“Neurosexism,” “populist science,” “neurotrash,” the problem with using terms like these to describe scientific investigations of sex differences is that their use may be interpreted as hostile. “Not fair!” claim the espousers of these terms, who argue that they only ever use such terms for pseudoscience and media distortions, not robust and replicable studies. In a recent op-ed for The Guardian, Cordelia Fine—the author who coined the term “neurosexism”—together with Rebecca Jordan-Young, argue that they have never been prima facie opposed to sex differences research. Their only concern is that of scientific rigour.
A great article about how science and feminism goes hand in handhttps://t.co/EKMoUDjhux
— AUforsker/Lea (@AUforsker) April 15, 2017
In 2005, the British philosopher Nicholas Shackel proposed the term “Motte and Bailey Doctrine” for this type of argumentative style. Taking the name of the castle fortification, the “motte” is strong and is built high on an elevated patch of land and is easy to defend. By contrast, the “bailey” is built on lower, more exposed ground, and is much more difficult to defend from attacks. Shackel used this metaphor to describe a common rhetorical trap used by postmodern academics, where a controversial proposition is put forward (a “bailey”) but is then switched for an uncontroversial one (a “motte”) when faced with criticism. In this case, the controversial position that has been proposed by authors such as Fine and Jordan-Young is that the scientific investigation of sex differences reinforce and legitimize harmful and sexist stereotypes about women. The uncontroversial proposition is that their concern is simply one of “[ensuring] the [maximum possible contribution] of neuroimaging research.”
An example of the controversial proposition could be seen in 2013, when Fine described a neuroscience study published in the Proceedings of the National Academy of Science (PNAS) as “neurosexist”. The study, which consisted of a large sample of 949 youths aged 8 to 22, found sex differences in the white matter, or connective tissue, of the human brain. Men were shown to have a greater number of white matter connections running from the front to back of the brain, while women had a greater number running between the left and right hemispheres.
The study’s authors also made speculations about what kinds of behaviours these structural differences might be related to. For example, in men, these differences translate to more efficient coordination between perception and action; in women, they facilitate better communication between analytical and spatial processing. It was this speculation which attracted the ire of Fine. It was “subtly neurosexist,” she said “[in] reinforcing and legitimating gender stereotypes in ways that are not scientifically justified.” The study was methodologically unsound (according to Fine) because it had not controlled for “gendered experiences (such as hobbies, subjects studied at school or higher education, or participation in sporting activities)”. However, gendered interests have been shown to be evident in children as young as 9 months of age which occurs before they are developmentally aware that gender exists, at around 18 months.
Despite Fine’s protestations this PNAS study is not pseudoscience. It has been cited 342 times since its publication and its results are indeed a reflection of the thousands of other research studies supporting sex differences in the brain and resulting behaviour. In 2016, researchers looked at the structural connectivity of the brain again and, using a sample of 900 participants, found what they described as a “reliable link” between motor, sensory and executive function sub-networks in males, and social motivation, attention and memory sub-networks in females. While it is perhaps too early to make causal statements, if we are to take Darwin’s theory of sexual selection seriously, the prior probability of there being no biological component to this link is extremely low.
As Fine and Jordan-Young point out in their Guardian op-ed, in late 2016, several news stories broke which reported that over the past three decades brain research had being hampered by a hostile political climate.
Sexism fears hamper brain research https://t.co/6lz3xm3kK9
— The Times IE (@thetimesIE) November 29, 2016
These news stories cited Larry Cahill, a prominent researcher who had written an editorial in the Journal of Neuroscience Research describing how the science of sex influences had progressed slowly and this had been partly due to scientists’ fear of being called “sexist”. Unfortunately for women, this slow progress has meant that women are at much higher risk of experiencing adverse reactions to drugs than men, due to the lack of research specifically teasing out sex differences. In a feature article one of us wrote for Commentary, Cordelia Fine was identified as one of the academics who had contributed to a precarious political climate. It is unfair however to single out Fine. Other academics have included Gina Rippon, who wrote an article about neurosexism for The Conversation as recently as October 2016, and Daphna Joel, who has said that research into sex differences makes her “blood boil”.
