Features 2 May 2019

Caster Semenya ruling: CAS helps sport put its blinkers back on

Yesterday, the Court of Arbitration for Sport (CAS) upheld an amended version of the Eligibility Regulations for the Female Classification (Athletes with Differences of Sex Development), promulgated by the International Association of Athletics Federations (IAAF). The CAS Panel dismissed Semenya’s requests for Arbitration, ruling that her legal team had failed to establish that the DSD Regulations (PDF below), as they have become known, were ‘invalid’.

The CAS has yet to publish its full 165-page decision. It has published an Executive Summary (PDF below) which provides some insight into its reasoning. This article is designed to examine that insight, and assess what the ruling will mean for Semenya and other athletes with a DSD, with input from experts in endocrinology, sports policy, and human rights.

The DSD Regulations

The DSD Regulations are extremely limited in their scope. They cover athletes with one of five listed DSDs competing in international events run between 400m and one mile in the IAAF female category, if their testosterone levels are above 5nmol/L and have an ‘androgenising effect’ (i.e. if that testosterone is taken up by their androgen receptors and boosts their physiology). The DSD Regulations only apply to XY women:

• That have one of five DSD conditions outlined by the IAAF in Article 2.2 of the Regulations;
• with naturally occurring testosterone above 5 nmol/L that compete in the Restricted Events;
• That have sufficient androgen sensitivity for naturally occurring testosterone at levels above 5 nmol/L to have a ‘material androgenising effect’.

XY females covered by the Regulations must:

• Be recognised by law as either female or intersex;
• must use hormonal contraceptives to reduce testosterone levels to below 5 nmol/L for six months prior to competing;
• must maintain naturally occurring testosterone at below 5 nmol/L.

If a XY female athlete covered by the DSD Regulations fails to take these steps, they will be ineligible to compete in the international Restricted Events, or to set world records. However, they would be eligible for all non-international events, male category events, or any ‘intersex’ events offered in the future.

The DSD Regulations that exist today are not the ones that Semenya challenged, as the CAS allowed the IAAF to amend them after the hearing. The original DSD Regulations covered seven, rather than five, DSD conditions. The amended Regulations carve out Congenital Adrenal Hyperplasia (CAH) from their scope, as well as 3β‐hydroxysteroid dehydrogenase deficiency. 

In very general terms, females have two types of the same chromosome (XX) and males have two distinct chromosomes (XY). However, a number of variations on this basic theme are possible, and these variations are understood to include athletes with the five DSDs covered by the IAAF’s Regulations.

The CAS has clarified that the amended version of the DSD Regulations only seek to regulate 46 XY DSD athletes – i.e. athletes with an XY phenotype rather than an XX phenotype. ‘No individuals with XX chromosomes are subjected to any restrictions or eligibility conditions under the DSD Regulations’, reads the CAS Executive Summary. It would appear that the narrow scope of the DSD Regulations has been further narrowed by this change.

Exactly why the IAAF has made this adjustment is not immediately clear. The IAAF has denied that this change in focus classifies a DSD athlete as male. However, as it has narrowed the scope of the DSD Regulations to exclude XX athletes, it is hard to avoid the conclusion that this is exactly what the DSD Regulations do. And this conclusion is not helped by Doriane Lambelet Coleman, a former 800m runner and Professor of Law at Duke, who testified in Semenya’s case.

‘When we are told that 46, XY males with DSD who identify as female are simply “women with hyperandrogenism,” or “women with high T,” we aren’t fooled’, she writes. ‘We are just puzzled about why others are—or would want to pretend to be’. Her description of 46 XY women as ‘males’ is problematic, as the World Health Organization (WHO) specifically recognises 46 XY females.

The main piece of evidence produced by the IAAF in support of the DSD Regulations (PDF below) admits that testosterone has an especially potent effect on individuals with an XX phenotype. ‘There is also anecdotal evidence that the dose-response curves may be left shifted so that testosterone has greater potency in women than in men at comparable doses and circulating levels’, admits Circulating Testosterone as the Hormonal Basis of Sex Differences in Athletic Performance (PDF below).

