What is a “woman”?

This post is by Phil Price, not Andrew.

As we approach the Olympic Games, this seems like a good time to think about the rules for deciding whether a person is a “woman” when it comes to athletic competition. As I was doing some searches to find some information for the post, I found an excellent piece that puts everything together much better than I would have. Go ahead and read that, then come back here. (The piece is by Christie Aschwanden,  of whom I think very highly; she wrote a book I reviewed here two years ago).

The issue is: if you’re going to have separate divisions for men and women, then you need a way to define “woman.”  A good way of defining this might seem obvious: if the person has a vagina, she’s a she. That’s the way the international sports governing bodies used to do it, but then that was rejected for reasons mentioned in Aschanden’s piece. Well, how about “does the person have two X chromosomes?”  After nixing genitalia as the criterion, this is what they switched to…but then that was rejected for reasons also mentioned in the piece. Currently,  according to the article, “Female athletes who [have] functional testosterone (in other words, not just high testosterone levels, but also functioning receptors that allowed their bodies to respond to the hormone) above a threshold number [are] not eligible to compete unless they [do] something to reduce their testosterone below the threshold.”  (The original sentence is in the past tense, but this is pretty much the current situation too). 

I think it’s safe to say that most people with an interest in how “woman” should be defined for sporting purposes are unhappy with any of the past standards and with the current one. An issue with the current article is, as it says in the article: “People who go through male puberty are taller, have bigger bones and develop greater muscle mass than those who go through female puberty, said William Briner, a sports medicine physician at the Hospital for Special Surgery in Uniondale, New York, during a session on transgender athletes at the American College of Sports Medicine meeting in early June. Men also have more red blood cells than women, and their hearts and lungs are bigger too. Some of these advantages are irreversible [even if testosterone levels are later reduced].”  

Furthermore, although the Olympics focuses only on the best athletes in the world, there’s a need for competition rules that apply at other levels too. High school basketball, for example, needs a way to determine who is eligible to play on the girls’ team. Is it fair for a 17-year-old 6’4″ high school athlete who went through puberty as a male, but then transitioned to female, to be allowed to compete against girls who went through puberty as girls? Viewed one way, being a woman who went through puberty as a male is just another genetic advantage, and we let athletes use their genetic advantages, so there’s no problem with such a person competing on the girls’ team. Viewed another way, it’s not fair to let someone compete as a girl if they got their body by growing up as a boy. 

So what do I think the rule should be? I have no idea. Indeed, I’m not even sure how to think about what the rules are trying to achieve. We have separate competitions for men and women because it would be “unfair” for women to have to compete against men: in most sports the best women wouldn’t stand a chance against the best men, and the average woman wouldn’t stand a chance against the average man. A female sprinter, for instance, would have no chance of reaching the elite level if competing against men. OK, fine…but what about me? I’m a man but I would also have no chance of reaching the elite level if sprinting against other men. Indeed, I would have no chance of reaching the elite level if I were sprinting against women! The fact is, very few people have the genes (and other characteristics) to compete at the elite level. If the point of the rules is to give everyone a reasonable chance to be among the best in the world in a given category, well, that’s not gonna happen, because no matter how you define the categories there will be only a small fraction of people who have what it takes.  By having separate competitions for men and women we can’t really be trying to give “everyone” a chance. So what are we trying to do?

All of this puts me in mind of a statistical principle or worldview that Andrew has mentioned before, that I think he attributes to Don Rubin: most things that we think of as categorical are really continuous. For some purposes (and with some definitions) male/female is indeed categorical — no male can bear a child, for example — but when it comes to innate ability in a sport, what we have is one statistical distribution for men and another statistical distribution for women, and (for most sports) for any reasonable definition of “man” and “woman” the high tail for men will be higher than the high tail for women, but  the bulk of the distributions will overlap. If  Caster Semenya is a woman for sporting purposes, in spite of her XY chromosomes and typically-male testosterone level, then she is at the very top of the sport. If she is a male for sporting purposes, then she is not remotely competitive at the elite level. (Welcome to the club!). Sporting ability is continuous but we have to somehow force people into two categories, assuming we want to continue the current male/female division in competitions. 

It’s said that “hard cases make bad law” but this seems like a sphere in which all of the cases are going to be hard.

This post is by Phil.

197 thoughts on “What is a “woman”?

  1. I basically agree with Martina Navritilova’s take on this: At the elite (i.e. professional) level, athletes who’ve gone through male puberty should be excluded from women’s competition. Otherwise, if you want to compete as a woman then sure, whatever – the really good female athletes will appreciate the competition although young Becky who’s lost her place on the high school volleyball team to a 6’4 trans-woman may beg to differ.

    • And Greg the 6’2 volley ball player gets the vacated spot on the male team. Do men appreciate the downgrade in competition?

      If Greg and Becky are academically equal then which one is going to look better to the college recruiters?

      • Yes perhaps wrt Greg & Becky. A more likely scenario in elite athletics is Laurel Hubbard, who won a spot on the New Zealand women’s weightlifting team for the Tokyo Olympics. She had long ago aged out of competitive weightlifting in the men’s division, so there was no Kiwi Greg who got a spot on the men’s team by virtue of Hubbard joining the women’s team.

        That’s a more likely scenario because the overlap between women’s and men’s performance distributions is very limited in most elite athletics involving strength, speed, or endurance. The searchable database at worldathletics.org has many examples.

        In 2019, Sifan Hassan ran the fastest 1500 meters by any woman (3:51.95); in the same year, men ran faster than that over 5000 times in competition. A median time from the men’s distribution (~3:54.00) would have been faster than the second-fastest woman in 2019 (Faith Kipyegon, whose fastest time was 3:54.22). IOW a random draw from the male distribution of elite performance is likely to be the best/fastest/strongest female performance if that male is considered to be a woman. That leaves a lot of scope for unfairness to females.

        That doesn’t speak to the question of how best to mitigate the unfairness to trans women as well as to females. In the Caster Semenya case, the CAS agreed that the testosterone rule was unfair to Semenya but that this limited unfairness was justified and necessary in order to treat females fairly. Of course, that wasn’t a case involving trans women, but it’s related in the sense that Semenya has some male biological traits but a difference in sexual development. It’s relevant because the rules adopted out of Semenya’s case (involving DSD) are now being applied to other cases (trans women).

      • Greg. Girls/women have higher grades etc and are more likely to attend college, but at elite schools they want to ensure that there is a gender balance for social reasons (which many of the women appreciate).

  2. I was talking with a friend who coaches high school. He was saying how in the discus throw, in high school, men start out at 1.6 kg and women at 1 kg. In college, men move to 2 kg, women stay at 1 kg. Professionally, men are still 2 kg, women remain at 1 kg. By the time women hit pro, it actually becomes a problem – it’s too light and too small to get a proper grip on it, apparently most female discus throwers wish that they used the men’s sizes.

  3. I second the point that the international contests are for extremely elite individuals. How about if we define eligibility for events the way automobile racing does; they have specific criteria for Formula One, Indycars, stock cars, etc. I am personally a high mileage junker in such a system, but I’m happy to watch Daytona without thinking that Scuderia Ferrari is being unfairly excluded from that race.

  4. > I think it’s safe to say that most people with an interest in how “woman” should be defined for sporting purposes are unhappy with any of the past standards and with the current one.

    I suspect that the number of people with an interest in the (re)definition of “woman” is much higher than it used to be. And it’s possible that most of them were actually happy with the “obvious” standards of the past.

  5. Interesting and timely topic Phil

    “We have separate competitions for men and women because it would be “unfair” for women to have to compete against men”

    At the lowest level at any rate we have separate competitions because that allows people to compete against others who are of approximately equal ability, regardless of the reasons for the differences (age, and ability like varsity and JV in addition to gender). This is done more for educational/developmental reasons than because of the fairness of the outcome: it’s more beneficial to compete against someone of approximately equal ability than to compete against someone of much lesser or much greater ability.

    How that should translate into the highest levels is more difficult to say, since the stakes are much higher, the competition is professionalized, and the underlying reason for it does become more about “fairness” of the outcome.

    “the bulk of the distributions will overlap. ”

    Is this true? This is a difficult comparison because performance reflects both ability and training, and probably the only way to gage the innate differences between men and women would be to compare people with no training at all, since otherwise it would be difficult to control for the amount of training. Surely Phil if you had trained in tennis from age six, you’d be at least much better competition for Serena Williams than you (presumably) are today, (presumably) with little or no training. If you had trained equal to her, would you be able to beat her? We’ll never know.

    I’m willing to bet a modest part of the distribution would overlap – very roughly comparable to the differences in height of men and women. Quickly looking for data, on height it looks like about 35% overlap.

    • >Surely Phil if you had trained in tennis from age six, you’d be at least much better competition for Serena Williams than you (presumably) are today, (presumably) with little or no training. If you had trained equal to her, would you be able to beat her? We’ll never know.

      I’m pretty sure Serena would win that one (No offense Phil!) The odds are in her favor given that she’s already shown she’s top of class in women’s tennis, plus tennis seems like a sport where the best women have much better odds of beating the men than something like discus or weightlifting.

      • Yeah this isn’t even close to true. She’s an excellent tennis player, but (like most female athletes) would lose to a great high schooler and mediocre college player. Turns out testosterone is a helluva drug, and the athletic advantages of differently angled hips/femurs (since men have no need to bear a child), denser bones, and harder-pumping heart conspire (with other stuff, too) to give men significant advantages in athletics.

        Coming from the gymnastics world, the governing bodirs have never been shy about embracing the differences in the two sports. Women have a lower center of gravity (and rotate differently), have proportionally stronger lower bodies, etc, and the divergence in the sport is a wonderfully beautiful thing!

        This all matters because women deserve to have sports that are competitive but fair. I’m not sure what the best solution is, but I don’t think the best solution is a world where all the best women went through male puberty. I wish this was easy.

        • Perhaps you are too young to remember the Billy Jean King/Bobby Riggs match? Of course, it was featured in a recent movie, but perhaps you don’t go to the movies either.

          • Not really the best example – Bobby Riggs was 55 when he played Billie Jean King. And he had beaten Margaret Court a few months earlier. What t-r-t writes is spot on.

          • You’ve actually made my argument here (and highlighted part of why many are upset about the NZ trans powerlifter). Prime athletics age women can beat boys and old men on what is (only then) a level playing field. The only older women who can compete in powerlifting against younger women went through male puberty.

            None of this was controversial until what feels like minutes ago. I think we all want what’s fair to athletes while also respecting that the line between male and female isn’t as bright as was once thought.

          • You might not believe me, but I really do care about women’s sports (and when my daughter is old enough I hope she can participate).

            I’ll use gymnastics as my reference frame again because that’s the sport I know most intimately. Before I stopped due to injury I was mediocre, I’m not going to pump myself up. But even at 14-15 I was performing vaults that were winning medals at women’s World Championships. The risk in having a poorly defined women’s category is that the ambiguity never goes the other way. Simone Biles is a generational talent, but comparing her to men diminishes her excellence (and I wasn’t trying to do that regarding Serena).

            The incentives around this matter. If girls risk being crowded out by athletes who experienced male puberty, we might not see the inspirational women that have been discussed here. “The best way to be a woman is to have been a man” goes against about 120 years of feminism.

          • +1 Ridiculous comment. Put any mediocre college tennis player in Flushing Meadow or the main court at the Australian Open. They would soon learn that the game at that level is about a lot more than strength.

        • trt –

          > She’s an excellent tennis player, but… but would lose to a great high schooler and mediocre college player.

          Can you provide evidence to support that statement?

      • I think Serena, at her peak, would have been about the thousandth-best tennis player in the world. (She played against some guy ranked around 500, and he was substantially better than her. But perhaps with a small amount of practice against someone with his speed and power she would have been able to adjust; that’s what I would guess). Maybe she was #800 in the world, maybe 2000, I’d think almost certainly in the top 3000. If I had dedicated my life to tennis, could I have made the top 1000 or 2000 or 3000? Mmmaybe. I’m fairly tall, fairly athletic, was fairly fast, had fairly good hand-eye coordination. But, well, read David Foster Wallace’s article “The String Theory.” There are about 90 million tennis players in the world and at least 90,000 of them are young and fit and take it seriously. Basically I agree with Jessica on this: to be better than Serena you don’t have to be an elite male player but you do have to be very very good. It’s not out of the question that I could have been that good, but it’s definitely not the way to bet.

        To look at something less speculative: I was faster than any of my Ultimate Frisbee teammates over several years (and most opponents), including a guy who had run a 52.2 400m a few years earlier. I was fast enough to be proud of how fast I was, but I would still have lost to the fastest women by forty meters or so..

        • There is a semi-famous story about Serena and Venus claiming they could beat men ranked outside the top 200 in the ATP. They tested the claim by each playing a friendly one-set match on the same day in 1998 against Thomas Braasch (then ranked #203). Serena was only 16, but Venus was older and made the US Open final that year. In the story, Braasch is said to have played a round of golf in the morning, drank a few beers in the afternoon, then won each set 6-1 or 6-2. Again friendly, not competitive, and anecdata.

          Not sure I agree that “probably the only way to ga[u]ge the innate differences between men and women would be to compare people with no training at all, since otherwise it would be difficult to control for the amount of training.” Seems better to compare elite athletes who have all been highly trained and are all highly motivated to win, otherwise it would be difficulty to control for degree of interest and motivation.

          • Yes. But Serena was in her teens at the time and was a much better player at her peak. And although Braasch was ranked around 200, he had been as high as #40, just way more experience.

            In any case, whether Serena would be the 1000th-best or 10,000th-best player in the world, she’d still be better than the vast majority of male players, even if not nearly good enough to be a male professional.

            I think a better way to address the general question of male/female differences in athletic performance is with track and field (and road racing), where we can simply compare actual times and distances. Elite women are nowhere near elite men but easily better than Joe Schmoe. To give an example: according to runrepeat.com, using data from amateur road races, a 40-minute 10k puts you in the top 0.8% of female runners age 20-29, and beats more than 96% of male runners in that age group. So, yes, some men beat the fastest women, but the fastest women beat the vast majority of men.

            If you only look at elite events then you’re comparing the high tail of men to the high tail of women, and it’s hard to judge the overlap of the distributions that way.

            • Yes agree with all those qualifications about Serena vs. Baasch. It’s an imperfect comparison.

