BINARY CATEGORIES, MESSY INDIVIDUALS

Sex is often thought of as a straightforwardly binary categorical variable. From biomedicine to ecology, from textbooks to cutting-edge research, most work in the life sciences understands and operationalizes biological sex as a difference between females and males. But reality often does not fit into neat boxes labelled ‘M’ and ‘F’. Scientists are constantly dealing with variation in the would-be sex traits of the organisms they encounter, including genitals, morphology, hormones, neurology, and behaviour.

Feminist research has been especially good at teaching us that variation is rife when it comes to sex. For example, Joan Roughgarden ([2004]) has collected evidence for diversity in all traits commonly associated with the sexes. Whether hermaphroditism in fish, masculinized female genitals in spotted hyenas, or male pregnancy in seahorses, there is a great variety of ways to mix and manifest large and small gametes for reproduction. Similarly, Anne Fausto-Sterling ([2012], [2020]) points out how traditional characteristics for males and females, such as genitalia, chromosomes, or hormone levels, often come apart in human individuals. Fausto-Sterling also argues that these sorts of traits develop in interaction with social gender categorization practices; so, for instance, bones are affected by nutrition, which in some cultures is different for different genders. Even defenders of more traditional definitions of sex typically recognize variation in sex traits, sex-related characteristics, and sex roles.

Given such vast and complex variation, how do life science researchers nevertheless construct sex as binary and categorical? What strategies do they have for dealing with variation in sex traits? There are a bunch of debates happening right now about how to conceptualize and define biological sex. Instead of telling biologists what concept of sex to use, in our article we look at how sex gets conceptualized in practice. In the process, we point out how a binary, categorical understanding of sex doesn’t go without saying.

We start with an example from ornithological field guides. Field guides contain lots of information about how to identify not just different species of birds but also things like their sex. They do this by emphasizing differences between the sexes. Take, for instance, the hooded warbler, a small bright yellow bird found in Central and North America. In a popular online guide, Birds of the World, females and males are depicted separately, emphasizing the male’s distinctive black hood around its face and the absence of a hood in the female. However, reading further, the guide tells us that female hooded warblers do sometimes have a hood! Depending on age, as well as individual differences, some female hooded warblers can even have a hood that is ‘almost identical’ to the male. What’s going on here? How can a black hood be a sex-distinguishing characteristic that neatly divides males and females, even while some female birds have a black hood just like males? We argue that this guide performs what we call ‘the individualizing variation strategy’.

The individualizing variation strategy is an approach taken by researchers that involves excluding variation from a phenomenon by considering it to be a matter of idiosyncrasies or individual differences, rather than a part of the phenomenon itself. Think, for instance, of statistical analyses, which typically lump individual variation with measurement error and then separate out this variation to get at an association or experimental effect. What we show is that researchers do something similar when it comes to sex and variation in sex traits.

Let’s come back to the hooded warbler. The guide tells us that males have a black hood and females don’t. Although in actual fact some females do have a black hood, this is treated as individual variation in an otherwise sex-distinguishing characteristic. This is an example of the individualizing variation strategy at work.

Hooded warblers differ in more than just their hoods. For example, there is variation in the amount of white in their tail feathers. Tail whiteness actually correlates with sex, but it’s not listed as a characteristic to distinguish the sexes. Why not? The answer is, again, the individualizing variation strategy. It turns out that even though sex is correlated with tail whiteness, it explains very little of the variation in this trait. So, most of the variation is put down to individual differences. This highly variable trait is then excluded altogether from the set of possible sex-distinguishing characteristics for the species.

The individualizing variation strategy works to stabilize a binary and categorical conceptualization of the sexes in the face of variation. Variation may be considered a matter of individual differences in an otherwise sex-distinguishing trait. It might also mean that a trait does not get counted as a sex-distinguishing trait in the first place. Regardless, the individualizing variation strategy assigns non-binary, non-categorical variation to individual differences. This stabilizes the idea that biological sex is binary and categorical.

Of course, individual scientists don’t typically go around explicitly deciding to treat some bits of variation as individual differences and other bits as sex differences. Rather, the individualizing variation strategy is a structural feature of scientific practice. This means it is about learned patterns of behaviour throughout the scientific community. Common goals, values, standards, methods, infrastructures, and so on contribute to making the individualizing variation strategy a widespread part of scientific practice.

Is it always like this? In short, no. When we look at another example, we find a different strategy used to deal with variation in relation to sex. Our second case study concerns the ‘brain sex’ debate. At stake in this debate is whether there is a male and a female brain.

Some brain researchers argue for the ‘brain sex thesis’, which is the idea that there are important differences between the brains male and female humans. This is the basis for the widespread talk of male and female phenotypes of human brains, or sexual dimorphism in brain phenotypes. In contrast, other brain researchers argue that there are no ‘pink’ and ‘blue’ brains, that average differences do not amount to dimorphism of male and female brains. They are what we call ‘brain sex critical’ researchers.

Much could be said about the brain sex debate, but we focus on the role that variation plays and how it’s conceptualized. For example, Daphna Joel ([2012], [2021]), a brain sex critical neuroscientist, distinguishes four sexes, where the largest groups are 3G males and 3G females—people whose genes, gonads, and genitals are concordantly assigned as male or female. Joel recognizes slight statistical differences between brains of 3G males and 3G females in a studied group. For instance, two traits—say, larger overall brain size and larger hypothalamus—may be slightly more common among 3G males than 3G females.

