Some Thoughts on Gender, Part 1: Social Construction and First-Person Authority

One of the recurrent, and predictable, responses that my last post got was ‘but what is gender?’ or ‘but what is gender identity?’ People suggested that it’s evasive or disingenuous to reject ‘gender-critical’ ideology as transphobic while not offering my own account of gender.

Conveniently, I do have a bunch of thoughts on the topic, but I felt it was important to keep these questions separate, because I think their epistemic and political status is quite different. That is: how exactly to understand gender is a really open, interesting, subtle and complex question, and what I have to say about it is fairly speculative and I expect smart people to disagree a lot. By contrast, that we should reject ‘gender-critical’ feminism as a divisive reactionary movement is more like a baseline position that I think should be (and, I think, in fact is) widely shared among feminists and LGBTQ activists.

(This open letter here by a whole bunch of feminist philosophers makes, I think,  precisely the same point: we can and should explore the complexity of gender, and admit our uncertainties and disagreements about it, while loudly rejecting attempts to leverage that complexity against trans people.)

So I wanted to state my case against the gendercrits by itself. But since so many people demanded an account of gender, and since I think it’s an interesting topic, I’m now going to make some remarks on that topic in this separate series of posts. It’s coming much later than initially expected, because life is chaos and time-management is difficult and there’s a global pandemic, but whatever, time isn’t real so here goes.

I think one particular reason why people raise questions about the nature of gender is that they worry about apparent conflicts between different commitments that seem to be equally insisted upon by progressive activists. So the question isn’t just ‘what is this thing called gender?’, it’s also ‘how can feminist commitment X and trans-inclusive commitment Y both be true together?’

For example, one quite basic question is whether gender is fundamentally a social or a biological phenomenon. I know that some people think of trans identities in terms of a biological ‘brain sex’ that doesn’t match the biological sex of the rest of the body. That’s not the view I’m inclined towards: I think of myself as very much a ‘social constructionist’ about gender. But I think for some people, it’s hard to make sense of a social constructionist account which doesn’t construe trans identities in terms of something like conformity to stereotypes. I think part of what I can usefully do in this post is to just illustrate one way that it’s possible to occupy that bit of logical space.

And I think there are other puzzling combinations of views – like wanting to ‘abolish gender’ while taking individuals people’s genders seriously, or wanting to recognise feminine and masculine roles as asymmetrically oppressive to women, yet not wanting to denigrate femininity and valorise masculinity in the very same way that traditional sexism does. I think that, again, something useful I can do is to just illustrate how a position could combine some of these seemingly conflicting ideas. That position may well turn out to be wrong, but hopefully showing that it’s consistent can undercut certain arguments that such-and-such feminist commitment conflicts with such-and-such trans-inclusive commitment.

So I’m going to organize this series in five sections:

Part 1: How is social constructionism compatible with first-person authority?

Part 2: How can gender be a product of social norms, if individual gender identities can defy social norms?

Part 3: How can monosexual (i.e. gay or straight) desires be oriented towards gender, as opposed to sex, if gender depends on an invisible internal identity?

Part 4: How can the gender system be an asymmetric oppression of women, if femininity is something desirable and valuable in itself?

Part 5: How can I both support getting rid of gender, as a long-term objective, and also see it as important to support individual’s genders unconditionally?


Part 1: Social Construction and First-Person Authority

So first: I’ve seen people say that, since gender is ‘a social construct’, belonging to a gender must depend on social facts like how someone is perceived by those around them. That is, if it’s not about individual biology, it must be about the way that people in your society do or would categorise you. If people see you as a woman, you’re a woman, is the idea. And while that does imply that some trans women are women, it also means that some aren’t (if people don’t perceive them that way) and it’s certainly not very friendly to nonbinary identities. This kind of view doesn’t, as I’ll put it, uphold ‘first-person authority’ (FPA) for gender: it doesn’t say that people are the authorities on their own gender. But the principle of first-person authority is a central part of the shifts in social norms that pro-trans folks are pushing for.

I think FPA is a good principle, and I also think gender is a social construct. I think a lot of people think these two things together! But I can see why there might look like there’s a tension: how can gender be both up to the individual and also created by society?

