Science, Sport, and Sex

In “Science, Sport, Sex, and the Case of Caster Semenya” (Issues, Fall 2019), Roger Pielke Jr. and Madeleine Pape systematically misrepresent the International Association of Athletics Federations’ position on female athletes with differences of sex development (DSDs). The IAAF is neither targeting athletes whose appearance is “insufficiently feminine” nor trying to “question and reclassify the sex of such athletes.” It does not use testosterone levels “as the basis for … sex classification.” In fact, its regulations expressly state that they are “not intended as any kind of judgement on or questioning of the sex or the gender identity of any athlete.”

On average, men have physiological advantages (including bigger and stronger muscles and bones, and more hemoglobin) that give them an insurmountable performance advantage over women. Therefore, women can excel at sport only if they compete in a separate category. The primary driver of the sex difference in sport performance is the testosterone levels that men’s testes produce post-puberty (7.7-29.4 nmol/L) compared with the levels produced by women’s ovaries and adrenal glands (0.06-1.68 nmol/L). Therefore, the IAAF has to address, rationally and fairly, the two categories of athletes with a female gender identity (46 XY DSD, and trans-female) who have testes producing testosterone in the normal male range that gives them the same physiological advantages.

Biological sex is not a spectrum. In 99.98% of cases, all aspects of sex (genetic, gonadal, hormonal, anatomical) are aligned, making classification as male or female straightforward.

Biological sex is not a spectrum. In 99.98% of cases, all aspects of sex (genetic, gonadal, hormonal, anatomical) are aligned, making classification as male or female straightforward. Complete alignment is lacking in only an estimated 0.02% of births. For example, XY babies with 5-alpha reductase deficiency (5-ARD) lack the hormone responsible for normal genital development, and so are born with undescended testes and ambiguous external genitalia. Endocrinology nosology classifies them as undervirilized males, but in some countries they may be assigned a female legal sex at birth. On puberty, however, their testes start producing normal male levels of testosterone, causing the same androgenization of their bodies as XY individuals reared male. A reported 50%–60% of 5-ARD individuals reared female therefore transition upon puberty to a male gender identity, which is why current medical advice is to assign a male sex to 5-ARD babies.

Whether given a legal female sex from birth (some XY 5-ARD) or later (trans-female), such individuals have enormous physiological advantages over XX females, making competition between them unfair. The IAAF nevertheless permits 46 XY DSD and trans-female athletes to compete in the female category, provided they reduce their testosterone levels below 5 nmol/L, whether by surgery or hormone therapy (the recognized standard of care in such cases). This is because a woman with ovaries, even if she has the hormonal disorder polycystic ovary syndrome, will not have testosterone above 5 nmol/L unless she is doping or has a serious medical disorder. The data gathered by the researcher Richard V. Clark confirm this. The fact that 5-ARD individuals have testosterone below 5 nmol/L before puberty is entirely expected, and Pielke and Pape’s attempt to discredit Clarke’s paper is irrelevant.

On the sports field, biological sex must trump gender identity. References by Pielke and Pape to “biological sex as assigned and maintained at birth” serve only to confuse the two and distract from the facts that have to be addressed if competition in the female category is to remain fair.

Director, Health and Science Department
World Athletics (formerly known as the International Association of Athletics Federations)

Partner, Bird & Bird


(Both authors appeared for the IAAF in Court of Arbitration for Sport cases: C. Semenya and ASA vs IAAF, and D. Chand vs AFI and IAAF)

Richard Clark and his coauthors have already responded to Roger Pielke Jr. and Madeleine Pape’s criticism of their paper elsewhere. Here I want to address Pielke and Pape’s decision to play the race card as part of their disinformation campaign against the International Association of Athletics Federations’ regulation regarding differences of sex development (DSDs). Their campaign is based in two related arguments, neither of which has legs. The first is that the regulation discriminates against women of color based on their appearance. The second is that it mostly targets women of color.

Stripped of its misleading vocabulary and veneer, the argument that the regulation discriminates against women of color based on their appearance goes like this: athletes with testes, male testosterone levels, and male secondary sex characteristics look male only to those of us who privilege white female features and have a racist view of the black female body. Racism exists and should always be exposed for the scourge that it is, but calling someone a racist because they can distinguish an androgenized body from a nonandrogenized one is wrong. We shouldn’t have to spill ink making the obvious point that discriminating on the basis of race and distinguishing on the basis of sex aren’t the same thing. Racism and sexism sometimes intersect, and this results in special burdens for black women, but that’s not what’s going on here.

We shouldn’t have to spill ink making the obvious point that discriminating on the basis of race and distinguishing on the basis of sex aren’t the same thing.

Male secondary sex characteristics develop in early adolescence when testes—but not ovaries—begin to produce increasing amounts of testosterone that, in turn, cause the androgen-sensitive human body to go through male rather than female puberty.

Caster Semenya’s own experience is illustrative. Growing up and competing in local competitions, opposing sport teams would contest and then test her sex in the crudest sense of that term. Her principal has said that he didn’t realize she was a girl for most of the years she attended his school. Her father has said that she sounds like a man, and an early coach tended to use the male pronoun when referring to her in conversation. She has been described by African people in African publications as having a “masculine phenotype” and “man-like physical features.” Perhaps most telling, the first time they met in the girls’ locker room, the woman who eventually became Semenya’s wife thought she was a boy and questioned her presence in that space. The suggestion that these reactions are anything but universal is both offensive and destructive.

