Are there more than two sexes? This is a question that has caused an enormous amount of social and political debate in recent years, but at its core, it is a scientific one, and I want to treat it as such. In other words, what we do with the answer to that question certainly has social and political ramifications, but the question itself is one of biology, not politics. Therefore, I am going to try to answer it in this post from a strictly scientific standpoint. I am not going to make any statements about politics, morality, religion, etc. Instead, I am going to talk only about the biology. As always, if you are going to read this, then all that I ask is that you lay aside any ideologies and views you might hold and look solely at the facts. Political and social positions must be based on facts, not the other way around. So, in this post, all I am going to do is present the facts.
Terms, definitions, and critical background information
On topics like this, it is always a good idea to define the opposing positions at the outset. In this case, there are basically two camps. One holds that sex is strictly binary and is determined by the presence or absence of a Y chromosome (sometimes stated more explicitly as XX = female, XY = male). The other position argues that sex is more complicated than this binary and follows more of a spectrum rather than a clear dichotomy. Some people misunderstand this and construct a straw man about people arguing for the existence of “third sex.” The argument is not that there is a third sex, but rather that sex cannot be adequately defined by two discrete categories because there are many people with both male and female traits. In other words, one position argues that sexes can be defined as two distinct boxes into which all individuals fit. The other argues that the situation is more complex and there are some individuals who do not fit cleanly into either box and are actually somewhere in the middle. I think part of the confusion arises over the way that we talk about this, and I fully admit that I have been guilty of this as well. We often say things like, “there are more than two sexes” as convenient shorthand, but what we really mean is that sex cannot be adequately defined using a simplistic dichotomy in which all individuals with a Y chromosome are males and all individuals without a Y chromosome are females. It is more complicated than that, and there are many intersex individuals that do not fit neatly into traditional categories of males and females.
Note: I opened with the question, “Are the more than two sexes?” simply because that is terminology that is familiar to most readers and introduces the topic.
Note for clarity: based on the comments thus far, I want to clarify that I don’t have a problem with a more nuanced position that says something to the effect of, “based on reproductive physiology most individuals are binary in that they either have the physiology for producing sperm or the physiology for producing eggs, but there are a variety of exceptions to this. So the binary classification is useful in some contexts, but we should acknowledge that there are exceptions and situations where the binary classification does not work.” Indeed, that is more or less what I am arguing (also note that I am only talking about biological sex, not gender or gender identity). Also, please see the addendums at the end before accusing me of a straw man or bringing up the “10 fingers” argument.
Next, we need to talk about how we define sexes, and before we get specifically to humans, it is really important to look at the biology of sex more broadly, because this gives us context and important background information. So, let’s start with the general definition of male and female. If you were to ask professional biologists to provide a general definition of “male” and “female,” the one answer you are not going to get is, “if a Y is present it’s a male; if Y is absent, it’s a female.” There’s a very good reason biologists don’t use that definition. Namely, because it doesn’t work for a very large number of organisms. You see, many organisms don’t have sex chromosomes; instead, male vs female is determined by some environmental factor (I’ll come back to that in a minute). Further, even for species with sex chromosomes there are lots of exceptions and atypical situations (again, more later).
Because of these problems biologists have historically defined sex based on the production of gametes (sperm and egg). The sex that produces small (usually mobile) gametes is considered to be the male, and the sex that produces the large, stationary gametes is considered to be the female. Thus, it is the production of gametes that defines sex, not the presence of a particular chromosome. To put that another way, sex is defined by gamete production, but in some cases, it is determined by chromosomes. In others, it is determined by environmental factors. This may seem like pointless semantics, but it is actually really important (as will become increasingly clear as we go), because the biological definition of sex is not about chromosomes. This already puts the “Y = male” position on shaky ground (it’s also worth noting that in many species it is the female that has two different sex chromosomes, not the male).
Having said all of that, there is a caveat that needs to be explained. Namely, the broad definition of male vs female that I have given can run into trouble at the individual level because some individuals are sterile, so by this definition, it seems like they simply shouldn’t have a sex. In reality, we define sex practically based on the physiology that would result in the production of a particular gamete under normal circumstances. This is important, because physiology is rarely binary. There aren’t, for example, two distinct groups of people with regards to metabolism: high and low. Rather, there is a whole spectrum of metabolic activity.
