It’s hard to think of an area of medicine more controversial today than the treatment of gender dysphoria in youth. Proponents and opponents of the new “affirming” paradigm of treatment routinely accuse each other of politicizing medicine, promoting dangerous ideologies, and abusing vulnerable children.
In the United States, civil rights discourse has come to overlay—some would say distort—these debates. In 2016, Attorney General Loretta Lynch likened North Carolina’s “bathroom bill” to the “dark days” of Jim Crow, when states “had signs above restrooms, water fountains and on public accommodations keeping people out based upon a distinction without a difference.”
A handful of federal circuit and district courts have ruled that schools must defer to the gender identity of their students or risk being found guilty of sex discrimination. Most of these cases have dealt with access to restrooms, and courts have consistently held that schools may not require transgender boys—most plaintiffs have been natal females who identify as male—to use the girls’ restrooms or some alternative unisex facilities. Additional lawsuits have raised questions about who may participate in girls’ sports, whether schools may facilitate gender transitions for students without notice or consent from their parents, and clashes between religious liberty and transgender “inclusive” policies. If the litigation campaign over restrooms is any indication, we should expect most or all of these cases to be resolved in favor of transgender students.
The Supreme Court's 2020 decision Bostock v. Clayton County held that Title VII of the Civil Rights Act of 1964 protects gay and transgender people from employment discrimination under the “ordinary public meaning” of “sex.” Justice Neil Gorsuch, who wrote the opinion, went out of his way to explain that the Court’s decision didn’t turn on whether “sex” means reproductive traits or gender identity. Thus, he emphasized, the ruling did not reach controversies over access to sex-specific accommodations. Yet lower courts have since interpreted Bostock to mean exactly that, ruling that schools may not rely on the conventional definition of “sex.”
At the heart of these legal controversies is the question of what makes people male or female. Without exception, courts in the education lawsuits have based their answer on what they have learned from the medical experts who testify or file amicus briefs on behalf of transgender students. Yet the arguments that these experts present to federal judges are highly partisan interpretations of an already limited, and often methodologically flawed, body of research. Their interpretations seem geared to producing a desired legal outcome rather than faithfully reporting on an ongoing medical debate.
More broadly, transgender student lawsuits illustrate the deeper difficulties involved in using law to settle scientific controversies and, at the same time, expanding science beyond its sphere of competence to settle, through legal means, moral and philosophical questions. Civil rights discourse relies on abstract analogies and moral absolutism. It avoids talk of ambiguity and trade-offs on principle, treating opposition to desired legal outcomes as rooted in ignorance and bigotry. Scientific inquiry, by contrast, calls for self-doubt, skepticism, and genuine openness to being proved wrong. In theory, these are two wholly different enterprises guided by different values and incentives.
Two arguments underwrite judicial rulings on transgender students. According to one, gender identity is rooted in the brain, fixed by age three, and immutable. When that identity conflicts with the sex that one was “assigned at birth,” the only medically appropriate treatment is to alter the body to align it with the mind. According to the experts who testify in court, the transition process includes living full-time according to one’s felt gender, using puberty blockers to prevent the onset of undesirable physical traits, and injecting hormones to spur the onset of desirable ones. Courts have declared this treatment protocol, known in professional circles as the “affirming” or Dutch model (due to its country of origin), as having the unanimous endorsement of “all major medical organizations.” To them, it is “settled science.”
But even if the gender feelings that plaintiffs have are sincere and persistent, why should they trump reproductive traits and capacities in determining maleness or femaleness? To answer this question, plaintiffs and judges have needed a second argument, one meant to call into question the belief that humans are, by nature, divided into two sexual categories according to reproductive capacity. This second argument, which need not have anything to do with gender identity, asserts that sex in humans is not binary but actually ranges on a “spectrum,” with up to 1.7 percent of the population falling between the male and female poles. Because many students are “intersex,” goes this argument, schools could not consistently enforce their definition of “sex,” even if legally allowed to do so.
Together, these two arguments make up a philosophical anthropology—a theory about the fundamental basis of human nature and sexuality. According to this anthropology, “biological sex” is a social construct, while gender identity is real, universal, and scientifically demonstrable. Both arguments suffer from empirical difficulties. More important, the premises that make the second argument plausible make the first impossible.
