Are There Two Sexes? Yes and No, But Mostly No (and Gender is Something Else Anyway – More or Less)

3.2 Sex(?) Chromosomes, the Development of Primary and Secondary Sex Characteristics, and Diversity

One might wonder how the so-called “sex chromosomes” work into all of this. We must start by recognizing that not all sexual organisms have “sex” chromosomes. Due, perhaps, to our anthropocentric biases, there appears to be a tendency to assume that the genes producing differences related to sex in humans are shared widely across taxa. This is not, in fact, the case. Certainly, all mammalian species have X chromosomes and most of them have Y chromosomes that are involved in developmental pathways that typically produce primary and influence secondary sex characteristics. (It is important to note that humans and the members of other mammalian species come in more than just two chromosomal types – XX and XY – though we will leave the details until later.) Birds and a variety of other animals have different sex chromosomes, the W and Z. While, in mammals, having two X chromosomes is associated with producing large gametes and the XY combination is associated with small ones, in birds it is the doubling of the Z chromosome that typically gives rise to male animals, with the ZW combination typically giving rise to females. Many sexual species do not have sex chromosomes, either because their members typically produce both types of gametes or because sex is determined by some other means, like a chemical signal, the ambient heat surrounding the fertilized egg, or even how many sets of chromosomes an individual has. Historically, of course, chromosomes are a latecomer to sexual classification, as chromosomes were only discovered at the end of the 19^th century and the link between X and Y chromosomes and sex was only identified around the turn of the 20^th century (Richardson 2013, 23ff.).

Some species are hermaphroditic, meaning that there is only one morph (at least so far as sex is concerned) and it contains the gametes and associated sex organs of both sexes. It is important to distinguish this meaning from the use of the term that has been applied to certain humans. This latter use identifies a developmental variance found within the species (sometimes called “gynandromorphism” when applied to nonhumans; Fusco and Minelli 2023), not a species-typical morph. Though it may seem natural to assume simultaneous hermaphroditism, where an organism produces both male and female gametes at the same time (as is common in flowering plants and snails), there is also what is called “serial hermaphroditism” in animals and often “dichogamy” in plants (Stout 1928). This is where the organism changes from producing one type of gamete to another, though the extent to which this is associated with a change in primary or secondary sex characteristics varies across species. In flowering plants it may just mean that the stamen and pistil ripen (and so release their gametes) at different times, but in some animals there is a significant morphological and physiological change (see Griffiths’ discussion of the Eastern Blue Groper, Achoerodus viridis; Griffiths 2021). So far as the gametes are concerned, however, it’s the same thing.

Here is one of the obvious places where a decision, rather than the biological facts, determines how we describe things. It makes just as much sense with serial hermaphroditism to classify organisms into one hermaphroditic sex as into two sexes. This is perhaps easiest to see when we notice that the sex of serially hermaphroditic organisms is entirely developmentally contingent. Sex in these species is thus like maturity – a life stage – rather than something that renders the organism one sex or another throughout its life (though unlike life stages, some species can switch back and forth; Todd, et al. 2016). I suspect that how exactly biologists navigate identifying sex for serially hermaphroditic organisms has as much to do with the history of their subdisciplines as anything else (though, so far as I can tell, the scholarly work exploring this history is yet to be done). However, one principled way of deciding the matter might be based on the extent to which a phenotypic change accompanies the change in gamete production, which takes us back to primary and secondary sex characteristics. Likely, how serially hermaphroditic individuals are actually classified in the lab or the field, as male, female, in some transitional state, or what have you, is often not based on gamete size (as gametes are rarely easy to directly observe) but instead on primary and secondary sex characteristics. Although the importance of development and life stage for determining sex are particularly obvious for serially hermaphroditic organisms, this is no less true for any other sexual organism – a consideration that has led Paul Griffiths to suggest that we should treat all sex morphs as life stages (Griffiths 2021). Of course, every physiological, morphological, or behavioural trait of a multicellular organism depends not only on their genetic starting point but also on the contingencies of development, to which we now turn.