              “Elite women are nowhere near elite men but easily better than Joe Schmoe. ”

              Yes for sure. But Flo-Jo and Joe Schmoe differ in many ways other than female vs. male physiology. Comparing them doesn’t lead to a measure of sex differences in athletic performance. The clearest comparison is between male and female elite athletes with similar training and motivation. The performance difference between the elite groups is more likely to be mainly caused by sex differences in physical ability. At least I think so.

  6. I have followed this topic and related ones with detached interest and find distressing the heat and vitriol that seem to develop very rapidly among discussants. Gender identity makes people very, very angry and they accuse others of denying or erasing other’s identities or, worse, of threatening their lives. I hope this comment thread can be different.

      • What bothers me about the attention this conversation is getting that there’s issues in groups that have a larger proportion in, for example the US population with clear inequity in income, that aren’t being discussed. :) I definitely agree that non-binary people should be allowed to choose gender identity, and should be allowed proportional representation in employment. But there’s bigger fish to fry :). By bigger fish, of course, I mean more people being affected. Of course, the subgroup I’m vaguely to earlier doesn’t exclude the community referenced in this blog post.

        • For almost any issue we could discuss, you could say “there are other issues that are much more important.” And you’d be right. I suppose one could argue that people should only discuss the very biggest issue, or the five biggest, or the ten biggest. But in fact people like to discuss all sorts of things, like which movies are best, and whether baseball games take too long, and why the sea is boiling hot, and whether pigs have wings.

        • With this reasoning is relevant to consider that frying smaller fish as you say, may potentially help in preparing bigger fish too. It does not have to detract or be detrimental to other egalitarian goals; it could actually open the door for other changes.

  7. For individual events, I suggest we use the “Robust Beauty of Improper Linear Models”. Decide on a number of criteria that are associate with increased sporting capacity in each sport. Find the average of these measures in the general population, then create a score sum(M[i]/mean(M[i])) for each of the measures i. Then segment that measure into deciles and offer categories in each decile (obviously deciles should be enough to take care of nonprofessionals, but for professionals you’d probably only use the upper 4 or 5 deciles.

    For team sports, you could offer a budget for the team or a similar strategy (like, limits on the max/min values or the variance or the mean absolute deviation)

    • obviously in the super-elite olympic range you probably would have to segment it differently because probably all athletes would be in the top decile. But you’d still just look at the distribution of the score among these elite and sub-divide it into reasonable subcategories. This probably would result in some medium elite men competing with top elite women… but I don’t think that’s necessarily a major problem.

    • This is not a workable solution, for obvious reasons. Persons who participate in sport do so to win. If we have multiple categories, the simplest way to win is to cheat on the segmentation exam. The only approach that could work is to segment on a component of person NOT related to the competition.

      The clear example in male sport is wrestling. Large males easily beat small males. So, there are divisions based on weight. Within the weight division, skills and strength are the determiner of winning. It would make no sense to segment on strength. Weight is a simple and easily measured segmentation variable. It is also only of limited ability to manipulate – you can drop 1-2 lbs easily, but it’s pretty hard to drop 15.

        • We could also add in some of the relevant sex hormone measures. such as “had at least one functioning testicle at some time between the ages of 10 and 25”, which we could have rules that simply presume a yes answer if you don’t currently have a functioning ovary or a documented medical history.

          For the most part this is a 0/1 indicator divided by essentially 0.5. Suppose we’re using 10 measures, then this means normal “men” would increase their score by 0.5 over normal “women” on a scale that we could imagine would have a mean of 10 and a standard deviation of say 2 and be fairly normally distributed. An extra 0.5 on this measure would on average move you about 7-8 percentage points, enough that it would distinguish between the top decile and the second top decile and do most of the work of separating “normal” men and women without actually needing to do much else

        • length from shoulder to finger tip, length from hip to heel, overall weight, span from thumb tip to little finger tip, circumference of the bicep, circumference of the thigh, length of the femur, heart rate when running at 3 different speeds on a treadmill…

          there are tons of useful objective measures that are very hard to cheat which we could use to build such a score.

      • Wrestling is in fact an interesting case. It is stratified by a continuous variable, weight, as you mention.

        Yet nowhere do I see you advocating routine testing for testosterone levels across sports for cis and trans people alike. Surely if transwomen have such an advantage due to hormones, ciswomen who naturally have somewhat elevated testosterone levels have an advantage over those who do not as well?

        Being that we’re commenting on a blog which often comments on the misuse of p-values by treating them as binary rather than continuous in order to lend credence to a pre-decided conclusion, I hope the irony is not lost on my fellow readers that you’re committing the same error here.

  8. I think this is key: “to think about what the rules are trying to achieve.”

    It occurs to me that the rules for sports are for the benefit of the whole of society, not just for the competing athletes. If that’s the case, what’s “fair” in sports can be different from what’s “fair” in a general competition for status and power. So, I don’t know what the rules for sports are trying to achieve, but surely you’d have to be very careful to not automatically transfer fairness arguments that apply to the role of gender in society, to sports and especially elite sports. And vice-versa, off course.

  9. This is an unpleasant topic to discuss, because it’s an area where people who actually do hate trans people astutely recognize that the demands of (some) trans women athletes can be offputting to people who don’t otherwise have strong opinions on trans issues.

    • I think it would be a mistake to let the haters squelch the conversation. This blog has a great record for commenters making fair-minded, relevant comments; I’m expecting this post to maintain that tradition.

    • So, to placate the miniscule number of trans bullies, we are to not discuss this, and simply allow these cheaters in to dominate whatever sport they wish?


      And it is not an “unpleasant topic” to discuss. It is important. What is unpleasant is the denial of reality and the complete confusion being DELIBERATELY introduced by the trans bully lobby.

      There are 2 sexes, and females are lower in ability. A simple solution is 1) stop calling them “trans women”, and use the correct term “gender-dysphoric males” 2) deny males the ability to participate as females.

      It’s not complicated. It’s unfair to the females. Period.

      • Paul:

        Just because you prefer a term, that does not make it correct. Just because you say “it’s not complicated,” that doesn’t make it not complicated. Phil’s post above has a long discussion of how it is complicated. Also, it’s hardly a “simple solution” to not call people what they want to be called.

        • Sorry, the term “gender-dysphoric male” IS the correct term. This is the medical term for this condition. The term “trans woman” has been devised by this group, and is the main reason for the confusion now evident in this question. Use the accurate term.

          • Paul:

            This particular discussion is going nowhere. Language is a social and even a political process, and fighting linguistic changes can be like trying to push aside the tide.

            I’m not saying that one should never try to change language. I’ve had many impassioned takes over the years regarding what I consider confusing conventionally used statistics terms such as “bias,” “random effect,” “noninformative prior,” and many others, and sometimes I think my linguistic activism has made a difference. But using all caps, insisting that your preferred terms are “accurate,” and using an imperative mood . . . that won’t take you far. When you want to call people something that they don’t want to be called, you’re kind of coming in with two strikes against you. I’m not saying that I’ll always call people what they want to be called, it’s just that these sorts of things can be complicated, which again is related to Phil’s post.

            • The “problem” begins with language. Let’s say you have a horse that you name “Big Dog”, and you wish to enter the “Big Dog” into a dog race. Those who argue for “fairness in dog racing” will rightly say that your attempt to name the horse a “dog” is not enough to get the animal into the dog race.

              The entire issue of the gender-dysphoria is with us do to the language that is being used. When a person says “I am a woman”, they wish entry into female spaces. So, we end up with Wi Spa in CA (physiological male exposing himself to minor children). Because he claims to be a woman, the spa employees are unable to keep him out of female spaces. If the correct term is used of “gender-dysphoric male”, the ability to keep him out would be obvious.

              We do not allow self-identification of “physician”, “lawyer”, “airline pilot”, “policeman”. In our society, we have a credentialing system. When I obtained my PhD, this gave me the ability to call myself a “Dr.” within limited contexts. It did not give me the ability to perform as a physician, which is another form of “Dr.”. I am able to operate on “data”, not “knees”. So terminology IS important. It’s important in this area as well.

          • Although gender dysphoria is a related subject, it is not (at all) the only relevant one. I don’t think that term applies to María José Martínez-Patiño or Dutee Chand, for example.

            • Again and again, I have made the point that conflating different things does not clarify. There are at least 2 situations: 1) persons with gender-dysphoria 2) persons with DSD. Rather than actually addressing that issue, you further confuse the situation by bringing in a 3rd situation. Putting 3 different things together and pretending that they are all the same does not lead to clarity. That’s one of the main reasons this issue is “confusing”. Because it is not one issue, rather it is 3 or more.

        • You note that “Phil’s post above … how it is complicated”. Phil (and you) should read some of the voluminous literature in this area:




          These discuss the issue in a more careful and more physiologically informed manner.

      • Paul – You seem to care an awful lot about how the rules govern how people chase a ball around or pick up and hold heavy things. Why? What purpose does it serve to get as hysterical as you are being over sports? You say it’s important, but… why? It’s sports. It’s not intrinsically important. It’s also not about fairness because you don’t seem to care about unequal advantage within a group (why aren’t you arguing for divisions in sports the way something like boxing has weight class divisions?). What purpose could it possibly serve you to get this emotional about making sure the rules governing organized athletics keep people separated on the basis of a complex collection of topics that you are almost assuredly not an expert on?

        • You said it better than I could. Generally I prefer not to speculate about motive, but what drives someone to make 16 mostly combative/inflammatory posts on a single comment thread discussing what is ultimately a fairly minor point, (one that is entirely academic for the overwhelming majority of everyone alive), in the broader conversation on trans issues?

  10. As other commenters have discussed, this is a fraught issue given how little the average person knows about trans people, the biology of sex, and the psychology of gender. However it is my experience that this blog’s comment section thankfully is above-average!

    I’ve personally been rather persuaded by Sophia Hottel’s argument here (https://transliberalism.substack.com/p/trans-athletes-revisited). Now, Sophia is a trans woman which you could argue introduces bias, but I would argue that it’s a bias that is totally subsumed by (a) the fact that we *know* the bias, whereas we do not know if specific medical researchers or the IOC or parents of the kid who came in 2nd place have an agenda to push (potentially even subconsciously), (b) the fact that no one knows more about how transitioning affects their body’s physiology than the people undergoing it, feeling it every day.

  11. As Daniel Lakeland alludes above, the real problem is that there are only two categories: men and women, and a continuum of abilities in a host of dimensions. This problem is sort of solved in poker by having different rooms operating at different stakes with individuals self-selecting into the stakes that reflect their abilities. Though the best poker players in the world could clean up at nickel poker, cleaning up isn’t worth their time at such low stakes. Horse racing operates similarly.

    What you would ideally like to have (and I can’t think of a way to make this practical) is enough divisions that people will self-select into divisions in which they are competitive amongst each other, but moving up in class is bad because you can’t win, and moving down in class is bad because the returns to winning (financial, personal, or what have you) are too low. In equilibrium, there will be divisions with substantial numbers of men and women (by whatever definition) competing. In some sports (tennis is a perfect example) this would be very, very bad for women’s pay, since as Phil and others point out, there would be little prize money in the division with the 1000th best men’s player in the world, the division which would capture elite women. Which raises the question, do we have prize parity in tennis for the quality of play, or the quality of the competition?

    Of course, at the top of each distribution is the competition to be “The Best,” which will always attract entrants outclassed. But this isn’t the gender problem…

  12. All of the women whose testosterone levels are being regulated are biological males, with XY sex chromosomes and internal testes: https://www.sportsintegrityinitiative.com/dsd-regulations-call-out-athletes-as-biologically-male/

    Virtually none of the press coverage of this issue notes this crucial fact. No female athletes are affected by these rules. For nearly a decade, media accounts of Caster Semenya’s disqualifications described her as someone who was born female or as a cisgendered woman. Neither claim is true.

    • You have a point that reporting has been misleading, at least some of the time. However, I think it’s misleading to say without qualification that she’s “biologically male” because it fails to distinguish intersex individuals assigned the female gender, who generally have a physiological basis for their gender assignment and may not be aware of their “biological” sex. Yes, the medical definition of “biological male” is based on having one X and one Y chromosome, but there’s no necessity that we give special power to the medical definition. Her chromosomes aren’t even strictly the basis of her disqualification, so it’s relevance is questionable.

      Certainly, those who made up these rules weren’t like, “Argh, we really wish we could call her a woman, but this medical book says we can’t!” They went searching for a definition that would *allow* them to call her a man, based on their sense of sporting fairness. Sporting fairness is a social construct, even if it involves biological traits like hormone levels and sensitivity, or physical realities like rate of development or relative performance. Calling her a man for competition purposes is therefore by definition a gender assignment being made by a governing committee, not an objective sex determination being detected by the committee. I find that far more disturbing than the prospect of a “biological female” being disqualified on the basis of her hormone levels.

    • Your logic is circular and self-serving.

      Biological sex is not sharply well-defined. For most people, of course, it’s clear, but intersex people absolutely exist.

      Moreover, this is orthogonal to the issue, because trans people and trans advocates reject the notion that being born with XY chromosomes or testes necessarily decides that that person is not “female” for the rest of their life.

      This is a physiological discussion. Linking to the “Sports Integrity Initiative” betrays a woeful bias and lack of curiosity in actually understanding the issue.

  13. Maybe the answer is just whatever policy maximizes medium term expected profit for the IOC. Should we imagine it has ever been otherwise?

    Related issues include runners with prostethic legs (cf. Oscar Pretorius) and the free-throw shooting robots that ended up winning lucrative endorsement contracts in Kurt Vonnegut’s imagination. So, what counts as human? Body temperature between 33 and 41C for one thing? This will bring the “hot-hand” issue to a new realm.

    Maybe it’s interesting to contemplate women’s and men’s Bridge tournaments. While men dominate the top-level of the game, I guess it is not clear that they have any advantage a couple of tiers below that. For some, Bridge is ideally played in isolation, over computer terminals, to prevent subtle signalling bewteen partners. Does that change your analysis?

  14. The “problem” of “trans” persons in female spaces is not as complicated as it is being made out to be.

    There are 2 situations, and they should not be conflated:

    1) “trans woman” participation: The real problem begins with the term “trans woman”. These persons are not women. They are gender-dysphoric males, and have multiple dimensions of male advantage – muscle fibers, heart, lung, testosterone, bone density, hip structure. Call the males “gender-dysphoric males”, and it is pretty obvious that male persons should not be in female sport competitions.

