Five Myths about Gender Inclusive Sport

Common falsehoods that get in the way of pragmatic policy

Lausanne, Switzerland

This week I am in Switzerland at a major conference on Inclusive Gender Equality in Sport, organized by the Global Observatory for Gender Equality & Sport, which is supported by UNESCO. I expect this conference to provide important input to international federations who are seeking to implement gender classification regulations consistent with the International Olympic Committee’s Fairness Framework.

Like many areas of policy making in 2023, misinformation is not in short supply when it comes to discussions of gender classification. This post highlights and corrects five myths about gender inclusive sport that often get in the way of the development and implementation of pragmatic regulations for gender classification across elite sport.

1. FALSE: Trans women and women with variations of sexual development are identical, as far as athletic performance is concerned.

World Athletics, in its new proposed regulations that were leaked to a few reporters earlier this month, claims:

“As far as athletic performance is concerned, there is no significant difference between a 46 XY DSD individual, a cis male and a trans female prior to transition”

This claim is unproven and, in fact, available evidence suggests that it is simply false. For more than a decade, World Athletics (formerly IAAF) has been waging a campaign against women with variations of sexual development (who are called DSD females in the World Athletics regulations). Their proposed new regulations are reportedly focused more on these women than they are on trans women.

Ben Rumsby @ben_rumsby

WA's plan to harmonise its trans and DSD rules is also intriguing. Indeed, the document sent to member federations shows just how much more of an issue athletics currently has with DSD athletes than trans ones on the international stage.

Trans women who wish to compete in elite sport necessarily have changed their legal gender status from male to female. Females with variations of sexual development are female from birth. So from a legal standpoint they are different with differing implications for gender classification in elite sport — akin to citizenship by birth versus citizenship via naturalization, for which elite sport has different regulations.

There are important biological differences as well. It should be obvious, even to a non-medical expert, that the biology of trans women and females with variations of sexual development who maintain that gender identity from birth are necessarily different in their biology and physiology. That’s in the very definition of variations of sexual development. You can read a lot about those differences in the judgment of the Court of Arbitration for Sport in the Semenya case.

Further, there has been no study of the athletic performance of females with variations of sexual development that indicates that they perform similarly to cisgender males. In fact, there has never been a single documented case of such a woman achieving performances in line with top males. It is a myth.

World Athletics claims that they will “follow the science” — in this case they appear to be inventing the science.

2. FALSE: Males and females have non-overlapping testosterone distributions

It would be great if there were a simple marker that could be used to differentiate men from women. There is not — for instance, there are XX males and XY females. As Alice Dreger has written, “Humans like their sex categories neat, but nature doesn’t care. Nature doesn’t actually have a line between the sexes. If we want a line, we have to draw it on nature.”

A common myth is that testosterone levels in an individual can always and reliably indicate whether that individual is male or female. Typically, testosterone levels can be used to identify an individual’s gender, but not always. And that’s the problem.

The figure below summarizes the results of a literature review of testosterone ranges of males and females. In the left column are women and on the right are men, with grey zones indicating typical levels of testosterone for each. Trans individuals are not included in these reviewed studies. You can see in the figure that not everyone is typical, as there are a lot of people who fall outside the grey zones. In fact, you can see that there is an overlap of testosterone levels of XX and XY individuals, and of men and women.

There is an interesting back story to the figure above. It is the result of me finding a major error in an important literature review of testosterone levels among men and women that has been used to justify sport regulations based on testosterone levels. That error involved reclassifying women as men for purposes of the study. Rather than classifying people according to their actual gender, the study classified them by gender according to testosterone level — no wonder the study initially found non-overlapping distributions, that was defined into the analysis!

The reality is that most males and females fall into typical, non-overlapping distributions of testosterone. But not all do. Thus, testosterone levels are not a fail safe indicator of an individual’s sex or gender.

3. FALSE: An unfair advantage in competition can be identified by comparing bodies

Elite sport organizations seek to regulate “unfair advantage” in competition. This makes obvious good sense. But what, exactly, is an “unfair advantage”? You can’t regulate something unless you can define it precisely.

Fortunately, there is well-established legal precedent and sports arbitration cases that have defined “unfair advantage.” It is an advantage measured in competition, and the magnitude of advantage matters in determining if it is unfair or not. For instance, in the Dutee Chand case the Court of Arbitration for Sport found that a 10% advantage would be unfair, but a 1% advantage might not be unfair:

While a 10% difference in athletic performance certainly justifies having separate male and female categories, a 1% difference may not justify a separation between athletes in the female category, given the many other relevant variables that also legitimately affect athletic performance. The numbers therefore matter.

