familiarity breeds contempt

tl;dr people don't use "gender" consistently as a concept, and I'm not sure how to deal with this

Gender (as in being a man, a woman or nonbinary) is clearly an important concept that affects lots of people's lives, mine included. But can we define it in a coherent way that matches real-world experiences? Here are some of the definitions I've seen, and why I don't think they make sense. I'm not trying to argue that these definitions couldn't be used to define gender or categorize people as men, women, or nonbinary, but just that these definitions don't reflect how we actually use those categories, and therefore aren't very useful for thinking about gender in today's world (for instance, figuring out how to combat misogyny, question one's own gender, relate to people of a different gender, etc.).

I'm very open to pushback or argument in favor of any of these definitions, or proposals of new definitions that are more workable.

I. The Dictionary

Merriam-Webster defines "gender" as either:

(1) grammatical gender (not what I'm talking about) (2a) "sex": either of the two major forms of individuals that occur in many species and that are distinguished respectively as female or male especially on the basis of their reproductive organs and structures, (2b) "the behavioral, cultural, or psychological traits typically associated with one sex", or (2c) "gender identity", in turn defined as "a person's internal sense of being male, female, some combination of male and female, or neither male nor female."

This actually does a pretty good job of capturing different ways I've heard people talk about gender, as defined principally by biological state, adherence to gender roles/norms, or internal state of being. I've also seen a few more ways of looking at this, also flawed (eg societal treatment, desire to be X, circular argument).

II. Why not define gender based on behavioral or cultural traits?

In my opinion, this is the strongest of the Merriam-Webster definitions, but I believe that these traits are not in fact sufficiently correlated with the things we mean by 'gender' in everyday speech. The straw-man version of this argument is that the "cultural traits" associated with gender roles, while a less regressive version of this argument holds that gender is some type of 'performance' or 'presentation' which may not adhere to traditional stereotypes.

II.a. Why not define gender based on adherence to gender roles?

Traditional gender roles are not universally endorsed by each gender, and typically one recognizes that a cis male person who enjoys stereotypically feminine interests like painting his nails or fashion is not a woman, and a cis female person who enjoys stereotypically masculine interests like hunting or engineering is not a man. In fact, we usually take such claims as misogynistic when applied against women, and celebrate women in stereotypically masculine activities, such as women in weightlifting. Most women and men have some traits that do not align with their gender's role, and nearly everyone has some deviation. Gender roles come from gender, not the other way around.

II.b. Why not define gender as a performance or presentation?

This is at least less obviously unworkable than gender roles, since the notion of what behaviors constitute the 'performance' or 'presentation' gender is underspecified; but it seems like in general it comes down to some combination of personality, aesthetic, and fashion. Maybe that's really how people use the labels man and woman, and we should just accept that...but I really don't want to. I feel like a cis female person with an assertive, argumentative personality; traditionally masculine fashion, like cargo shorts or suits; no interest in makeup; and a relatively androgynous appearance (short hair, breasts and/or hips not accentuated by clothing, etc) is definitely still a woman. Also, lots of people like this exist in the real world, and don't generally get lumped in with men.

Maybe there's something else to performing gender or presenting as a gender besides these aspects, but I haven't thought or heard of it. Please let me know if so.

III. Why not define gender as an internal sense of maleness or femaleness?

Because not everyone has such a sense innately without knowing what to look for, and proposals of what to look for are flawed. (Also, it's difficult to verify claims of internal sense that don't lead to any other actions or traits, so adopting this definition of gender might mean we have to avoid any situations where people might lie about their gender.)

It's unfortunately very hard to find any kind of academic surveys of whether people have this internal sense of gender, and if I see a survey that >90% of people have this sense then I'll accept the internal sense definition as legitimate. But until then, we'll have to make do with shoddy sources of data and anecdotes. I and several people I know — at least a quarter of the people I've spoken to about this — don't have such a sense; this Reddit poll and this Tumblr poll both find that people are split on the matter. Again, I'd love some real data on this! But so far I'm not convinced that this is true. You can just bite the bullet here and admit that a large chunk of the population is agender or nonbinary; that may in fact be true. But probably lots of the people you're applying that label to haven't heard the word "agender," so I'm not sure this really reflects how people use the concept of gender in practice.

Another option is that the "internal sense" doesn't refer to innately feeling that one is a man or a woman or neither, but some other gender-related feeling, like preference for particular sex characteristics, preference for being a certain gender, or anti-preference for either of those things (gender dysphoria).

III.a. Why not define gender as preference for being a certain gender?

Because (like the definition analyzed in section V.b.) this is sort of circular. If you don't know what a man or a woman is (aside from someone who wants to be a man/woman), how can you know you have a preference for being that gender? (If you think this is a strawman, here's a website that endorses this, saying "it’s your desire to be a girl that makes you a girl, nothing more and nothing less." Maybe this is a weakman, though - I'd welcome stronger versions of this argument.)

III.b. Why not define gender as a preference for particular sex characteristics?

Just as with the internal sense of gender, some non-obviously-agender people seem to lack such a preference: the cis-by-default, who are contented with their body but wouldn't object to having the opposite sex's body. Also, some people might rather have other sex characteristics for reasons that don't necessarily seem related to *gender identity*, like preferring to be seen as a man to avoid sexism, or wanting female sex characteristics to attract men (or lesbians).

I think that gender dysphoria is mostly the inverse of III.a. and III.b., so I won't really be addressing it here. I think it's a pretty serious thing and I hope that people with dysphoria are able to transition and potentially change their gender, but at the same time I'm not convinced that "would have dysphoria about not being X gender" is a good definition of X gender.

IV. Why not define gender as biological sex?

In short, because of transgender people. You have basically three ways of construing biological sex, depending on how you want to classify various trans and intersex edge cases: chromosomes and/or sex assigned at birth, genitalia, or hormones.

IV.a. Why not define gender as having XX or XY chromosomes, or being assigned female/male at birth, or what gametes their body produces?

I think probably 95% of the readers of my blog already understand why this is a bad idea, but for those who don't: transgender people change their genders to not match the ones they had at birth nor their chromosomes. It's generally considered transphobic to categorize trans people as their birth gender, and even if you don't care about that, putting fully-transitioned trans people whom everyone assumes to be a man/woman into the opposite gender is just confusing. If you doubt me, Google images of Buck Angel (an XX AFAB) or Hunter Schafer (an XY AMAB)*. This would be very confusing as well as transphobic.

IV.b. Why not define gender based on what genitalia someone has?

This is probably the most defensible of the biological/physical definitions; unlike chromosomes/ASAB/gametes, trans people often do change their genitalia. However, you still have a similar issue when it comes to trans people who are functionally members of the gender they transitioned to, but haven't had surgery yet or don't plan to; in particular, lots of trans men get called men or treated as men even though they haven't had phalloplasties.

IV.c. Why not define gender based on what hormones someone has?

Children don't have very high levels of sex hormones: typically, male 9-year-olds have 0.45 ng/dL or less of testosterone, and 9-year-old girls have... 0.46 ng/dL or less. You can't use estrogen for this, because after menopause, female estrogen levels drop down to or below (!) typical male levels.**

Yet gender is very salient in childhood, and in fact one often hears about how gendered socialization into what a boy or a girl 'should' be can affect us well into adulthood. So, it doesn't make sense to adopt a definition of gender which labels 9-year-olds as indeterminate gender.

V. Why not define gender as something else?

V.a. Why not define gender as how society perceives someone?

This was my preferred definition of gender for quite a while, since it bypasses both biology and internal mental states. I still think it's probably my favorite of these options, even though almost no one else seems to use this definition when asked! (Even if people's behavior and attitudes seem to match it.) However, after arguing with one of my friends about it, I've come to see two problems with it.

Firstly, genderedness of social treatment varies. If we are saying that being treated as women (or cis females) are treated makes one a woman, then it follows that there are some ways that society treats women. Suppose person X experiences such treatment occasionally, perhaps because they rarely experience sexism and are mostly not treated according to their gender, while person Y experiences such treatment very often (maybe they, idk, get catcalled a lot because of a female appearance, or their family or friends heavily pressure them to adhere to gender roles). This model says that person X and person Y are both women, because they are both perceived by society as women. But why shouldn't we say that person Y is more of a woman than person X, because they experience more such treatment?

You can say "because that doesn't make sense, either you are a woman or you aren't," which is fine but feels arbitrary (other gendered experiences lack this problem because they can't vary in intensity - but it is true that people who define gender as chromosomes don't usually say that someone with XYY is "more of a man" than someone with XY); or you can bite the bullet and allow some people to be more of their gender than others. However, I worry that in practice this will end up falling into many of the same issues as defining gender as adherence to stereotypes or as presentation (see section II).

The second objection is that social perception is inconsistent; perhaps person Z is treated as a woman in some situations and as a man or as an indeterminate gender/nonbinary person in other situations. The theory would then have to conclude this person was in some kind of in-between category, which may be inappropriate when it comes to some categories of people who are sometimes treated as one gender and sometimes as another, like extremely butch women, crossdressing men, and transgender people early in transition. I think in some sense these people are in between genders more than most people, but I'm not sure if that justifies categorizing them as in between genders, rather than outlier examples of a specific gender.

A third objection is that this is still sort of circular, like I argue against in section V.b., because how does society (which is just a collection of people making decisions) know what gender to treat someone as, without being circular? One, admittedly slightly self-obsessed, response is that lots of people don't use this definition, and if everyone adopted it then it would be circular. But if everyone adopted this definition we could just stop caring about gender altogether, and it wouldn't exist as a concept and therefore wouldn't have to be defined. If you are against gender abolition in even hypothetical glorious futures, as many people seem to be, then this definition doesn't work.

V.b. Why not define a person's gender as being the gender they identify as?

Because this is a circular argument. Elaborating on my argument against this sort of thing from section III.a., this claim is essentially that a person X is a gender G if X identifies as a G. But claims of identification must come from somewhere! How should one know which gender to identify as? If it's just some sort of voice in your head, then we're back at internal senses of gender, as in section III. If it's not, X must make a decision of which gender to identify as***, and they've chosen gender G. But being gender G is just defined as being someone who identifies as gender G, so it's not possible to know if one is gender G before figuring out what one identifies as.

Another way to think about this is that if you swapped the meanings of the words "men" and "women," so that everybody who identified as a man now identified as a woman and vice versa, under this definition of gender...absolutely nothing would be incoherent.

Perhaps a steelman of this argument is that "woman" refers to a set of people who are all more similar to the "woman" cluster than the "man" cluster, and vice-versa, so one's gender is defined relative to other people and where they stand relative to these clusters. This has many of the same issues as gender roles in my opinion (see section II), but is also probably in contention with section V.a.'s gender-as-social-perception for my preferred way to think about gender.

VI. Why did you just write 2500 words about definitions? And where do we go from here?

Because words should mean things, and gender is important to many people's lives. Also because I want an answer to the question of what gender I am - not that I seriously plan to change my gender, but I have no good explanation of why I'm [gender I am]**** other than "I guess that's how it is, people say I'm that, I'd maybe be unhappy if I were something else? Does it matter?"

As far as where to go, although I really hope I find some universal coherent and useful definition of gender, for now I think we have to muddle along with some strange mix of these poor definitions of gender. But I think it's important to be aware that gender isn't a well-defined concept, and to be mindful of what we mean when we talk about "gender," "men," and "woman"; I think some conflicts are caused by differing definitions of these words rather than other disagreements.

I would really appreciate anybody reblogging this post, or replying as a comment, if they have anything to say on the matter, especially disagreements with the post, defenses of definitions I don't like, and suggestions for better definitions.

Footnotes

A male person with feminine interests may or may not be interested in identifying as trans, but for a male person who prefers to be called a “woman” and “mom” and referred to as “she” (and all the rest), it makes a lot of sense to describe them as having (or “identifying with”, whatever) the female gender. And opting in to that description doesn’t imply any more or less about their interests and clothing choices than it does for female people who use that description.