    2) Persons like Semenya, with DSD (disorders of sexual development): Many of these persons are XY persons with developmental abnormalities. Again, they are male in chromosome, and have many if not all of the male advantages. They should be excluded.

    • You mention you have a PhD upthread. Therefore I assume you are familiar with the notion of axioms in a philosophical or mathematical argument?

      You are operating under a different set of axioms than trans people and their advocates, yet are trying to say their axioms are wrong using arguments *that follow from your own axioms*.

    • Is Michael Phelps a ‘male’? a ‘human’? He certainly possesses characteristics that are well outside the scope of normal human male. This likely applies to many elite level male athletes. To use your words, they have…”muscle fibers, heart, lungs, testosterone, bone density, hip structure” that, taken together, exist well outside the most extreme percentiles of the normal male spectrum. Your dichotomy is just that…yours.

      There have been no good objective criteria developed to address the issue of sex and gender in sports participation. Coming up with one would almost certainly have some interesting unintended consequences.

  15. I’ve read the piece by Aschwanden. It is not a well-written piece. It confuses 2 very different situations, and this leads to “the perplexing situation”. I clarify the issue in the comment which you did not address.

  16. Those who wish to read articles which discuss this from the perspective of physiology and kinesiology should examine the following articles:

    https://jme.bmj.com/content/45/6/395 “Transwomen in elite sport: Scientific and ethical considerations”


    These have clear and well-written discussion of the multiple dimensions which should lead to the exclusion of gender-dysphoric males from female competition. Sport involves bodies. Gender is not important. It is bodies which determine capability and capacity.

  17. Andrew has pushed the Aschwanden piece. Well, there are plenty of reasons to be skeptical of that piece.

    Here’s an obvious one: What does Caitlin Jenner say about this?

    Aschwanden: I asked Jenner whether hormones were a reasonable standard and if it was fair for her to compete as a female. “I have seen no indication to this point that trans people, male or female, have any advantage whatsoever at that level,” she said. “There’s no trans person out there, male to female, that’s out there dominating. It just doesn’t happen.”

    Jenner: Caitlyn Jenner, the former Olympic champion and reality TV personality now running for governor in California, said she opposes transgender girls competing in girls’ sports at school.

    Jenner, a 1976 decathlon gold medalist who came out as a transgender woman in 2015, told a TMZ reporter on Saturday that it’s “a question of fairness.”

    Jenner is saying that gender-dysphoric males SHOULD NOT compete against females, and that it is a matter of fairness. Aschwanden has either misquoted Jenner or quoted Jenner out of context.

    • Paul –

      I’m assuming you agree with Jennifer on this

      > “a question of fairness.”

      If I”m right about that…there is a choice among an infinite number of criteria that could be used as an exclusion criterion.

      What is the conceptual basis on which you determine which criteria are “fair” to use and which aren’t?

  18. My solution would be to get rid of dividing sports by sex/gender and just let everyone compete together. This is just one more bastion of segregation.

    Part of my problem with this discussion is how asymmetrical it is. The discussion is always focused on trans-women participating with cis-women. This has always struck me as a huge red flag that the primary reason we’re having this discussion in the first place is because cis-straight men want to continue ogling women athletes and not feel kinda gay about it.

    By centering the discussion on women, trans and cis, the discussion completely ignores so many crucial aspects of the discussion that show how incoherent and inconsistent the arguments against trans participation in sports are.
    -Rarely is it ever even acknowledged that trans-men are participating with (or are trying to participate) with cis-men, and proportionally speaking have won just as many competitions as trans-women have; which is to say, very very few in raw numbers, but approximately proportional to the number of trans athletes participating (this is a nice challenge to the biological essentialism arguments about advantages and disadvantages).
    -What about non-binary people who want to play organized sports? Why the heck are we expecting people to publicly lie about who they are for the purpose of athletic competition? In what universe is this anything other than an assault on a person’s agency and autonomy?
    -We have no problem with unfairness within other arbitrarily defined groups of people, so it’s not really an issue of a level playing field either. At one point in the 90s, the height difference between the tallest and shortest men’s NBA players was over 2 feet tall (Manute Bol and Muggsy Bogues, who both coincidentally happened to play for the same team) and we never had national or international discussions about having different height divisions in basketball in the interest of fairness.

    We just want to keep reifying certain abstract groups of people that we’ve decided matter for no other reason than because they’re the easiest to implement systemic injustices around. Let’s maybe just stop it altogether and let people play together.

    • Emilio, I think I understand where you’re coming from here. Like you, I don’t want trans people (or anyone else) to be treated unfairly. I disagree with your suggestion but don’t want to contribute to a negative tone to this thread. I’ll just say that I don’t think “women” is an abstract group of people, and I don’t agree that women matter only because it’s easy to implement systemic injustice around them. I think there are good reasons to have women’s athletic events that exclude biological males.

      • I might point you to Abby Thorn’s recent Philosophy Tube video on Social Constructs, then, as it is pretty relevant here despite spending basically zero time on discussing directly the idea of trans athletes. And I might point out as well that even your response is 100% centered on trans women and cis women, with no acknowledgement of trans men or non-binary athletes at all. This is part of the problem with the whole discussion, particularly with regard to the incoherence on the side arguing to keep trans-women separate from (and unequal to) cis-women in the athletic arena. If you cannot discuss the idea of fairness or justice without ignoring entire segments of the population (whether unintentionally or not), that’s a serious problem! And it is a problem either with the underlying motivation for wanting to frame the discussion in terms of fairness/justice in the first place, or with the proposed solution(s) that people are proffering to address the issue.

      • Mike –

        > I think there are good reasons to have women’s athletic events that exclude biological males.

        What are those good reasons? I”m trying to get a sense of how people determine which exclusion criteria are “fair” or “good,” except in what IMO, is an arbitrary manner.

    • Emilio –

      This is my view, also:

      > -We have no problem with unfairness within other arbitrarily defined groups of people,

      Essentially any choice is arbitrary (not in the sense of random), and IMO, thinking otherwise because that’s just how they way it is, is part of the problem.

      On the other hand, you say…

      > …maybe just stop it altogether and let people play together.

      In some abstract world, or maybe in a future world, that might make sense. But I’m not sure how that’s a practical solution given where we are now. Is there any non-zero sum way to get a hold of this issue?

      • Mike,

        I have no problem with the idea of using arbitrary categories to separate divisions of competition, per se. We do it for boxing, where we have over a dozen weight divisions, each separated by roughly 3-5 pounds of weight. That’s pretty arbitrary, but it seems to work well enough. We do it for kids athletics, where we create divisions on the basis of age, which can create its own kinds of unfairness because the oldest X-year-olds will tend to be a little bigger than the youngest X-year-olds. We aren’t having huge discussions about that. It seems we’re only really having a discussion about this with regards to specifically women’s sports (not sports in general, but women’s sports specifically), which casts suspicion on the purpose of the entire discussion in the first place.

        I agree that insofar as we create divisions they will be arbitrary. No getting away from that. But maybe we could stop with certain kinds of arbitrary groupings that seem to naturalize/essentialize socially constructed categories that are used for primarily discriminatory ends. It’s not like these categories are static. There’s nothing immutable about these concepts, as evidenced by the very article that Phil linked to. The very concept of woman has been redefined by the Olympics committee multiple times, and weaponized to bar elite athletic women (who are disproportionately non-white, let’s not forget that aspect to the entire discussion) from competing. But I disagree with the assumption you’re making that eliminating the divisions of “mens” and “womens” X-event is a zero-sum affair. I also don’t think this can only be solved in some abstract future world. Why can’t we be a bit more reflective about the categories we create and use to help us navigate/understand the world now? What is the possible harm that could come from getting rid of “mens” and “womens” sports divisions and letting all the competitors play together that could possibly match or supersede the current harms that are being inflicted on womens athletes (cis and trans) today?

        • At the non-elite level, men and women often do compete together. But at the elite level, if you eliminate men’s and women’s division, you’re basically excluding women from competition.

          I would also add the women’s elite sport has never been in a stronger position although there’s still a lot of work to be done.

        • > What is the possible harm that could come from getting rid of “mens” and “womens” sports divisions and letting all the competitors play together that could possibly match or supersede the current harms that are being inflicted on womens athletes (cis and trans) today?

          It’s pretty clear if you make an “everyone” sport unless it’s ultra-endurance races or something then essentially 100% of the pro, semi-pro, or olympic sport, will be “normal” men. Thus shutting women out entirely. I think that’s definitely a worse harm than whatever is going on today.

          On the other hand, if you make divisions based on objective predictors of performance (so basically body morphometrics) and let people compete who are close enough in score to have very little predictable difference in performance, then it doesn’t matter whether they have that score because they’re a normal woman, or a smaller sized man, or a trans woman or a trans man or a mildly obese person, or a person with a genetic disorder of bone growth or whatever.

          • Daniel,

            Perhaps I wasn’t as clear as I wanted to be on this aspect, but I’m kind of arguing on the same side you are. At least in the abstract. We need to change the arbitrary groupings we use to cluster together competitors away from sex/gender to something(s) else. I’m not saying we don’t have boundaries around levels of competition at all (again, weight divisions in boxing seem to be working fine). I’m just saying that some of those boundaries are being used as weapons to deny people the right to compete at all and we need to move away from those. Especially when they have the added effect (which is perhaps also a cause as to why they’re being used in the first place) of endorsing a genuinely pseudoscientific form of biological essentialism that has been the basis for literal centuries worth of discrimination against multiple groups of people (I would recommend the book “Not In Our Genes: Biology, Ideology, and Human Nature” by Lewontin, Rose, & Kamin as a nice overview of how various forms of biological essentialism/determinism have been upheld by the same shoddy science and statistics practices Andrew rails against all the time here in service of nothing more than maintaining existing injustices and inequalities).

            In other comments to this post, you’ve proposed an alternative. Awesome! That’s a way to get a meaningful conversation started. It’s fantastic! Even if at the end of day I may not fully agree with it (which I can’t say I do now because I haven’t devoted a lot of time to looking into it yet), that’s generally a good direction I think we should be moving the conversation towards: Acknowledge the current system is imperfect, propose fixes or alternatives, work towards implementing ones that survive scrutiny.

            • For someone who said “What purpose does it serve to get as hysterical as you are being over sports? (…) It’s sports. It’s not intrinsically important.” you have really strongs opinions about what needs to be changed and how anyone who might disagree can be motivated only by hate against one minority or another.

              • Having strong opinions is not the same as being hysterical (a corollary to the Javert paradox, perhaps?). Contrary to the strawman of me you are creating, I have zero opinions as to the motivations of any specific individual participating in this thread, whether I agree with them or not – although that Paul fella via his deliberate and insistent use of dehumanizing language does make it straightforward to assume he likely has an ulterior bad faith agenda in play here. What I was speculating about, and was very clear about, that you’ve decided to strawman was about a primary reason the entire national/international discussion on this has always been centered around specifically trans women participation, as opposed to any of the myriad other equally relevant issues with regards trans and non-binary people in athletic competition. Does it not strike you as strange that there is so much insistence that the conversation start and stay anchored to trans women? Is that really the only worthwhile aspect of this topic worth focusing on?

              • Ok, so it’s not the only reason. But it’s also hard to take seriously the claim that it’s a primary reason.

                > Is that really the only worthwhile aspect of this topic worth focusing on?

                It’s the natural aspect to focus on given the title and text of Phil’s post.

        • Emilio

          Not Mike here, but you were responding to me actually…

          > It’s not like these categories are static. There’s nothing immutable about these concepts,…

          Well, of course. It certainly isn’t lost on me that for the bulk of our history, having a vagina was a biologically-based, and “objective,” legally enforced exclusion criterion for voting, and even some women’s rights activists thought that women voting was too radical a concept in the 1840s.

          So it’s important to remember that context when we consider what is or isn’t too radical, or whose ox is or isn’t being gored, or what is or isn’t arbitrary.


          > But maybe we could stop with certain kinds of arbitrary groupings that seem to naturalize/essentialize socially constructed categories that are used for primarily discriminatory ends.

          One person’s discrimination is another person’s protection against “unfairness.” So what are the philosophical outlines of where lines can be drawn? I think people on both sides of that issue should be able to articulate their working framework.

          > But I disagree with the assumption you’re making that eliminating the divisions of “mens” and “womens” X-event is a zero-sum affair.

          Zero-sum in the sense that people who feel that certain biologically-based distinctions are a matter of “fairness.” They will see this as a zero sum context. That you don’t agree with them doesn’t make it non-zero sum from their perspective.

          > Why can’t we be a bit more reflective about the categories we create and use to help us navigate/understand the world now? What is the possible harm that could come from getting rid of “mens” and “womens” sports divisions and letting all the competitors play together that could possibly match or supersede the current harms that are being inflicted on womens athletes (cis and trans) today?

          I think it’s much like the “harm” that some feel, and many more felt, would result from same-sex marriage.

          I come at this from the world of education, where people think there’s harm in eliminating arbitrarily-derived distinctions among students. If you think this issue is thorny, consider talking to people about eliminating tracking of students based on standardized test scores.

          > …letting all the competitors play together that could possibly match or supersede the current harms that are being inflicted on womens athletes (cis and trans) today?

          I’m just saying that getting to that from what we have now is a big ask; there are those who see this as a zero sum framework, just as there were those who saw women’s suffrage in that frame, or civil rights, or same-sex marriage. And sometimes you make advances by just saying “fuck-it, you’re just going to have to suck it up.”

          • Sorry for misidentifying you in my response to you Josh (which, kinda ironic, but I do sincerely apologize for it) and sorry to Mike for misattributing to them comments made by Josh. I have no excuse. Bumbling on my part.

            > One person’s discrimination is another person’s protection against ‘unfairness’
            As the entire conversation is centered on trans women, and ignores trans men and non-binary people in athletic competition entirely, I think we can safely conclude the issue isn’t one of fairness and is primarily if not exclusively an issue of finding reasons to discriminate against trans women (who are mostly non-white). I have, in my original comment, speculated as to a possible contributing reason why that is (i.e. cis-het men wanting to ogle women athletes without the risk of feeling kinda gay about it). As you’ve noted, context is important. And here, the context includes what we’re not talking about. We’re not talking about, for instance, what is a man. You’d think it’d be a natural thing to discuss when the topic of thread is “what is a woman” but that’s gone completely ignored in this highly commented on post. That’s kind of curious, don’t you think? Isn’t it kind of strange that we’re not talking about trans men athletes, despite trans men athletes winning competitions roughly proportional to the amount of competitions won by trans women athletes (which is to say, a very very very low percentage, but considering that literally hundreds of thousands of athletic competitions happen every year around the world it does happen enough that one could collect enough anecdotes to make it seem like it’s happening all the time)?