Biologists have a way of talking about the contingencies of development; they call them “norms of reaction.” Basically, this just picks out the range of possible outcomes of the development of an organism from a given starting point. The classic experiment here is to take a group of genetically identical saplings (clones) and plant them in different environments, perhaps at different altitudes on the same mountain, to see how the different developmental resources available in those places affect the growth of the plants (see, e.g., Clausen and Heisey 1958, 123–56). The idea here is that the differences in outcome cannot be genetically driven – genes can’t be the difference makers – because the plants are all genetically identical. Of course, this is a bit misleading, as some differences will depend on different genes being expressed or being expressed at different times. The outcome (and, indeed, every developmental outcome in all life on our planet) is thus really a product of genotype, phenotype, and environment interactions.

For sexual species, we often want to know about the possible range of outcomes (and, often, their frequency) for a whole species, population, or morph (i.e., body type) within a species. In the serially hermaphroditic organisms mentioned above, the norm of reaction from the fetal state of a typical individual will include both male and female morphs. For non-hermaphroditic species, however, the primary and secondary sex characteristics develop, in some cases, after some kind of trigger or the initiation of a scaffolded process, be it something in the environment or the expression of a gene, such as the SRY gene (usually found on the Y chromosome), which plays a significant role – along with other genes – in the development of the morphology of most humans who are identified as male at birth (and most, though not all, other mammals; Terao et al. 2022). In other cases, there is a kind of default development that happens unless some causal factor initiates or scaffolds a different developmental pathway, which is how typical female development in mammals is sometimes described, despite significant flaws in this analysis (Richardson 2013, 127–46). Norms of reaction remind us that there will always be a range of possible outcomes from any given starting point.

It is important to note that norms of reaction are empirical and not normative. If there is an observed developmental outcome in a member of a species, population, or morphological type, then it is part of the norm of reaction for the species, population, or morph. It would be nonsensical to claim otherwise. Of course, some outcomes are common and others are unusual, given the starting point. Within our own species, we can take height as an example. There are relatively few people who are seven feet tall and there is likely to be a complex set of genetic and developmental factors that explain why a given person is seven feet tall. For any particular seven-foot-tall individual, their height could have been otherwise if they had matured in a different environmental context. The statistical “abnormality” of this unusual outcome has no normative content until it is put into a particular social context, where difficulties buying clothes and the call of professional basketball might make this trait a good or a bad thing.

So, when we come back to thinking about primary and secondary sex characteristics, it is unsurprising that from any given genetic starting point for the fertilized egg there are a range of different possible outcomes. Certainly, some are more common than others – so much so that they might be considered species-typical. So, for instance, height between five feet and just over six feet is species-typical for humans (though this is not true of all human subpopulations); being over seven feet tall is not typical. For any given species, we can say a similar thing about any given set of primary and secondary sex characteristics and how they tend to clump around having certain gonads that produce gametes of a certain size.

With a firm appreciation of developmental contingency in hand, we are in a better place to consider the significance of the range of possible genetic and chromosomal starting points. Again, this varies significantly across taxa. In some sexual organisms, extra sets of chromosomes can be added either during reproduction or in regular cell growth so that, rather than two (diploid) sets of chromosomes, the organism has three or more. This polyploidy is quite common in plants (Comai 2005). In some organisms, including humans, one can often get an extra copy of one of the chromosomes, ending up with three (or more) rather than two – thus trisomy 21, the chromosomal cause of Down syndrome in humans. Whether and which extra chromosomes make an organism unviable (effectively killing it in an embryonic state) can only be addressed species by species (or sometimes at a higher taxonomic level like genus or family). In animals, often an extra chromosome is developmentally deleterious, leading to various health challenges, as is common with trisomy 21. However, the seriousness and extent of these developmental outcomes vary significantly as is, again, exemplified by people with trisomy 21 (for an interesting example see Pradhan et al. 2006).