If the IAAF evidence is correct and testosterone has a more potent effect in the XX phenotype, then logic would appear to dictate that XX women with testosterone levels above 5 mol/L would hold an even greater advantage than XY DSD athletes. Presumably, this apparent arbitrary targeting of XY DSD athletes will be explained in the full CAS Decision, which has yet to be released.

Performance advantage

Has the IAAF proven that testosterone levels of above 5 nmol/L provide 46 XY DSD athletes with such a performance advantage in events run between 400m and one mile that it is necessary to exclude them from international competition? The CAS certainly thinks so, but other experts disagree.

‘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’, reads the CAS’s Executive Summary. ‘Having carefully considered the expert evidence, the majority of the Panel concludes that androgen sensitive female athletes with 46 XY DSD enjoy a significant performance advantage over other female athletes without such DSD, and that this advantage is attributable to their exposure to levels of circulating testosterone in the normal adult male range, rather than the normal adult female range. The majority of the Panel observes that the evidence concerning the performances and statistical over-representation of female athletes with 46 XY DSD in certain Relevant Events demonstrates that the elevated testosterone levels that such athletes possess creates a significant and often determinative performance advantage over other female athletes who do not have a 46 XY DSD condition.’

As mentioned, the majority of the Panel agreed. In other words the decision was supported by two of three Panelists. Other experts disagree with the assessment that the IAAF has proven that endogenous testosterone levels provide 46 XY DSD athletes with such an unfair performance advantage that it is necessary to exclude them from international female events run between 400m and one mile.

‘Long before the CAS proceedings, we identified flaws in IAAF research used to justify the regulations, writes Roger Pielke Jr. who acted as a pro bono expert witness in the Semenya case. ‘Our findings have since been published in the peer-reviewed literature. On the basis of these findings, and these findings alone, we argued that the IAAF research should be retracted and the regulations should be put on hold until more rigorous research was done by independent researchers. The scientific issues that we identified have not been contested by IAAF – indeed, many of the issues that we identified have been acknowledged by IAAF.

‘The fact that a majority of the CAS Panel voted to uphold the regulations indicates that these issues of scientific integrity were not considered to be critical in its decision […] the press release indicates that CAS still considers the flawed IAAF research to be authoritative. This alone is stunning.’

The main piece of research underpinning the DSD Regulations, is Circulating Testosterone as the Hormonal Basis of Sex Differences in Athletic Performance, a PDF of which is available above. It was not published when the DSD Regulations were promulgated, and the evidence it uses in support of the DSD Regulations is discussed in detail here.

As Pielke Jr. points out, the main piece of research referenced in support of the DSD Regulations, Circulating Testosterone as the Hormonal Basis of Sex Differences in Athletic Performance, was also referenced in the Dutee Chand case. ‘The evidence does not go so far as to equate, or correlate, the level of testosterone in females with a percentage increase in competitive advantage’, found the CAS Panel in the Chand case. ‘The evidence does not, for example, establish an advantage of the order of 12% rather than, say 1% or 3%. Once the degree of competitive advantage is established, the IAAF would then need to consider, if the degree of advantage were well below 12%, whether that justified excluding women with that advantage from the female category.’

The CAS Panel added that the IAAF’s previous Hyperandrogenism Regulations would be declared void unless the IAAF could provide evidence proving the actual degree of performance advantage enjoyed by hyperandrogenic females due to higher testosterone levels. On 5 July 2017, two years after the CAS ruling in the Chand case, the IAAF submitted evidence, however it was based on studies performed at the Daegu 2011 and Moscow 2013 IAAF World Championships, where significant doping is alleged to have taken place.

The IAAF’s evidence maintained that female athletes with elevated testosterone enjoyed an advantage of between 1.78% and 4.53% in five athletics disciplines, none of which were Chand’s. This was below the 10-12% performance advantage that males have over females, as outlined at the CAS. The authors re-submitted the Study (revised version here) after flaws were found in the data. 