It is not enough to say that one group of athletes has an unfair advantage in competition over another group if the first group is, for instance, taller, has more muscles, can jump higher, has a stronger hand grip or via some other metric. What matters is what the athlete does with their body in competition.

Making things more complex is that advantages will be different in different types of competition. For example, judo is different than archery and both are different than synchronized swimming, and so on. To identify an unfair advantage in actual competition requires measuring competitive performances, not body characteristics.

There is further precedent for identifying unfair advantage involving the rules for inclusion of athletes who run on prostheses in mainstream competition. Just because an athlete runs on so-called cheetah blades does not mean that they have an unfair advantage over other athletes — some do and some don’t. We tell the difference based on evidence in actual athletic performance.

Simply being different or atypical does not by itself provide proof of “unfair advantage” — in fact, regulating differences rather than performances is a good recipe for inappropriate discrimination and worse.

4. FALSE: Gender categories in sport were originally created to regulate biology

Men’s and women’s categories in elite sport were originally created because women fought (and continue to fight) for equality starting more than a century ago. These categories were not created because enlightened and science-loving sports administrators of the early 20th century decided to classify sport according to biology. They didn’t think about science at all. Testosterone wasn’t even synthesized until the 1930s. Testing of chromosomes didn’t start until the 1960s.

Elite sport categories were originally created to reflect the broader society in which sport took place. That society was one dominated by men, typically to the exclusion of women in many important settings like politics and sport. At the time was there a correlation between biology and gender roles in society? Of course. But that does not mean that sport gender categories were originally based on biology.

The invention of such a founding myth is instrumental — after all, if gender categories in elite sport were originally based on biology, then implementing strict biological criteria for classification in 2023 could be justified based on historical precedent. Some make exactly this argument. But that story is indeed a myth. Elite sport was divided into men’s and women’s categories long before biological sex testing became a thing in elite sport.

An accurate story of the creation of gender categories in sport is that they were originally based on gender — which of course is typically correlated with biology, but not always. Biology matters of course, but it is not all that matters. As society’s perspective on gender has changed over time, sport has needed to keep up, since sport is a part of society.

Longing for a mythical past — where men were men and women were women — may resonate in some contemporary political settings, but it is not a sound basis for the regulation of gender categories in sport.

5. FALSE: Fairness means exclusion

Sebastian Coe, the head of World Athletics, was once fond of describing fairness in absolute terms. Speaking on trans women athletes Coe stated last summer:

“If we ever get pushed into a corner to that point where we’re making a judgment about fairness or inclusion, I will always fall down on the side of fairness”

The implication was that fairness and inclusion were mutually exclusive categories, implying that fairness required exclusion. An Orwellian shift in meaning to be sure.

Since that time the views of World Athletics, if not Coe, appear to have evolved considerably. Now the federation is apparently proposing regulations for the inclusion of trans women athletes based on testosterone suppression and a waiting period after transition. Apparently there is a middle ground to be found between fairness and inclusion after all.

Perhaps Lord Coe spent some time with the lawyers for World Athletics who may have told him about the regulation of shoes with carbon plates, athletes who run on prostheses, athletes who take banned substances under a therapeutic use exemption, athletes penalized for doping who return to competition after their ban, or maybe of athletes from a country banned from competition who are allowed to compete.

What unites all these cases is the idea of “reasonable accommodations” for inclusion. More specifically, the outright banning of a class of individuals based solely on their legal status (including gender) is a losing proposition in sport regulation. Sport organizations has tried over the years to ban black people, women, disabled, and others, and courts and arbitral bodies have consistently rules against such bans, and for good reason.

Of course, the notion of “reasonable accommodations” in any setting is one that is negotiated, and it may also be arbitrated and adjudicated. Those favoring the banning of certain individuals will push for “reasonable accommodations “ that are as onerous as possible, while those seeking maximum inclusiveness will press for the opposite. Pragmatic regulation must negotiate this type of conflict, and hopefully arrive at a compromise position that will evolve over time as values change and experience unfolds.

Neither fairness nor inclusion are absolutes. They are part of an interwoven tapestry of societal values, often contested, that when forming regulations for elite sport result in a stabilization of shared expectations for the right balance — a balance that we should expect will change over time.