The next thing we need to talk about is genotype vs phenotype. The genotype is what a person is genetically. In other words, what their genes code for, whereas the phenotype is the physical characteristics of the individual. This is important to understand because different genotypes can lead to different phenotypes, but also the phenotype does not always match the genotype. This becomes particularly true when we start talking about epigenetics. An epigenetic effect occurs when something other than genetics affects the expression of the trait. In other words, the phenotype is determined not only by the genotype, but also by the environment, enzymes, etc. and in some cases, those factors can override the genotype.
Sexes in the animal kingdom
With all of that background in place, let’s look at the animal kingdom and see what sort of variation exists for the sexes, because there is a lot we can learn from this broad perspective (I promise I will talk about humans later). Even a cursory knowledge of zoology will quickly tell you that sex is complicated. There are, for example, many species that are hermaphrodites. This means that they simultaneously have the physiology to produce eggs and the physiology to produce sperm. They are not “male” or “female;” they are both.
Many other organisms can switch between the sexes, and in many cases do so obligately (i.e., all individuals start out as one sex and switch later in life). This is one of the places where epigenetics comes in. Anemonefish (aka clown fish) are a good example (Todd et al. 2016). Anemones are inhabited by a male-female pair, where the female is larger and dominant. Individuals start off life as males and pair up with a female, but if that female dies, this causes epigenetic changes in the male, resulting in it changing sexes and becoming a female. Thus, if Finding Nemo was biologically accurate, when Nemo’s mother died, Marlin (his father) should have changed sex and become Marla.
In many other species, individuals do not change sex as adults, but their sex is determined by the environment as they develop. Some (but not all) turtles provide a good example of this (as do crocodilians, some lizards, etc.). They are what we call temperature sex determined (TSD), and the temperature at which the eggs are incubated determines the sex of the offspring. I don’t want to get too technical here (and indeed there are important pieces of information that we don’t have yet), but I do want to briefly walk through some of how this works because it is instructive (see a more detailed overview here: Lance 2009). During early embryonic development, sex has not been determined (this is true in humans as well) and whether an embryo becomes a male or a female depends on the hormones present. Under many conditions, the embryo will develop as a female, and this seems to be largely driven by the hormone estradiol, which is made from testosterone via the enzyme aromatase. At certain temperatures, however, aromatase stops converting testosterone into estradiol, ultimately resulting in the development of male characteristics.
I went through all of that info on TSD because that background knowledge lets us look at some import questions. For example, what happens if we raise eggs at a male-producing temperature, but we supply them with estradiol? The answer is usually that females develop (Lance 2009). In other words, even though temperature usually determines sex we can over-ride that and produce a different sex. Further, the fun doesn’t stop there, because in at least some cases, we can take turtle species that do have sex chromosomes, paint the eggs with estradiol, and get hatchlings with female physiology even if they are genetically male (Freedberg et al. 2006)! In other words, we can make turtles that have male sex chromosomes develop female phenotypes, including the ability to lay fertile eggs. This is why I’ve been arguing that chromosomes sometimes determine sex, but they don’t define it. We can change the sex to be something other than what was determined genetically. To put that another way, even though chromosomes usually determine sex in these species, we can override that and make the estradiol treatment determine sex.
Similarly, there are some lizards that are usually genetically sex determined (i.e., sex is based on chromosomes) but at certain temperatures, there is an epigenetic effect and the temperature overrides the genetics and determines the sex of the hatchlings. In bearded dragons, for example, at high temperatures, animals that are genetically male (based on chromosomes) develop as females and produce fertile offspring (Holleley et al. 2015). So, if you want to insist that chromosomes define sex, rather than determining it (under normal circumstances), then you must claim that these lizards who are running around laying fertile eggs are actually males. This is a notion that any biologist would scoff at because, again, that’s not how we define sex. If you are going to claim that males are laying eggs, then you have invented your own definition of “male” that biologists do not accept.
Finally, you may be wondering, given all this complexity with TSD and chromosomes, can you ever get intermediates? The answer is, yes! There are situations where individuals don’t develop entirely as male or entirely as female and instead end up developing partially as both (Ewert and Nelson 1991), which makes it pretty impossible to maintain a view that sex is binary. In other words, up until this point, you could have tried to make a post hoc change to the original argument and claim that, “there are only two sexes, and it is determined by physiology,” but that doesn’t work, because some individuals have aspects of both male and female physiology.
