I Still Don’t Understand The Point Of These Arguments About Biological Sex. By Jesse Singal

jessesingal.substack.com

13 - 16 minutes

Over the weekend, a couple people posted a new paper about neuroendocrinology to Sexnet, a sex-research listserv I’m on. It was published in the journal Hormones and Behavior, and it’s called “Deconstructing sex: Strategies for undoing binary thinking in neuroendocrinology and behavior.” The authors are Megan G. Massa, Krisha Aghi, and M.J. Hill.

Massa and her colleagues describe their paper as “a call-to-arms” against what they view as an over-simplistic, rigidly binary view of sex in the field of behavioral neuroendocrinology. The abstract nicely captures what they’re going for:

    The scientific community widely recognizes that “sex” is a complex category composed of multiple physiologies. Yet in practice, basic scientific research often treats “sex” as a single, internally consistent, and often binary variable. This practice occludes important physiological factors and processes, and thus limits the scientific value of our findings. In human-oriented biomedical research, the use of simplistic (and often binary) models of sex ignores the existence of intersex, trans, non-binary, and gender non-conforming people and contributes to a medical paradigm that neglects their needs and interests. More broadly, our collective reliance on these models legitimizes a false paradigm of human biology that undergirds harmful medical practices and anti-trans political movements. Herein, we continue the conversations begun at the [Society for Behavioral Neuroendocrinology] 2022 Symposium on Hormones and Trans Health, providing guiding questions to help scientists deconstruct and rethink the use of “sex” across the stages of the scientific method. We offer these as a step toward a scientific paradigm that more accurately recognizes and represents sexed physiologies as multiple, interacting, variable, and unbounded by gendered preconceptions. We hope this paper will serve as a useful resource for scientists who seek a new paradigm for researching and understanding sexed physiologies that improves our science, widens the applicability of our findings, and deters the misuse of our research against marginalized groups.

The first sentence of the introduction explains that “Neuroendocrinologists have long known that ‘sex’ is a specious category,” and the authors then continue: “Much of our research relies on identifying mechanisms that produce differences in brain morphologies and behaviors, including how factors like hormones, chromosomes, and life experiences differences across ‘the sexes.’ This work makes evident that ‘sex’ is not a biologically coherent concept (Karkazis, 2019; Roughgarden, 2013) but is instead a constructed category reliant on several biological criteria that do not always align (Ainsworth, 2015).”

At different points in the paper, the authors employ different terms to describe the problem with using sex in the allegedly oversimplified manner they are describing: sex isn’t “internally consistent,” it isn’t “biologically coherent,” and it lacks “internal coherence,” they argue. The concept is just a mess.

In my view, this paper is the latest example of what has become a yearslong effort on the part of some left-leaning scientists, activists, and journalists: fuzzing up the concept of biological sex as much as possible, attempting to turn it into a troubled, subjective, unknowable mystery. 

This movement seeks to distract from the fact that more than 99% of people can be described as straightforwardly male or female, solely (as I see it) in the service of a single goal: making sure no one can ever claim someone who says they are a man or a woman or nonbinary isn’t really that thing. That is, if it’s not only cruel or mean but scientifically inaccurate to point to a trans woman and say “That’s really a male!” then — the thinking goes — this will advance the cause of justice.

I’ve argued in the above-linked post that this causes a lot of shoddy and confused thinking, and is unnecessary. There are all sorts of moral arguments for treating trans people how they would like to be treated that don’t rely on the belief that they are really in some biological sex, X or Y or Z. You see the same logical leaps over and over again. In this paper, for example, Massa and her colleagues argue that “the use of simplistic (and often binary) models of sex ignores the existence of intersex, trans, non-binary, and gender non-conforming people,” without explaining what this means. But it doesn’t necessarily follow. If you point out that someone is biologically male or female in a context where that matters, you’re of course not denying the possibility that they don’t identify with that sex or might seek to change it to the extent biomedical technology allows. In fact, until recently this was the Trans 101 definition of. . . well, being trans! It described the condition of having a gender identity at odds with your biological sex.

Let’s just nibble one other part off Massa and her colleagues’ paper that, I think, shows how strange and muddled and strained this line of thinking is:

    Our collective overreliance upon the simplistic heuristic of “sex” in data collection, hypothesis formation, analyses, and data interpretation leads to inaccurate and underspecified scientific knowledge. Uncritically dividing subjects into “female” and “male” categories (or other hypersimplistic models) obscures relevant physiologies and precludes the possibility of more specific (and more accurate) analyses. This obscuring effect of “sex” was elegantly demonstrated in a recent neuroendocrine study, where researchers analyzed their data by “sex” and then by estrous stage. Their analysis revealed that collapsing sex category hid the dynamic nature of ovarian hormones (Rocks et al., 2022): if the experiment had assumed the internal coherence of sex categories (as is common practice), important dynamics within these categories would have been overlooked to the detriment of our scientific understanding.

“Rocks” is a very strong last name, is my first impression. The Rocks in question is Devin Rocks, a recently matriculated (dissertated?) PhD student at Fordham University who is now doing a postdoc at Weill Cornell. He also ran Division I track there as an undergrad! (I hate successful young people.) Alongside Drs. Heining Cham and Marija Kundakovic (listed as the corresponding author), Rocks published a 2022 paper in Biology of Sex Differences (differences, hmmm. . . ), which is indeed titled “Why the estrous cycle matters for neuroscience.”

