This is a version of a piece published in the current Summer edition of The Philosopher, which contains loads of great stuff, and can be bought here. Thanks to Anthony Morgan for commissioning it.
The word “gender” is a buzzword of our times. People talk about the “gender pay gap”; fighting “gender norms”; raising children “gender-free”; and choosing a “gender presentation”. Notions of gender are also invoked in some pretty toxic arguments. In particular, there’s a furious public argument around whether you can change your gender. This isn’t just the question of whether one can legally change sex. In many countries, one can; but doing so doesn’t literally involve any change of status other than a legal one. The possibility of changing sex, as it relates to law, is what is known as a legal fiction: an assertion treated as if true, for defined legal purposes. The question of whether one can change gender is about literal change, not fictional change for legal purposes.
In order to assess candidate answers to our question, we first need to establish what gender means. The unfortunate answer is that, over its history, the word has come to be used in relatively many senses, each referring to different things. This is part of the reason arguments are often so toxic: people talk angrily past each other, each using different senses of gender as they argue. Imagine if two people were discussing whether or not to go to the bank, but one meant financial institution and the other land alongside a river. Something similar happens in many arguments about gender. The participants are each referring to separate things, but haven’t noticed, because each is using a different homonym. So, for purposes of clarity, I’ll differentiate five — yes, five! — possible meanings, and consecutively ask, of each of them, whether one can change gender, in that particular sense.
As the Oxford English Dictionary relates, until the mid-20th Century, gender was mostly used, either as a word for a set of grammatical categories, or, more relevantly, as a polite-sounding word for two biological sexes, male and female. Relatedly, man was, and still often is, treated as a synonym for adult human male, and woman for adult human female. The word gender is still often used this way, as a straightforward synonym for biological sex — as when a UK passport form asks you to fill in your gender as male or female. It’s arguably also what economists are referring to when they talk about the gender pay gap: the fact of members of the female sex getting paid less, on average, than those of the male.
So: can a human change gender, in the sense of changing their biological sex? Can a male ever become a female, or vice versa? Here’s an argument that might be thought relevant. First, it’s pointed out that information about people with Differences of Sex Development (“DSDs”, sometimes also called “intersex”) challenges what’s traditionally considered standard for the categories of male and female. Occasionally, what seem to be mixtures of “male” and “female” characteristics can co-occur in a single person. For instance, someone with Complete Androgen Insensitivity Syndrome can have a combination of XY chromosomes, undescended testes, vulva including labia, and clitoris. Someone with Classical Congenital Adrenal Hypoplasia can have XX chromosomes, ovaries, and external genitalia resembling a small penis with fused labia. Someone with an ovotesticular DSD can have gonads which contain both ovarian and testicular tissue.
What such cases are taken to show is that biological sex, as we traditionally understand it, isn’t determined by any single, unitary set of essential criteria. Perhaps there are more specific, lower-level kinds of sex, each governed by distinctive essential criteria: e.g. “chromosomal sex” or “gonadal sex” or “genital sex”. A person could, for instance, be chromosomally and gonadally male, but genitally female. But at an overall level, there is no single set of features a person must have in order to count as male or as female. The challenge is to get from this reasonable-looking conclusion to any claim about being able to change sex.
It is sometimes said that people with DSDs get their sex “assigned” at birth. This carries an implication that sex-classification happens relatively arbitrarily, and perhaps in a way that creates a new fact about a person. In fact, the majority of people with DSDs are straightforwardly identifiable as male or female. This is demonstrated by the fact that medical textbooks describe most DSDs as having different presentations in males and females, respectively. Equally, in the rare cases where the process of classification is difficult, classification is still an attempt to do justice to a pre-existing biological state of affairs. Where medical professionals suspect a baby might have a DSD, they are keen to identify the precise nature and aetiology of the DSD, via tests. They don’t arbitrarily “assign” anything. They rely upon an established methodology, aimed at capturing pre-existing biological facts.
Some people with DSDs undergo surgical or hormonal intervention. However, this isn’t an attempt to change their underlying biological nature, but to give an appearance more typical of one category. Generally speaking, artificial interventions such as hormone treatment and surgery can make someone’s body look more typical for a particular sex, but this is not the same as changing the body on a deeper structural level. For instance, thousands of differences in gene expression persist, in spite of surgery or drugs.
