Biological Science Rejects the Sex Binary, and That’s Good for Humanity

If you guys would just say “teleost” rather than “fish”, all this quibbling could be avoided.


I think you mean genotype, not phenotype.

1 Like

I’m going based on a podcast interview from…oh geez I listened to it after teaching a class one day, so it was pre-pandemic, and then a book I read afterwards, and that musta been 2019, so I apologize but I don’t recall the interviewee/author.

In terms of “frequently”, I don’t mean to imply that it’s a common thing - single digit % or higher. Just that it’s a non-trivial number of people. That would have been a better way to say it.


Correct - in androgen insensitivity the genotype is XY, the phenotype is female. Thanks for catching that.

After some back and forth in my head on this subject, I’ve found myself somewhere in the middle. I have found a resting place there (in the middle), by recognizing that we are seeing a messy example of the evolution of language. On the one hand, there is the traditional biological sex which is determined by what gametes one produces. Egg or sperm. Female or male. The occasional hermaphrodite, and the lability of clown fish or the multiple sexes of slime molds does not provide us an escape hatch to say that there are multiple sexes in our species. It is a classical naturalistic fallacy to say that because of those other species, we too have multiple biological sexes, or that “sex” is only a social construct. Those claims are not entirely true.

Languages evolve, however, and an old-fashioned biologist cannot prevent it even if they wanted to. By example, several decades ago when we talked about a “mouse” as an object, we meant the rodent kind. But now we have computer mice. And so the meaning of that word has evolved to include a new meaning. The old meaning is not now wrong. It just must share space with the newer meaning, and we have to sometimes be clearer about what context we are referring to when we say “mouse”.
So too with the sexes, I think. We still have biological sex, and in that context there is still male and female based on gonadal tissue. But now we recognize that sex is also a personal identity AND sex is a social construct. Our brains tell us what sex we identify as, and the social construct is about how we present ourselves and how others see us. Its all pretty fascinating, really, and I think about it a lot.

So in my view, the old-fashioned binary of biological sex is not wrong, but neither is it wrong to assert that sex is at the same time a matter of personal identity and it’s a social construct. And these things can appear in different combinations, so we have cis- and trans- individuals, and even individuals that claim intermediate genders for themselves. The only people who are wrong in this debate are those who say the other side is wrong. In my opinion.

1 Like

There are many problems with that definition. For one thing it leaves out prepubescent males and post-menopausal females; one might in fact argue that it leaves out all female humans, as their gametes are produced before birth, depending on just what you mean by “gametes”.

My opinion:

Coming up with a definition of “male” and “female” (even for us humans) will always run into some problems. It’s akin to the species problem.

Essentialism (as far as biology is concerned) is dead people!

It also leaves out the castrati, but I don’t see these as useful talking points for describing biological sex. We can resolve most of these cases with a few added words about biological males being those that have the potential to make sperm, and biological females can be similarly described with their eggs. Surely we should be able to usefully describe this aspect of our biology with essentially the same terminology that we would use while describing the biology of zebra fish or fruit flies. Else we will get too bogged down in exceptions and semantics.


That just adds an undefined term. Who has the potential to make eggs? How do you judge potential? And what do you know about oogenesis in placental mammals? The point is that it isn’t as easy as you’re trying to make it.

1 Like

It’s often times remarkably difficult to come up with a rigorous definition of some category without leaving some sort of outlier or exception out, yet somehow we’re nevertheless able to easily distinguish members of the categories at a glance.
I can reliably tell the difference between cats and dogs in almost every case, yet I would have a hard time listing the defining characteristics of each without there being major areas of overlap between them, and there’s probably also going to be exceptions where I would still recognize them as cats or dogs and yet they’d somehow fail to abide by the definition.

I think a big source of the sex/gender controversy comes from this apparent conflict between our everyday experience, and more sophisticated and academic definition.

It’s extremely hard to convince people that something they find they have no trouble doing every waking moment of their lives (telling males and females apart, or cats and dogs) is somehow an illusion and there’s this sophisticated philosophical argument that purports to do away with the whole thing.

