Something catches the eye—no, not something, somebody—but the passage from something to somebody takes a fraction of an instant. The familiarity of the person’s outline, and how they move through space, makes my inference to their personhood all but guaranteed. Yet, I have no grasp of the particularities of their person and, therefore, no knowledge of their determinate personhood. I have no idea what makes this person the unique individual they are. What else can I infer about this person? Well, I am watching the Olympics, and this person is competing—they must be an athlete—and they are competing in the women’s competition. I take these cues to justify an inference to her being a woman.
However, there was a great deal of speculation online about whether this athlete had a legitimate claim to compete in women’s competitions. Aside from unverified claims about a gender test, most of the outcry seemed to center on the fact that when people look at her, they thought she looked like a man. The relatively sharp jawline, the relatively bulky musculature, and the fact that she is significantly taller than her competitor; these phenotypic traits, some people thought, gave them perceptual access to the falsity of her claim to be counted as a woman.
What is perhaps most striking about this episode is that both boxers at the center of the scandal are, by any standard, female-bodied cisgender women. The core of the outrage is then just the fact that, to some people with massive online platforms, they look like men.
It would be easy to reiterate the moral and ethical rebukes that many have taken in response to this situation, and these are key elements of the discourse about this public episode of misogyny. But what further philosophical lessons can be salvaged from the wreckage? Is it possible that this shipwreck lies downstream from a peculiar epistemological framework?
It is sometimes said that gender is obvious—that one sees it in phenotypic traits—and those who claim that gender might be something other than biological traits of individual human bodies are obscurantists, unfair to facts, or worthy of some related derogatory dismissal. But what exactly is a fact? And is seeing the same thing as knowing? In a key text of analytic philosophy, “Empiricism and the Philosophy of Mind” (EPM), Wilfrid Sellars argues against what is called the “myth of the given,” or the “myth of givenness.” Givenness or “the given” is a phrase that Sellars takes as a piece of philosophical jargon for the “idea that empirical knowledge rests on a ‘foundation’ of non-inferential knowledge of matter of fact.”
After reviewing Sellars’ argument, I’ll return to the story above to consider what, if anything, mid-twentieth-century analytic philosophy can tell us about current public discourse surrounding gender identity and sports.
Sellars begins his essay by considering what he calls “sense-datum theories.” Sellars’ objective in EPM is to pry knowing away from sensing, and he does so by laying out the inconsistent triad of positions that sense-datum theorists endorse. Any two of these positions, he claims, are compatible, but the attempt to hold all three leads to contradiction:
- Sensing an object entails having non-inferential knowledge of that object.
- The ability to sense objects is unacquired (it does not involve learned concepts).
- The ability to know facts is acquired (it relies on learned concepts).
If sensing an object is a form of perceptual (non-inferential) knowledge, then the ability to know facts is unacquired. But if paradigmatic knowing involves using learned concepts, knowing cannot be unacquired in the way that sensing is. It isn’t at all clear what it would mean to know something that I can’t articulate and describe using concepts. Therefore, seeing cannot be knowing.
For Sellars, sensing objects is different from knowing facts because knowing a fact means articulating the determinate character of what is known, as being some specific way as opposed to some other way, and its connection with other related facts. So, I know the fact that some fruit is an apple if I can say something about its color, its shape, the stem coming from the top, perhaps the firmness and resistance it presents to my attempts to squeeze it in my hand, the kind of tree it grew from, and so on. This is why, for Sellars, describing some statement as a statement of knowledge is not “an empirical description” of that statement but a way of “placing it in the logical space of reasons, of justifying and being able to justify what one says.” This epistemological position, taken by Sellars and developed by philosophers such as Robert Brandom (whose 2023 seminar on Sellars’ work I am thoroughly indebted to for any extent to which I understand Sellars), is known as inferentialism.
An approach to the epistemology of gender that takes cues from inferentialism offers a useful way of reframing debates about third-personal ascriptions of gender identity.
First, suppose it is acknowledged that third-personal ascriptions of gender identity contain at least a minimal inferential step rather than a simple perception of “given” gender attributes (whether they be phenotypical or social-symbolic). In that case, it is easy to make sense of how such ascriptions are open to challenge and correction. Rather than taking mistaken ascriptions of gender as incorrect seeings, we can take them as mistaken inferences involving seeings.
Learning to think more carefully about how one makes gender ascriptions could look like Sellars’ myth of Jones. In this myth, the titular Jones learns that he is no longer a reliable reporter of green ties after electric lights are installed in his tie shop. Rather than taking the green appearance of a tie as a good reason to state that “the tie is green,” thereby endorsing the claim as true, Jones learns to withhold his endorsement. Now, when confronted with the green appearance of a tie, Jones says, “The tie looks green,” which involves a suspension or withholding of endorsement so that all Jones is doing is reporting that he is tempted to call the tie green but has good reason to doubt that temptation.
Similarly, when presented with the temptation (because of phenotypic traits) to say that someone is a man but given good reason to think otherwise (for example, they might be competing in the Olympics in a women’s category), we might withhold our endorsement and express, if anything, the temptation to call them a man. Turning to Sellars’ “logic of looks” may seem treacherous if the suggestion is that people should say, “That person looks like a man.” Still, I think this indicates that in everyday usage, “’looks’ talk” does not usually indicate a withholding of endorsement. So, I want to emphasize that the point here is the suspension of endorsement—recognizing that inferences about gender identity are provisional—rather than using “looks” language. Norms of respect and public discourse still apply, and whether “looks” claims of the type “that person looks like a man” also legally constitute harassment is beyond my scope here.
Further, as I alluded to in the first couple of paragraphs, an epistemology of gender that is informed by inferentialism also allows us to consider some of the metatheoretical dimensions of public debates about gender identity. If a third-person ascription of gender is an inference, then we need to clarify what types of evidence and conceptual relationships can be used to justify such an inference. A biological essentialist claims that phenotypic and genotypic traits are necessary and sufficient conditions for inferring gender identity and that talk of gender as something other than, or over and above, sex is mistaken and misleading.
On the other hand, one committed to a variety of performative or sociological approaches, such as West and Zimmerman’s “doing gender” framework, would argue that instead, inferences about gender identity can be drawn more reliably from social-symbolic elements such as the individual’s participation in women’s sports, or her expression of claims to be counted as a woman. This position does not require total rejection of the obvious role that phenotypic traits play in third-person ascriptions of gender, just the weaker rejection of the idea that such traits stand as necessary or sufficient conditions for such ascriptive inferences.
Even though we often use phenotypic traits as heuristic devices to make provisional inferences about the gender identity of people we meet and interact with in day-to-day life, we ought to prioritize the social-symbolic elements of gender identification. We should be ready to revise those provisional inferences when new information is presented. This seems to be a helpful way to avoid the notion that observations of phenotypic traits are necessary and sufficient to infer gender identity (and sex), a mistaken and harmful way of approaching the epistemology of gender that was put on full display during these most recent Olympic games.
Willow McElderry
Willow McElderry (she/they) is an undergraduate student at Arizona State University pursuing an Anthropology major and Philosophy minor. Their publications include “The Health of the People is the Supreme Law: COVID-19, Disability, and Ethics” (2023) and “Logic and Reason? On the Possibility of a Positive Account of their Relationship” (forthcoming) both written as honors contract papers while completing their AAs in Liberal Arts and Sciences at Las Positas College. Her research interests include prosocial behavior among nonhuman primates, bioethics and disability, inferentialism, the metaphysics of logic, gender theory, and the problems generated by naturalistic accounts of normative symbolic behavior.