            > I think it’s much like the “harm” that some feel, and many more felt, would result from same-sex marriage.
            so, none?

            > I’m just saying that getting to that from what we have now is a big ask;
            Oh sure, that’s true. It is going to take a lot of work to overcome the current systems of discrimination and prejudice in any arena, but it’s still a goal worth pursuing. It always has been and always will be. But it’s a bigger ask when the biggest barrier towards that goal is people giving credence to bad faith arguments and trying to find some sort of reasonable middle ground that still allows for and enshrines some forms of discriminatory practices and pseudoscientific/anti-scientific prejudicial ideologies.

            > And sometimes you make advances by just saying “fuck-it, you’re just going to have to suck it up.”
            Such as…? I’m drawing an absolute blank as to when any meaningful societal progress has been made as a result of a “fuck-it, you’re just going to have to suck it up” message for members of marginalized groups.

            • I think you’re misreading the meaning of “fairness”. You think of it as somehow “treating people equally” and so the lack of concern about trans female to male athletes shows that no one cares about fairness.

              But that’s not what people mean by “fair” in a sports contest. What they mean is “it would be unfair to force normal women to compete against people who have organs in their body that produce tremendous amounts of testosterone because it’s well established that that drug gives enormous benefits for athletic performance, and because it’d be disallowed for normal women to inject that drug.”

              No one is concerned about trans women to men because they don’t have testosterone producing organs and they don’t have a large advantage in performance.

              • Daniel,

                As part of routine hormone therapy for trans individuals, trans men take hormone supplements to boost testosterone production, and trans women take hormone suppressors that drastically reduce testosterone production. On average, trans women athletes produce less testosterone than cis women (the word you’re looking for is cis, not “normal”) athletes do because the hormone therapy reduces production to the average range for cis women in general, as opposed to average range for cis women athletes (which is higher than for cis women who aren’t athletes). As a result, trans women do not have a large advantage in performance on the basis of testosterone levels. And because trans men do take hormone supplements that cis men are prohibited from taking for purely competitive reasons* perhaps the conversation should be centered instead on trans men instead of trans women if we really are concerned about fairness.

                *exceptions are made for medical reasons, a la medically necessary blood transfusions versus blood doping in cycling

              • I think the point is that testosterone levels today are not the important question for athletic performance, the real question for athletic performance is body morphometrics. people who go through male puberty are taller, have longer reach, more lean body mass as a fraction of overall mass, larger hearts, etc. The “memory” of that exposure is long lasting if not permanent. Taking testosterone suppressors today for a trans woman is just not as relevant to performance as the fact that they can reach farther and hence hit a wider variety of tennis shots, or have mechanical advantage on the pedals of the bike due to femur length relative to leg length or whatever it is.

                If however we focus on body morphometrics, then it doesn’t matter why your body is the size and shape and density and such that it is, it’s what you can do with the body you’re given that matters, because you compete with people who have similar bodies. The mistake is to think that the essential category for fairness is “whether you’re a man or a woman” rather than that being essentially a noisy binary predictor for a bunch of continuous variables.

              • Yes the key here with androgens is area under the curve. Going through male puberty indeed shapes bodies in powerful ways so that a snapshot of current T is nearly useless. Moreover, for doping and athletic enhancement the purpose of exogenous T isn’t so much it’s instantaneous impacts on performance – negligible within physiologic ranges – but because it seriously augments recovery and anabolism enabling a higher volume of higher load training to be successfully absorbed and adapted to productively —> greater performance on the field, lifting platform, boat, bicycle, whatever. Essentially, it raises the ceiling on the fraction of genetic potential you can reasonably hope to express.
                This is why sexual dimorphism is a very real, profound statistical observation across sports to the extent they rely on muscle cross sectional area, left ventricular stroke volume, and mechanical leverage. As an example, I’m a genetically pretty normal male – reasonably athletic but not remarkably so. Moreover I’m currently a mid 30s academic. Nevertheless, with 6 months of concerted training I could make a serious run at women’s world records in either rowing or drug tested raw powerlifting and maybe both concurrently (given 15 hours per week or so to train and ability to sleep 10 hours which I do it have :). this is not to brag, but to offer a reality check in this conversation, where I so often these days see people say things about androgens and sport that seem hugely misguided.
                Daniel may be on to something with stratifying on a batch of anthropometric markers rather than imposed dichotomous sex. It’s worth exploring and thinking through carefully IMO.

              • Thanks Chris, finally someone seems to get my point!

                I would say that the thing that stratifying on a morphometric score does NOT do is deal with sociological and psychological issues. For example sexual harassment, or psychologically supportive team-members or freedom from jerks in the locker-room or constant bickering in the news media about whether “Joe Schmoe” will beat “Jane Schmane” and show “once and for all” that “men win the battle of the sexes” or some horrible garbage like that.

                Nevertheless, when it comes basically to compensating for a major role of androgens in secondary sexual characteristics, stratifying a score based on those major characteristics would keep “like competing with like”. What you’d probably wind up with is one or two groups that are almost entirely men (let’s call it elite men, and athletic men), two or three groups that are more or less mixed (moderately athletic men and elite women, less athletic men and athletic women, amateur part time men and moderately athletic women), and one or two groups that are almost entirely women (amateur women, less athletic women). The men who would fit into the “amateur women” type groups would be mostly small and non-athletic. A man with a 20% body fat is basically pudgy, whereas a woman with 20% body fat can be reasonably athletic for example.

                Given men’s strong ability to change their body morphometry through training particularly compared to women, you might wind up with some unintuitive situations where men training too hard would cause them to shed fat, add muscle and move up in the categories making them **less** competitive, but hey nothing’s perfect, and if they keep it up they’ll eventually wind up competitive in their final category most likely.

            • Emilio –

              > Sorry for misidentifying you in my response to you Josh (which, kinda ironic, but I do sincerely apologize for it) and sorry to Mike for misattributing to them comments made by Josh. I have no excuse. Bumbling on my part.

              Please. It happens. No big deal.

              > As the entire conversation is centered on trans women, and ignores trans men and non-binary people in athletic competition entirely, I think we can safely conclude the issue isn’t one of fairness…

              I don’t know how to come to an objective evaluation of whether it’s an issue of fairness. That seems, necessarily, like a subjective evaluation. I’m inclined towards agreement with you on much of this, and to agree that much complexity lies behind the PERCEPTION that this is about fairness, but I also think that people that disagree with you (or me) can certainly be firmly focused on what they consider as a fairness issue.

              I just think that “fairness” kind of missed the point – again as if “fairness” here is some kind of binary determination, as if one answer is “fair” and the other isn’t. Fairness to whom? is the necessary follow up question. How is “fairness” to be measured also needs to be answered and agreed upon, and who’s to determine whether any particular “fairness” criteria are fair criteria? Who makes the measurement? Looks like a mess to me.

              To make it worse, while I agree that the very question at hand, of “What is a woman” is problematic, so would be “What is a man?” and I think the whole question category is off base. A better question, IMO, should be “What are the criteria we think should be used to determine who should compete with whom at elite level sports, and what are our shared goals here, if any? I think that part of what’s complicated issue here is that people mixing together huge societal and political questions about gender and sex snd discrimination and mistakenly looking at sports s an appropriate venue for working on these issues. As for the more immediate question – maybe we need to have different kinds of competitions to meet different people’s goals and then people can choose. In a way, the “Special Olympics” is a kind of interesting model to think of.

              I come at this from the world of education – where the question of choosing criteria for evaluation has long been effectively a proxy for perpetuating the status quo of existing social hierarchies. But on top of that, our artificial and arbitrary educational paradigm likewise rests on a malignant framework, IMO, as you’re getting at. Shouldn’t education be simply based on helping all students to achieve their potential rather than assessing which students fit into what kind of hierarchy based on an essentially arbitrary set of standards?

              But that doesn’t map onto sports, where competition is more legitimately the whole point? And yet we’re mixing in to this question of sports, much more far-reaching questions.

              > and is primarily if not exclusively an issue of finding reasons to discriminate against trans women (who are mostly non-white). I have, in my original comment, speculated as to a possible contributing reason why that is (i.e. cis-het men wanting to ogle women athletes without the risk of feeling kinda gay about it). As you’ve noted, context is important. And here, the context includes what we’re not talking about. We’re not talking about, for instance, what is a man. You’d think it’d be a natural thing to discuss when the topic of thread is “what is a woman” but that’s gone completely ignored in this highly commented on post. That’s kind of curious, don’t you think?

              See above.

              > I think it’s much like the “harm” that some feel, and many more felt, would result from same-sex marriage.
              >> so, none?

              Yes, that was my point. Over time, a lot of people came to see that their fears were not borne out.

              > But it’s a bigger ask when the biggest barrier towards that goal is people giving credence to bad faith arguments and trying to find some sort of reasonable middle ground that still allows for and enshrines some forms of discriminatory practices and pseudoscientific/anti-scientific prejudicial ideologies.

              That’s largely what I was getting at – except I’m not as ready as you are to determine what is or isn’t good or bad faith.

              > Such as…? I’m drawing an absolute blank as to when any meaningful societal progress has been made as a result of a “fuck-it, you’re just going to have to suck it up” message for members of marginalized groups.

              I think you may have misunderstood my point there. I’m saying that these aren’t questions unique to this issue. I’m asking a question about the efficacy of various strategies to promote change. In other words, do you determine with confidence that people are arguing in bad faith and just say to THEM fuck it, suck it up or get left behind? At some point, even if I were to agree with your confidence that the arguments are being presented in bad faith I would question whether just telling them to jump in a lake is the best way forward.

  19. One point not made about the Aschwanden piece is that it was written in 2016. That’s 5 years ago.

    Since that point, there have been the following events:

    1) Semenya and others with DSD were denied the ability to compete in distances shorter than 800 M. I think this Olympics will see the complete ban but who is to say.
    2) The articles that I reference above were published
    3) The CT situation developed. 2 natal males became gender-dysphoric, and competed as female. In 2 years (2016 and 2017), they rewrote the CT HS record book, setting numerous state records. The suit to stop this was dismissed as the issue was moot (all persons had graduated from HS). The issue remains.
    4) Numerous states now disallow natal male participation in female categories, and more do so every year.

    • With regard to (4), the Texas Senate just passed two bills that would “allow student-athletes in K-12 schools and in public colleges and universities to participate only on sports teams that correspond with the sex listed on their birth certificate at or near the time of birth.” So people like Caster Semenya and Martínez-Patiño would compete as women in Texas schools, but would not be allowed to do so in international competition. That seems poorly conceived on Texas’s part.

      • I don’t know what TX would do about persons with DSD (Semenya) and persons with hyperandrogeny. Since BOTH Semenya and Martinez-Patino are chromosomally male, I think TX would bar them from female competition. That is, BTW, a road out of the “confusion” – chromosomal testing. I have checked to see what Chand’s chromosomal situation is, and it is not noted in a number of discussions. So who knows?

        • The law seems unambiguous. They were both listed as ‘female’ on their birth certificates, so they would both compete as girls/women (if the laws pass. They have passed the Senate but not the Assembly).

  20. It comes down to the purpose of having women’s sport as a separate category. What do we try to achieve by separating men and women at the elite level? If it is just about fairness, why do we care about giving women a chance to compete at the highest level but not giving short people a chance to compete in pro basketball? Why do we need to separate men and women even for non-physical sports such as chess?

    • > Why do we need to separate men and women even for non-physical sports such as chess?

      That’s an interesting example. If I’m not mistaken there separation is not complete. There are tournaments open to anyone and tournaments for (some definition of) women only.

    • Yes, that’s what I’m asking: what are we trying to achieve, exactly?

      Maybe it’s as simple as this: people are interested in knowing who is the best man, and who is the best woman.

      (I think that for many women who play chess, there’s an additional reason for women-only tournaments: at most tournaments, girls and women are a small minority and have to put up with the same kinds of comments they have to put up with anywhere they are a small minority.)

      • I don’t know about the “best” I think its interesting to watch soccer games even if it’s not last years champion team against last years second run champion team or whatever. I think a lot of the times people just want to watch something where the competition is not ridiculously one sided.

        I haven’t watched in a long time, but my impression of mens tennis since the invention of modern large ultra-light rackets in about 1980’s was it was basically “Ace, Ace, Ace, Ace, Ace… switch serving… Ace, Ace, Ace, return, return, out, Ace, Ace…” etc.

        The women’s tennis is usually a lot more rallies a lot more about tactics. If you put Serena against some high end college male he might win, but it’s just because he can slam the serves at higher speed.

        As I say, I could be wrong, but basically the reason people want to watch women’s sports is because it’s no fun to see one group clobbering another group. Same reason we have wrestling or boxing weight classes, and we run qualifying rounds in car racing and brackets in the world cup and lots of people ignore the group stages and round of 16.

        The basic purpose of sports is really entertainment. Relatively few people are brain damaged enough to want to see slaughter after slaughter.

        • You say “basic purpose of sports is really entertainment” – is this a serious statement? That the only reason people go into sports is the entertainment of others?

          There’s something called a “competitive impulse”, and this is an internal motivation which leads someone to wish to excel. This has nothing to do with “entertainment value”. If that was the only reason, pursuit cycling would never be a sport.

            • You guys are talking past each other a bit:one of you is talking about sports, the other is talking about professional sports.

              There would be organized sports even if there were no pro sports.

              • I see amateur local soccer league games that are mixed gender all the time, and the Masters Swimming groups don’t care either, let’s face it we only impose these gender restrictions on pro sports and on children

              • Daniel that’s not even close to being true! There are sports leagues for women in many many sports, at all levels.

              • Right, but adults who don’t have enormous economic interests in excluding trans male to female or intersex people just handle this stuff on the basis that they’re adults and can figure it out. Do you know of any women’s sports leagues at less than the level “I make money at this or am in training to make money in the future” who require karyotyping or testosterone tests or etc? Or do they just have some people in charge who adjudicate the issue and if you aren’t accepted you have to go find a league that will accept you?