To understand the empirical problems, consider the evidence that the transgender experts themselves have presented to the courts. In the Florida case Adams v. School Board of St. John’s County, in which a transgender boy sued his school for being denied access to the boys’ restrooms, Diane Ehrensaft, a clinical psychologist and one of the leading proponents of “affirmative” care, told a Florida district court that gender identity is “primarily dictated by messages from our brain.” Thus, she argued, the only medically appropriate treatment for gender dysphoria is gender transition. The plaintiff’s use of the boys’ restrooms was crucial in this regard, since being recognized by others as a boy was essential to his medical treatment for gender dysphoria. The court agreed that “neurological sex and related gender identity are the most important and determinative factors” in who counts as male or female and ruled that the school district had relied on unlawful “stereotypes.” Yet in her own peer-reviewed research, Ehrensaft notes a “lack of evidence” for the biological explanation of gender identity. When it comes to why some children identify as another gender, “the question of nature versus nurture,” she writes, is “yet to be settled.”
An amicus brief submitted by reputable “medical organizations” in the same lawsuit claimed that, though researchers were “not certain” what causes cross-gender identification, “[s]ome research suggests . . . biological influences.” Amici provided two citations. One is an article whose authors recommend “assigning” a female gender identity to females born with congenital adrenal hyperplasia—a condition in which overexposure to masculinizing hormones makes the body appear more masculine—even though 5 percent of females with this condition experience “serious gender identity problems.” The other citation states unequivocally that “the (patho-) biological basis of [gender dysphoria] is still poorly understood” and that diagnosis of gender dysphoria “relies totally on psychological methods.” This is important, the authors emphasize, considering that “80–95 percent” of the prepubertal children diagnosed with gender dysphoria will desist from it later on and feel comfortable in their own bodies. In simple terms, socialization seems to play a major, if not fully understood, role in gender-identity development.
In the Virginia case G. G. v. Gloucester County School Board, the first federal lawsuit over restrooms in high schools, the American Civil Liberties Union also asserted that gender identity is rooted in the brain. The ACLU cited a law review article that it claimed “summarizes the research” on human gender development. The article’s author is M. Dru Levasseur, a lawyer who, at the time of writing, was director of the Transgender Rights Project at LAMBDA Legal. Its revealing title is “Gender Identity Defines Sex: Updating the Law to Reflect Modern Medical Science Is Key to Transgender Rights.”
Levasseur bases his claim that “all [experts] agree” that gender identity is “hardwired into the brain” on two sources. One, an article by another legal scholar, merely suggests that neurological explanations for gender identity might be true on a “balance of probabilities,” while emphasizing that “there will be no conclusive ‘scientific proof’ of the causation of transsexualism until medical science can identify and ratify the sexual differentiation of the human brain and/or genetic identifiers for transsexualism in living human beings.”
The other source is a single sentence—“the organ that appears to be critical to psychosexual development and adaptation is not the external genitalia, but the brain.” This sentence comes from an article published 20 years earlier by William Reiner, a physician with expertise in child psychiatry and urology. Levasseur leads the reader to believe that this quotation summarizes Reiner’s position, but Reiner’s argument is only that reproductive (genital) anatomy alone cannot account for gender-identity development in humans. In the same article, moreover, Reiner stresses that “gender identity, like sexual orientation, is a complex process in only the initial infancy of scientific understanding.” In a later article addressing the misuse of his earlier work by activists, Reiner clarified his position: “Trans-friendly, and, in general, patient-friendly, evidence for [the neurological explanation] is generally lacking.” In fact, he says, that explanation may be “too patient friendly” if it reinforces clinicians’ “faith in following the child’s lead” (i.e., self-diagnosis). Reiner published these comments a few years before Levasseur’s article, yet Levasseur doesn’t mention them. Still, the Fourth Circuit agreed with the ACLU that the meaning of sex “in 2016” is psychological rather than reproductive.
To date, researchers have not been able to trace cross-gender identification exclusively, or even primarily, to biological causes. Postmortem studies on the brains of adults have shown that transgender women (natal males) may have brain structure more typical of natal females; but, given a phenomenon known as neuroplasticity, these studies cannot refute the possibility that the resemblance is a result, rather than a cause, of lifelong cross-gender identification and therapy. Studies of identical twins suggest a possible role for genetic factors in gender-identity development but are otherwise inconclusive. In sum, very little is known about why some children are gender-dysphoric. The consensus among researchers seems to be that cross-gender identification is neither entirely a result of biological factors nor entirely a result of social conditioning but is instead some combination of the two. Nevertheless, these inquiries can tell us only about the causes of sincere and persistent cross-gender identification. They cannot tell us how much weight society should give to sincerity—or, to use the preferred nomenclature, “authenticity”—in determining what male and female mean.