3.3 Intersex Syndromes (With a Distinctively Anthropocentric Bias)

It is, then, unsurprising that although most living humans have either XX or XY chromosomes, not all do. While missing one of the “sex” chromosomes, 45X (Turner Syndome) or having an extra, 47XXY (Klinefelter Syndrome), 47XXX, and 47XYY are relatively common, there are rare cases where folks have two or three extra. As noted, having any extra chromosome is rarely a good thing for health and typical functioning, though, again, for 45X, 47XXY, 47XXX, and 47XYY individuals, there are a wide range of outcomes, and for some their chromosomal anomaly does not end up being clinically significant (O’Connor 2008) and may go undiagnosed. The point is, from each of these genetic starting points, there is a range of developmental outcomes. We typically call those with XX chromosomes female and XY chromosomes male, though certainly not always. Moreover, it’s worth noting that this only became an option after the biological role of chromosomes was discovered; so, prior to the 20^th century, we could only use primary sex characteristics to identify sex. We still primarily use this old-fashioned method, except in places that have normalized genetic screening. (As for other species, it’s worth noting that if sex is not significantly influenced by genes or chromosomes because sex is determined by some environmental factor, then primary and secondary sex characteristics will be the only way of identifying sex, absent specific knowledge of the developmental history of the organism.)

Most of these chromosomal differences are considered “disorders” of sexual development and certainly some of them have serious deleterious health effects. Many of them would be considered intersex syndromes. Again, here we are presented with a messy, complicated biological reality and how we classify it is largely a matter of decision. Before addressing the classificatory question, however, it is worth considering that there are many more intersex conditions (though how many will depend on how we classify them; Witchel 2018; Fausto-Sterling 2000b, 52). Here, we will just consider two relatively common cases that exemplify the developmental contingency of primary and secondary sex characteristics for those who develop from XX or XY embryos.

The first is androgen insensitivity syndrome (AIS). This is where an individual who is XY is insensitive to the androgens produced by their gonads and so develops primary and secondary sex characteristics that are typical of XX individuals. People with Partial AIS visually appear to be intersex insofar as they have anomalous genitalia at birth that may be considered more or less paradigmatically male or female and, even if they are tending toward a more typically male presentation as a child, they may develop breasts at puberty. People with Complete AIS are not typically identified as intersex until puberty or adulthood when it is found that, despite their having typical secondary and external primary sex characteristics of females, they have undescended testes, not ovaries, and sometimes other anomalous internal sex organs (Gottleib and Trifiro 2017 [1999]; Fausto-Sterling 2000b, 52, 64).

Congenital adrenal hyperplasia (CAH) is a condition affecting both XX and XY individuals. Caused by a recessive gene, CAH is a serious endocrinological disorder where the adrenal glands don’t produce particular hormones. The details are complicated, but the hormonal imbalance produced can lead to various serious health conditions, some of them life-threatening. In XX individuals, CAH can cause the overproduction of androgens, leading to the presentation of primary and secondary sex characteristics that tend towards those typical of XY individuals. As with AIS, CAH comes in degrees, and many cases cannot be identified until later in life or not at all, though in this case it is the mild case that often escapes notice (Hannah-Shmouni et al. 2017; Fausto-Sterling 2000b, 52–58).

Clearly, these differences in sexual development (and there are many others), whether chromosomally driven or not, produce morphs that are neither standardly male nor standardly female (however we define these). How exactly we classify the range of sexual morphs is a question of choice. In 1993, Anne Fausto-Sterling suggested that there were five human sexes – male, female, merm (intersex but with predominantly male features), ferm (intersex but with predominantly female features) and herm (intersex with neither dominant male nor dominant female features). However, only a few years later she suggested that “sex and gender are best conceptualized as points in a multidimensional space” (2000a, 22; a suggestion brilliantly realized by van Anders 2015).

Some might be tempted to take a gamete-centric hard line and assert that gamete size is necessary and sufficient for identifying sex within the human species. This would suggest that only those who have both ovarian and testicular tissue could be intersex, which is rare. However, it would also have the result that the reaction norms for human males and females would significantly if not entirely overlap, as, for example, morphologies typical of individuals with AIS (or relevantly similar syndromes) would be part of the reaction norm for males. It would also simply not align with how we ordinarily identify sex, which is on the basis of primary and secondary sex characteristics. It is, after all, not easy to get access to the gametes of, say, a newborn child. Reliance on primary sex characteristics in medical and intimate sexual contexts is how we currently do identify and historically have identified people by sex. Even beyond the medical sciences, there are various research programs, most obviously in archaeology and anthropology, that ascribe sexes to human remains despite it being impossible to use gametes to do so (see e.g., Rösing et al. 2017). Ultimately, it is difficult to see what would be gained by revising the concept of sex to conform to a gamete-centric hard line.