  21. How to give everyone a reasonable chance?

    Sure, basic sex stratification is not perfect and I guess that no stratification scheme will ever be completely satisfying.

    Can we see it as a model building problem?
    Too few features lead to under-fitting, but too many lead to over-fitting…
    Keep adding features, and you’ll soon be the only member of your competing category (data quickly become sparse!)

    Sex stratification probably led to a big improvement in the goodness-of-fit, and we are now asking how we can do better without over-fitting.

      • I’m completely unclear as to what this means. Does this mean we get them to run the race, and then stratify on basis of how they did? Or what? This “stratification” is not workable, as it is easily and obviously manipulated.

        There are 2 time-tested successful stratification methods:
        1) sex
        2) weight for wrestling, boxing, and powerlifting

        Other stratification approaches involve classes of accomplishment, like in judo and tai-quon-do. But this is not at the Olympic level, where I believe they stratify on basis of weight.

          • OK, set it up. This is your proposal. Hint: it won’t work. You can’t use performance to stratify on performance. You can use an orthogonal characteristic. That is why weight classes work.

            I hear this proposal frequently, but no one explains how to do it.

            • Hint, if you weren’t so busy telling me it can’t be done and instead just looked up thread you’d find that none of the categories I suggested were themselves performance on the task, they were all things that were objective facts about the bodies of the people involved such as the size of certain long bones or the lean body mass or the heart rate when performing a set task… But I’ve had enough of this conversation. I don’t think you actually care you’re just pushing your political view.

      • I just don’t think any of this addresses the real issue. People are still going to want to know “who is the fastest woman” and so on, no matter how many strata there are that mix men and women.

        • It’s for the competitive sports where you play against someone else that the problem lies rather than for the individual sports where you have an objective outcome like “100m sprint time” or “shot put distance” or whatever

          Suppose there are two strata, 1 and 2 where 2 is a “higher ranked” group. Suppose that group 1 is say 40/60 women/men, and group 2 is 3/97 women/men… Clearly this is near some kind of “cutoff” where “women” really don’t score higher than the cutoff here except rarely.

          So, suppose group 1 has a “champion” who’s a woman, and group 2 the two women rank 40 and 47th percentile.

          Is the champion woman in group 1 “better or worse” than the 47th percentile ranking woman in the higher group? If it’s a sport like tennis, we won’t really know. But we’ll still know that woman A was group champion of group 1. Is that not enough?

          • Good luck convincing the world to think this way.

            I predict that fifty years from now there will still be separate championships for men and women in most sports, because I think that’s the way people think about this stuff and that that’s not going to change.

        • There is a lot of “not answering the question Phil asked” and “not answering the question society wants answered” here, interesting though it is. We have men’s and women’s competition and will for the foreseeable future. This never seemed “discriminatory” until recently, but now raises deep issues that seem to be mostly philosophical and not empirical.

          • Well, I think that’s fairly deliberate. I mean, I don’t really care what society wants, I personally think I’d enjoy a number of sports more if I saw male/female interactions placed on a relatively competitive playing field, so it’s good enough to say that I prefer we answer the question “who is a woman” with, “it doesn’t matter, let’s watch some excellent competitive sports”

  22. My suspicion is that most people do not really care about this issue. Rather those who make that claim likely have a different agenda from sport competition.Even interest in Olympics is to a large extent about group affiliation. If in fact competition excellence is the salient point then logically it should be between the most talented and as fairly as possible.
    For those who want some justice/restitution or whatnot,create other categories but don’t diminish quality of the sport competition.
    The whole thing is largely irrelevant and maybe ridiculous to most.

  23. Re: ‘All of this puts me in mind of a statistical principle or worldview that Andrew has mentioned before, that I think he attributes to Don Rubin: most things that we think of as categorical are really continuous.’

    continuous or discontiuous?
    Having played football with boys only, as a kid, I doubt that, while playing, I was mindful that I was a girl. I was labeled as a tomboy by adults. And who knows why I had a kind of a macho stride. LOl. But honestly I was aware that the boys could overpower me more easily than I could them. Ohhhhh what fun!

    • Most things that we think of as categorical are really continuous.

      Sometimes that’s pretty obvious — like, you’ll see a newspaper article about how Generation X differs from Generation Z, and obviously these are pretty arbitrary categories created by drawing a dividing line based on the continuous variable of age. But sometimes even things that are seemingly categorical would be better thought of as continuous. For instance, the effect of drugs is often quantified separately for men and women, but for physiological reasons it might make more sense to quantify them based on hormone levels or something. “Race” is an even better example.

  24. Maybe thinking of the categories as “men” and “women” is not the best way to conceptualize the issues.

    Maybe it is better to think of the categories as “unrestricted” versus “restricted to women”. Anybody can compete in the “unrestricted” category. For many sports (but not all) this will effectively mean “men”.

    But, we recognize that a lot of human beings are “women” and some “women” like to compete in athletics. To make this possible, we create the second category, and to qualify, you have to meet certain criteria. That used to not be too difficult, but as medicine has progressed, there are more grey areas. To avoid opportunism, it is easiest to keep the criteria strict and require that athletes in the “restricted” category meet all the previous requirements and perhaps some new ones.

    We do this with age (by establishing a “senior” category) and by weight (in boxing).

  25. I would encourage reading “We celebrated Michael Phelps’s genetic differences. Why punish Caster Semenya for hers?” by Monica Hesse in the Washington Post May 2, 2019.

    To those claiming “but it’s unfair to women!” I say – Michael Phelps being a disfigured “human male?” with features outside typical human characteristics is unfair. Should Phelps have been required to undergo fusing of his joints and surgeries to make him more of a normal ‘male’ and avoid unfair competitive advantage? Or taken pills to reduce his naturally high testosterone to ‘average’ levels? These arguments apply to a huge number of elite-caliber athletes who possess biological features outside of the normal ‘male’ range, and might apply at lower levels too.

    • If you want to have a separate competition for women, you need a definition of “woman.” It may be obvious to you and to Monica Hesse that Caster Semenya is a woman, but a reasonable person can look at the XY chromosomes and the high testosterone level and say she is not a “woman.” Me, I’m not sure what I think, which is why I posted this post in the first place.

      In any case this is not just about the single case of Caster Semenya. I think we need a definition of “woman” that we can apply to determine whether someone is a woman. You haven’t proposed a definition. Do you have one to propose?

      • I think saying “we need a definition of woman that we can apply to determine whether someone is a woman” is like saying “we need a definition of statistical significance that we can apply to determine whether something is significant”. It’s false, and non-helpful, and it only seems obvious because most of the women in the world are women p < 0.000001 but there are some who are women p < 0.13 and 0.05 is just a convention that isn't serving anymore as we begin to acknowledge the complexity of biology.

        • Professional biologist who studies the evolution and ecology of sex (including in humans). So not a Javert, just interested in discussions about the biology of sex.

          Yes the biology is complex but only on one of two levels.

          There’s an epistemological question here (how does one identify or categorize or define individual people as “men” and “women”?), but there’s also an ontological question (how many sexes are there?).

          In organisms with sexual reproduction, there are only two sexes and they are categorically or qualitatively different because they have one or the other gonad type (ovary or testis) and one or the other gamete type (egg or sperm). These are the primary sex traits, and they define two sexes. There isn’t a third gonad type and there are no other gamete types. There is no continuous variation between the phenotypes of eggs and of sperm within a species (although there are lots of between-species differences that have evolved by natural selection on the traits of eggs and sperm – one can think of individual gametes to be organisms with their own brief, complicated lives in which they don’t feed or do much of anything else except try to find and fuse with another gamete).

          In some organisms, individuals can have both testes and ovaries; this isn’t a third sex, but those individuals can have both male and female function (in animals the adjective is hermaphroditic, in plants it’s monoecious; potato, potahto). In some species, all individuals are hermaphrodites; in other species, some individuals are hermaphrodites and some individuals are only male; in other species, individuals change sex from one to the other partway through adult life, and sometimes are hermaphrodites during the change from one to the other.

          For all practical purposes, humans don’t do any of those other things. It is vanishingly rare for an individual human to have both gonad types or make both gamete types. And unlike hermaphrodites or sex-changers in other species, these extremely rare humans are not capable of reproducing as both sexes afaik. It’s also vanishingly rare for an individual to lack gonads (and not be either sex).

          So ontologically there are two human sexes, each individual is one or the other sex, and there is no continuous variation between their primary sex traits.

          The two human sexes (and the two sexes in many other species but not all other species) also differ in lots of secondary sex traits. These include body size, lung and blood volume, muscle mass, shape and articulation of bones, and the many other traits summarized by other comments, including some and that are relevant to athletic performance, plus lots of others (external genitalia, distribution of hair, length of vocal cords, etc.) that are not relevant to athletics. Human male and female brains also differ in interesting ways, some of which might be relevant to athletics (and where females in some ways might perform better, and where the consideration of trans men in sport might be important as suggested upthread). Those secondary sex traits together contribute to the perception and social construction of gender.

          The distinctions between women p < 0.000001 and women p < 0.13 (or between women and men) are epistemological (how do we know what's a woman? as in the title of the OP) and focus on secondary sex traits and gender. This is where all of the complexity lies. The distinction between males and females is ontological (what sexes exist?) and focused on primary sex traits, and there is no biological complexity there, at least not in humans.

          In this sense, the distinction between human sexes (males and females) is not arbitrary (as repeatedly claimed elsewhere in these comments) because it is not quantitative or continuous, and doesn't require an arbitrary statistical convention like p < 0.05. The distinction between human females and males is binary.

          But yes the differences among human genders are quantitative, there is lots of continuous variation between women and men, different combinations of values for many secondary sex traits in two individuals can lead to the same gender identity for those two individuals, some individuals may not identify as any gender, and individuals may change genders within a lifetime. Lots of complexity.

          IDK whether women's athletics should be circumscribed by sex or by gender. Distinctions based on sex are simple and binary and non-arbitrary, but require intrusive assessment of the gonads. Some individuals like Semenya who were raised and identify as one gender might not know that they have gonads (and are members of the sex) that are typical of a different gender. To the extent that such individuals occur more often in some ethnic or racial groups, these distinctions based on sex could lead to a racial bias in outcomes of circumscribing women's athletics based on sex (as suggested upthread). One would need data on the racial distribution of such cases (as well as the racial distribution of gender dysphoria leading to trans identities). The reluctance among some commenters to even discuss these issues suggests it may be difficult to get those data and talk about their meaning.

          Distinctions based on gender are not simple or binary, and require either making arbitrary distinctions like p < 0.05 or multilevel modelling of performance with post-stratification into many divisions (I hope I used those two phrases correctly?), but can be based on less intrusive assessment of circulating testosterone combined with some other readily assessed traits.

          And IDK whether women's athletics should exist as a separate competitive division. Feminists seem to have advanced good arguments for a separate women's division, but other commenters here (seemingly all males, maybe including some women) disagree.

          • To someone who studies reproduction, the essential question is “do you produce egg cells, or sperm cells” and yes that’s a mostly binary question. Of course, some people produce neither or at least neither viable. But so far “getting pregnant and giving birth to a viable child” is not a sport. the ONLY thing that matters for sports are the *secondary* characteristics normally **associated** with “producing sperm” or “producing eggs” and in that sense, those are 100% continuous measures.

            If you have two people both of whom are 140cm +- 2cm tall, both have a mass of 65kg +- 1kg, both have 15% +- 1% body fat, both have the same heart rate of 140bpm +- 3bpm while running at 10km/hr, both have the same length of arm, both have the same ratio of leg length to height… then whether they have organs that produce sperm or organs that produce eggs is really mostly irrelevant. There are some effects of testosterone that can be short-lived during the competition etc which can matter, but for the most part the real effects of testosterone that produces differentiation are the long-term ones that lead to different body morphology, and since we’re controlling for body morphology, the competition between those two people is very near “fair” regardless of gamete production.

            Note that those two people are going to represent a “fairly fit athletic woman” and a “moderately fit small man”

              • That’s a good point, and I acknowledge that this is an important factor, though even as it is, women who are on their cycle will have to compete with women who are off their cycle. So it’s no more of a handicap than it already is.

            • Daniel, we get it. You support the idea of implementing a model-based score method to sort individuals into competitive classes to make competition “fair.” I feel your description of strictly mapping non-modifiable physical domain characteristics to performance comes off a little disturbing, eugenic-lite, and would not be accepted in the real-world. I don’t think people would accept that because of their unique combination of height, weight, body mass, body ratios (and whatever) give them a score below a certain threshold, they would not be eligible to compete in a given class. Would such a model even really be able to accurately separate individuals well into competitive classes? I’m not so sure. At the very least, I think it would be miscalibrated for the upper echelons of competition.

              Why should we pit 140cm +/- 2cm tall individuals with roughly the same body mass, body fat, and heart rate against each other in competition (or individuals with similar model-based scores using weights for each attribute)? There is a lot of heterogeneity within a given sport for different body types. At the very least, position-based characteristics would need to be considered. For the NBA, Muggsy Bogues, Spud Webb, and Isiah Thomas come to mind. For MLB, position may not matter as much–what do we do then? There are players of all sorts of body types for all different positions. Compare the physicality of John Kruk or CC Sabathia to Aaron Judge or heights of Randy Johnson (6’10”) to Marcus Stroman (5’7″) Jose Altuve (5’6″) or Mookie Betts (5’9″). Based on pure, measurable physical domains such as height, mass, arm length, body fat, heart rate, run speed, I think a lot of great athletes would fail your quantitative Übermensch score. So perhaps mapping of controlled, physical performance measurements should be used. Except…

              For the NFL, it’s been very clear for years that great NFL combine scores, where players are put through a battery of measurable physical ability tests, do not necessarily translate into a player becoming a decent pro! https://www.tandfonline.com/doi/full/10.1080/08959285.2011.555218

              Predicting competitive performance based on measuring non-modifiable physical domain characteristics or physical ability tests doesn’t necessarily mean individuals will be well-sorted to competitive strata.

              I also think you are taking an overly simplistic view of such a model and that certain items in your model could be modified or gamed (e.g. heart rate, body fat, weight.) So I see all sorts of problems!