In a February 2019 article published by the Asser International Sports Law Journal, it is alleged that the revised Study remains the bedrock to the DSD Regulations, despite many of the flaws not having been corrected. Researchers found that the times of athletes disqualified for doping had been included in the Study, as well as duplication of some athletes and times.

It is significant that the CAS allowed the IAAF to further amend its evidence after the Semenya hearing in an attempt to correct flaws in the problematic data used from the Daegu 2011 and Moscow 2013 IAAF World Championships. Footnote 3 of the amended DSD Regulations references ‘Serum androgen levels are positively correlated with athletic performance and competition results in elite female athletes’, a December 2018 Study not referenced in the original DSD Regulations.

The CAS Executive Summary references the fact that the IAAF failed to produce the scientific evidence required by the CAS, but doesn’t provide these details. ‘The IAAF was given the opportunity to provide further evidence to validate those regulations, which had set the maximum level of testosterone for an athlete in female competition to 10 nmol/L, this being well above the maximum level in the female population and slightly above the minimum level in the male population’, it reads. ‘In March 2018, the IAAF informed the CAS that it intended to withdraw the Hyperandrogenism Regulations and to replace them with new Regulations’.

In its Executive Summary, the CAS also warns about the ‘paucity’ of evidence justifying the inclusion of the 1,500m, which Semenya also runs, and one mile in events covered by the DSD Regulations. In yesterday’s media statement, the CAS Panel suggests ‘that the IAAF consider deferring the application of the DSD Regulations to these events until more evidence is available’.

The IAAF confirmed that is has considered this request, and has decided to ignore it. ‘We have enough evidence from the field across all the disciplines covered by our Regulations so they will remain included in the Regulations’, confirmed an IAAF spokesperson in an email. ‘We may have more data in the 400m and 800m, but CAS’s decision clearly states that our regulations, as they stand, are reasonable, necessary and proportionate so we will keep them as they are and administer them with care and compassion. Like all our rules and regulations they remain under review.’ 

The IAAF’s 2018 Study references a number of studies in support of the proposition that DSD athletes benefit from increases in circulating testosterone that increases circulating haemoglobin, which in turn translates to an increase in oxygen transfer. The first is a 2017 Study examining women with Congenital Adrenal Hyperplasia (CAH), a condition in which the adrenal gland can produce more testosterone. The Study found that in women with CAH, erythropoiesis may be driven by androgens. The proposition is that as DSD athletes have higher levels of testosterone (an androgen), they benefit from increased erythropoiesis (production of oxygen-carrying red blood cells).

The second is a 2017 Study which sets out that women with CAH can require glucocorticoid replacement therapy, but exhibit widely varying levels of hormonal control. ‘The degree of poor control is associated with increasing levels of circulating testosterone ranging from normal female concentrations up to 36 nmol/L, and these levels correlate closely (r = 0.56) with levels of circulating hemoglobin’, reads the 2018 Study. ‘Interpolating from the dose-response regression, increases in circulating testosterone measured by LC-MS from 0.9 nmol/L to 5 nmol/L, 7 nmol/L, 10 nmol/L, and 19 nmol/L were associated with increases in circulating hemoglobin of 6.5%, 7.8%, 8.9%, and 11%, respectively, establishing a strong dose-response relationship’.

The eagle-eyed will have spotted a potential problem with both of these pieces of evidence. The amended DSD Regulations no longer cover CAH, so the two pieces of evidence referenced above are redundant in proving any performance advantage enjoyed by athletes subject to the DSD Regulations. 

Others have previously pointed out that testosterone is only half the story. They argue that when attempting to assess performance advantage, the receptiveness of an individual to elevated endogenous testosterone must also be taken into account. The DSD Regulations do state that in order to them to apply, elevated testosterone levels must be shown to have a ‘material androgenising effect’ on the 46 XY DSD athlete. 