It’s also wrong to suggest that the IAAF’s DSD regulation mostly targets women of color from the Global South. As the African National Congress itself has conceded, albeit incompletely, it affects athletes from “East Europe, Asia and the African continent.” These are the facts: The regulated conditions, including 5-alpha reductase deficiency, exist in all populations. Sometimes differences that have a genetic basis cluster in particular geographical areas. And because different cultures place different value on the appearance of boys’ external genitals, when a male child presents ambiguously at birth, preferences for their legal sex assignment may vary based on those norms. Finally, because individuals in different regions have different access to medicine and medicine itself has a cultural basis, DSDs are sometimes addressed surgically in infancy and sometimes children are left to grow up naturally. But neither the conditions themselves nor the approaches to their management are racially bound.

The Olympic Movement has had an eligibility rule for the women’s category for decades. Across those decades, affected athletes have come from different countries and continents, including from Western Europe and the Americas. The rule was not put in place for Semenya.

It’s important not to be blinded to the facts by politics and incendiary rhetoric. In the sport of track and field, black and brown women from around the world and across the tonal spectrum are not only ubiquitous, they are properly celebrated as beautiful, as strong, as winners. As is often the case in our sport, many of these women grew up in economically challenged circumstances, and they beat many who were more privileged. The IAAF’s DSD regulation hasn’t changed this. Since it’s been in place, as expected, all 12 of the medals in the affected events have gone to women of color. These athletes are also deserving of consideration.

I agree with Pielke and Pape that Caster Semenya is perfect just the way she is. This doesn’t change the fact that she and others with 46 XY DSD have the primary and secondary sex characteristics the women’s category was designed specifically to exclude.

Professor of Law
Duke Law School

I was both pleased and disturbed to see the article by Roger Pielke Jr. and Madeleine Pape. I was pleased because it adds further argument to the worldwide campaign to have the most recent version of the IAAF’s sex test, the “Eligibility Regulations for Female Classification” (aka the “Caster Semenya Rules”), abolished on the grounds that they lack scientific merit, violate human rights and medical ethics, and unfairly target women from the Global South.

Pielke and Pape do us all a favor by exposing the manipulation of data that the IAAF employs to justify the persecution of Caster Semenya and other outstanding athletes.

But I was disturbed because the article further exposes the extent to which the IAAF architects of the Caster Semenya Rules manipulated the categories of analysis to justify the regulations. As an Olympian in athletics myself, I have always hoped that my sport would be governed in the spirit of fairness and best practice in policy-making, drawing on impartial, widely vetted science. But as Pielke and Pape make clear, the drivers of the Caster Semenya Rules chose categories that would give them the nonoverlapping distribution of testosterone that they used to construct the regulations, presenting “what (they felt) ought to be (as) what is.” Pielke and Pape offer a much more realistic and inclusive approach to classification for the purposes of organizing sports competition, but there is nothing in the IAAF’s behavior to suggest that it would ever be open to such an approach. It’s been that way for more than 50 years, as leading geneticists and ethicists have urged the IAAF to abolish the test only to be ignored.

It’s hard to know how to right this wrong. In the decision that upheld the IAAF regulations, the Court of Arbitration for Sport (which serves as the highest appeal body in international sport) said that it did not have to consider human rights, medical ethics, or the quality of the science. It’s impossible to argue today that sport is not in the public realm nor that it should be free from the obligation to enforce the protections that exist in other spheres of society, especially human rights. In every country in the world, sport is enabled by public funding and publicly created facilities, and the best athletes and teams are held up as exemplars of universities, cities, and even entire countries. If the international sports bodies do not voluntarily accept responsibility for human rights and accountable policies, I am confident that a strategy to bring them under such standards will soon be found.

In the meantime, Pielke and Pape do us all a favor by exposing the manipulation of data that the IAAF employs to justify the persecution of Caster Semenya and other outstanding athletes. Their article is one more step in the campaign to the complete abolition of the sex test.

Professor, Faculty of Kinesiology and Physical Education
University of Toronto

In their attempt to distinguish transgender athletes from those labelled as intersex/DSD, Pielke Jr. and Pape err in their suggestion that appearance of external genitalia would prove an acceptable basis for determining sex for athletic purposes.

Although in social situations, gender identity should be used to categorize humanity, it seems likely to be the case that in sport the biology that is important for performance should be considered also, including hormone exposure and its consequences. To date, testosterone level is the most well-established correlate with athletic performance and the best basis for distinguishing individuals in sex-based categories. In the Semenya case, the Court of Arbitration for Sport panel of judges all agreed that “on the basis of the scientific evidence presented by the parties, the Panel unanimously finds that endogenous testosterone is the primary driver of the sex difference in sports performance between males and females.” Unless and until data suggest a better method, we advocate for continued use of testosterone to separate male athletes from female ones.

Loughborough University

Mount Sinai Hospital, New York

(Harper was a witness for the IAAF at the Semenya and Chand trials, and Safer has served as an adviser to the IAAF)

Cite this Article

“Science, Sport, and Sex.” Issues in Science and Technology 36, no. 2 (Winter 2020).

Vol. XXXVI, No. 2, Winter 2020