              • Unanon:

                I don’t think your particular objection holds up, as in Daniel’s plan there’d still be a top tier, in which Muggsy, Spud, and Isiah would continue to play, along with Michael, Larry, Kareem, etc. The size classes would just be for the lower tiers. But, yeah, I agree with you that Daniel’s proposal wouldn’t work for pro sports. The whole point of the NBA is that it’s a women’s league; mixing in various short retired male players, high school boys, etc., would pretty much destroy the rationale.

              • Evidently I’m not doing a particularly good job of communicating. You are taking my particular example which was meant just to illustrate how controlling for body type the male or femaleness of a person is relatively unimportant and instead taking it as an example of my proposal which it is not.

                My proposal instead is to collapse a multitimentional measure down to a unidemsional measure. Most of your objections wouldn’t hold as people could very widely in individual dimensions so long as they had a compensating variation in some other dimension also the individual score of an athlete would only matter for individual sports. For team sports you could do a number of different strategies based on the score for example have a point buy system where the team’s total score has to be lower than some number or where such a total score is combined with a range for individuals or the like

                Furthermore you would only really use score based adjudication in leagues where people care to try to create a fairness situation. Right now no doubt the existence of women’s leagues is primarily about excluding men from those leagues so that women can compete since there are so many men who are more capable athletically than even elite women athletes however ultimately the point I’m making is it’s not whether you’re a man or a woman that matters for these issues but rather what the outcome is in terms of your body’s strength and power and reach and so forth.

                In the end for example in tennis you might find someone like Serena playing against 65% other women and 35% men who are relatively smaller or slightly less athletic than the elite men in the men’s tennis world. Th is whould enable more variation and a wider range of sporting opportunities for a wider range of people I consider that to be a good thing

              • Daniel:

                What I would like to see better explained is the “why” behind your proposal.

                Don’t you first have to understand why we have sports and then explain how your proposal would better accomplish the purpose of sports?

                Also, your proposal sounds like handicapping, which lets competitors of different abilities compete with each other. Handicapping in golf and horse racing come to mind. If so, then is it really a way for each person to compete against themselves and the winner is the one that scores a “personal best”?

              • Hi Terry, absolutely, and thanks for asking.

                My presumption is that the purpose of restrictions in who can compete in sports is to do a number of things, some of which are good and some I think illegitimate such as:

                1) Make a sport safer (wrestling, boxing, etc)
                2) Make competition between pairs of people more interesting in the sense that the outcome won’t be obvious (ie. LeBron James vs my 10 year old son in 1v1 basketball)
                3) Further political goals, such as nationalism “showing that Argentina is a better soccer country than the rest of the world”
                4) Make competitions between teams of people safer and more interesting (Under 11 year old soccer, Under 12 year old soccer, Under 13 year old soccer etc)
                5) Increase the range of people who can participate meaningfully (if my kids always had to play LeBron James why would they ever play basketball?)
                6) To settle gender/sex specific questions: “who is the best women’s tennis player in the world”

                I personally think goals like 3 and 6 which further nationalism or sexism are illegitimate reasons for these kinds of restrictions. I enjoy watching the World Cup and such because it’s supposed to be good soccer, not because “Argentina is the best” or whatever. I also think that restrictions like preventing Simone Biles from doing certain maneuvers because they’re “mens only” or whatever is a mistake, or that preventing Caster Semanya from calling herself a woman and participating with women on the basis of “not being a woman” is a mistake. But these other goals are all important.

                We should WANT people to be safe and not permanently injured from doing sports. We should WANT the competitions to be between individuals or teams which are a-priori not obviously imbalanced. We should want as many people as possible to participate. Etc.

                I think that my proposal works better than most ways we do this already. Stratifying kids sports by age leads to kids hitting puberty earlier and suddenly they’re within 1 year of each other but one boy is 5ft 2 and weighs 97 pounds and the other is 6ft 1 and weighs 180. That was me (puny) against a guy in my 8th grade class (enormous) for example, I got steamrolled in a flag football match in PE class, unsafe, unfair, unfun.

                The point is that improper linear models such as the one I’m proposing are virtually guaranteed to be better than simple rules based on one category, because they combine multiple sources of information. The article by Dawes that I link in my original post on the subject is a classic. The method can be used all the way from kids sports at age 6 through the Olympics, and with adjustments for the various levels of competition, would inherently satisfy the main goals.

            • Mike –

              > In this sense, the distinction between human sexes (males and females) is not arbitrary (as repeatedly claimed elsewhere in these comments)…

              I’m not sure if that’s in reference to anything I said, but in case it is, to clarify…

              I wasn’t saying that the distinction between human sexes is arbitrary. I’m pretty agnostic on that point as I don’t really know enough about it and I’m open to arguments.

              I was saying that the choice of sex as an exclusion criteria for participation in various sports (as opposed to gender, say) is an arbitrary choice.

        • I agree with Daniel. Not sure I would draw the analogy to statistical significance. Biology is complex.

          In watching many kid shows these days, I find that the female characters are shown to have the same powers as the male characters. So that maybe we are preparing future generations to accept this narrative of equal powers.

          • Sorry, I meant to be clear that I was responding to Daniel’s analogy to arbitrary distinctions between significant and nonsignificant based on p values.

            Sure, the biology of secondary sex traits, sexual dimorphism, and reproductive behavior is complex. But the biology of primary sex traits is not very complex. And in lots of animals there are few secondary sex traits and there is no sexual dimorphism between males and females: in those species sex = gender. But of course not so in humans.

            In discussions like this one, the diversity and complexity of secondary sex traits and gender is often used to imply that sex is similarly complex and diverse. I’m just here to point out how (some) biologists think about some of these ideas and questions, and why biologists tend to prefer to keep the meaning and definition of sex distinct from gender.

            Great discussion I have to say again!

        • Mike, that’s a great summary, thank you.

          Daniel, I should have said: IF we are going to have competitions just for women, then we need a definition of “woman.” Obviously if we are giving up on the whole idea then sure, who cares.

          Of course, if you’re going to have divisions of the sort you are championing, you’ll need definitions to determine who is allowed to compete in Category B. So if you object to thresholds for making decisions, well, physician, heal thyself.

          To me, I just don’t see your proposal as remotely appealing or workable or desirable, for more reasons than I care to list. In the near term, by which I mean the next two hundred years, we are going to continue to have competitions for women and will need a definition of “woman.”

          • I object a lot more to taking something which varies over a wide range and trying to dichotomize it into a yes or no variable than I do taking something which varies over a wide range breaking that range down into a set of sub ranges over which the variables are relatively constrained and calling each sub range fairly homogeneous.

            I actually don’t have any real opinion on whether over the next 200 years we will continue to have women’s and men’s sports or we will change the concept to something more like what I’m talking about but I think it’s fairly straightforward for you to say that you care about the definition of woman so what do you think that definition should be

            • People love categories. Biologically there is nothing whatsoever that is special about what state you’re from, or your nationality, but we have state championships and national championships. There are separate competitions for juniors and for seniors. And of course there are separate competitions for men and women. Of all of the distinctions, the one between men and women seems the most natural to most people. And I think there’s a sense in which it really is the most natural: animals, including humans, have evolved for a loooong time to be able to recognize potential mates. If there’s a categorical distinction in life that can truly be said to not be arbitrary, it’s the distinction between men and women.

              One can argue, and I think you (Daniel) would argue, that just because these distinctions are important in a non-sporting context, there’s no reason in the world that they should be important in sports. I understand your point, and at an intellectual level I’m largely in agreement with it, but for me it has no emotional power whatsoever. Men and women are different, and having separate competitions for men and women seems way more normal to me than having separate competitions for people from California and Arizona…no that anything really seems abnormal about the latter either. I could be wrong, but I just don’t think you’re going to get anywhere with the suggestion that we do away with women’s sports.

              This does not mean we can’t _also_ have divisions based on other characteristics. Many sports have skill-based categories: you can play in, say, a squash tournament as an A, B, C, or D-level player. There are basketball leagues for players under 6 feet tall, or at least there used to be. I used to play Ultimate Frisbee at tournaments that had an “A” and “B” division, each with its own winner. Such things are common in amateur sports but also occur at the pro level: baseball has the minor leagues, pro cycling has ‘World Tour”, “Elite Continental”, and “Continental” tiers, and so on.

              I favor a continuation of women’s sports — doesn’t matter whether I do or don’t, I’m convinced we’re going to have them anyway! — and that means I want a definition of “woman” that can apply. The 99.9% of people who are clearly male or clearly female can easily be put into the right category, so it’s only in the fuzzy middle that we need to be careful about the definition. The definition doesn’t necessarily have to be the same at all levels of competition: I’m on board with the idea that 99% of sports is just fun and games and the rules can be modified to find workable compromises. If a male transitions to female, including reducing testosterone and so on, and wants to compete in a local women’s tennis league or whatever, I would prefer a world in which that is allowed even if such a person would not be allowed to compete on the women’s pro tennis circuit.

              Having mulled this over more, read the comments above, and read some of the things people have linked, my inclination is: for sporting purposes, a “woman” is someone who does not have a Y chromosome.

              That means Caster Semenya would not be allowed to compete as a woman, which I recognize would strike many people as unfair (including Semenya) but you’ve gotta draw the line somewhere. Some people say Semenya shouldn’t be ‘punished’ for having the genetics she has, but I don’t really get that: I can’t compete as a woman either, because of my genetics, but there’s nothing unfair about that. If we have a separate competition for women, half the population isn’t going to be allowed to compete in it, that’s just the way it is.

              For me, a harder case than Semenya is Martínez-Patiño: she has a Y chromosome but has Complete Androgen Insensitivity Syndrome and thus developed as a woman. If that is literally true then she really is a woman with a Y chromosome. I could make a strong argument that such a person should be allowed to compete as a woman. For simplicity it would be nice to just look at the Y chromosome, but if androgen insensitivity really can be truly complete, and if there’s a reliable test for it, then “Either has no Y chromosome, or has CAIS” seems like a good definition to me.

              I’m not saying there aren’t plenty of other ways to think about it, and if someone disagrees with me I’m not going to say they’re “wrong.” I don’t even have a firm definition of what the rules are supposed to accomplish. This really is just one man’s opinion.

              • Phil, maybe it’s just that I’ve got a different perspective because I am married to a PhD in developmental biology from Berkeley, but I think “doesn’t have a Y chromosome” is not a great definition of woman.

                For example, there are XX people with functioning testicles who you would never suspect aren’t men. https://en.wikipedia.org/wiki/XX_male_syndrome

                Also there are chimeras who are **both** xx and xy depending on which portion of their body you sample from. It’s entirely possible for such people to have vaginas, ovaries, and testicles all together.

                My basic saying in biology is “if you can imagine it, it’s happened to at least hundreds of people since the ice age”. There have been on the order of 100B people born since the dawn of humans, so we’re talking hundreds of times that a one in a billion chance has occurred. A lot of these abnormalities are more like 1 per 100k or 1 per million, so with 7B people alive it’s 7000 to 70000 people who have the abnormality.

                In my opinion the problem is regression too far back in the causal chain. The problem is body type, the root cause of the body type is something about androgens and development, but the proximal cause of the unfairness or whatever is the body type, not the unusual genetics or whatever. Whenever you’ve got unusual genetics, there are some people for whom it produces “Male like bodies” and some people who get “Female like bodies” and some who get some in-between stuff, and if you condition on the root genetic issues, there will be no clear separation.

              • Basically the consequence of your definition is that eventually we’ll get an XX male, with testicles and a penis, who weighs 225 pounds has 4% body fat, can do one handed inverted push-ups and can beat any woman’s world record in whatever the sport is and the world record will forever go to this guy or whatever XX male comes after him. I don’t think that makes any of this better.

              • Daniel,
                Well, that hasn’t happened yet. All of the famous cases seem to involve women with XY. But just as there are people with XY who develop as women, I can believe there are XX people who develop as men.

                But to me, to say that all kinds of different things can happen that make sex ambiguous in rare cases, therefore can’t have women’s sports at all, is throwing out the baby with the bathwater. I want to continue to have women’s sports, so do hundreds of millions of other people (possibly billions), let’s find a way to make it happen. Yes there may always be corner cases, but if they only come up once per decade at the highest level of a given sport then we should be able to live with that. It would be nice to anticipate as much as possible, to avoid extremely unpleasant and decisions that end up targeting a single individual, but if definitions are going to have to keep reacting as new cases crop up, maybe we’re just going to have to accept that.

              • To me the “purpose of women’s sports” isn’t really “to exclude men” so that “we can find out who the best woman is”, but rather “to create a realm of sports in which a wide variety of women are competitive”. excluding men is just the way that was implemented because it seemed simple.

                I honestly think that “women’s tennis” would be much more interesting as “tennis for people whose bodies are similar to women tennis players”, similarly for golf, archery, baseball, soccer…

                I also think it’d be interesting to have a professional or semi-pro category of “basketball for people who are athletic but not dramatic outliers in body shape”. I’d like to see a college category for something like that for example. I’d also think it’d be really interesting to have college teams for basketball where everyone is say within mean +- a standard deviation for men on the score I’m suggesting, but that would include some of the more athletic women. I honestly think these would be far more interesting or watchable than womens college basketball today.

                Up thread Andrew says it’d never help to add a bunch of high school men or old guys or whatever. But although those people might technically be eligible, right now Andrew is technically eligible to play for the NBA and yet he isn’t, and neither is anyone like him. Just because a short 65 year old guy might qualify for body-type to play women’s tennis, doesn’t mean he’d be competitive or even want to play against Serena or whatever.

                Anyway, I don’t see why we can’t have **both** systems. Yeah, we have “women’s tennis” and then we also have “group W1,W2,W3,W4” tennis where we take something like my proposed score, we calculate it for current women tennis players, find the range, divide it into 4 groups of equal size, and then let anyone who has similar scores play in their group and stop worrying about whether they have a vagina or a penis or two testicles and an ovary, or complete androgen insensitivity or whatever… none of those are really that important for sports.

              • Daniel:

                It might actually be fun to see some WNBA games where they throw some high school boys and retired NBA players into the mix. I’d watch it!