3.4 Decisions and Clinical Interventions Determining Sex and Their Cultural Context

Whether someone who is, say, non-standardly male – due to having hypospadias (a condition where the urethral opening is not at the tip of the penis but somewhere along the underside of the shaft) – is intersex or not is a matter of decision (Lee et al. 2016, 159). There is no obvious biological fact to determine the matter, and where a biologist or clinician falls in their judgment will depend on their training and the severity of the condition, i.e., how anomalous the penis is when compared to some stipulated (though not entirely arbitrary) norm and whether the condition is associated with other serious medical problems, like a difficulty urinating. Interestingly, those who defend a traditional dimorphic sex ideology tend to define this group very narrowly, rendering intersex people rare through their definition. The idea seems to be that if intersex conditions are sufficiently rare, we can ignore them (like people who are seven feet tall).

Often, decisions about whether someone is intersex (and what to do about it if they are) are decisions made about infants. Infants, obviously, have a huge range of possible developmental outcomes that will be significantly affected by environment, physical interventions on their bodies, and – especially for biologically cultural animals like humans – their cultural milieu. Whether a difference in sexual development is retained as something that is accepted or even celebrated, or whether it is treated as tragedy to be corrected and covered up – a mark of shame for the individual and their family – is a question of cultural norms. In North American society, this second attitude has been dominant and has led to dangerous surgeries being performed on infants who subsequently often experience long-term medical intervention and surveillance throughout their childhood, puberty, and into adulthood (Fausto-Sterling 2000b, 86–91; Jordan-Young 2010, 235–55; Mulkey, Streed and Chubak 2021). Recently, there has been a push to move away from surgical interventions prior to a child’s expressing some gender identity or being able to consent to surgery, though this is controversial (Lee, Mazur and Houk 2023).

Fausto-Sterling explains the decision-making process by medical teams that seems to inform and justify these interventions (2001, 45–61), and Stephanie Kapusta offers a useful analysis of its conceptual underpinnings (2017). The first step is to discover whether the child is “really” male or female (Fausto-Sterling 2000b, 50); Kapusta describes this as the “precisification” of genital sex through various standardized measures (2017, esp. sec. 3). The second step is to create what Kapusta calls a determinate “sex of rearing” (2017, esp. sec. 4) through various medical interventions to change the child so that they meet the accepted standards of what male or female should be. The sex-specific goals of such interventions include things like giving the child the capacity to direct urine while standing up (Fausto-Sterling 2000b, 57), a penis that can “achieve penetrative intercourse” (Lee 2016, 169), and fertility in a manner typical of the chosen sex (Fausto-Sterling 2000b, 57). In this way, medical practice effectively imposes “determinate sex predicates” on children (Kapusta 2017), identifying intersex infants only to surgically eliminate them.

Even without critically examining the standards for successful surgeries, there is some extremely peculiar reasoning going on here. If a medical team determines that an infant is, say, male, then whatever their external genitalia is, it is part of the range of naturally occurring male morphology. So, it’s difficult to see how their own reasoning can justify these often dangerous surgical interventions. If clinicians are confident in their claim that a child with anomalous genitalia is male, why not just educate their parents, and indeed, the public, about the range of genital morphology for males – the many ways, morphologically and physiologically, of being a human male. This natural diversity appears to be assumed by their own reasoning. Of course, another option is to allow that the child is “really” intersex. The medical intervention is then designed to transition the child from being intersex to male or female. This, in effect, imposes a gender-affirming surgery (and, potentially, a lifetime of medical surveillance and care) on infants who have not yet expressed any gendered behaviour, let alone developed their own gender identity. The shift away from the term “intersex” to the more explicitly normative “disorders of sexual development” or its bowdlerized cousin “differences of sexual development” serves to elide this reality and to justify “corrective” surgery through semantic sleight of hand.