It has been pointed out that in their co-authored paper for the open-access journal Frontiers in Neuroscience, Fine, Jordan-Young and Rippon called on neuroscientists to become familiar with Intersectionality Theory, a theory which derives from Critical Race Studies in the U.S., and which has little in the way of scientific validity (“validity” is scientific jargon-speak for the ability of a construct, model or concept to make predictions about the real world). Again, advocating that scientists take remedial courses in this kind of theory could be described as a “bailey” position. It is a controversial argument, which is difficult to defend from a scientific point of view.
Unsurprisingly, Fine and Jordan-Young have responded to criticism by retreating to their “motte” argument—that they are only concerned with scientific rigour. In The Guardian they write:
[W]e, together with other colleagues, made recommendations in the peer-reviewed journal Frontiers in Human Neuroscience on best practice in sex/gender neuroscience. Some of the errors and traps we identified included human neuroimaging studies with small sample sizes, and the common “snapshot” approach, which interprets neural associations with sex as a matter of timeless and universal male and female essences, without taking seriously the fact that biological associations might as easily be the effect of social differences as the cause of them.
In the same article they proceed to state that:
Ambien is actually an example of the importance of considering variables that correlate with sex. As Lise Eliot and Sarah Richardson recently pointed out in the Journal of Neuroscience, bodyweight differences mediate most of the male/female difference in how quickly Ambien’s active ingredient, zolpidem, is cleared. “Hence, the Food and Drug Administration (FDA)’s revised drug safety advisory about zolpidem actually recommended a lower dose in both women and ‘many men’.”
However when looking at the original study which is referred to by Eliot and Richardson, it is clear that “most of the difference” is actually “half of the difference,” in a sample of 24. The authors of the original study state clearly in their abstract that body-size does not account for the gender difference in drug metabolism and that females have “greater intrinsic sensitivity” to zolpidem. Fine and Jordan-Young’s dismissal of the sex effect is simply premature. Nevertheless, they go onto argue:
What does this mean, in medical practice? It means that women and men don’t have distinctly different responses, but instead have different responses on average – mostly a result of differences in body size. So if treatment is based on sex, then smaller-than-average men, and larger-than-average women, will be given the wrong treatment. Treating by sex, instead of by body size, actually results in more people being given the wrong treatment.
Yet conflating average group differences with outliers is an obvious fallacy. In the two essays which Fine and Jordan-Young critique in their op-ed, The XX Factor and Equal ≠ Same: Sex Differences in the Human Brain, sex is not proposed as a variable that cancels out all others. It is not argued that there will be no outliers or that there will not be significant overlap in distributions, or that there will not be significant variation within the sexes. It is simply argued that sex is an understudied variable that has suffered scientific neglect for varied and complex reasons.
The Motte and Bailey Doctrine has been a successful rhetorical device for anti-sex difference academics and authors for some time now, but it is beginning to fray at the edges. The radical social constructionist viewpoint which posits that all differences between men and women have been socialized or created by culture has not led to any reliable predictions or explanations of human behaviour. It has not created a body of robust and replicable scientific work. By contrast today’s sex neuroscience is booming. In the January 2017 special edition of the Journal of Neuroscience Research, around 70 research groups from across the globe submitted papers on the state of the understanding of sex differences in the brain.
There will always be flawed studies in any field of science and there will always be media distortions of research on the topics of sex and gender. We agree that efforts to improve methodological integrity and reduce media distortions are vital and worthwhile causes. But science cannot progress if speculation is taken off the table due to political concerns. Gender equality should not (and cannot) rely on the false notion that the sexes must be the same in order to be equal, and dishonest rhetorical tactics deployed by critics only sows unnecessary confusion. Ultimately, politicizing science is a losing strategy, and given enough time, unfettered science will always win.