                I don’t think it’s ever gonna happen, though. Consider this: back in the 1970s, those battle-of-the-sexes matches got tons of publicity and good TV ratings. But they didn’t do more of them. More generally, there are lots of cool “exhibition sports” ideas that don’t get done. I’m not talking about Battle of the Network Stars or whatever, more like potentially interesting matchups such as a touch football game between two MLB teams, or a soccer game with both men and women on the field, or a one-on-one matchup between two retired NBA stars, or a matchup of college softball teams playing kickball . . . all sorts of things. I think some of these could get good ratings, and they could be fun, but they don’t happen. Risk of injury might be one reason not to do these, but that can’t be the whole story. I feel like there’s some way in which people prefer sporting matchups that are “official,” and exhibitions don’t count.

              • I think you’re thinking of my proposal as an “exhibition” or “unofficial” sort of thing. But I think we should have official categories. Like Wimbledon would have 4 different categories that would in essence serve as “all small women”, “women and some smaller men”,”men and some larger women” and “all men” but would be defined in terms of body stats. Similarly I think it’d be a good idea to take the MLS and add another category to each team franchise for “teams defined by total body score”… Teams could take some lower scoring women (who score lower because of their body shape/size not skill) and then use the points to “buy” some high scoring men for example a very tall goalkeeper, or whatever.

                I just think this is a good way to do **official** mixed gender sports, and I think people would find mixed gender sports perhaps more interesting than say womens soccer or womens golf or womens basketball all of which struggle today.

              • Daniel:

                Yes, I understood what you were proposing. My point was that, given that nobody instituted regular battle-of-the-sexes matches in any sport, despite the popularity of the exhibition matches that have been done, this suggests to me that there’s a real resistance to the idea. You or I might enjoy seeing the WNBA augmented by some hight school boys and retired NBA players, or Wimbledon augmented by throwing in some high school boys and retired male tennis stars, but there’s a resistance to the idea. I’m not sure why, exactly, but the fact that it’s not already been done, that nobody followed up on the Battle of the Sexes etc., suggests to me that it’s not going to happen.


                I did watch that Big 3 thing once on TV and it was really boring! It was kind of the worst of both worlds: it had the endless stoppages and inane commentary of pro sports but without the athleticism. In general, I find sports more interesting when I have some kind of rooting interest, so I guess it didn’t help that I hadn’t heard of either of the teams or any of the players on the court. And, despite the enthusiasm of the announcers, the outcome of the game didn’t seem to matter to anyone. I guess this gives one hint as to why exhibition sports aren’t more popular. The popularity of events such as Riggs vs. King might be due to their novelty value more than anything else. Similarly, the Olympics are popular but that again could be due to their rarity. Indeed, I find Olympics TV boring for the usual reason that there’s very little action and lots and lots of waiting time. It’s not just the commercials. Even when they’re telecasting the event itself, it seems like mostly commentary and waiting. I guess that’s what most of the viewers want!

              • Andrew –

                As someone who enjoys watching women’s tennis or women’s Olympic events, I have to admit to an irrational reaction that mixed events as you describe would somehow diminish my enjoyment of those women’s events. Even though I feel a curiosity that would make the mixed events interesting to watch, somehow (again, I acknowledge irrationally), I geek it would diminish my admiration for Serena’s brilliance. Both feelings can exist at the same time.

                Of course, there’s no reason why different types of competition category would be mutually exclusive, but some part of me that’s resistent to change, that’s wedded to how the world is now. doesn’t want to be challenged. Maybe I’m not a total outlier in that regard.

                I just ran across this interesting overview article through a totally independent channel:


              • “To me the “purpose of women’s sports” isn’t really “to exclude men” so that “we can find out who the best woman is”, but rather “to create a realm of sports in which a wide variety of women are competitive”. excluding men is just the way that was implemented because it seemed simple.

                I honestly think that “women’s tennis” would be much more interesting as “tennis for people whose bodies are similar to women tennis players”, similarly for golf, archery, baseball, soccer…”

                Daniel, I just wonder the issue is more complicated than you are making it out to be. Serena (5’9″) and especially Venus (6’1″) Williams could have decent, but maybe not great, physical dimension scores. Setting physical dimension scores aside, and to make the issue even more complicated, they both have outstanding (taking male performance into account) performance on some dimensions of tennis play; for example, both have serve speeds just as fast (even faster) than men (say, Michael Chang, for example). Top women tennis players can serve just as fast as good men players. https://slate.com/culture/2014/09/sabine-lisicki-record-serve-science-explains-why-female-tennis-players-can-serve-as-fast-as-men.html

                Michael Chang and Dave Ferrer were 5’9″. Presumably, they’d have a physical dimension scores that would allow them to play alongside Serena and Venus. Maybe this is ok. I don’t know. I wonder if Chang/Ferrer-like players were to dominate in such a league, if it would diminish the accomplishments of Serena/Venus. Would they dominate? There may be some precedence to get an idea; these types of exhibition “Battle of the Sexes” matches have already occurred!! A short-ish 5’11” male tennis player outside the rank of male top-200 (Karsten Braasch) defeated both Williams sisters in back-to-back single sets (1998 Australian Open), after a full round of golf earlier in the day! https://www.tennisnow.com/Blogs/NET-POSTS/November-2017-(1)/The-Man-Who-Beat-Venus-and-Serena-Back-to-Back.aspx

                I think the idea of some type of non-performance-based physical-attribute multidimensional “score” is super neat. I don’t know how we’d get around certain perverse incentives it could create, where mid-tiered pro male players with low physical dimension scores could be allowed to compete, and dominate, against female peers for high stakes. What would you propose?

              • Phil – you say ‘Some people say Semenya shouldn’t be ‘punished’ for having the genetics she has, but I don’t really get that’. The aspect of this that I find difficult is that due to her genetics she will get a lifetime ban from the sport (through no fault of her own) while people who take performance enhancing substances will get a 2 year ban. That seems quite an incongruous situation. Added to that – the drugs program was put in place, ostensibly, to protect athletes from the negative effects of these substances while in the Caster Semenya case the sporting bodies are saying that she would be able to compete if she took drugs to alter herself. Again, these concepts seem at odds with each other.

              • Hi Unanon,

                > Michael Chang and Dave Ferrer were 5’9″. Presumably, they’d have a physical dimension scores that would allow them to play alongside Serena and Venus.

                Presumably they might be allowed to play against Venus at 6″1 but would be unlikely to be allowed to play against Serena at 5ft 9. Why? Because they would probably have lower percentage body fat than Serena, and so being “more muscle” would be given a higher score than her even though their height is the same.

                Athletes are different than “normal” people, but typical athletic men are in the 6 to 13 percent body fat range, and athletic women are sort of 14 to 20 percent. The difference between the two would automatically make men about 1 point higher on my scale most likely. Venus’ height advantage might compensate for her body fat disadvantage enough that she could play Chang for example. That seems like a fairly fair competition. Karsten Braasch was 5ft 11in which gave him height advantage over Serena and was probably 7% body fat which gave him enormous muscle advantage over both.

                Furthermore, I would want to include measures of recovery. So you’d do something like force people to run on a treadmill at different speeds and then measure their heart rate after a certain time. So you’d have them do a mild sprint, and then back to a jog, and watch how quickly their heart rate declines. Most likely men would have an advantage there as well, so again Chang would probably not compete with either Serena or Venus on these score bases.

                I don’t have an answer, only a suggested process. If someone were to give me a huge database of body measurements of thousands of humans, and to include many athletes, and info on their performance in various athletic tasks, we could put together a score and see.

                I don’t think scores would completely separate athletic men and women, but I think the overlap would be small. Women competing with men their size would be competing with overweight unathletic men, and would easily dominate them most likely. Instead the athletic men would be smaller than them to compensate for their higher percentage muscle and higher cardiac output.

              • Daniel:

                I don’t know how may fans would want to watch the top female athletes in the world play overweight elderly out of shape guys. I mean, sure, once in a while would be fun (Riggs vs. King) but maybe not on the regular. And I’m pretty sure that the world’s top female athletes would much rather play other top female athletes than overweight elderly out of shape guys.

              • Andrew, I don’t understand why you think they’d be playing older overweight guys. They’d be playing WOMEN, and then occasionally a highly athletic but small man, or a medium size man nearer to retirement. But 90% of the time they’d be playing WOMEN.

              • Daniel:

                I dunno, I guess you can set up the league and see who shows up to play! If it’s a rec league, then it depends on how many men women are out there wanting to play the sport. In recreational tennis, for example, people find their own level and nobody will really care if they’re playing a man or a woman. If Olympic medals are at stake, who knows??

              • Maybe the part we’re not mind-melding on is that just because you get into the same “body score” bracket as Venus or Serena doesn’t mean you get a seat at the table. A 6ft 1in male with 18 percent body fat would be totally outclassed by Venus and would never qualify in earlier rounds, because for a man to have 18 percent body fat he’d have to be a pretty unathletic person who hadn’t been training at all, and his heart rate recovery score would drop him way down so that he didn’t have the same score as her. Instead she’d play some athletic but smaller guy, who had a noticeable reach disadvantage to compensate for his higher strength to weight ratio and higher cardiac output. That’d be an interesting match!

            • RE: Daniel

              “Anyway, I don’t see why we can’t have **both** systems. Yeah, we have “women’s tennis” and then we also have “group W1,W2,W3,W4” tennis where we take something like my proposed score, we calculate it for current women tennis players, find the range, divide it into 4 groups of equal size, and then let anyone who has similar scores play in their group and stop worrying about whether they have a vagina or a penis or two testicles and an ovary, or complete androgen insensitivity or whatever… none of those are really that important for sports.”
              My view as well.

              • We already have this, informally: there are competitive leagues that are open to both sexes and where both sexes do in fact participate.

                But this doesn’t change the fact that if you -do- want to have a competition for women only, you need a definition of “woman”.

              • Phil, sure you have to define a woman. I’d suggest perhaps something like the following:

                A woman is a person who is over 15 years old, and whose total exposure to testosterone is less than 100ng/dL * their age

                A man is diagnosed with low testosterone if they have less than 300ng/dL so this should probably work well.

              • Daniel:

                In USA ultimate frisbee, “every athlete should select the gender identity option that best aligns with their gender identity, so the definition of woman is if you select the “woman” option. No physiological measurements required. But I guess that would not work in an environment such as the Olympics where the East Germans would do just about anything to get a medal.

              • Daniel Lakeland said, “Phil, sure you have to define a woman.”

                Some people think that women are undefinable!

              • > A woman is a person who is over 15 years old, and whose total exposure to testosterone is less than 100ng/dL * their age

                That is a great definition. Who doesn’t know their total exposure to testosterone? And in case of doubt total-exposure-to-testosterone-meters are widely available.

              • A good definition is one that describes the underlying construct of interest, it’s a bad idea to make your definitions based on some trivially measurable thing. It takes something like $100M to buy the equipment to measure a kilogram to the accuracy used by NIST, but we did that because it gets at the underlying construct, rather than “1000 cm^3 of water at 4C” or “whatever this one lump of platinum-irridium alloy weighs” or whatever.

                The inference for whether the definition had been met would probably involve a number of rules which allow you to establish the definition presumptively… Such as “has XX chromosomes, does not have a functioning testicle on examination by ultrasound, and daily measurements of testosterone for 10 days in a row while not taking any testosterone affecting drugs show less than 100ng/dL in at least 9 measurements.”

                then there’d be some more complex situation for someone like Caster Semenya who has functioning testicles and presumably daily testosterone exceeding 100 ng/dL

                but the point would be at each step along the way, you’re trying to establish the truth of the underlying definition of the construct. The definition doesn’t change when someone discovers a biological marker for testosterone exposure for example, it’s just a matter of calibration. Just like the definition of a kilogram doesn’t change when someone discovers a new electrical circuit that can measure the resistance of a strain gauge bridge more accurately.

              • Well, again, it comes down to what “women’s sports” are trying to achieve. Daniel, you don’t think there’s any point to them at all, and there are other commenters here who agree with you. But I think I’m more typical of hoi poloi: I am a big supporter in women’s sports, perhaps partly because of how much my wife’s life has been enriched by them but for other reasons too. And the idea that instead of men/women the divide should be (high testosterone)/(low testosterone) just seems ridiculous to me. A man with a testosterone deficiency is not a woman, and including such men in the women’s division doesn’t seem right to me at all.

                Different people like different things. You heard it here first!

              • Phil, I looked up the normal range for women’s testosterone levels and it’s something like 6 to 80 ng/dL, and for men you have a diagnosis of low testosterone at 300ng/dL. By choosing 100ng/dL it says essentially every person you currently think of as a woman would qualify, and essentially zero of the people you think of as a man, even a somehow low testosterone man would qualify.

                You yourself have said that Caster Semenya shouldn’t qualify on the basis of being XY and having testicles, and then you said that someone else, I don’t remember the name who had complete androgen insensitivity should probably qualify but your criterion wouldn’t do it, and that would just be tough for that person.

                Now I offer a definition that gets at exactly the issue of women’s sports, which is that people who are exposed to a bunch of testosterone wind up with a body that’s nothing like a normal woman, and which puts them at a huge advantage in competitions and makes normal women non-competitive. What kind of benefit would there be to your wife or similar people if there were a whole bunch of XX people with high testosterone who have penises and testicles who dominated all the women’s sports (fortunately that’s not the case, but hey under your definition it COULD be).

                so, what is the purpose of women’s sports to you? is it to keep things delineated so that women have competitive events, or is it to exclude anyone from participating if they have a Y chromosome even if for example the Y chromosome is nonfunctional and they develop to within epsilon of the average woman’s body type with a vagina and two ovaries and a functioning uterus?

              • Also Phil, it is absolutely an incorrect interpretation of my view that “there is no point in women’s sports”. To me there’s no point in trying to define “woman” when we could enable women to play competitive sports in ways that make it competitive without worrying about “whether they are a woman or not”.