The point that I’m trying to get at here is that sex is complicated. It is clearly not as simple as a binary state determined strictly by chromosomes, because we know that you can have reproductive “females” who are genetically “males.” We know that there is more to sex than simply the chromosomes. and we know that environmental factors can override the genetics. Now, you may protest to this because I have been using examples from non-human animals, but that counterargument misses the point. The point is that traits are more complicated than a simplistic understanding of genetics would lead you to believe, and there is no reason to think that sex is only complicated in non-human animals. Indeed, as I’ll explain in the rest of the post, sex is extremely complicated in humans. To put that another way, using non-human animals is a good way to get people to lower their biases and look at the evidence, and as you’ll see, the bizarre situations in other animals are highly analogous to what happens in humans.
Sexes in humans
Let’s being by looking just at the sex chromosomes. In humans, you have probably heard that there are two possibilities for sex chromosomes: XX and XY, but that is not correct. In reality, there are many possible combinations, and it’s not that uncommon for someone to have an atypical number or arrangement of sex chromosomes. Indeed, one large study found that 1 out of every 426 people (2.34 out of 1,000) had one of these conditions (Nielsen and Wohlert 1991).
For example, some people get extra X chromosomes. When this is associated with a Y chromosome, it is known as Klinefelter syndrome, and people with it can be XXY, XXXY, or even XXXXY. These unusual genotypes are associated with a combination of male and female phenotypes (with female traits being more prominent when more X chromosomes are present). People with this condition have male genitalia, but they are often have small testes and are sterile or have reduced sperm counts, they have less body hair and often no facial hair, they have lower testosterone levels, and in some cases they develop breasts (Visootsak and Graham 2006). So here, we have people who have two X chromosomes, but also a Y chromosome, breasts but also a penis, testes but low testosterone levels, etc. They simply don’t fit neatly into the discrete boxes of “male” and “female.”
Extra X chromosomes can also occur without the presence of a Y, and you can have someone who is XXX (sometimes called “superfemale”). People with this present mostly as normal female phenotypes, but they are taller on average, and often have learning disabilities (Tartaglia et al. 2010). Things often become more severe when there are four X chromosomes (“tetrasomy X”; XXXX). Some people with this develop normally, but others do not experience normal puberty, don’t develop a normal female phenotype, and are infertile. Beyond this, some individuals actually have a full 5 X chromosomes (XXXXX) and experience even more severe symptoms. Here again, we have atypical chromosome arrangements resulting in different phenotypes.
There can also be unusual numbers of Y chromosomes. For example, some people are XYY. These individuals have mostly normal male phenotypes and are usually fertile. Others may have XYYY or even XYYYY. These conditions are quite rare making it hard to generalize, but behavioral problems such as aggression have been reported in several cases (Abedi et al. 2018).
Additionally, there are XXYY individuals. These individuals are largely similar to XXY individuals, though there are some differences (Tartaglia et al. 2008). Like XXY individuals, they are generally sterile, and have reduced male features (e.g., small testes).
Finally, there is a condition known as an X monosomy (Turner syndrome; XO). This occurs when an individual has a single X chromosome and either no Y or sometimes a partial Y. These individuals appear female, but are generally infertile and do not have properly developed gonads (Fryns and Lukusa 2005). I want to pause here for a second to note that you can get a situation where someone has part of a Y chromosome. So if your definition of sex is based on the presence or absence of a Y, how do you define someone who has part of a Y? Are they only partially male?
By this point, it should be abundantly clear that sex in humans is far more complicated than XX vs XY, and there are lots of genotypes and lots of phenotypes. It should be obvious that chromosomes determine sex rather than defining it, but there are still more layers of complexity that we haven’t gotten to yet. What if I told you, for example, that it is possible to be born with normal female genitalia, even though you have a Y chromosome? This is a condition known as Swyer syndrome, and it’s often a result of a mutation on the SRY region (aka testis-determining factor) of the Y chromosome, but many other genes can cause it as well (Thomas and Conway 2014). These genes often play key roles in activating the right chemicals for an embryo to develop into a male (think back to the turtles earlier for an analogous situation), so when they are modified, those chemicals don’t get produced at the right amounts. As a result, people with Swyer syndrome have a predominantly female phenotype, but instead of having either testicles or ovaries, they have “streak gonads” which are undifferentiated pieces of tissue that can produce neither eggs nor sperm. People with this condition typically don’t go through puberty and require hormone treatments to develop secondary sexual characteristics such as breasts. However, people with this condition can usually carry a child and give birth if an embryo is artificially implanted. I want you to stop and think for a second about just how complex this is. Here we have people who have a Y chromosome, but also have vaginas, don’t have either testes or ovaries, but have all the other female reproductive physiology and can carry a child if implanted with it. The line between male and female is really blurred in this situation.