There are a few different ways to resolve this tension, but here’s the way that I find myself most attracted to: society constructs categories, and part of what we do as members of society is verbally put individuals into those categories. The categories are not essentially and necessarily tied to any criteria: the criteria are a matter for social agreement and contestation, and FPA is the best criterion if we prioritise individual autonomy and social equality.

I think a lot of feminist philosophers are working on views in this broad vicinity: drawing a distinction between, on the one hand, a single ‘meaning’ for all uses of gender terms, which relates them to the broad social history of a gender category, and then, on the other hand, a diversity of alternative ‘meanings’ or ‘criteria’ for applying the term to particular sets of individuals. Some examples include:

The challenge for such a view is, I think, to give a good account of how the diversity of criteria co-exists with the unity of meaning – what makes it true that even when different people gender others using different criteria, they are doing the same thing (‘gendering’), and not just using the same word with different, unrelated, meanings?

In the rest of this post, I’ll sketch out my preferred approach, which develops the thought that gendering is an action, not a statement. This position can also be put in semantic terms, as a claim about gender-terms like ‘woman’, ‘man’, and in many contexts ‘male’ and ‘female’. The position says: these terms do not describe social categorisations (‘X is a woman’ doesn’t report how society actually categorises X), they effect and express social categorisations – they socially categorise.

I sometimes call this position ‘expressivism’ about gender-terms: these terms express something rather than describing something. One key implication is that there isn’t really a correct way to complete a sentence like ‘the word “woman” means…’. There isn’t a correct definition of gender-terms, and gender-ascriptions don’t really have truth values. Gendering someone isn’t a way of identifying an objective fact about them, it’s a way of constructing social reality.

Or to put it using the useful terms used here by Ozy Franz:

…[the statement] “you’re a woman if you identify as a woman!” is not a definition of womanhood. It is a criterion for who should be a woman. It states that our social genders should be fully consensual: that is, if a person says “I would like to be put in the ‘woman’ category now,” you do that. Right now, this criterion is not broadly applied: a trans person’s social gender generally depends on their presentation, their secondary sexual characteristics, and how much the cis people around them are paying attention. But perhaps it would improve things if it were.

Since it is not, properly speaking, a definition, the decision of who should be socially gendered male or female, and how many social genders we should have is not an epistemic decision. This decision can and should be made on purely utilitarian grounds.

(In place of the last line, I might say ‘on purely ethical-political grounds’, because I’m not sure utilitarianism is the right ethical theory to use here, but I am otherwise in complete agreement with the basic point that that there are two different sorts of ‘meaning’ for gender-terms, one that’s common to all uses of them, and one that varies between people who use them in different ways.)

What’s common to everyone’s use of ‘man’ is its function, that it’s a word for assigning people to a certain social category, the category ‘man’. What varies is the criteria people use to guide such categorisations: someone who categorises anyone born with a penis as a man is using different criteria from someone who categorises anyone who identifies as a man as a man. In one sense, these people are ‘defining’ the word ‘man’ differently (different criteria); but in another sense, they’re ‘defining’ it the same way (same function, same social category).

I think this is actually somewhat important, because it means that although pro-trans activists are clearly trying to change something about how we use the word ‘man’ (to change the criteria), they’re not simply ‘redefining it’ in a way that would just substitute a new meaning onto the same sounds – the function of the word is exactly what it’s always been.

The criterion/function distinction also lets me accept the obvious fact that, often, saying that X is a man/ is a woman does convey some factual information: it conveys the information ‘X meets my criteria for the man-category/ for the woman-category’. If you tell me X is a man, and I know that you use ‘man’ for all and only assigned-male-at-birth (AMAB) people, that tells me that X is AMAB (or at least that you think they are); if I know that you respect FPA about ‘man’, that tells me that X identifies as a man. If I don’t know how you use the word, your statement doesn’t tell me any objective non-social facts about X, it just tells me how you categorise X, and invites me to follow suit. But in all these cases, I claim, the word ‘man’ has fundamentally the same meaning.

I think expressivism invites two obvious questions. The first is: ok, but what is being expressed? What is the category we’re assigning people to? I think this sort of question lurks behind the worry people sometimes raise, that gender-terms are going to end up circular, or empty, or ‘vanish into meaninglessness’.