                I absolutely think it’s a worthwhile goal to include the widest variety of people in sports as possible. So that would include children, women, men, trans individuals, intersex, developmentally disabled, etc etc

                Also my wife the developmental biologist tells me there’s a largish literature on how you can chemically convert an XX embryo so that it develops penis and testicles by exposure to testosterone during development at the right time. So really “being a man” is not really a genetic thing, it’s a developmental thing that happens to only happen spontaneously most of the time because of a Y chromosome, but it’s like saying “a fuel pump defines what it means to be a car”

              • Dovetailing on Daniels ranges, T among doping bodybuilders not uncommonly in the supraphysiologic ranges like 2000-4000! Still, a very common use case is exogenous T up to around 1000-1200 or the upper end of reference range, usually in concert with other hormones. If you watch Icarus, you can see that type of approach deployed for cycling, principally because it enables much faster recovery from high intensity work…

  26. Conceivably of interest:

    Maggie Heartsilver (pseudonym), “Deflating Byrne’s ‘Are Women Adult Human Females?’.” _Journal of Controversial Ideas_ 2021, 1(1), 9; doi: 10.35995/jci01010009 [Open source, available at:
    Abstract: The primary aim of this paper is to show that Alex Byrne’s arguments in “Are Women Adult Human Females?” provide no reason to doubt the truth of the proposition that trans women are women. Byrne’s conclusion is that women are adult human females. However, it is safe to say that much of the interest in his article is driven by the assumption that it is a short step from that conclusion to the further conclusion that trans women are not women. If Byrne is understood to be defending that further conclusion, however, then some of his arguments are dialectically ineffective. The others commit an evidential fallacy or rest on a false premise.

    – the most recent paper in a series that began with

    Alex Byrne, “Are Women Adult Human Beings?” _Philosophical Studies_ vol. 177 (January 2020): 3783–803.

    (associated with some controversy; see: https://web.mit.edu/abyrne/www/GenderGate.pdf and https://leiterreports.typepad.com/blog/2020/06/still-more-on-phil-studies-and-the-dembroff-byrne-debate.html).

    Louise Antony, “Feminism Without Metaphysics or a Deflationary Account of Gender,” _Erkenntnis_ (2020) 85:529–549 https://doi.org/10.1007/s10670-020-00243-2
    Abstract: I argue for a deflationary answer to the question, “What is it to be a woman?” Prior attempts by feminist theorists to provide a metaphysical account of what all and only women have in common have all failed for the same reason: there is nothing women have in common beyond being women. Although the social kinds man and woman are primitive, their existence can be explained. I say that human sex difference is the material ground of systems of gender; gender systems serve to enable male control of female reproductive capacities. This explains the fact that most women are female, but it does not entail that all women are female or that all females are women. To clarify my position, I draw an analogy between the kind woman and the kind parent. While it is difficult to come up with necessary and sufficient conditions for being a parent, it is clear that the social institution or practice of parenting has its material ground in the biological facts about human reproduction together with facts about infantile dependency. Saying this does not entail that all and only biological progenitors are parents.

  27. A few random thoughts late to this discussion:
    Most comments and discussion on this topic (whether sports related or more general) start from the position that women and/or females are inferior, therefore we must protect them and isolate them from the better men. I’m surprised more women don’t find this line of reasoning incredibly offensive as the entire premise of the discussion is ‘women are lesser’.

    All these ‘tests’ and other definitions used in sport and discussed in this thread are designed to define ‘female’. Why has nobody proposed to define male instead?

    Do all athletes (starting in middle/high school, or just olympic/elite athletes) need to undergo highly invasive assessment of their physical traits? Is testosterone enough (keeping in mind that a lot of those identifying as men with low testosterone may get re-categorized)?

    • Anon:

      It depends on the sport. The controversies that I’ve heard of have to with people objecting to someone competing in the women’s division, hence the need for a definition of “woman.” I haven’t heard of examples of people objecting to someone competing in the men’s division, but if it happens, I guess there would be a need for a definition of “men.” Spectator sports are pretty much male centered. The point is not that women are inferior at basketball, say, but rather that the best women basketball players are not as good as the best men. A sport based on endurance or flexibility would favor women rather than men. It’s also possible to define “man” and “woman” based on athletes’s self-identification as with USA Ultimate. There doesn’t seem to be any general solution that satisfies all the stakeholders involved, hence Phil’s post.

      • Having a definition of “woman” is really only necessary at the highest levels of sport where there is something important (usually money) at stake. Ultimate (being the Berkenstock of sports) can take a very laissez faire attitude to gender identification although I imagine that there would be some grumbling if a group of cis-males decided to identify as women in a bid to dominate the sport.

        At lower levels, people tend to stratify themselves into groups (e.g. leagues) where competitors have more or less equal ability (which typically has little, if any, relationship to their physical or physiological measurements) and very often this means that men and women compete together, and no one thinks anything of it.

        • My wife has competed for her whole life in women’s sports. She was a three-sport varsity athlete in college, she played club-level women’s lacrosse, she played Ultimate Frisbee on a women’s team (and a co-ed team on which I also played), she plays in women’s squash tournaments. The idea that “having a definition of ‘woman’ is really only necessary at the highest levels of sport” is just wrong.

          • I agree with you, except that at lower levels, you’re relying on some sort of honour system rather than policing things strictly.

    • Anonymous

      Very good thoughts.

      I think that there is a presumption that women are physically less capable [strong] than men. But that is an oversimplification, to begin with. As a 12-year-old, I could run faster than the boys with whom I played football. Recognizing that they could tackle me if they reached me, I began to watch how each of them played football. Of course, nearly all were physically stronger than I was. So the point I’m making is that developing compensatory skill sets to offset weaknesses and bolster strengths are part of the calculations in competing with men. It goes without saying that as an adult I would not have qualified in either an informal football or professional team b/c of my height and weight. But in other sports, I may have advantages because of my size.

      What I’m trying to get across is that times have been changing b/c women have not as likely to let men then physically assault them, for example. By learning to protect to themselves from assault, they have been demonstrating that they are not physically and mentally inferior. I believe that women have had overcome the psychological challenges posed by cultures.

      Having a bit of a ‘I’ll kick your ass’ can be helpful in some contexts. Smile.

    • Anonymous (the first one), you say “Most comments and discussion on this topic (whether sports related or more general) start from the position that women and/or females are inferior, therefore we must protect them and isolate them from the better men. I’m surprised more women don’t find this line of reasoning incredibly offensive as the entire premise of the discussion is ‘women are lesser’.”

      If women were better at sports than men then I think there would still be different competitions for men and women, and we’d be discussing how to define “man” instead of how to define “woman.”

      I heard once that women are better at standing handgun shooting than men, and there’s some possibility that they are better (by which I mean faster) at extreme running events such as 100-mile races. But in most sports — all of the sprints, runs up through 50 miles or so, jumps, weightlifting, and many others — it’s an empirical fact that the best men are better than the best women. Should we ignore that fact, do you think?

      Maybe you could make a concrete proposal, such as “Since it is offensive to suggest that women are inferior to men at sports, we should eliminate women’s sports”, is that your suggestion?

        • Operating on an adult without their consent is barbaric and of course should be illegal in every country. But I don’t think that’s what you’re asking about here.

          I don’t think there’s any good definition of “female” that for sporting purposes won’t have a few people right up against the line on both sides, no. That’s unfortunate (if true) but I don’t think it’s “unfair” in the sense of “unjust.”

          • Phil –

            I think that there’s clearly a problem when you bar a person from competing in the gender category with which they identify (unless you can determine they’re trying to game the category determinations for the sake of a competitive advantage).

            As for “fair.” I don’t see a way to distinguish the (un)fairness of doing that from allowing that person to compete against others who are at a competitive disadvantage because of innate attributes. So basically, I don’t think that “fair” applies one way or the other.

            Having people self-select to compete in whichever type of league they choose would be the best answer – you could have one league where people choose to compete in an open competition and another where people choose to participate in a league where particular “sex” categories are enforced.

            But short of that, I”m unable to see a clear way through this, unless someone can propose some kind of principle that makes sense to me – on which to base these decisions. I guess one principle would be to pick the rules paradigm that would negatively impact the least number of people but that seems pretty unsatisfactory to me. Particularly in such a solution, the people most negatively impacted are likely to be those who are most likely to be discriminated against more generally in life.

            As near as I can tell, that’s kind of what you think is the best way to go? Is it a matter of choosing the least sub-optimal route?

            • “(unless you can determine they’re trying to game the category determinations for the sake of a competitive advantage)”

              But that’s Phil’s point. The vast majority of men (however defined) will do better in women’s categories than in men’s categories. (Of course, that’s taking Phil’s point that the vast majority of them won’t be competitive in the women’s categories either, so the problem is confined to a fairly small subset of men.) So for some subset of men, the choice is being completely uncompetitive in the men’s competition, or (potentially) world-class in the women’s competition. The incentives to self-identify (for this small set of men) as a woman are large, and have only grown as society (slowly,slowly) comes to accept (or at least not instantly reject) self-identification as determinative. So the benefits are large and the costs are dropping…. How are you going to judge sincerity? Phil wants objective standards to obviate that problem, recognizing that any definition will have edge cases that are uncomfortable.

              • > The incentives to self-identify (for this small set of men) as a woman are large, and have only grown as society (slowly,slowly) comes to accept (or at least not instantly reject) self-identification as determinative.

                I’m not sure that incentive to self-identify applies very broadly. I suppose it’s true for some, but it seems to me that not many people are out there forming their whole life around a sexual identity based on a competitive payoff. Maybe I’m being naive about that.

                The issue that I have is where those who might be incentivized in some way are mixed in with those who are just being who they are, just like everyone else, but who are effectively being discriminated against because of their innate characteristics.

                > How are you going to judge sincerity? Phil wants objective standards to obviate that problem, recognizing that any definition will have edge cases that are uncomfortable.

                But we haven’t really obviated a problem. We’ve chosen between problems.

                I don’t necessarily have a problem with that: There is no free lunch. But I like it better when there’s some clear principle guiding the choices rather than basically a choice that’s arbitrary. If a principle can be articulated it’s possible to see whether there’s a common or overlapping interest – like if we said the principle is based on numbers and causing an injustice to the least amount of people.

                Sometimes there is no actual working principle – but I think that (1) it’s a good idea to explore whether there’s some sort of principle lurking in there somewhere that isn’t immediately apparent and (2) it’s good to be straight forward if you can’t find one and just acknowledge that the choice is arbitrary or just one of personal preference for some reason or another.

    • “I’m surprised more women don’t find this line of reasoning incredibly offensive ”

      Well that’s funny!

      Why is it that all people are talking about here is how women can be allowed in “male” activities? Shouldn’t we also be talking how we can integrate men into “female” activities – knitting, makeup, hairdressing etc?

      Shouldn’t it be offensive to at least *someone* that males don’t even care to be involved in “female” activities, while females are fighting one another to get respect in “male” activities?

      There seems to be an “offended” bias

      • I don’t get the impression from my female friends that they are really interested in the topic. They play baseball, volleyball, and run with males. At least they are may not get into as much detail as many males do on this blog.

        I would venture that there is more emphases now being placed on rehabilitation of cumulative injuries over an pro-athletic career. This situation could over time alter the types of sports engaged. Maybe not in the next 20 years, but in the next 50. Just a speculation on my part after watching and reading about the aging process, more generally; and specifically reading accounts about & by gold medal athletes. We don’t hear much on that score as we are not gold medal athletes. Chronic pain follows them in more cases than acknowledged.

  28. Jonathan (so) thank you, that puts it well. But action don’t think “gaming the system” is the only problem. If someone has testes, produces sperm, and (more to the point, sporting-wise) has the testosterone level of a typical male, and is also genetically gifted in the manner of gifted athletes of either sex, they’re going to be completely dominant compared to non-intersex women. Such a person might genuinely identify as female, no gaming involved, but I think it’s reasonable to argue that this person has a male body and should compete as a man. It might also be reasonable to argue the contrary. Either way the decision will be “unfair” to some people: either the woman with testes and a high testosterone level is being “unfairly” prevented from competing as a woman, or the women without testes and a high testosterone level are being unfairly forced to compete against women who have testes and high testosterone levels. I could argue either way, although not with equal conviction. If I thought the general version of this question had an easy or obvious answer, I wouldn’t have made this post!

    As for which sex someone identifies with: I think society has made huge strides recently in recognizing that when it comes to one’s sexual identity, it’s the mind that counts, not the body. I have several friends and relatives who identify as the opposite sex as their body (I’m not sure the right way to say that but I trust you know what I mean); most of them say they’ve known since they were eight years old. For almost all purposes I think you are whatever sex you think you are. But that doesn’t make sense to me when it comes to sports. The defining characteristic of a sport is that it’s a physical contest. I think if you have a male body you should compete as a male. If it helps, we could re-name the competitions: instead of “men” and “women” we could say “men’s bodies” and “women’s bodies.” But I don’t think that would help.

    • Phil –

      See my comment above related to much of what you wrote in this comment.

      > I think if you have a male body you should compete as a male.

      I don’t get the “should.” I don’t see some kind of principle or framework for creating a moral or ethical hierarchy among the choices. It seems like an arbitrary choice to me. As you laid out – this is a tradeoff. In one direction, the outcome that one particular group of people desire is manifest. Or you an reverse that. I could see the relative size of the groups as one way of thinking about this – generally fewer people being unhappy is a better way to go.

      But I think there’s good reason to be extra sensitive here about leaning towards a status quo, particularly when that status quo has long perpetuated a form of injustice and prejudicial discrimination.

      >But that doesn’t make sense to me when it comes to sports. The defining characteristic of a sport is that it’s a physical contest. I think if you have a male body you should compete as a male. I

      Without meaning to be dismissive, honestly, that looks like reverse-engineered logic to me. Sports is a contest between people, not bodies. Part of what I like about sports is the overlay of the people who are involved, their life stories. I think of Jesse Owens or Jim Thorpe or Jackie Robinson.

      I’m not saying that your preference is “wrong” in some sense. Just that I think that as a society we should be careful about clearly articulating a principle for making these kinds of choices so it can be discussed – or we should be clear that the choice is essentially an arbitrary (or maybe I should say purely subjective) one.

      • There’s no way to even discuss this issue without angering people. Some women are going to be told they aren’t “really” women. But at the same time, if you’re going to have a women’s division then you need a definition of woman. As to whether that should be based on mind or body or something else, I can see arguments either way…but if we have someone with a penis, testicles, a beard, an Adam’s apple, and a y-chromosome competing in the “women’s division” then I don’t think I’m alone in saying I don’t think the division is serving it’s purpose.

  29. “But I think there’s good reason to be extra sensitive here about leaning towards a status quo, particularly when that status quo has long perpetuated a form of injustice and prejudicial discrimination.”

    The extra sensitivity here ought to go in both directions, as with Chesterton’s Fence. What you see as a mere perpetuation of injustice and discrimination also made the Williams sisters really rich.

Leave a Reply to Phil Cancel reply

Your email address will not be published.