Why, then, do medical professionals who participate in transgender lawsuits assure judges that the neurological explanation is, if not quite “settled science,” then close to it? One reason is that this belief carries important therapeutic implications. If gender identity is innate and unchangeable, then the only way to resolve gender dysphoria is to bring the body into alignment with the patient’s felt gender. Note that this is a purely technological argument. We must defer to self-identification because changing the body is easier, technologically speaking, than changing the brain. If a technology existed that allowed us to change the brain, the experts would have no principled reason to object to it—and perhaps strong reasons to support it. But in any case, it is easier to change the body before the patient has gone through full puberty.
Another reason for the medical professionals’ insistence is that “brain sex” resonates with a legal culture shaped by the civil rights movement. The Supreme Court has long recognized that a trait’s immutability is relevant to its eligibility for constitutional protection. In the final stages of the Gloucester litigation, the Fourth Circuit based its equal protection analysis on the claim that gender identity is, like race, an “immutable characteristic.”
This brings us to the second argument shaping federal regulation of schools—namely, that intersex conditions prove that sex in humans is not binary but ranges on a spectrum between male and female. Intersex refers, broadly, to a range of conditions in which both male and female physical attributes of sex (gonads, external genitalia, secondary sex characteristics such as breasts) appear in a single individual. In the past, these conditions were referred to as hermaphroditic or as disorders of sex development. In recent years, these terms have fallen out of fashion in favor of “intersex.” The argument is that there are no disorders but only differences of sex development, which are just as natural and normal as “typical”—that is, statistically common—male and female phenotypical alignments.
Gender-rights advocates insist that intersex and transgender are distinct phenomena. In practice, however, the experts who participate in transgender litigation argue that “neurological sex” is one of several biological components that make up an individual’s sex. Intersex is defined as misalignment of biological traits; so by this logic, students with gender dysphoria are intersex.
By far the leading scientific authority on intersexuality is Anne Fausto-Sterling, a professor of evolutionary biology and gender studies at Brown University. Having originally claimed that up to 4 percent of the population is intersex, Fausto-Sterling downgraded her assessment to 1.7 percent in 2000, after conducting independent research with colleagues. Academic publications, advocacy literature, and corporate diversity training seminars widely cite this figure today.
Several scholars have taken Fausto-Sterling to task for sloppy empirical work and politically motivated definitions. On the first front, University of Toronto political science professor Carrie Hull accepts, for the sake of argument, Fausto-Sterling’s definition of intersex as “any deviation,” however slight, from the “Platonic ideal” of male and female but finds serious errors of interpretation and computation. Using data from the studies on which Fausto-Sterling relies but with “correct math,” Hull finds that a more accurate estimate of intersex incidence is 0.373 percent. Even this, Hull emphasizes, is likely a “dramatic overstatement.”
The primary driver of Fausto-Sterling’s exaggerated assessment was her inclusion of a condition known as late-onset congenital adrenal hyperplasia (CAH). The classic version of this condition occurs when hormonal production structures are defective in utero, causing the overproduction of male sex hormones, which impairs anatomical development in females. Practically speaking, this leads to a female infant being born with genitals that appear more masculine than feminine. In the condition’s late-onset version, however, hormonal malfunction occurs much later in life, and because it is often asymptomatic, those who have it tend to discover that fact only incidentally, in the context of treatment for infertility.
The study on which Fausto-Sterling relies for her assessment of late-onset CAH reports an incidence rate of 1.5 percent. Hence, this single condition represents 88 percent of all intersex conditions in her 1.7 percent figure. Yet the original study (1985) did not say that late-onset CAH occurs at 1.5 percent across the general population. Rather, it sampled four high-risk demographic groups: Ashkenazi Jews (3.7 percent), Hispanics (1.9 percent), Yugoslavs (1.6 percent), and Italians (0.3 percent). The 1.5 percent figure seems to be an average of these four groups, adjusting for their respective portion in the general population. Among Caucasians, the frequency is about 0.1 percent; among blacks and Native Americans, the figure is close to zero, according to the authors of the original study. Fausto-Sterling has conceded that this study is nonrepresentative, but her 2000 article leads the reader to believe that 1.5 percent is universal. That has certainly been the way lawyers and judges have used her work.
