Friday, 9 November 2018

On whether sex is binary

Anne Fausto-Sterling is a developmental biologist who has spent decades exploring the biological and social factors that work together to produce maleness and femaleness.
She has detailed the myriad ways that 'normal' sex development can be disrupted and is famous for arguing that we need to revise our sex categories in order to better accommodate biological reality. Fausto-Sterling details ways that the sex characters, which we treat as co-occurring (testicles go with chest hair go with XY chromosomes go with wide shoulders go with penis) can and regularly do come apart, so that the biological facts about a person can give them an indeterminate position in our folk male-female binary. The existence of these 'intersexed' people, she argues, show us that the folk-idea that humans come in only two kinds - male and female - is not biologically sound. In 1993 she proposed five sex categories, but in 2000 she revised this and said that we need to think of sex as a multidimensional space rather than as a set of options or even a continuum.

Fausto-Sterling recently wrote a piece for the New York Times, reiterating her position in the context of attempts by the Trump administration to legally define a person's gender as corresponding to their birth sex, which is to be established by examination of a newborn infant's genitalia, with chromosome testing to be used as a back-up if this first step is inconclusive. One reason people are objecting to this move is that it prohibits anyone from adopting a gender that is different from the one they were assigned at birth, which denies the existence of people who are transgender. Another reason is that it presupposes that there are always empirical facts of the matter about what sex someone is, which seems to deny the existence of people who are intersex. I'll focus here on that second problem. Fausto-Sterling was one of 1600 scientists who wrote to the Trump administration informing them such an approach has no scientific basis, using the tag line #not-binary.

Alex Byrne, a philosopher of perception at MIT, has written a response to Fausto-Sterling, defending the notion of binary sex, and accusing Fausto-Sterling's article of being more confusing than it is enlightening. Byrne admits on his website to being a newbie to the sex/gender discussion, and although I'm not much less of a newbie myself, I think that Byrne has got a few things wrong. Here is my attempt to explain.

I'll assume Byrne recognises that the idea of defining sex by genitalia as birth, with chromosome testing as a back-up, is problematic, because he doesn't consider it. One problem is you might get people who have ambiguous genitalia and whose chromosomes are ambiguous as well - such as if they have XXY or XO chromosomes. Another problem is that people's genitalia can mismatch with their chromosomes.

Bryne argues instead that we can turn to a standard biological definition of sex to show us that sex is binary.  He suggests we use the standard biological, multi-species sex distinction which says that the female of the species is the one that produces the largest gametes. Then we can say that "females are the ones who have advanced some distance down the developmental pathway that results in the production of large gametes - ie ovarian differentiation has occurred, at least to some extent." A male, meanwhile, is someone who has advanced some distance down the developmental pathway that results in the production of small gametes - sperm.

Byrne argues that if we use this definition then there will be no humans who are neither female nor male, thus proving that sex is, after all, binary.

There are several problems with Byrne's argument.

First of all, the definition  wouldn't be easy to apply - we would have to come up with some rules about exactly which developmental events constitute the right amount of ovarian differentiation, for one thing. And Byrne would have to accept that any fetus younger than 6 weeks has no sex at all, which flies in the face of normal intuitions about sex.

Second, Byrne allows that there will still be humans who count as both male and female, because they have advanced some way towards production of both types of gamete (such individuals have standardly been called 'true hermaphrodites' but the Intersex Society of North America is campaigning for such genital-focused terms to be avoided). This, he says, is in line with the fact that definitions in biology are never perfectly precise, but always admit of exceptions. I agree, but I fail to see why his desired conclusion - that sex is binary - is not threatened by such exceptions just as it would be threatened by cases which are neither male nor female? Byrne seems to think it is fine if the sex binary is inclusive rather than exclusive, as usually assumed. Indeed, he crafts his definition deliberately to avoid getting cases which count as exceptions to an inclusive-binary sex concept. But it is hard to see why we should be upset if there are exceptions to binary sex rendered as an inclusive binary - ie cases that are neither male nor female - but not get upset about exceptions to binary sex rendered as an inclusive binary - ie cases that are both male and female.

Third, notice that there are other ways we could define sex which would be both simpler to apply and which would avoid the problem of allowing anyone to be both male and female. For example, we could define as female anyone who lacks a penis at least two inches long. Or we could simply define as female anyone with a functioning uterus (I'll assume there is a way to settle what we would mean by 'functioning').