Equally, it cannot reasonably be concluded that the existence of DSDs shows us that there are no limits to who may count as male or female. In fact, what DSDs seem to show is that maleness and femaleness are two clusters of non-essential characteristics (chromosomal, gonadal, genital, hormonal). The majority of people have all of the characteristics in one cluster or other; a smaller proportion of people have most or many of the characteristics in one cluster or other, but not all of them. So, for instance, you don’t need to possess all of the “female” sex characteristics to count as female. However, you do still need to possess some of them. This is a real, material condition upon sex-category-membership. (This isn’t a radical suggestion, by the way: some philosophers think that many natural-kind concepts pick out clusters of properties, in a way compatible with realism about their referents).
I turn now to a second, non-synonymous meaning of gender. This refers to the collection of stereotypical characteristics, behaviours, and attitudes, culturally approved of as “normal”, “right” or “appropriate” (etc.) for male- and female-sexed people, respectively. Those stereotyped features normatively associated with maleness are summed up as masculinity; those associated with femaleness, femininity. Though details will vary, in many cultures males are supposed to be strong, ambitious, competitive, repressed, aggressive, and logical; and females weak, domestic, self-effacing, emotional, passive, and kind. Those who don’t fit these norms are socially disapproved of for their failure. When people talk about rejecting “gender norms” or raising kids “gender-free”, this is the sort of thing they mean.
Can you change your gender, in this sense? What you can do, certainly, is intentionally or unintentionally adopt an appearance, behaviour or attitude, culturally approved of as normal-for-the-opposite-sex. Males can wear make-up, work as nurses, knit, cry, whisper; females can work as car mechanics, avoid personal grooming, belch, or shout. What you can’t unfortunately do, as a single individual, is change the cultural fact that such characteristics are deemed abnormal for your sex by others, and perhaps on some level even by yourself. That would take a much larger sociocultural shift.
I turn now to a third sense of gender, again not a synonym for either of the previous two. On this view, gender means womanhood or manhood, where simultaneously these are each respectively defined in terms of the fitting of a “social role” in relation to a given set of sex-based stereotypes (i.e. feminine or masculine). What is it to fit a social role, in relation to a set of sex-based stereotypes? At least two interpretations look relevant. Taking woman as our example: on the first interpretation, a woman is defined essentially as “anyone who appears and behaves as a female should appear and behave, according to prevalent feminine normative stereotypes”. On the second, a woman is defined essentially as “anyone to whom feminine normative stereotypes are widely applied, whether or not that person appears or behaves as those stereotypes say she should”. So: on the second but not the first version, a woman could fail to meet feminine stereotypes, but stay a woman, as long as she was also socially disapproved of for doing so. That is, she could be an unkempt, loud, car mechanic, but still be a woman as long as society judged her deficient because of it.
Either way, on both readings, woman and man are defined, as such, in terms of the fitting of social roles, in some sense. This is a step beyond the previous view, which defined gender only in terms of a set of stereotypes, but with no accompanying essentialist claims about womanhood or manhood. On this approach, then — unlike the first one which defined woman synonymously with female — the referents of female and woman potentially come apart. Sex is what determines femaleness (and maleness); fitting a social role, in one of the senses given just now, is what determines womanhood (and manhood). Womanhood and femaleness potentially come apart because, for instance, a male could, in certain circumstances, socially transition to the point of being perceived as either enacting, or as becoming subject to, the stereotypes associated with femininity, or both. Then this male, who was formerly a man, would become a woman. Gender — in this sense — would have changed, though sex remained constant throughout.
Some don’t accept these claims about the meaning of womanhood and manhood, insisting that, for the majority of language-users, woman refers exclusively to an adult human female, and man to an adult human male. Some also reject the essentialisation of a connection between womanhood and norms of femininity, and manhood and norms of masculinity, even as an aspiration. They worry that it’s a reactionary means of further imprisoning females and males in a set of harmful stereotypes. I share these worries. But whether or not such criticisms are right, let’s accept that, within the terms of current discussion, males can become women, and females can become men, as long as they fit the relevant social role, on either of the readings just outlined.