The inverse of Sewyer syndrome is “XX male syndrome.” This condition produces individuals with typical male genitalia despite the fact that they do not have a Y chromosome. The cause of this is usually a mutation that resulted in the SRY region ending up on an X chromosome (Anik et al. 2013). Much like Sewyer syndrome, individuals with this condition are generally sterile and often have reduced testes.
There are other situations that are even more bizarre. For example, there are documented cases of people developing “ovotestes.” These are gonads that have some of the features of a testis and some of the features of an ovary. This often occurs in people who are XX but have a mutation on the RSPO1 gene (Tomaselli et al. 2011), which results in ambiguous gonad development. Others actually have both an ovary and a testis and were historically referred to as “true hermaphrodites.” This can occur in both XX and XY individuals (though XX is more common) as well as individuals with some of the chromosome abnormalities described earlier. Further, some individuals with this condition are actually fertile and have children (this usual happens when one gonad is developed and the other is an ovotestis; Krob et al 1994). In other words, there are people who are reproducing even though they have both ovarian and testicular tissue (this is more common in mothers but there also people who are fathers despite this condition). You may remember from the beginning of this post that biologists have typically defined sex based on the physiology required for producing sperm vs eggs. So how are we supposed to classify these individuals who have both physiologies?
There are also cases of individuals who are chimeras. In other words, they have two sets of DNA, and in some cases, one of those sets is XX and the other is XY. In some cases, this has little effect on individuals, and they can reproduce, but in other cases, it results in the development of either ovotestes or other odd combinations of gonads as described earlier. Nevertheless, some of these individuals can still reproduce (Verp et al. 1992). To put that another way, there are people who have a Y chromosome, and have testicular tissue, but still produce eggs and give birth. Now, if you are going to insist that things are as simple as, “if you have a Y you are a male,” then you must argue that these people are males, even though they have mostly female phenotypes and give birth. This is, again, not something that any of the biologists I know would accept.
Beyond all of that, we know that there are epigenetic effects at play in sexual development (Gunes et al. 2016). There are, for example, epigenetic effects on the expression of the SRY region. Exactly how this plays out in developmental sex disorders (DSD) is still poorly understood because epigenetics is such a new field, but we know that there are epigenetic effects that influence the development and expression of male and female traits (phenotypes), and as this field expands, it is likely that we are going to discover that sex is even more complicated than we currently realize (we’ll have to wait and see).
As you can hopefully now see, the topic of sex is extremely complicated, and there is far more to it than simply XY = male, XX = female. There is a whole suite of genotypes and phenotypes, including individuals that are XO, XXX, XXXX, XXXXX, XXY, XXXY, XXXXY, XYY, XYYY, XYYYY, and XXYY. Further, there are individuals who are XX yet develop mostly as males, and there are individuals who are XY but develop mostly as females. There are literally people who give birth, despite having a Y chromosome. There are people who have both ovaries and testicles. There are people who only have part of a Y chromosome, etc.
So, if you are going to insist that Y = male, you are going to have to make some bizarre claims. For example, you are going to have to say that XY individuals with an SRY mutation are, in fact, males, despite the fact that they were born with vaginas, lack testicles, and, if implanted with a fertilized egg, can carry a fetus to term. You are literally going to have to say that a male can give birth. Similarly, you are going to have to say that some XX individuals are females, despite the fact that they have mostly male physiology (including penises). Those are, of course, nonsense positions that biologists don’t accept. Biologically, sex is defined by the physiology needed to produce particular gametes (eggs or sperm), not by sex chromosomes, but recent years have shown that this simply is not a binary situation. There are many individuals that have aspects of both male and female physiology, thus making it impossible to use binary categories.
Let me put that another way. Given the existence of individuals with conditions like XXY who have some female traits and some male traits, the existence of individuals who appear female despite being XY, the existence of individuals with both an ovary and a testis, the existence of people who give birth despite having a Y chromosome, etc., which of the following descriptions seems more accurate, “sex is strictly binary; if you have a =Y you are a male, if you don’t you are a female, no exceptions” or “sex is a complex trait with many genotypes and phenotypes as well as epigenetic factors. It is a spectrum of traits and cannot adequately be described using strictly binary categories.” Which of those does a better job of describing the enormous variation that I have discussed in this post?
Again, to be clear, I’m not making any political or social arguments here. What you do with this information and how it affects your views is up to you, but you must accept facts, and the facts clearly show that biologically, sex is more complicated than a simple binary dichotomy.