Leonard Sax, a physician and psychologist with extensive clinical experience in treating children with intersex conditions, believes that behind Fausto-Sterling’s dubious empirical claims lurks a politicized definition of intersex. That definition is clinically useless, he argues, because it makes no distinction between symptomatic and asymptomatic conditions, or between types and severities of symptoms, and it fails to recognize that the vast majority of intersex people have variations of sex development so subtle as to be imperceptible to the untrained eye—or even to the person with the condition. “A definition of intersex which encompasses individuals who are phenotypically indistinguishable from normal,” Sax cautions, “is likely to confuse both clinicians and patients.”
A good example is Klinefelter syndrome, which accounts for roughly 5 percent of intersex conditions in Fausto-Sterling’s final estimate. Individuals with Klinefelter are essentially males born with an additional X chromosome (X-X-Y). The features that distinguish them from other males become detectable after puberty and include smaller-than-average testes, larger-than-average breast tissue, and lower-than-average bone density and muscle mass. Genitals are normal for such males and capable of both erection and ejaculation. The majority of Klinefelter males are infertile. Affected individuals are usually so indistinguishable from other males that their condition may go undiagnosed until fertility problems arise; those who are fertile may go their entire lives without knowing that they have it. Yet the Fourth Circuit in Gloucester suggested that the existence of students with Klinefelter makes it impossible for schools consistently to enforce a policy that adheres to the conventional meaning of male and female. The Obama administration’s definition of “sex” as a matter of gender identity, the court explained, would at least “resolve ambiguity” by applying a fail-proof criterion (the assumption being that a student’s asserted gender identity is accurate simply by virtue of being sincere).
True intersex conditions, according to Sax, are extremely rare, representing fewer than two out of every 10,000 births, or 0.018 percent of the population. Put into perspective, this means that a medium-size high school is likely to have one intersex student every 15 years or so, making utterly implausible the claim by federal judges that intersex conditions create an administrative headache for schools that choose to enforce the conventional definition of “sex.”
It is essential to distinguish between the intersex-rights movement and the use of intersex conditions by feminists to advance the idea that sex—not gender, but sex—is socially constructed. Fausto-Sterling explicitly cites the latter as the goal of her scholarship. A self-described lesbian and feminist, she regards intersex phenomena as having “profound” implications for women, gays, and lesbians. “If nature really offers us more than two sexes, then it follows that our current notions of masculinity and femininity are cultural conceits. Reconceptualizing the category of ‘sex’ challenges cherished aspects of European and American social organization.” To achieve this objective, one must understand intersex conditions as mere differences, rather than disorders, of sex development. Fausto-Sterling wants us to think of the typical as merely “statistically common” rather than “natural” (in the sense of “normal”) and, for that matter, of “perfectly dimorphic” males and females as less frequent than is commonly supposed.
But if the statistically common is no guide to the natural/normal, why should it matter whether intersex conditions occur in 1.7 percent or 0.018 percent of the population? Either way, there would be a naturalistic fallacy, or inferring an ought from an is. Is it possible that Fausto-Sterling’s insistence on higher-than- expected numbers reflects deeper insecurities over whether what is common in nature does, in fact, give us clues as to the normal? Moreover, does it matter that about seven in ten people with intersex conditions regard themselves as males or females with “disorders of sex development” and do not find this terminology offensive?
In response to Hull’s criticism of her numbers, Fausto-Sterling herself has conceded possible “mistakes in interpretation and imperfect judgment” but has allowed her erroneous findings to stand uncorrected. “I am not invested in a particular final estimate,” she has written, but “only that there BE an estimate. . . . Beyond getting the numbers right . . . our article suggests a different approach to conceptualizing sexual difference, one that allows for human sexual variation to be considered socially normal, albeit infrequent.”
Hull, a self-described feminist, is disheartened to find “many feminists and activists” who “unquestioningly” repeat Fausto-Sterling’s inflated statistic. Feminist-oriented law reviews are perhaps the worst offenders in this regard and give us a good example of what philosophy professor Peter Boghossian has called “idea laundering” (scholars citing dubious research and then one another, thus creating the impression of a well-established empirical tradition). In one recent example, a legal scholar noted that statistical assessments of intersex range from 0.018 percent to 4 percent (a figure that Fausto-Sterling has long disavowed). After calling 4 percent “extreme,” the author described 1.7 percent as a “conservative” estimate.