The most obvious and deeply serious thing wrong with my suggestions is that they would inflict unnecessary suffering on people. There will be many men who for one reason or another do not have a penis that measures two inches, but who would be deeply upset if I categorise them as female.  There will be many women who don't have a functioning uterus, but to whom their identity as female is a precious component, one which I would cause psychological hurt in denying. And there are women who likewise fail Byrne's definition of a female. One such woman is the athlete, Maria Jose Martinez-PatiƱo, who was born and raised female, who lived her life as a woman until it was revealed in the process of gender testing by sports ruling bodies that she has Androgen Insensitivity Syndrome. AIS occurs in XY individuals who bodies go a long way down the developmental pathway of producing sperm - far enough to develop testes. But at a subsequent point in development their bodies permanently stop recognising the hormones that push the body down this pathway. The individual then develops what are standardly taken to be paradigmatically female external genitalia- a vagina, clitoris, labia. At puberty these individuals develop paradigmatically female secondary sex characters - breasts, narrow shoulders, hairless chins.

Such individuals don't constitute difficult cases for Byrne's definition - they are classified unproblematically as male. But Byrne's definition would pose a problem for these individuals - at least some of them get hurt by being classified as male. Patino was hurt because she lost her career as an Olympic hurdler. Some are hurt because they feel like females. Some are hurt because they feel like neither males nor females.

Byrne's attitude towards such harms is plain: We don't mitigate such pain by lying to people. He says "To those struggling with gender identity issues, it might seem liberating and uplifting to be told that biological sex in humans is a glorious rainbow, rather than a square conservatively divided into pink and blue halves. But this feel-good approach is little better than deceiving intersex patients." We better honour the autonomy of such individuals by telling them the truth, Byrne says.

But the appeal of that argument vanishes once you pull the truth out from under it. Byrne and I stand in no disagreement, I assume, about what the truths of the matter are. We can compile an assortment of cases of individuals whose development has proceeded in ways that deviate from what is deemed 'standard' or at least what sits in the centre of the bell curve when it comes to human development. But such a collection of facts no more forces us to accept Byrne's definition of the sexes than it does either of mine, nor a definition which refrains from categorising humans into sexes at all. The truth sometimes hurts, yes, but when the hurt is caused not by facts but by a normative agenda then people become culpable. We don't have to define sex as binary any more than we have to define indigo as separate from violet.

As Georgia Warnke wrote in 2001"The idea that we just are essentially male or female is thus less an idea about nature than it is an interpretation of natural properties". An interpretation is something over which we have control. "We can ask whether this interpretive framework is consistent with our fuller conceptions of ourselves and our aspirations."

If we are committed to being dispassionately honest about facts, why spend time trying to think of a way to alter our definition of sex so that it looks binary? I think Warnke has the answer. "We possess expectations about the look, activities and capacities of individuals based on their gender and we employ these ideas to understand their bodies." (Warnke 2001, p. 133).

The primary mistake I think Byrne makes is to think that the argument is about whether or not it is possible to come up with a definition according to which sex is binary. It's easy to invent definitions which are true to empirical facts - just think of cloud types, or the way we categorise wind speed into hurricane, storm, etc. We don't just accept any category that we are able to come up with - we try to only utilise well-motivated ones. For example, we consider whether grouping a bunch of things under a label is going to help us make effective predictions about how things will behave in the future.Because generalisations by their very nature go beyond finite sets of facts, our proposal and acceptance of them is always motivated by pragmatics.

So the argument regarding sex isn't about whether we can define it as binary, but about whether we should.  There are biological matters of fact about particular bodies...but they don't force any definition of sex. There is a wrinkle, in that social kinds - by which I mean categories of people -  can end up being a  self-fulfilling prophecy. We can ask whether its useful to generalise that people who wear dresses lack penises? This might be a useful generalisation, but not because there is any underlying biological law connecting dress-wearing to male biology. The predictive power comes about purely because males who wear dresses often get verbally and physically attacked.

So then a further question must come into play. Not just 'Is this generalisation useful?' but 'Is this generalisation harming people?'

Why is Byrne keen to find a definition of sex which allows no post-fetus human to be neither male nor female? I venture that it's because he is motivated to preserve the status quo - the folk concept. But most of us would agree, I think, that intuitions should be updated if their preservation causes human suffering.

Parallel arguments to Byrne's are common in the context of discussions about race. For example, Robin Andreassen argues in a 2005 article that "Statements about biological differences are descriptive statements of empirical facts. Assertions of racial superiority are normative claims, the result of imposing a value system upon the fact of biological variation." But this call to separate facts from values is naive, because our values always guide our selection of facts.

