In defence of Semenya et al, many argue: 1. athletes with 5ARD are female; 2. features associated with 5ARD are normal female variation; 3. these athletes should be included in female sports.
The first claim is incoherent.
To understand 5ARD, let's look at healthy reproductive development.
Both male and female development are well-understood.
Male development 1. Y chromosome carrying functional SRY that directs testes development 2. testes produce hormones, notably testosterone (T) 3. T first drives male internal genitalia development 4. T>DHT conversion drives male external genital development
Female development 1. No Y chromosome and no SRY, so ovarian development is triggered 2. low/no testosterone production from ovaries 3. female internal genitalia can develop in low/no T environment 4. female external genitalia can develop in low DHT environment
I often describe reproductive development as:
Sequential | structures develop in a known chronological order
Coordinated | reproductive anatomy develops as a system, with each part linked mechanistically to others
The "cluster account" of sex attempts to classify individuals by the sum of their parts.
"These Male Bits plus Those Female Bits equals This Sex."
This completely ignores developmental biology, which describes the sequential, coordinated development of a functional system.
So given a functional account of sex that actually reflects biology, let's look at the disorder of sex development that is 5ARD.
The sequential development of 5ARD is mapped against healthy sex development here.
Keen eyes will spot that I omitted the word "coordinated" when describing 5ARD.
That's because it's not coordinated. There has been a break in the chain of development.
And that's because people with 5ARD are missing the enzyme needed in a healthy male to convert T>DHT to make a penis.
External genitalia differentiate from a bipotential tissue field.
A DHT signal (in healthy males) pushes the development of that field towards a penis and scrotum.
No DHT signal (in healthy females) pushes the development of that field towards a clitoris and labia.
People with 5ARD have mutations in a gene called SRD5A2 and this means the enzyme converting T>DHT doesn't work properly.
As a result, people with 5ARD have defects in external genital development. We say they are undervirilised (virile = manly 😉) compared to healthy males.
People with 5ARD might have some, low or no DHT.
This means that external genital field might develop as clearly male but with issue like a micropenis, but it might also develop to look like that of a female, or ambiguous between the two.
Consider a person with 5ARD and male external genitalia.
Now work backwards along their developmental pathway. Mostly typical, except for the mutation that has affected penis growth.
Of course, as a male, they have XY chromosomes and an SRY gene and testes and so on...
What about those argued to be female?
Well, they've got a ton of stuff wrong with them.
Their genetic information is all backwards. How odd.
And their hormones? OK, they are up in male range and that's unusual for a female.
And their anatomy? Apart from their external genitals, it's all atypical, unusual, aberrant.
None of the "female" schema makes any sense in terms of developmental biology.
Why would you anchor the endpoint metric as "sex" and therefore have to label everything else about the system "wrong", rather than acknowledge that the endpoint metric may be the issue here?
Especially when you know precisely why that endpoint metric has emerged.
But further, advocates of 5ARD as "female" must face up to the implications of their claims.
A male with 5ARD has a mutation in SRD5A2, and this gene is required for penis growth. This is simple, and maps perfectly onto well-established developmental processes.
"This is the chain, here is the break, this is the result."
A female with 5ARD however? The genetics world will be shook, I tell you.
Because she can't have a mutation in SRD5A2, can she? After all, the sequence she carries is perfectly concordant with being female.
Further, SRD5A2 must now be considered the master switch of sex determination in humans.
Not only that.
Possession of what was previously thought to be a perfectly healthy XY karyotype is now a chromosome defect, and possession of what was previously thought to be a perfectly healthy SRY sequence is now a mutation.
The special pleading is bonkers.
What they desperately try to argue is female with a ton of disparate features is actually, quite simply, male with a vulva.
Learn some dev bio, would be my suggestion.
Even further, according to their schema, Semenya doesn’t, in fact, have 5ARD.
😂
I’m reminded of this passage from Eve by Cat Bohannon.
Genetically male, but external appearance of girl. “They have two testes where their ovaries would normally be”.
Their ovaries wouldn’t normally be anywhere. Ovaries are not part of their developmental sequence.