For most males, this will involve significant alteration of their original sexed appearances. It won’t be enough just to put on certain clothes or make-up because even so, others will usually still be able to easily read them as male, and so be unlikely to apply feminine social stereotypes in this case, or to think of them as enacted. Equally though, even for those people who have major surgery and hormone treatment, inhabiting a social role goes beyond simply “looking female” or “looking male”. Though an objectifying culture encourages us to think of women, in particular, as a set of shallow appearances, in fact looking female is not the same as either “acting” female or “being treated as” female. A male who looks exactly like a female might still exhibit many masculine traits, and be approved of for them, as females who possessed the same traits would not be.
To be clear: none of this is any endorsement of what’s sometimes called “biological essentialism”: the view that being female and male irrevocably determines distinctive behavioural characteristics for each group. It’s rather the claim that society supplies us with ready-made ways of differently coding females and males. Many of these ways of coding go beyond surface appearances, and can be hard to shake off, despite one’s best efforts. Nonetheless, it does seem true that, within this paradigm (which, as I’ve made clear, I have reservations about), males can become women, and females can become men — at least with sustained commitment.
With our fourth sense of gender we come full circle, to find it used again as a synonym for sex. This time, however, sex is understood as something completely socially constructed. That is, there are no discernible material facts about biological sex, detached from the ways in which society codes it. As Judith Butler tells us in her celebrated book Gender Trouble:
Gender ought not to be conceived merely as the cultural inscription of meaning on a pregiven sex. … [G]ender must also designate the very apparatus of production whereby the sexes themselves are established.
The first use of gender we looked at said that it was synonymous with sex, and that sex (and so gender) is characterised in terms of material biological facts. The third differentiated gender, as something socially constructed, from sex as material and non-social. This fourth use returns us to thinking of gender as synonymous with sex, but this time, gender (and so sex) is understood as wholly non-material and socially constructed.
Given that the human species continues to reproduce, one might well look askance at any conclusion saying there are no underlying material facts about maleness and femaleness to which humans can have reliable epistemic access. But even if Butler was right about this, could a person ever change gender in this sense? I don’t see how. Social facts are not things an individual has control over. As individuals, we might play with social norms, parody them, attempt to subvert them: but we can’t thereby control how the wider collective is culturally primed to read us.
A fifth and final sense of gender refers to “gender-identity”. Though many quasi-mystical statements are made about this, the most reality-based way of characterising it is that a gender-identity is a person’s individual feelings of ease, or lack of ease, with their sex, and with the sociocultural expectations placed upon that sex. It seems that gender-identity appears most saliently to a person when they feel ill-at-ease and dysphoric. Whether actually male or female, a dysphoric person might potentially have a gender-identity of male, female, or non-binary (neither male nor female). In cases of severely distressing, prolonged dysphoria, a person can reasonably seek social and/or legal transition as a therapeutic strategy to alleviate such feelings, by acquiring an outward presentation which “matches” their gender-identity. In non-dysphoric contexts, the concept of gender-identity rarely comes up, so that actually, it would seem false to say that everyone has one.
Since gender-identity is grounded in a set of feelings, it’s detached from the question of one’s actual sex. It’s also detached from the issue of one’s gender, in the socially constructed sense outlined above. Can one change one’s gender-identity? That’s effectively a question about whether one can change one’s feelings. And as with feelings generally, there doesn’t seem much one can do directly to alter them: simply telling yourself to feel something different rarely works. The best way of changing distressing feelings, generally, is to do something to alleviate them indirectly, by doing something else first — even if that thing is just waiting for feelings to change. Some dysphoric feelings are temporary, particularly when they occur in childhood and teens, and can be changed simply by the acquisition of new experiences, altering one’s self-conception. In the case of severe dysphoria, there are a number of things one can do under the supervision of a trained professional. These include talking therapies, medication, and, as just reported, social or legal transition as a final resort in severe cases. The success of each of these strategies for a particular person will depend on individual factors, such as personality, history, and how deeply psychologically-rooted and prolonged the feelings are.
To sum up: the answer to the question “Can an individual change their gender?” depends on which of the many different meanings of “gender” is being invoked by the questioner. For two of these — gender as fitting a social role and as gender-identity — the answer is yes, sometimes. (This conclusion doesn’t entail whether these are notions of gender a society should legislate around; that is a separate question). For three of these — gender as sex, as sets of normative stereotypes applied to sex, and as socially-constructed sex — the answer is no.