A capacity to produce gametes is an undeniable mind-independent fact about a person. But the positioning of this fact as more significant and definitive than any other fact about the person - whether they want children, whether they are currently pregnant, what colour their eyes are, how many fingers they have, what their fingerprints look like, what kind of novel they like to read - is certainly not mind-independent.

But if anyone thinks that the existence of facts about ovarian differentiation can themselves tell us whether sex is binary or not, then they have rather missed the point.


  1. ''But most of us would agree, I think, that intuitions should be updated if their preservation causes human suffering.''

    I would like to see a source on that. I, for example, disagree vehemently with it. Avoiding human suffering is a worthwile goal, and I agree that we shouldn't expose people needlesly to it, so that it would be insensitive to try to convince a person with gender identity issues that they are in fact male or female. But that's different from saying that I would have to take their suffering into consideration in trying to reach a conclusion about their gender identity. Their suffering does not seem a relevant factor in the latter project.

    Also, your claim that we could find a number of categorisations which would neatly divide people into either male or female is a bit disingenuous. Yes, we could, but not all those are equally relevant to the question of whether one is male or female. For example, the length of the penis is not relevant for the question of whether someone is male - men with small or large penises are men. Nor is wearing dresses relevant for that matter.

    But producing sex cells of a certain type is relevant, and is in accordance with the use of those notions (''male'', ''female'') in biology, for the most part.

    So Byrne's categorisation does not seem arbitrary, while yours does.

  2. Thanks for your comment, it is illuminating to me to hear your disagreement about the status of intuitions in the face of suffering. I suppose there are two considerations about which i should have been more explicit. One is that, in the case of the sex, the intuition that sex is binary is the *only* piece of evidence on the table in support of the binary. If there was supporting empirical evidence, or indeed if there were no other sort of evidence available at all, then i might be inclined to give some weight to intuitions. Second, i take it that people have widely differing intuitions about sex. Some cultures recognise a third sex and arguably, english people assumed a one-sex model previous to the 18th century.
    I'll assume more broadly that we have different feelings about intuitionns! Coming from a philosophy of science background, i'm alert to the fact that intuitions, even when very strong and entirely universal, can turn out to be utterly false. I'm alert to the fact that intuitions often function to enforce ideology that serves dominant interests. I'm alert to the fact that intuitions can and do change radically. In general i think of them as something to be overcome, to be mistrusted and challenged, not to cling to!

    Regarding your point about penis length - i mentioned the two-inch definition, because apparently one inch is a rule of thumb that doctor use when deciding whether an intersex infants penis is 'adequate' or whether, instead, they should surgically construct a vagina (this is mentioned in Warnke 2001).

    And while i'm well aware of the biological definition of sex in terms of relative gamete size, its worth pointing out that this definition is fairly empty - there are not thought to be any interesting generalisations that correspond to it. for example, the sex that makes the smaller gametes sometimes has the larger chromosome (in birds). and then there are tons of species that have more than two sexes, or that don't have sexes at all, or in which individuals are all hermaphrodites, or are all capable of switching sex. so the idea that male/female are deep or important categories in biology generally is illusory.

  3. Thank you for your interesting analysis. I'd like to point out that, since ISNA closed in 2008 (and even before that) intersex advocacy has grown to become a worldwide movement. That movement has produced a range of human rights institutions and declarations that have important things to say about how we are treated, and how we are classified. Examples include the 2013 Malta Declaration - - and (in my region) the 2017 Darlington Statement -

    Looking at these, it turns out that we have different perspectives on who we are, how we define ourselves, and how we should be classified. I think these come from being at the pointy end of the debate, but also an understanding of history and the medicalisation of sex and intersex traits. This tells us, for example, that the legal concept of sex predates the kinds of genetic and histologic analyses currently prevalent and has allowed for some recognition of diversity in biology (in the West) and identity (in some non-Western societies). I've written on this, and good sources on the medicalisation of intersex include the work of Elizabeth Reis (notably in her book Bodies in Doubt), and the work of Morgan Holmes.

  4. Sorry for the very long delay Morgan (my research time is punctuated by university semesters at the moment!)
    Thanks very much for the comment and the suggestions - i've just enjoyed your article 'Impossible "choices": The inherent harms of regulating women's testosterone in sport' (
    This might be an unreasonable question to expect an answer to, but do you have an opinion about what *should* be done about sex-segregation in sport? Ie, if you were on the IAAF, what would you have them do?