But Joel argues that these statistical differences translate in a ‘mosaic’ way to individual brains. This means that any given individual can have a combination of brain traits associated with both 3G males and 3G females (and, of course, traits that aren’t correlated with 3G sex at all). For example, an individual 3G female may have an overall brain size more associated with 3G males, combined with a hypothalamus size that is more associated with 3G females. Thus, Joel argues that there are no female and male human brains.

What is happening here? Joel and other brain sex critical researchers see variation among individuals as a form of evidence that challenges the idea of pink and blue brains. This variation is treated as a part of how brains and sexes relate, rather than being disqualified as individual differences. This reveals a second strategy that life scientists sometimes use to make sense of variation. We call this the ‘de-individualizing variation strategy’. The de-individualizing variation strategy treats observed variation that could have been attributed to individual idiosyncrasies as, instead, a part of the phenomenon or category of interest.

In using the de-individualizing variation strategy, brain sex critical researchers take variation as evidence that the relation between brains and sexes may not be best understood in terms of male and female brains. They de-individualize variation that the individualizing variation strategy would discard. This has effects on how these researchers conceptualize what their object of study is like. In other words, it affects the ‘ontological picture’ of how brains and sexes relate.

On the one hand, the de-individualizing variation strategy creates room for alternative concepts and models of sex that may fit the phenomena better. Some brain sex critical researchers drop the categorical understanding of sex, while maintaining that there are two sexed types of brains that blend into each other. Others develop pluralist conceptions of brain–sex relations that are neither binary nor categorical, such as Joel’s mosaic metaphor.

On the other hand, the de-individualizing variation strategy affects those researchers who argue for the brain sex thesis. These researchers now need to work hard to make variation concordant with binary, categorical ideas of brain sex. The evidence that the de-individualizing variation strategy makes available cannot simply be discarded; it pushes researchers to modify and nuance their accounts. This leads to more refined approaches to the brain sex thesis, in addition to the alternative approaches to brain–sex relations developed by brain sex critical researchers.

This means that the de-individualizing variation strategy can play an epistemically beneficial role. It can help researchers to spot anomalies in those theories that construe sex as binary and categorical. In contrast, the individualizing variation strategy may play an epistemically harmful role. It can sometimes mask or explain away important counter-evidence to a dominant theory.

This insight has further ramifications. For example, it provides reasons to caution against mandates such as the NIH Sex as a Biological Variable policy: prescribing a specific concept of sex might pressure researchers to make their findings congruent by using the individualizing variation strategy in a detrimental way.

We argue that researchers need to exercise caution when using the individualizing variation strategy, especially for sex. An uncritical use of the individualizing variation strategy might make the binary categorical picture of sex more immune to counter-evidence. This would be particularly problematic when binary and categorical sex concepts interlock with oppression based on a naturalized picture of sex differences. Fortunately, there is a way to counter the individualizing variation strategy: actively employing the de-individualizing variation strategy can open up the conceptual space, allowing researchers to develop alternative pictures of sex.

Noticing the dynamic between the individualizing and the de-individualizing variation strategies offers fresh insights into how researchers perceive variation to maintain or change their ideas of sex. This takes us one step further in our understanding of how researchers travel across the sometimes vast space between concepts, theories, and models on the one hand, and variation among real animals on the other. In particular, we think that recognizing the role played by these two strategies can help create space for reflection and undogmatic conceptual debates. Ultimately, we hope that this will lead to a plurality of ways to conceptualize and study sex, without confusing the map for the territory.

We also have a few more concrete suggestions for scientists. First and foremost, researchers should not simply assume that a binary categorical concept of sex is the only one or the most appropriate one in their study. Instead, they should consider what concept of sex is most suitable—or even whether they need to work with a concept of sex at all. Among others, Sarah Richardson ([2022]) offers advice in this regard.

Second, researchers could consider that the individualizing variation strategy is not the only way to deal with variation. With strategies like the de-individualizing variation strategy, individual variation might be recognized as part of the phenomenon of interest. This change of perspective could have consequences for many of the phenomena studied in the life sciences.

Finally, researchers should aim where possible to be explicit about conceptual choices and strategies. This would prevent the individualizing variation strategy from operating implicitly, helping to highlight where it may be suppressing counter-evidence or limiting research directions.

Acknowledgements

This publication is part of the project ‘The Reconceptualization of Sexual Difference: Exploring Enactivist Approaches and Pluralist Conceptions to Sex-Gender’ (project no. 019.221SG.009) of the research programme Rubicon, which is financed by the Dutch Research Council (NWO).  This project received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant no. 101001145). This paper reflects only the authors’ view and the Commission/Agency is not responsible for any use that may be made of the information it contains.

References

Fausto-Sterling, A. [2012]: Sex/Gender: Biology in a Social World, New York: Routledge.

Fausto-Sterling, A. [2020]: Sexing the Body: Gender Politics and the Construction of Sexuality, New York: Basic Books.

Joel, D. [2012]: ‘Genetic-Gonadal-Genitals Sex (3G-Sex) and the Misconception of Brain and Gender, or, Why 3G-Males and 3G-Females Have Intersex Brain and Intersex Gender’, Biology of Sex Differences, 3, p. 27.

Joel, D. [2021]: ‘Beyond the Binary: Rethinking Sex and the Brain’, Neuroscience and Biobehavioral Reviews, 122, pp. 165–75.

Richardson, S. S. [2022]: ‘Sex Contextualism’, Philosophy, Theory, and Practice in Biology, 14, available at <journals.publishing.umich.edu/ptpbio/article/id/2096/>.

Roughgarden, J. [2004]: Evolution’s Rainbow: Diversity, Gender, and Sexuality in Nature and People, Berkeley, CA: University of California Press.