Rules for commenting on this post
As explained, this post is solely about the science. If you think I am wrong about the science, feel free to explain, but I do not want the comments to divulge into endless political and social debates. As I said, for the sake of this post, I am just presenting the science. What you do with that is up to you. Comments that are not about biology or that tack political arguments onto biological ones will be deleted. Similarly, if you think I am wrong, please actually explain why rather than just saying, “no, Y = male.” Actually deal with the points I raised and evidence I presented. Also, be civil (see the Comment Rules for my more general policies).
Addendum 1 (Straw man accusation): Quite a few people have accused me of a straw man fallacy on the grounds that no one actually thinks that there are strictly two sexes. This is empirically incorrect. There are a lot of people (particularly in American evangelical circles) who argue adamantly that there are only two sexes and everyone is either male or female, no exceptions. I have many family members and former friends who I have frequently heard make this exact argument, and I have seen it made countless times by others (including in response to this article and by politicians). So I have not committed a straw man.
Addendum 2 (the “10 fingers” argument): The other frequent response is that we shouldn’t define things based on relatively uncommon variations, usually accompanied by the claim that, “there’s nothing wrong with saying that humans have 10 fingers, even though exceptions exist” (sometimes I get numbers of limbs or other analogous arguments). There are two points to make here. First, if someone insisted adamantly that there is only one number of human fingers (10) with no exceptions, they would clearly be wrong. Even so, it is clearly wrong to say that there are exactly two sexes, with no exceptions. That is the position I am arguing against. Second, as explained in the article, there is nothing wrong with saying that most individuals are male or female, but that strict dichotomy does not work in all cases, and the situation is more complicated and nuanced than a simply binary. Some individuals cannot clearly be defined as male or female. That is all that I am arguing here, and it is completely analogous to the correct position that, “most individuals have 10 fingers, but there are exceptions and the situation is more complicated than a simple rule that all individuals have 10 fingers. There are some who do not follow that.” So if your argument is that the binary criteria works in most cases, but there are individuals who cannot be clearly defined as male and female, then great! You and I agree, and you are claiming exactly what I was arguing.
(see this post if you have trouble accessing these for free)
- Abedi et al. 2018. Rare 48, XYYY syndrome: case report and review of the literature. Clinical Case Reports 6:179–184.
- Anik et al. 2013. 46,XX Male Disorder of Sexual Development: A Case Report. Journal of Clinical Research and Pediatric Endrocrinology 5:258–260.
- Ewert and Nelson 1991. Sex determination in turtles: diverse patterns and some possible expliantions. Copeia 1991: 50–69.
- Freedberg et al. 2006. Long-term sex reversal by oestradiol in amniotes with heteromorphic sex chromosomes. Biology Letters 2
- Fryns and Lukusa 2005. Monosomies. Encyclopedia of Life Sciences.
- Gunes et al. 2016. Genetic and epigenetic effects in sex determination. Birth Defects Research Part C Embryo Today Reviews 108:321–336
- Holleley et al. 2015. Sex reversal triggers the rapid transition from genetic to temperature-dependent sex. Nature 523: 79–82.
- Krob et al 1994. True hermaphroditism: Geographical distribution, clinical findings, chromosomes and gonadal histology. European Journal of Pediatrics 153:2–10
- Lance 2009. Is regulation of aromatase expression in reptiles the key to understanding temperature-dependent sex determination? Journal of Experimental Zoology 311:314–322.
- Nielsen and Wohlert 1991. Chromosome abnormalities found among 34,910 newborn children: results from a 13-year incidence study in Arhus, Denmark. Human Genetics 87:81–83.
- Tartaglia et al. 2008. A new look at XXYY syndrome: Medical and psychological features. American Journal of Medical Genetics A. 146A:1509–1522
- Tartaglia et al. 2010. A review of trisomy X (47, XXX). Orphanet Journal of Rare Diseases 5
- Thomas and Conway 2014. Swyer syndrome. Current Opinion in Endocrinology & Diabetes and Obesity 21:504–510.
- Todd et al. 2016. Bending genders: The biology of natural sex change in fish. Sexual Development 10.
- Tomaselli et al. 2011. Human RSPO1/R-spondin1 Is Expressed during Early Ovary Development and Augments β-Catenin Signaling. PLoS One 6:e16366
- Verp et al. 1992. Chimerism as the etiology of a 46,XX/46,XY fertile true hermaphrodite. Fertility and Sterility 57:346–349
- Visootsak and Graham 2006. Klinefelter syndrome and other sex chromosomal aneuploidies. Orphanet Journal of Rare Diseases 1.