That’s a good question, and the rest of this series will try to go some way towards answering it. To prefigure that partial answer: I think gender-terms express certain ‘ways of seeing’ a person, forms of perception which relate their body to the bodies of the other people who are perceived in that way. I think these ways of seeing – ‘seeing as a man’, ‘seeing as a woman’ – are psychologically deep-rooted, with close connections to a lot of our emotional lives, especially our emotional reactions to people’s bodies. But, crucially, I don’t think these ways of seeing are truth-evaluable: they can’t be true or false, there isn’t an objective fact that they’re trying to accurately represent. So in this first post, I want to focus attention not on the content of what’s being expressed, but on the structural shift that comes from treating gender-talk as expressive, as talk that does something, rather than as talk that reports or describes something.

The second obvious question that expressivism invites is ‘so why use gender-terms in one way rather than another?’ To put it another way, expressivism doesn’t by itself imply FPA, or any other policy about how to gender people. According to expressivism, the semantics of the how gender-terms function isn’t by itself going to tell us how to arrange our society. The case for FPA is, as Frantz says, ethical and political: we should arrange our society to give people as much autonomy as possible in how they enter into it, and to respect the choices they make, and to avoid unfairly privileging some people’s social identities over others. The way to do that, in most cases, including gender, is to socially categorise them in accord with how they wish to be categorised, and with how they categorise themselves.

Obviously the ethical case I just outlined was super-quick, and contained lots of phrases (like ‘avoid unfairly privileging’ or ‘in most cases, including gender’) that could do with a whole lot of elaboration. I’m not going to do that here, partly because it would require its own whole other series, partly because I don’t need to for my aim in this post. That aim is simply to show the compatibility of social constructionism and FPA, and for that it’s enough to simply exhibit one view that combines them, the way that expressivism-plus-an-ethical-case-for-FPA does.

I could understand if some trans folks were unhappy with expressivism for precisely this reason. It doesn’t provide an apolitical case for respecting trans identities, in the sense of a case that didn’t rely on controversial claims about ethical values and social organisation. You might worry that this ‘politicises’ trans identities, makes them into political claims rather than simple personal facts.

But note that it doesn’t ‘politicise’ trans identities more than cis ones: if expressivism is right, then calling anyone a man or a woman, whatever the criteria we use, is participating in a social practice that is not governed by any objective facts, and can be justified only by pragmatic, ethical, arguments.

I’m drawn towards expressivism in part because it seems to me the best way to reconcile the obvious sense in which pro-trans folks are pushing for a change in our use of ‘man’ and ‘woman’, with the idea that they are still associating those words with the same concepts, that they are advocating a change in our criteria for how we use these same concepts we’ve been using, and not just offering up new, unrelated concepts to use instead. But as I say, that may not seem as important a desideratum to everybody, and so expressivism may not be everyone’s preferred theory, and I’m by no means certain this is the right way to view things. I’m open to feedback on this point.

Finally, someone might mount a different objection here:

‘Debating the right criteria to use only makes sense if we can control our seeing-as-a-man and our seeing-as-a-woman, but in fact we can’t. If someone looks a certain way (e.g. ‘like a man’) people will see them in a gendered way that reflects that, automatically and involuntarily. So this whole discussion is pointless.’

There’s obviously a germ of truth here: gender-perceptions are not entirely voluntary. But they’re not entirely involuntary either. The contours of how the voluntary and the involuntary connect here is probably too complicated for me to do justice to, but here are some preliminary observations.

First, we could distinguish ‘bottom-up’ and ‘top-down’ influences on gender-perception, with the former being, roughly, ‘perceptual’ (the way someone looks, moves, sounds, laughs, smells, smiles, dresses, etc.) and the latter being, roughly, ‘conceptual’ (what I know about this person from the past, or what other people have told me, or what they themselves have told me). I think that top-down influences have some, limited, force: even if the person before me ‘looks like a woman’ according my brain’s automatic processing, the fact that other people are calling them ‘he’, or that I’ve been told they’re ‘really a man’, also has some tendency to push me towards seeing them that way.