‘The variation in sensitivity for testosterone is neglected’, repeats Dr. Dick Swaab, Professor of Neurobiology at the University of Amsterdam, in another email. ‘Testosterone levels only give half the story’. Swaab has previously pointed out that in order for testosterone to work on muscles, bones and brains, an androgen receptor and a molecular cascade stimulated by the receptor is needed. Therefore a receptor complex that is more sensitive to testosterone could be viewed as unfair, just as a receptor complex that is less sensitive to testosterone needs more of the hormone to function in a ‘normal’ way. 

His comments have previously been supported by Peter Sönksen, a former Professor of Endocrinology at St. Thomas’s Hospital and King’s College. As illustrated by this article, scientific opinion is divided on whether the IAAF has demonstrated that 46 XY DSD athletes hold such a performance advantage in events run between 400m and one mile that it is necessary to exclude them from international competition. This raises questions as to why the CAS upheld the Regulations, which will presumably be clarified when the full judgment is published.

“Testosterone is crucial”, Ross Tucker, a sports scientist and expert witness in the Semenya case, told PA Sport. “As a concept the IAAF has a very strong position and its final legal argument in Lausanne was excellent. But their evidence is really weak and that was exposed quite early – Roger, Erik and I were calling for the retraction of that paper last year [As previously mentioned]. What they managed to do was sidestep the weakness of their evidence and make a conceptual argument. CAS has clearly decided to weigh the concept and theory more highly than the evidence.”


The CAS Statement and the Executive Summary confirm that the CAS Panel considers the DSD Regulations to be discriminatory. ‘The Panel found that the DSD Regulations are discriminatory but that, on the basis of the evidence submitted by the parties, such discrimination is a necessary, reasonable and proportionate means of achieving the legitimate objective of ensuring fair competition in female athletics in certain events and protecting the “protected class” of female athletes in those events.’

Interestingly, whilst upholding the DSD Regulations, the CAS Panel is at pains to point out that the IAAF may be required to amend them in the future. This is despite it highlighting ‘serious concerns’ about the future application of them. ‘The side effects of hormonal treatment, experienced by individual athletes could, with further evidence, demonstrate the practical impossibility of compliance which could, in turn, lead to a different conclusion as to the proportionality of the DSD Regulations’, reads yesterday’s CAS media statement.

The statement accepts the IAAF’s assertion that the DSD Regulations are a ‘living document’ and may be subject to change. ‘The majority of the Panel observes that it may be that, on implementation and with experience, certain factors, supported by evidence, may be shown to affect the overall proportionality of the DSD Regulations, either by indicating that amendments are required in order to ensure that the Regulations are capable of being applied proportionately, or by providing further support for or against the inclusion of particular events within the category of Restricted Events’, reads its Executive Summary.

In effect, the CAS appears to have ruled that it is OK for sport to discriminate against athletes despite a lack of conclusive supporting evidence regarding why such discrimination is proportional, and despite serious concerns about future application of such discriminatory Regulations, so long as such discrimination is necessary. It also clarified that sport is free to amend and add to such discriminatory regulation in the future. 

Human rights

For Brendan Schwab, Executive Director of the World Players Association (WPA), the CAS Decision highlights the narrow remit of the CAS, especially when attempting to adjudicate on human rights issues. “The reality is that this is a matter that has not been resolved by the CAS, because it is not competent to issue final pronouncements on matters involving human rights”, he said. “Taking a broader perspective, this Decision is so conditional and what the Arbitrators have said is couched in such reservation that it certainly does not create a reliable precedent to introduce similar rules in other sports. It is a very confined decision. And that, in many ways, is the tragedy of it. 

“It seems that it is so targeted at one elite athlete, rather than being a rule of broader application. That being the case, the World Players Association will make sure this decision is not seen as a precedent to create similar Regulations elsewhere.

“It seems that despite the CAS Panel having significant reservations about the practical proportionality of the rules, it’s given them a tick on a prima facie basis. If they had been given a burden to meet, it seems to have been very lightly met, and met despite the reservations of the Arbitrators. 

“This prima facie proportionality is really peculiar. It doesn’t explain how the Decision is proportionate in the way that I read it. Is it proportionate because the impact on the individual is not great? Is it because it only covers a targeted number of events?”