Oh yeah can you define x so it catches all cases and doesn’t overlap with y? Maybe I can’t, or it would take me a decade and it would become baroque, and yet somehow I can in almost every case I ever encounter tell them apart at a glance.

There’s a huge disconnect between sophisticated philosophical concepts, and this sort of everyday experience and intuitions.


You would appear to have led a very sheltered life.

That would not appear to be a very substantive response.


With respect to people, unless you are in a nudist colony I doubt that’s what you are doing. If you’re working in delivery, then you may well be telling males and females apart, but there is a decent percentage of intersex that complicates matters.

1 Like

It’s more of a hint at a response. If you prefer, I can belabor the point. How good at you at distinguishing babies by sex? How much of that is purely by social cues, pink vs. blue and such? How much of distinguishing adults by sex results from similar cues? What do you do in the absence of such cues, or when the cues are incongruent or deceptive? Have you ever seen an episode of Pose, for example?

I think I’m as good as most people, who also seem to have no trouble doing these things, and yes I think we all just go by some collection of the most obvious cues. But I don’t really see the point of these questions because I haven’t said there aren’t ambiguous cases where you can’t tell, and of course babies in particular are difficult exactly because most of those superficial anatomical traits you’d normally go by to tell them apart have yet to develop and can be easily masked by clothing.

I think being at a nudist colony would just make the hard to tell cases substantially more rare. But that would seem to be an implicit concession that visible anatomical traits really are highly reliable indicators of biological sex, if not among the most significant contributors to whatever we take to be the defining characteristics of the sexes.

Well again I didn’t say exceptions don’t exist. I just think they’re quite rare, and I don’t think they do anything to undermine the everyday experience that biological sex seems to refer to something physically real and that it has at least something significant to do with anatomical and physiological traits.


So it’s a question of what percentage of error would count as trouble.

1 Like

They certainly correlate fairly well. The thing is, sex is a bimodal distribution, not a binary distribution. This is an important distinction. But I think you missed my actual point, which is that you’re basing your not actually basing your common daily judgements on anatomical characteristics. You’re instead determining gender then assuming sex. This isn’t a bad strategy, since this also correlates fairly well (and is also a bimodal distribution), but still isn’t the thing you said you were doing.

Sure, obviously. As a bimodal, not binary, distribution.

1 Like

I definitely think that has something to do with it, yes. I think if rather than most people falling pretty neatly into one of two categories, that there was a flat distribution with equal amounts of individuals from everywhere on the anatomical spectrum, then I’m pretty sure this idea that sex is binary(or even bimodal) wouldn’t have developed in the first place.

I believe I am taking those into account where and if they are sufficiently obvious. Just to pick an example, most adult biological females visibly have breasts even when they’re covered, and I believe I can even usually tell them apart from manboobs on most clothed individuals. Then there’s facial hair and so on. Sure, things like fashion, makeup and so on are also things that count towards us putting someone into a category, and sure every one of those can in principle fail. But I just don’t think that happens all that often.

You seem to be saying that the fact that the correlations aren’t perfect means I will some times be making mistakes, and because I will be making mistakes I’m not doing what I think I’m doing. I think that’s a strange way to look at matters. No, I really do think I’m telling females apart from males, but I will concede I will some times do that erroneously.

1 Like

My attempt to take such factors into consideration was in the phrase “proper function”; @Mark_Sturtevant similarly used the terminology of “potential”. I.e., even in cases where no gametes are actually produced (due to level of maturity, or the presence of some malformation or disorder) we can often still tell what reproductive role the person’s body is in some sense biologically “supposed” to fulfill.

Of course there is still complexity and cases where it is difficult to tell, but it seems to me that this definition pins down the way “male” and “female” are actually used in biology. (Though perhaps instead of production of gametes, we should speak more generally of reproductive roles.)

To put it in other words: I think we can say there is some sense (based on biology alone) that human beings are “supposed” to have five fingers on each hand (despite the occasional occurrence of people with polydactyly), and in the same sense, there are two reproductive roles and human beings are “supposed” to fall into one or the other of them (despite the occasional occurrence of people with intersex conditions). And (again, in some sense) this explains why we see correlated bimodal distributions of various traits: it isn’t that we’ve been confused into erroneously thinking there are two reproductive roles by these bimodal distributions of traits; rather, there are these bimodal distributions of traits because there really are two reproductive roles.