That such interventions have been done without properly informing parents (Fausto-Sterling 2000b, 63–66), which means without informed proxy consent, is chilling. It is yet more disturbing when one considers the fact that the need for such interventions – some of which leave permanent scars and seriously negatively affect adult sexual function (Fausto-Sterling 2000b, 78–95; Jordan-Young 2010, 235–46) – is imposed by a culturally specific gender ideology that demands, against the evidence, that there are only two human sexes. When it comes to intersex conditions, the medical establishment has put itself in the position of correcting natural diversity (or, for the religiously minded, God’s work) by imposing morphological sex dimorphism on the population in their care. As Catherine Clune-Taylor (2019) has argued, this elimination of sex diversity appears to be squarely focussed on eliminating from our cultural consciousness the reality of folks having non-standard sex and gender identities. It is sex dimorphic ideology as standard medical practice.

This ideological commitment to sex dimorphism in places like North America is very odd given that the diversity of sex morphology and gender expression has long been widely understood throughout the history of European and European settler societies. After all, the term “hermaphrodite” is ancient – there’s even a myth, recounted by the Roman poet Ovid, to go along with it (see, e.g., Ovid c. 8/1922, Book IV) – and Jewish law refers to more than two sexes and has some rules about who is supposed to do what (though here we start moving toward gender roles; Fausto-Sterling 2000b, 33, 41; Strassfield 2022). The Pardoner from Chaucer’s medieval bestseller, The Canterbury Tales (1400/1987), uses “he” as a pronoun, but is described as “a geldyng or a mare” (I. 691) – i.e., a eunuch or a female, which, however we take it, is certainly gender-nonconforming. And so on.

There are plenty of societies that recognize more than two sexes beyond the European tradition, though here again sex is often mixed up with gender and sexuality. How these bodies and identities are understood varies from culture to culture and modern or colonial categories often fail to adequately represent this. For instance, “berdache,” as a colonial term, problematically lumps multiple culturally specific Indigenous gender identities together (Roscoe 1996), but to address post-colonial realities, people indigenous to North America have adopted “two-spirit” as a catch-all for the different ways these societies include those who didn’t and don’t fit sex or gender binaries (Davies-Cole and Robinson 2022). So, biologically, anthropologically, and historically we know there are more than two sexes within our species, and this is only if we restrict our gaze to intersex conditions. When we consider the ways in which sexual diversity has been created and sometimes diminished through various conscious interventions, the two-sex model becomes yet more suspect.

However we choose to classify the human population in terms of sex, at the end of the day it is clear that there are not just two sexes, if we mean by that two distinct and exclusive morphs correlated with gamete size that exhaustively describe the members of our species. Forcing everyone to be one or the other on the grounds not of their preference, but of a decision based on an ideological commitment to sex dimorphism, is simply scientifically and morally indefensible. The fact that most people morphologically and physiologically conform to one or another typical morph (more or less) is irrelevant. It would be ludicrous, biologically speaking, and ethically wrong, to force a person with size-14 feet to wear size-11 shoes on the basis that size-11 feet are more common in the population. Similarly, it is ludicrous to surgically modify a person who is intersex to fit a male or female sex category because intersex people are uncommon in the population. In both cases it amounts to a denial of a manifest biological reality – a refusal to address the real diversity in the population. Certainly, within a society that has a strong dimorphic gender ideology (like most societies in the European tradition), it will often be desirable, if not necessary, to identify with one sex or another for one’s personal safety, but this choice of identification and concomitant social roles is a question of gender, not genitals, chromosomes, gonads, or gametes. Gender is (in part) a public display of a socially meaningful identity, whereas genitals are typically covered up (though not in all societies), even if we understand that the public display of a certain gender is often assumed to be correlated with having certain genitals.

Before moving on to gender, it is worth spending some time with those sex differences that are imposed through surgical (or other medical) interventions. It is important to remember that many human bodies (and this will be true of other sexual organisms too) do not produce or house gametes. My mother had her ovaries removed in her 50s and, like my spayed cat, from that time on had neither gonads nor gametes. This seems to have had relatively little impact on other primary or secondary sex characteristics for my mother and no impact on her identity – though other women have other experiences, and certainly my cat’s behaviour and morphology would have been considerably different if she had kept hers.