Schwab argues that the Decision underlines the need for sporting bodies to embed internationally recognised human rights principles. “The reality is that the IAAF has only made very general commitments to non-discrimination in its Statutes. It hasn’t fully embedded recognised human rights commitments, as FIFA has. Therefore this case is going to be run on a much narrower legal basis than that on which it should be run.

“Given the inherent institutional bias of CAS towards the sports governing bodies, which comes out in the Chand decision where the CAS is at pains to point out that the IAAF has been acting in good faith – which is almost gratuitous obiter – we cannot be surprised by the Decision. It reinforces that two key planks are required. The first is that the IAAF should fully embed human rights; the second is that the CAS should be reformed so that it provides a human rights compliant grievance mechanism that meets the standards of the United Nations Guiding Principles [on Human Rights].

“When those two aspects are met, then athletes may have some confidence in bringing human rights related matters in front of the CAS. Until those two requirements are met, athletes must take enormous care in choosing which legal forum in which to fight these matters.

“What’s interesting is that had the human rights framework been applied to this case, then the essential due diligence would have required a deep understanding of the potential impact not only on Caster, but on all athletes that would be affected by the rules. Where those concerns exist, one shouldn’t move forward unless the risk is fully understood, and any risk mitigated or prevented. 

“In this instance, the fact that concerns exist should really be enough to prevent the Rules from going into effect. People should only be asking others to take drugs or undergo medical treatment whilst fully knowing what the effects will be. When that’s not done, it just seems foolhardy to persist. Athlete’s can’t be scientific guinea pigs.”

Caster’s future

Caster Semenya has never revealed her phenotype, or if the DSD Regulations apply to her. It is significant that the IAAF has clarified that the DSD Regulations only apply to the 46 XY DSD phenotype. It could be that Semenya is XX phenotype but has a condition such a congenital adrenal hyperplasia (CAH), which would raise her endogenous testosterone levels, in which case the DSD Regulations would not apply to her. However, this is nothing more than speculation.

If Semenya is covered by the amended DSD Regulations, and her challenge to the Regulations would suggest that she is, a legitimate question exists as to what would happen if she decided to change her running distance so that it falls outside of international events run between 400m and one mile. In both the media statement and the Executive Summary, the CAS Panel is at pains to point out that the IAAF is free to amend the Regulations in the future. If Semenya did change her distance, it would appear that the CAS has given the IAAF a green light to further target Semenya.

It is understood that Semenya plans to appeal against the CAS Decision. It will be interesting to see whether any such appeal takes place at the CAS or outside of it, given the concerns expressed by Schwab about whether the CAS is the correct forum to hear cases that involve human rights issues.

Prima facie failure to protect Semenya

Sport is a celebration of the physical diversity that enables some to run faster, jump higher and be stronger than others, to borrow phraseology from the Olympic motto. It could be argued that higher endogenous testosterone levels fall within a definition of physical diversity, and ought to be celebrated. At the 1996 Atlanta Olympics, a number of XY DSD athletes competed, but it is understood that they did not have undescended testicles or produce testosterone, yet they still managed to qualify and compete. If elevated testosterone is such an arbiter of sporting performance in XY DSD athletes how did these athletes, who are not sensitive to testosterone, reach the Olympic Games?

We now have a situation where sport is free to medicate athletes based on their natural physiology. Due to its association with doping, testosterone has become a dirty word in sport. As previously explained, if you add exogenous (external) testosterone to an athlete’s physiology, you are doping. The DSD Regulations regulate an athlete’s endogenous (internal) testosterone levels. Adding exogenous testosterone gives an athlete’s natural physiology something to build on which it didn’t have before. As previously explained, it is debatable whether elevated endogenous testosterone does the same thing, as an athlete’s physiology isn’t getting anything new.