(Note: under this definition in terms of potential to fulfill a given reproductive role, “male” and “female” are strictly speaking neither mutually exclusive nor jointly exhaustive, i.e., there’s logical space for the “both” and “neither” categories as well. It might be that cases where both capacities seem partially present, such as cases of mismatch between genome and genitalia, fall into one of those. Or it might be that we just can’t tell.)

I think that just because a definition uses terms (such as “potential” or “proper function”) that themselves lack precise definitions, doesn’t mean that the definition isn’t an advance in understanding. For one thing, it is still breaking the concept down and relating it to other concepts. For another, as @Rumraket pointed out, we can often judge where concepts apply and where they don’t even without precise definitions.

1 Like

Humans are very good pattern seeking, coming up with boxes and quickly putting things that we see in our boxes in an attempt to make sense of a very messy world. I think a good analogy here is is “colors”. Humans are able to recognize millions of different colors, and we have come up with numerous fancy names and color codes. However, most languages use only a few basic color categories. English has 11. Russian has 12 (having two distinct words for what in English are 2 shades of ‘blue’) and some languages have as few as 3 or 4. There is an interesting pattern with this. Languages that have 3 or 4 categories tend to categorize colors into ‘light/white’, ‘dark/black’, ‘red’ and (if 4) an additional name for green / yellow or both. ‘blue’ colors are still recognized, but they considered a shade of ‘dark’. When languages acquire new words, they tend to separate green and yellow first, and then a name for blue as distinct from dark. It’s not just that we have different name for different shades. The language has an affect on our perception of the world. For example, Russian speaking people (who have 2 words that cut blue into 2 categories) are better able to see the differences between shades of blue than those who speak English.

So we are very keen at grouping colors, and we apparently have a tendency to categorize colors in a similar manner (the difference is that some make more categories than others). Yet this doesn’t change the fundamental physical aspect of color: visible light is a spectrum.

The point that I am making is that, while humans may be very good at categorizing things, it may not tell us much about the fundamental nature the things in question.

I think you are stumbling upon the issues with the idealism of phenetics, as if taxa represent some ‘ideal’ phenetic form, with each group being defined by essential traits. However, as you have accidentally alluded to, biological essentialism is dead. One reason why cladistics has superseded phenetics. Ergo, we are fish (Sorry, not sorry). There is no ‘ideal’ form or ‘essence’ to what makes a dog a dog or a cat a cat. Similarly true for men and women or male and female.

I also would like to point out that you may be overestimating the reliability of people to recognize such categories. Like how hyenas are commonly mistaken for dogs, even though they are more closely related to cats. Bats as birds is another good example. This is also where our language influence our perception (like with colors). There are languages that don’t make the same distinctions. E.g. many languages don’t have different words for ape and monkey (although apes are monkeys, and humans are apes).

And as others have pointed out, when you guess sex of humans, you make indirect inferences based on their expression, and the way this is expressed depends on the culture (they way they are dressed, haircut, etc). Like when we meet other people we don’t infer sex (nor gender) by checking their genitals, or looking at their crotch and guessing based on the presence / absence of a bulge (at least, I don’t). So there is a preceding social and cultural layer when it comes to humans recognizing each other’s sex, which is often more important to us than inferring people’s sex from their physical anatomy, similar to how you are recognizing cats from dogs - or (more appropriate analogy) how we infer sex in dogs and cats is not the way how we infer sex in people.

Agreed. Try to argue that humans are fish. However, the difficulty of trying to convince people due to the backlash that the arguments provoke does not affect the validity of the arguments.

If I may elaborate the point CrisprCAS9 is making: He is not saying that you can’t infer sex based on anatomical features. He is saying that we rely almost entirely on the social cues of gender expression that we pick up on almost instantly when we turn our gaze at someone whom we have never seen before.

Sometimes a silhouette with little to no anatomical clues is enough:

Even when reducing these to mere abstract shapes:

1 Like