Human males also sometimes have their gonads removed – and concomitantly stop producing gametes. Thanks to modern medicine, this sometimes happens as a medical intervention for treating prostate or testicular cancer. Although there are often significant morphological and physiological changes, producing traits that are more typical of human females – secondary sex characteristics such as reduced body hair, loss of muscle mass, increased body fat, or growth in breasts – there appears to be a reluctance to think of these changes in terms of sexual diversity, at least, in contemporary societies in the European tradition (Wassersug, McKenna and Lieberman 2012). This is, presumably, because there is considerable stigma around the loss of one’s testicles for people who identify as men in this society. This is not merely encapsulated in various sayings and slang around “having balls,” but in the fact that “eunuch” is typically a term of derision and is often used as a slur. But this brings us, again, to gender, so perhaps it is time to address it.

What I hope is obvious from this section is the sheer diversity of expressions of sex, and that where we draw the line about who counts as what sex, even in the most biologically basic contexts, will sometimes be a question of judgment. Ignoring the diversity and the complex, contingent, multifactorial reality of sexed bodies is likely to be misleading in biological research. In biomedical research, it not only risks harming science but harming patients. This has led some to call for the elimination of sex as a category (Watkins and DiMarco 2025) and others to call for a contextual approach to sex in biomedical research that “recognizes the pluralism and context-specificity of operationalizations of ‘sex’ across experimental laboratory research” (Richardson 2021, 1; see also Clancy et al. 2024).

4.1 Eunuchs, Castrati, and Hijra

As noted, complexities and contingencies of gender/sex (and indeed gender and sex as well) are often difficult to grasp when applied to our own person. Similarly, there may be a tendency for us to mistake those norms and differences that are common in our own particular cultural and historical context as given, fixed, and universal, and so presume that they are in some sense biologically given. This is why reflection on eunuchs as a “third sex” with a distinct gender role is particularly instructive. Of course, in contemporary societies in the European tradition, few people identify as eunuchs (with some exceptions; Wassersug, McKenna and Lieberman 2012), despite the relative prevalence of orchiectomies in these societies. So, in this case we have a distinct sex morph with no corresponding gender identity, despite the fact that in other societies this particular sex morph has been associated with various distinct gender identities.

Although other societies have had eunuch gender identities (e.g., Byzantium; Ringrose 2003), I will focus, for the sake of brevity, on the eunuchs of Imperial China, the castrati of modern Europe, and the hijra of India. For two of these genders, the primary sex characteristics associated with them were considered necessary conditions due to their biological implications – in the first case, the impossibility of reproduction and in the second, the development of a distinctive, strong soprano singing voice. In ancient and medieval China, eunuchs had a distinct role as a kind of servant class in the imperial court and, intermittently, other high ranking households as well as the military and civil service. The status, number, and political influence of eunuchs varied significantly over time as did the source of eunuchs – some of whom were foreigners and others who were Chinese. Though some were castrated as a punishment, for others, choosing a eunuch identity was a path to economic security and a certain degree of political status and influence that was relatively easy when compared with the notoriously competitive bureaucrat education system (Tsai 1996, 1–9). During some periods, particularly when the emperor was suspicious of the bureaucrat class, the number and power of eunuchs grew (Tsai 1996) – as is exemplified by the story of Zheng He, the great Muslim explorer who, from 1405 to 1433, led a Chinese fleet around the Indian Ocean to Africa (Dreyer 2007).

There was not only a distinct social role for Chinese eunuchs, but also a normative gender identity, albeit an unflattering one. They were thought to be ruthless and treacherous and were often assumed to have many of the character flaws associated with women (Tsai 1996, 11–12). Doubtless, many eunuchs lived up to this stereotype, internalizing the characteristics that they were told they had and treated as if they had. However, despite the fact that the Chinese at the time would have understood these to be essential traits, effectively, just behavioural secondary sex characteristics of eunuchs, I doubt that many contemporary Europeans or North Americans would agree. After all, that would mean that men who have life-saving orchiectomies to treat their cancer would undergo the same change in personality. What, from the Ming Chinese perspective, would have been experienced and understood as gender/sex or even sex traits, most people in North American society now would simply consider culturally contingent gender norms.