The chain break is downstream of them being genetically male and having testes.
And the chain break has resulted in them having a vulva where their penis would normally be.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
I think the Andes hantavirus data is being misread right now.
Claims are circulating that the evidence doesn’t seem to support.
I want to walk through them carefully.
The 40% fatality figure: an artefact of who gets counted.
The 40% case fatality rate (CFR) figure comes specifically from hantavirus pulmonary syndrome (HPS), an outcome in a subset of hanta infections.
This is the severe cardiopulmonary presentation. It excludes subclinical infections that resolved without anyone noticing more than “a touch of flu”.
In Jujuy Province, Argentina, the seroprevalence (rate at which people have antibodies due to hanta infection) is 6.5%. Hospitalised HPS cases had a CFR of 13.3%, but most patients were described as having a mild clinical course.
Disclaimer: although an Argentine outbreak, this has not been confirmed as ANDV.
The 40% is the fatality rate among people sick enough to be diagnosed and hospitalised.
It is not the infection fatality rate.
These are not the same thing, and conflating them is causing significant confusion.
The true attack rate: barely any data.
The Boat had approximately 180 exposed individuals, and around ten cases have been detected. Three - soon to be four, I predict - have died.
Without antibody analysis of the full cohort, we don’t know how many mild or subclinical infections were missed entirely.
Disclaimer: I am a biology PhD, but not virology/epidemiology. Husbandman is a virology PhD. But I’m told I’m good at communicating science, so here’s my take.
#Hantavirus
Humans get hantavirus from rodents who carry it.
Some people went to Argentina birdwatching in a landfill, and were exposed to hantavirus because rodents like landfills.
Looks like one - if not two - people brought the virus onto their cruise boat.
So now we have an isolated boat with an index case: someone who is infected.
That’s not good for the index case. Hantavirus has a high fatality rate, and that’s scary.
A cell layer that has developed to protect your body from the outside doesn’t work like a cell layer that has developed to protect your body from the inside.
The cells lining my vagina are not the same cells, and they don’t have the same function, as the ones wrapping your penis.
There’s a name for what happens when you subject dry-adapted “outside skin” to wet-adapted “inside conditions”.
My vagina - “inside skin” - hothouses a healthy microbiome that promotes health and healing, and imparts immune function onto small humans that happen to come out of it.
Yours? Less so.
My vagina is a muscular organ, adapted to my healthy female function of receipt of peen, expelling menstrual products and pushing out small humans.
Let’s move the discussion from available techniques for sex screening and to matters of process.
Ross @Scienceofsport has described the need for detailed technical documents that inform sports federations in robust implementation of a sex screening policy. I’ll link to his video next.
But here, I’m going to take a wander through running an assay, highlighting standards and procedures.
First, this is Ross’ video of the overall process, highlighting the need for coherent implementation practices. He - correctly - evokes the reams of technical documents used by WADA in their anti-doping programmes.
Even the simplest of lab assays can have pages of instructions associated with it.
So, the assay for sex screening will be detection of the SRY gene. This is the ‘make male’ gene that is the master switch for testes-not-ovaries.
The assays out there are very sensitive and specific. That means they can detect SRY when it’s present, and they don’t give a signal when it’s absent. They aren’t 100% on either metric, but near as dammit.
In 2025, Jon Pike and I argued that exclusion of athletes with androgenising XY DSDs from female athletics is justified, because these athletes are male, not female.
@runthinkwrite This followed a 2024 paper where we, along with Ross Tucker, Tommy Lundberg, Cathy Devine and many others, argued for a return to sex screening to secure eligibility for female sport.
@runthinkwrite @Scienceofsport @TLexercise @cathydevine56 This followed another 2024 paper where we critiqued the (now former) IOC policy on inclusion of trans-identifying males in female sports.
Alvares 2025, n=7, fat mass is higher in females as both absolute and relative values. This is logged as "favours cisgender", which is kinda odd because high fat mass isn't usually considered favourable for sports, but whatever.
TIMS: 16.2 kg (24%). F: 19.5 kg (26%).
But Ceolin 2024 is also logged as "favours cisgender" when their values are: