85. The Clustering Theory of Gender Specification and Enumeration

(Epistemic status: a long-held theory/frame that I am finally feeling OK with putting on this blog; may now be somewhat obsolete; definitely incomplete. Descriptive in a very loose way, only barely prescriptive. Please do not cancel me about this. For PR, JM, and probably IL.)

 

I have this theory of gender, built on some obvious-seeming observations and frankly sorta facile charts as started to make the rounds on the internet circa 2014. Then I thought about it, and elaborated it. Then I thought about and elaborated it some more. The core is this: there is a high-dimensional (or perhaps hyperbolic) genderspace, individual people inhabit points in that space (mostly nearby to the perihuman gender manifold), what a "gender" is is a natural-seeming cluster in that space, and the number of genders you are willing to accept to exist affects which genders you consider there to be. Let's break that down, more slowly this time.

Consider those cute little two-axis gender graphs on Quadrant I of \(\mathbb{R}^2\). Amount of femininity on the x-axis, amount of masculinity on the y-axis, mark yourself somewhere on it, you probably know what I'm talking about. But this feels flat somehow even on the face of it, and I'm not talking about the part where it's literally 2D. There's no sense that any given location is privileged, and it feels like there's probably more axes of important variation than that. I won't even try to enumerate all the axes or even try to describe them yet, but I'll simply note that they sure seem to exist in multitudes.

 

But this immediately gives us two more pieces of the model: first, that clustering is of paramount importance. Think back to that overly simple 2D map of genderspace. Now scatter all the people you know reasonably well on it. I claim that you'll see some patterns in the noise. You likely marked a broad splotch of varying density of "feminine" and another similar one of "masculine". If you thought about it a little or if you have interesting friends, you might have marked a few points outside the quarter-annulus corresponding to some reasonable unit circle - people who cultivate extreme masculinity, say, or grandmothers who aren't at all masculine but perhaps not quite as feminine as they once were; delightfully genderfucked enbies towards the upper right and also actual literal babies near the origin. Some major clusters appear, along with some minor clusters and a scattering of isolated points. (The picture below isn't of people I know, just a suggestive sketch.)

 

But as I remarked before, this 2D picture is vastly oversimplified, and there's lots of axes it fails to take into account. To name just a few of many:

• How do they present themself? How do they want to be seen? 
• How much sex do they want? With who? How much do they display that level of desire?
• How many children do they want? What's their attitude towards raising those children?
• What do they consider their social role to be?
• What virtues do they value, and how much? Steadfastness? Self-knowledge and vulnerability? Curiosity? Accountability? Vigor?

And the sum total of these very very many possible axes of behavioral and presentational variability put you somewhere in some colossally preposterously high-dimensional space which I am not even going to try to depict here. Suffice it to say that high-dimensional space is spiky and lonely and - all the same - perihuman gender presentations seem to fall into a low-dimensional manifold within the space of all in-principle possible genders. Alternatively, I mentioned the possibility of a hyperbolic genderspace because it seems pretty straightforward to trace your way from one gender to another and consider all the points in between, all of which seem like plausible (if potentially rare or odd) gender presentations; that would instead indicate a low-dimensional hyperbolic genderspace most of which perihuman genders would then take up within some region.

So, so far:

• Gender can vary, and is a part of perihuman identity and presentation.
• Gender has a large number of distinct axes or qualities along which it can vary.
• As a simplifying assumption, any given person has at most one gender at any given time, and that gender is fairly stable over relatively short periods of time.
• As another simplifying assumption, we are treating differences in culturally-accepted gender presentation as out of scope, but they definitely exist and are relevant here; gender presentation is dependent on a cultural or social context. 
• Genders are natural clusters of points in genderspace.

So far, so anodyne, probably. Here's my spicy addition: the set of genders that you consider to exist depends largely on the space of gender presentations that you see and how many genders you are willing to admit.

• If you're dead-set on there being two genders, you probably think that there's just male and female and nothing else. You see no distinction in gender between the late Betty White and Kim Kardashian, and you probably want to put baby boys in "Womanizer" onesies.
• If you think there's maybe three genders, then you might divide things up as something like {male, female, fruity/nonbinary/other} or {male, female, neuter}.
• If you think there's surely four genders, then you might have a classification of {male, female, neuter, nonbinary} or even {cis male, cis female, trans male, trans female}.
• If you admit five genders, then you might have an ontology of gender something like (e.g.) {male, female, neuter, nonbinary, intersex}.
• From here on out it gets fuzzier but in the high single digits to teens and twenties, you might see additional genders showing up including "grandmother"/"grandfather", "demiboy/girl", "hyper-male/female", further refinement of nonbinary types, a smattering of gay, lesbian, and bi presentations (e.g. "butch" and "queen"), "androgyne", "bissu", "hijra", "two-spirit", "fa'afafine", "eunuch", "sworn virgin", "tumtum", and "baby". Notably a lot of these are extremely culture-dependent and might well show up much earlier!
• Further out you start getting xenogenders. "Catgirl/boy", the classic "puppygirl" archetype usually found among transfemmes, "voidgender", "faegender", "robogender", and "wizard". The sort of archetypes you might find largely in online spaces.

The exact ordering isn't the point, and will probably differ meaningfully for everyone past four or so. The point is that there are a few genders that most people observably agree on, and many more that only seem to arise for people who are willing to accept the existence of more genders; such further-on genders are often culturally-dependent, arguably tied to a sexuality, harder to understand, or have limited acceptance.

Backing this up, people tend to collapse their stated gender down to a more widely-accepted gender assertion with a larger clustering basin in systems that demand legibility, like work or official documents - someone might consider himself a foxboy wizard and say as much in more accepting spaces, but as far as work and passports are concerned, they're male. Even in less restrictive spaces, probably few people care that you're voidgendered, and you might largely round off to neuter or nonbinary; that foxboy wizard might decide to pass as a demiboy, nonbinary, or simply male.

And you, what's your gender? In what circumstances? And how many genders do you generally accept?