-board very feminine women, but the resulting male offspring would tend to be feminine, too.

The technical phrase describing the problem that prevents us from meeting our goal is sexually antagonistic selection, i.e., something that benefits one sex harms the other.  In our example, above average femininity favors women but hurts men, and above average masculinity favors men but hurts women.  There is no natural solution to this problem.  A quest for feminine women and masculine men will invariably produce some masculine women and some feminine men in every generation.  Under strong sexual selection, which can be conceptualized in terms of a strong push to meet our goal, whereas one would end up with a greater proportion of very masculine and very feminine women, one would also end up with a greater proportion of effeminate men and masculinized women, i.e., the average fertility of the general population would be somewhat diminished.  Weaker sexual selection would translate to a lower proportion of very masculine men and very feminine women, but also a lower proportion of effeminate men and masculinized women, corresponding to a somewhat higher average fertility of the general population.

The last point can be clarified in light of asking the question what would be a good reproductive partner choice for a man picked at random?  A good choice for a random male would be a woman that increases the likelihood of producing both male and female offspring with at least average fertility.  This good choice cannot be a feminine woman because a random male could be effeminate and effeminate man plus feminine woman increases the likelihood of effeminate male offspring.  The good choice cannot be a masculine woman either because if the randomly chosen man is masculine, masculine woman plus masculine man increases the likelihood of masculinized daughters.  The good reproductive choice for a randomly chosen male is obviously a woman who is in the neighborhood of average with respect to the masculinity-femininity range; see some examples below.

Yana P. from Met Art (a more feminine version of Gisele Bundchen).

Yanka from Hegre Art.

In summary, since one cannot find a reproductive partner that simultaneously increases the likelihood of producing sons with above average masculinity and daughters with above average femininity, in general, a good reproductive partner is one that simultaneously increases the likelihood of producing average male and average female offspring.  In this regard, feminine women would be a good reproductive choice for masculine men, but for the majority of men, a good reproductive choice would be an average woman and for some men a good reproductive choice would be a masculinized woman.

The purpose of this entry

This entry should hopefully clarify that this site isn’t arguing that feminine women are better, period.

The extreme contrast in the top picture was chosen to clarify the problem of sexually antagonistic selection and why mating preference scheme #4 above is a major tendency in the population.  In reality, most women would consider the muscularity of the bodybuilder repulsive.  However, by asking how one could make the sexes diverge, one considers a problem similar to the initial divergence leading to the sexes, and understands that mating preference scheme #4 is best suited to the task of increasing divergence whereas mating preference scheme #3 is best suited to the task of maintaining a given level of divergence.  As noted earlier, the practical implementation of schemes #3 and #4 is broadly similar.  Therefore, there will not be a strong push for scheme #3 to become a major tendency in the population; it will remain a minor tendency.

Some technical issues

Evidence showing that above average femininity in women corresponds to higher fecundity and fertility has been posted previously; see Fig. 3 on the “eating disorders” page and a study on estradiol and face shape in women.  For evidence that below average masculinity in men corresponds to reduced fertility, see one example.

In the top picture, the male physique shown corresponds to high androgen and low estrogen levels, whereas the female physique corresponds to low androgen and high estrogen levels.  If one considers the biosynthetic path of androgens and estrogens, one would note that estrogens (marked by red arrows in the figure below) are derived from androgens.

The biosynthetic paths of some steroids.

Testosterone levels in human males and females (left) and estradiol levels in human males and females.

Whereas the absence of testes eliminates a major source of androgens in women, more would be needed to produce feminine women than mere absence of testes.  Genes that convert more of the androgens to estrogens will favor women whereas genes that convert less of the androgens to estrogens will favor men.  Therefore, some of the molecular details underlying the problem of sexually antagonistic selection can be appreciated.