The power of top-down influences can be shown completely independently of anything about trans issues: suppose I meet a friend who I know as a cis woman, but who is very convincingly dressed as a man for some reason (she’s an actress, she’s a spy, she lost/won a bet, whatever). ‘My eyes tell me man’ (that is, the bottom-up influence of perception pushes me in that direction) but my background knowledge of this person reminds me that they are ‘really a woman’ – and the names and language I and other people use might go either way . If she’s adopted a full-on drag persona with a masculine name, and I keep using it, I’m more likely to see her as a man; if I keep referring to her by her normal name, I’m more likely to see her as a woman. I might swing back and forth between different perceptions, as these two influences battle (between seeing her as a man, and seeing her as a woman-dressed-as-a-man), or I might settle into one perception pattern, or some sort of in-between. (An experimenter might crudely gauge where my perception had settled by seeing what pronouns I used in unguarded moments.)

(Note how this case illustrates a distinction between the criteria of gendering I accept and the influences that affect my actual gendering. The example doesn’t specify whether I accept trans-inclusive, FPA-respecting, criteria, or trans-exclusive, assigned-sex-at-birth, criteria, in my judgement that this person is ‘really a woman’. But either way, that judgement can conflict with my spontaneous perception based on bottom-up influences.)

So there are top-down influences, which are ‘under voluntary control’ insofar as we have control over how we talk about and how we theorise about people, and how the others around us do as well. I think that there’s also a limited degree of pure voluntary control: we can simply will ourselves to see a certain person a certain way, in something like the way that we can will ourselves to see a cloud as a dinosaur, or to see the duckrabbit as a duck or as a rabbit, or more generally to see an object as falling under a given category. I’m not sure how much to lean on this sort of purely voluntary top-down influence, or how far it goes, or how much individual variation there is in it. But it is at least a phenomenon.

Finally, I think there’s a major source of voluntary control that isn’t top-down but rather comes from the way that bottom-up influences are trained by experience. The things that ‘automatically and involuntarily’ make me see someone as a man or a woman are learnt from experience: this must be true, because so many of them are free of any kind of biological basis (e.g. long hair vs. short hair, trousers vs. skirts, make-up vs. no make-up). Consequently, the more I get used to seeing someone as a man or woman, the more spontaneous and automatic it will get. And the more I get used to seeing lots of trans or gender-nonconforming people as their chosen gender, the more that in turn will influence my spontaneous perception of future people I meet.

Again, I want to recognise that all of these forms of direct and indirect voluntary control over gender-perceptions are limited. We don’t control our feelings or our perceptions, and I don’t think it’s useful to feel guilt or judgement over how we spontaneously see someone. But these forms of control do go far enough, I think, to allow us to form and follow conscious policies. Transphobes want us to follow one such conscious policy: even if a trans person looks to us like their chosen gender, they want us to ‘remember’ that they are ‘really’ their birth-assigned gender, i.e. to overrule our bottom-up influences on gender-perception and impose a top-down judgement based on what we know of their medical history. Trans advocates want us to follow another: even if a trans person looks to us like their birth-assigned gender, they want us to ‘remember’ that they are ‘really’ their chosen gender, and to overrule our bottom-up influences on gender-perception and impose a top-down judgement based on how they ask to be seen. Neither is in itself more psychologically possible or impossible, they just prioritise different things.


This entry was posted in Gender, Uncategorized and tagged , , , . Bookmark the permalink.

5 Responses to Some Thoughts on Gender, Part 1: Social Construction and First-Person Authority

  1. Pingback: Some Thoughts on Gender, Part 5: Gender Abolition and Gender Anarchy | Majestic Equality

  2. Siobhan says:

    It’s a bit ironic, after all the objections in your first post basically saying that your response had no value unless you took a crack at the ontology of gender, that you’ve not received any engagement on these pieces.

    • lukeroelofs says:

      Yeah, I think part of that is that the first post riled Stock enough to post about it, and she has a lot more readership than I do.

      • Siobhan says:

        True, but Stock has also made her campaign out to be a “dialogue” on gender, so I find her absence noteworthy.

  3. Pingback: Some Thoughts on Being Nonbinary | Majestic Equality

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.