Even if elevated endogenous testosterone does provide the athlete with an advantage, is it so unfair when compared to other physiological variables that it is necessary to exclude the DSD athlete from female international events run between 400m and one mile? If testosterone provides the huge advantage to DSD athletes that the IAAF asserts, then why does it only cover these events? Presumably, the full CAS Decision in the Semenya case will answer these questions when it is published.

Semenya’s battle with the IAAF goes back to 2009, when she was 18. Gender tests were conducted on her after she won gold in the 800m at the Berlin 2009 IAAF World Championships. “She is a woman, but maybe not 100%”, Pierre Weiss, General Secretary of the IAAF, told media at the time. “We have to see if she has an advantage from her possibly being between two sexes compared to the others”.

If Semenya had fitted sport’s visual perception of what a woman should look like, would she have been forced to undergo her ten year battle? Unless every female athlete is forced to undergo a test to determine if she has a DSD, can the Regulations be considered fair and proportionate?

‘The Panel also stresses that while much of the argument in this proceeding has centred around the “fairness” of permitting Ms. Semenya to compete against other female athletes, there can be no suggestion that Ms. Semenya (or any other female athletes in the same position as Ms. Semenya) has done anything wrong’, reads the CAS’s Executive Summary. ‘This is not a case about cheating or wrongdoing of any sort. Ms. Semenya is not accused of breaching any rule. Her participation and success in elite female athletics is entirely beyond reproach and she has done nothing whatsoever to warrant any personal criticism.’

Yet despite the CAS’s assertions, its Decision has failed to protect her from speculation that she is a man and not a woman. Despite the CAS Panel’s concerns about the DSD Regulations, it has given them the green light. As such, the media has already reported that she has ‘lost’ her case and has reported that she will have to take oral contraceptives in order to compete in the 800m and 1,500m at the IAAF’s Doha 2019 World Championships, even though Semenya has yet to confirm that this is the case.

Whether the CAS wants to admit it or not, this creates the perception that her past athletic achievements were ill gotten, because she enjoyed a performance advantage due to higher endogenous testosterone levels. Because she is somehow ‘male’ and not ‘female’, since the DSD Regulations only cover the XY phenotype. Again, we do not know for certain that any of this is the case, but it is easy to see how many might jump to that conclusion. 

As pointed out by Schwab, a danger is that the IAAF and other sporting bodies will become emboldened by this decision and will push forward with similar Regulations. For example, transgender and transitioned female athletes are covered by the International Olympic Committee’s (IOC) Transgender Guidelines, published in 2016 (PDF below). These mandate that in order to compete, transgender and transitioned female athletes must reduce their testosterone levels in serum to below 10 mol/L for 12 months before competing. It is understood that the IOC was waiting for the outcome of the Semenya case before deciding if these should be amended for the Tokyo 2020 Olympics.

It is interesting that in a case involving human rights, the CAS appears to have placed the burden on the athlete to prove that the DSD Regulations are ‘invalid’. The burden is also on the athlete to prove that concerns about the DSD Regulations ‘negate the conclusion of prima facie proportionality’, to use the CAS’s words. At first sight, it would appear that the IAAF faced little evidential burden in proving that the DSD Regulations are valid and proportional.

If this assessment is accurate, it supports Schwab’s conclusion that the CAS is unable to escape the narrow confines of its creation in order to rule on complex cases such as Semenya’s. Yet her legal team are now locked to the sports jurisprudence system. ‘The CAS award may be appealed at the Swiss Federal Tribunal within 30 days’, reads the CAS statement. The Swiss Federal Tribunal can only accept appeals on a limited number of grounds, including whether it wrongly accepted or declined jurisdiction; or if the award is incompatible with public policy. This could provide Semenya with a window to appeal, but it is a small one.

The CAS has prima facie failed to protect Semenya from suspicion that her athletics victories were due to advantages from raised endogenous testosterone levels due to her being XY phenotype, even though we do not know that this is accurate. The IAAF may think that they have won this case, but it has failed to learn from the broader sport and human rights debate that is currently underway. On a prima facie basis, it would appear that the CAS has helped sport to put its blinkers back on.

You may also like...

Pin It on Pinterest

Share This