Is “sex” socially constructed?

October 26, 2022 • 12:15 pm

This one-page paper from 2017 is what you get when two sociologists try to pontificate about human sex as not only not “binary”, but as “socially constructed”. They manage to make a number of errors, and of course it’s in the cause of ideology. If sex is “socially constructed”, so that there is no objective conception of what a “sex” is in humans, then that supposedly validates the identities all the people whose genders (sociocultural sex roles or feelings) don’t comport with the binary sex of male and female.  It’s a prime example of two errors:

a. conflating gender with sex, and

b. assuming that biological sex is socially constructed and forms a sort of spectrum, like gender.

Yes, gender does form a spectrum because there are a gazillion genders (more than 150 at this site!). But there are only two biological sexes in humans and most animals, with “sex” defined as “the category containing individuals whose bodies are organized around producing one kind of gamete.”  “Males” are all those in the class that produces small, mobile gametes (sperm), and “females” are all those in the class that produces large, immobile gametes (eggs).  These two sexes were produced by evolutionary processes, processes in which more than two sexes is an unstable condition.  The number of exceptions to the definition I’ve given is tiny–less than 0.05%, comprising individuals who are hermaphrodites or have disorders of sex development.  Thus, more than 99.5% of human fit neatly into one or the two sexes, and that’s about as “binary” as you can get. (One petulant reader told me that “almost binary” is not “binary”, and yes, I have  Ph.D. and know that.) But for all practical purposes sex is binary, and certainly wholly bimodal.

Biological sex is not socially constructed, either: it is an objective form of distinguishing individuals, a dichotomy that appears in nearly every species of animal, and fruit flies don’t socially construct their own sexes!

Read and weep at this “back page” from the American Sociological Association’s journal Contexts; click on screenshot below to go to the page, or find the pdf here.

I can barely bring myself to defend the biological definition of sex again, but humanities scholars keep popping up like Amanita mushrooms and making counterclaims, like this one:

Sociocultural scholars have explored the social construction of gender as a performative, fluid, and non-universal category for decades, but the notion that physical sex is also socially constructed has acquired far less exploration.

That’s because it doesn’t deserve exploration, because physical sex is not socially constructed—whatever they mean by “physical sex”. I presume they mean primary sex traits like penises, vaginas, and uteruses, and secondary sex traits like body hair and the presence of breasts. Even those are very highly (but not perfectly) correlated with biological sex, so even using genitals, or for that matter chromosome constitution, you can sort individuals into two modes with few intermediates.

The authors argue, however, that because some people in the past have used other definitions of sex than the gamete-based one settled upon by biologists, sex must be something slippery and undefinable, and therefore not very clear-cut.  But they mistake the refinement of our understanding (based largely on apprehending evolution) with biological confusion. To wit:

While we tend to rely on genital appearance at birth (more directly, the presence or absence of a phallus) as the basis of our sex assignment, what constitutes the essential sign of sex has varied over the years. Genital appearance, sex hormones, sex chromosomes, and the brain have each been used to sex categorize bodies at different points in time. Sex hasn’t always been a simple binary divide, either: pathologist Theodore Klebs, for instance, first classified anatomical sex into five categories in 1876, using the presence of gonads (ovaries, testes, or a mix of ovarian and testicular tissue) as his guide, and biologist and gender scholar Anne Fausto-Sterling further described these divisions in her influential 1993 piece, “The Five Sexes.”

Krebs’s view, and those other views, have been outmoded for well over 50 years.  As for individuals having both ovarian and testicular tissue, they can still be called “male” or “female” based on whether that tissue makes eggs or sperm. There are less than a handful of cases in history in which individuals make both eggs and sperm, and those might be called either a “third” sex or “intersex”, but most hermaphrodites are completely sterile.

Here’s another error: the authors conflate the hormonal criteria used by the International Olympic Committee to see who is allowed to compete in “men’s” or “women’s” categories with the definition of sexes themselves. This is not true: the Olympics is simply trying to place individuals into the category in which their participation leads to fair competition, not define “male” and “female”:

More recently, hormonal levels have been used to categorize sex, as is the case in sex testing conducted by the International Olympic Committee (IOC) and the International Association of Athletics Federations (IAAF). In 2009, South African runner Caster Semenya won the 800-meter race at the Berlin World Championships in Athletics. The media and several of Semenya’s competitors seized on her appearance and performance to pose stigmatizing questions about whether she was eligible to compete as a female. Semenya was temporarily banned from competition. In a purported effort to prevent another such fiasco, in 2012, the IOC and IAAF issued sex-testing policies centered on hyperandrogenism (a medical term describing, in females, higher than “normal” levels of androgen, including testosterone, and often associated with intersex traits). The groups claimed the guidelines were not about sex testing women athletes, but about ensuring fairness in elite athletic competitions. After years of scrutiny, Semenya (who has never self-identified as hyperandrogenic or intersex) was reinstated. She won silver at the 2012 Olympic Games. In the summer of 2015, the sex-testing policies were suspended after Dutee Chand, an Indian 100-meter sprinter, successfully appealed to the Court of Arbitration for Sport. Chand didn’t advance to the semi-finals in the 2016 Olympic Games, but Semenya won gold in the 800-meter race. Immediately following her win, the IAAF made a statement that they would consider the possibility of reinstating hyperadrogenism testing.
Semenya would be classified as having a disorder of sex development, perhaps as an intersex. Her chromosomes are XY but she has 5α-Reductase 2 deficiency, a mutation that causes elevated testosterone and produces rudimentary male gonadal tissue. These individuals vary in “bodily presentation”: while Semenya has male gonadal tissue, she is one of the individuals with this condition that has a vagina. I would call her an “intersex” individual: one of the 0.02% of people who don’t fall neatly into male or female. But she is not a member of a third “sex”, either, as these individuals are sterile.  As for her ability to compete as a woman, I have no judgment on that.  But the authors conclude this:


That one’s eligibility to compete as a female athlete is debatable and that the physical criteria used to judge femaleness have changed over time are evidence that the categorization of sex is a social, variable process.

Nope. Semenya was allowed to compete as a woman. This does not mean she was deemed a biological women.

It’s hard to make so many errors and promulgate so many misconceptions on a single page, but Dais and Preves have succeeded. At the end they underscore once again their incorrect view that sex is not a real phenomenon but a social construct:

Perhaps, then, we ought to ask parents “Who is it?” rather than “What is it?” when we meet a child. That way, the focus might rest more holistically on the newborn as a human being, rather than the predetermined product of a historically variable and socially constructed sex and gender system. Maybe then we can get to the root of why, as a society, we are so quick to categorize babies as “females” or “males” ascribed with “feminine” or “masculine” personalities. Doing so would require wrestling with, and perhaps unraveling, our widely held beliefs that both sex and gender are binary, neatly correlated phenomena. Simply changing the focus of the conversation seems a good place to start acknowledging the diversity of sex development.

Why do the authors do this? Well, perhaps they’re ignorant of biology, but I won’t accuse them of this. More likely they’re guilty of making a fallacious “appeal to nature”: because we see intersex individuals like Semenya, this means not only that biological sex doesn’t exist, but somehow vindicates the gender identity of those who see themselves as bisexual or one of the many other gender categories that don’t correspond with “male” and “female”.