If you read it, you’ll note that the authors did indeed reanalyze some of their old mouse data. They found that, as part of their abstract put it,

    . . . accounting for the estrous cycle significantly increases the resolution of the neuroscience studies and allows for: (a) identification of masked sex differences; (b) mechanistic insight(s) into the identified sex differences, across different neurobehavioral outcomes, from behavior to molecular phenotypes. We confirm previous findings that female data from either mixed- or staged-female groups are, on average, not more variable than that of males. However, we show that female variability is not, at all, predictive of whether the estrous cycle plays an important role in regulating the outcome of interest.

This is a very odd choice of paper for Massa and her colleagues to present as evidence that sex categories aren’t “internally coherent.” (For what it’s worth, I asked a philosopher if this term has an agreed-upon meeting and he said it didn’t. I also emailed Massa to ask her what definition of the term she was relying on, and didn’t hear back. The Rocks team didn’t return an email, either.)

It’s an odd choice because Rocks and his colleagues certainly view sex as internally coherent in their paper! They don’t even define the terms in question — male and female — because of course, anyone reading the journal Biology of Sex Difference already knows what they mean — the male mice are the one who produce the small gametes (sperm), while the female mice are the ones who produce the large gametes (eggs). There’s nothing incoherent here, and no, gesturing vaguely at (say) murine disorders of sex development doesn’t really change that. Just because there is some fuzziness at the edges of these categories doesn’t mean the categories themselves aren’t exceptionally useful — vital, even.

For example, Rocks and his colleagues also include charts like this. . . 

. . . and this. . . 

. . . and it just seems pretty clear that they somehow knew exactly which the male versus female mice were. They couldn’t have really run their reanalyses otherwise.

On top of that, Rocks and his colleagues included this paragraph:

    We want to emphasize that the increased resolution that we are seeing in our studies is because the hormonal status is a sex-specific factor that is more precise than sex. Multiple researchers have indicated that sex is a complex, multi-layered variable that should be used as a proxy rather than a variable that explains sex-based variation. We were warned that male and female populations are largely overlapping in both animal and clinical research and that over-interpreting sex differences can bring us further from truth both in experimental research and in medicine, and it hurts transgender and gender-nonconforming individuals. With our population graphs, we show that incorporating the estrous cycle stage, an exemplary sex-specific variable, increases our resolution so we can separate different populations and better interpret our data. The ovarian cycle stage is also gender-independent, so our findings are applicable to all individuals who experience ovarian hormone fluctuations, across gender, including cis women, non-binary individuals, and transgender men who menstruate. [footnotes omitted]

These obviously aren’t reactionaries — they know how to speak the necessary social justice lingo to ensure their paper gets published without much protest. But they’re also pointing out that however one identifies, there’s still a biological sex lurking underneath that has explanatory power. Yes, in this case, factoring in the estrous cycle provides far more data than merely factoring in sex, but the former depends on the latter. If you didn’t have the concept of sex, and if you can’t easily and accurately identify which are the male versus female mice, you wouldn’t be able to run these analyses!

The argument of the authors of the Hormones and Behavior paper, Megan Massa and her colleagues, seems to be something like If you only examine biological sex as a binary variable, but not the other things caused by being male or female, that will obscure potentially useful findings. Would a single defender of the old-school view of sex, anywhere, deny this? In some cases, yes, just biological sex alone can reveal important differences, as in this distribution of adult heights I found on Our World In Data:

But of course in other instances, you’ll get a lot more information from studying things at the level of the consequences of biological sex rather than just the differences between males and females. Again, who would deny this? This isn’t a new observation; in fact, there’s a lot of research on the effects of ovulation on human female behavior. My sense is it’s contested and maybe some of it has even been debunked — I don’t know enough about this area — but the point is that people have known forever that the terms male and female don’t always tell you everything you need to know about behavior or whatever else. Suggesting otherwise is a straw man argument.

Let me make one last very online point: after Massa posted a link to her article to Twitter, she got a lot of nasty and snarky replies and quote-retweets (not gonna link). I understand that Twitter is a free-for-all and if you study or write about a controversial subject, you’re going to catch some flak. But I still wish this wasn’t how public intellectual life worked these days, because of course the reaction Massa and her colleagues get, which is going to include some nastiness, some “LOL LOOK AT THESE DUMB WOKE MARXIST SCHOLARS,” etc., is not going to make them, or anyone, more likely to closely examine their priors and potentially be open to criticism. It’ll just make everyone double down.

Of course, I could be too doubled-down on my own views here! Maybe there’s something important that I’m missing, or too ignorant to understand. That’s why I reached out to the authors. But I will say that I’ve noticed a pattern where when I look into these claims about the supposed fuzziness and lack of importance of the basic male/female difference, I often find the arguments 1) are straightforwardly wrong or misleading, 2) collapse into something obvious, 3) are so hard to follow, in terms of their basic logic (rather than technical language), that I don’t know what to make of them, or 4) are quite philosophically muddled, such as by confusing and freely mixing distinct terms like sex, gender, and gender identity.

I just feel like there’s not very much there here, other than an awkward attempt to hammer some basic science into a more politically appealing shape, and I wish someone could explain to me what I’m missing, if I am in fact missing something. More dialogue would help.

Questions? Comments? Proof that I am not, in fact, male? I’m at singalminded@gmail.com. Image: “Gender Icons For Male And Female On Wooden Blocks On Table” via Getty.