This is not mere academic nitpicking. In a lawsuit filed by non-transgender students against their school after it voluntarily adopted a gender-identity policy, the plaintiffs argued that separating restrooms by male and female gender identity would be no less problematic considering that, if gender identity has nothing to do with reproductive traits, there would be no reason to suppose that only two gender identities exist—thus calling into question the school’s decision to maintain only male and female restrooms on the same “stereotyping” grounds cited by many transgender plaintiffs. Judge Edward G. Smith of Pennsylvania’s Eastern District Court agreed that the school’s policy was “unworkable” for this reason. However, since “1 to 4 percent” of students are intersex, the judge explained, the conventional definition of “sex” is equally unworkable, and it is not for the court to tell the school which of two unworkable solutions it should enforce.
Judge Smith took the “1 to 4” statistic from the first sentence of a law review article by Jennifer Relis, who took it from the first sentence of a law review article by Kate Haas, who took it from Anne Fausto-Sterling. On appeal, the Third Circuit commended Judge Smith for his “exceedingly thorough, thoughtful, and well-reasoned opinion.”
Thus far, we have discussed the distortions within the arguments about gender identity and intersex. It’s no less important that the premises that make these arguments possible are mutually exclusive. Fausto-Sterling credits Michel Foucault’s notion of “bio-power” as inspiration for her thinking on human nature (or on the lack thereof). Like Foucault, she regards objectivity as a charade and calls science “politics by other means.” In an interview with the New York Times a few years before the transgender issue hit the federal courts, she emphasized that choosing which attributes define a person’s sex is “a social decision given that a statistical or scientific way of deciding does not exist.” Intersex conditions are theoretically significant in this light because they prove that sex is really gender all the way down.
Fausto-Sterling thus agrees with Sax that her definition of intersex and, by extension, of sex is political rather than scientific; but this, to her mind, is a problem only for those who maintain a naïve faith in scientific objectivity. As for her views on gender identity: in her most important book, she contends—in an understatement suggesting a desire to avoid confrontation with transgender activists—that the “born that way” narrative of gender identity “does not necessarily describe the process by which the person develop[s] . . . a particular identity.” In effect, then, she agrees with critics of Diane Ehrensaft (on the left and the right) who argue that gender identity, perhaps especially when sincere, may be the result of internalized social conventions or “stereotypes.”
In short, the philosophical anthropology behind federal court rulings is a combination of two mutually exclusive philosophical positions. Perhaps the strongest evidence for this confusion is the criticism that court rulings have received from transgender advocates in the academy. Much to the chagrin of “critical theorists,” judges have, at the urging of plaintiffs and their experts, held that gender identity can only be male or female. (To suppose otherwise would be to concede that the brain is naturally and by default androgynous and becomes gendered through socialization, which would call into question the need for “affirming” medical intervention.) They have further insisted, to the alarm of Foucauldians, that being transgender is synonymous with having diagnosed gender dysphoria. The experts have dismissed identities like “nonbinary,” “queer,” and “genderfluid” as lacking any clinical basis and thus as not being “identities” in the relevant sense of the word.
Court rulings on transgender students are a fascinating example of how ideas travel from fringe academic theory into law and policy. The point is that they do not travel well. Yet to blame judges for these distortions is somewhat unfair. After all, judges are experts in law and, in their attempt to understand novel issues, must rely on what they learn from experts in a highly formalistic process ill-equipped to evaluate nonlegal knowledge. On the other hand, given judges’ busy schedules, lack of specialization, and reliance on self-interested plaintiffs and their partisan witnesses, one could expect them to refrain from declaring a novel and (as yet) inadequately tested treatment protocol “settled science.”
A highly complex, still-evolving field of medical research on a badly misunderstood phenomenon (gender dysphoria) now commands the moral urgency of a national civil rights cause. Court rulings on transgender students provide a powerful tool for activists to demonize critics, deter skepticism, and discourage careful thinking. For children and adolescents whose innocent confusions, insecurities, or playfulness about gender are medicalized and treated as gender dysphoria, “affirming” care is likely to exacerbate their suffering. It helps no one (save perhaps a tiny cohort of activists) that this issue has become as politicized as it has.
Top Photo: At the heart of the legal controversy surrounding transgender identity is the question of what makes people male or female. (ERIC AUDRAS/PHOTOALTO/ALAMY STOCK PHOTO)