This is not to say that there aren’t real biological changes for people who undergo orchiectomies, though what exactly these are vary considerably, and how any given individual experiences them will have a good deal to do with their own understanding of what their bodily changes mean (Wassersug, McKenna and Lieberman 2012). Most obviously, anyone who undergoes a complete orchiectomy, has no gonads and, concomitantly, no gametes. (Those who would make sex entirely dependent on gamete size should take note.) However, because of the role of the testes in producing hormones, they often have other physiological differences from cis-men without orchiectomies, such as the growth of breasts (Wassersug, McKenna and Lieberman 2012, 257). There is a developmental component here also, as males who have orchiectomies prior to puberty may develop distinctive voices and growth patterns. Not only do they not typically grow facial hair, but they also tend to have limbs that are rather longer, larger chests (associated with greater lung capacity), and higher voices than is typical for cis-men who do not experience orchiectomies (Skuse 2021, 17–18).

These high, strong voices are what led to the domination of castrati in Italian church singing and opera in the 16^th, 17^th, and 18^th centuries (Rosselli 1988). Here again there is a distinct social role – being a soprano or alto non-woman professional singer – and, as with the eunuchs of Ming China, the role is ascribed on the basis of primary sex characteristics, albeit with the intention to produce the desired secondary sex characteristic that was essential to the social role. For both Chinese eunuchs and European castrati a gender identity drove a surgical intervention, producing distinctive primary and secondary sex characteristics which made access to a certain social role and way of life possible for the people who underwent it. Is the castrati’s voice a gender difference or a sex difference? Well, it’s both – it’s a gender/sex trait.

Some social roles that are associated with orchiectomies do not actually require them. Consider the hijra, who are now legally recognized in India as members of a third gender (Boisvert 2020, 92). Although the ritual of nirvāṇ, in which not only a complete orchiectomy but also a penectomy is performed, is often an important step for hijra, their identity as hijra does not require it. Indeed, unlike the Chinese eunuch or European castrato, they become hijra prior to this intervention through an entirely distinct ritual (Boisvert 2020, 94–98). What is instructive for us is how clear it becomes through consideration of these examples that gender is entirely distinct from sex, while being intimately related to it. After all, all three of these gender identities fail to map onto contemporary North American gender identities associated with the same surgical intervention – orchiectomy – which, for some women, intersex folks, and gender diverse people, is an essential part of gender-affirming medical care and, for some cis men, is a cancer treatment.

Again, this diversity of the meaning and experience of orchiectomy is striking because orchiectomy removes that which some want to take as definitive of dimorphic biological sex (gametes and gonads) and changes primary sex characteristics (though some cis men choose to have prosthetic testicular implants to retain this; Haymon, Michael and Coward 2020) and secondary sex characteristics (though some cis men choose to undergo other medical interventions – for instance, hormone replacement – to lessen these; Fritz and Rienert 2024). So, the sex morphology definitive of eunuchs in a variety of historical cultures is not associated with any contemporary gender identity in North America (again, with some exceptions).

5 Conclusion

So, are there two and only two sexes? Yes and no, but mostly no. Yes, insofar as “sexual” as a type of organism is defined in terms of sexual reproduction and anisogamous organisms are identified as those that reproduce by joining large gametes with small ones. No, once we start looking at particular taxa where species vary from having one hermaphroditic sex to multiple different sex morphs with their own distinct developmentally contingent varieties. Even within our own species, where we have species-typical primary and secondary sex characteristics that typically align with gametic differences, there are many naturally occurring developmental differences that do not so align, and cultural and surgical interventions that sometimes make bodies align more with the morphology that is normative for the sex they were assigned at birth and sometimes don’t. These interventions are informed by culturally contingent gender norms. Gender is not sex, though because the social roles and norms associated with a particular gender are imposed, typically, on the basis of a sex ascription, gender is frequently experienced as inextricably entwined with sex. Moreover, in cultural animals, the traits that we see are frequently the result of the interactions between biological and social causes; thus the usefulness of coining a third term, gender/sex. One thing is clear, the idea that there are two and only two sexes in our own species and that gender can be reduced to secondary sex characteristics is clearly false. In the face of this reality, our moral and political obligations are obvious – working toward a world where all people, regardless of their physiology, morphology, or gender identity, can flourish.