Why We’re Turning Psychiatric Labels Into Identities

To name something—to separate it from the rest of existence and bestow a label on it—is a foundational act. It is the beginning of understanding and control. In Genesis, the first thing God did after splitting light from darkness was to call the light “day” and the darkness “night.” After Adam was created and let loose in the Garden of Eden, his original job was human label-maker. God brought him creatures “to see what he would name them; and whatever the man called each living creature, that was its name.”

If Adam was like most people, he probably set about attaching names to “natural kinds”—groupings seemingly dictated by inherent features of the natural world. Referring to a group of animals as “pigs,” he would have assumed that the critters so designated all shared properties that differentiated them from every other non-pig animal. Psychologists say that we intuitively treat categorical distinctions—whether among fruits, emotions, or ethnic groups—as if (in Plato’s famous metaphor) they carved nature at its joints.

No sector of human activity is as serious about naming, or as intent on respecting natural kinds, as science. Across centuries of debate and revision, fields such as physics, chemistry, and biology have refined nomenclatures to better align with the natural order. Psychiatry, at first, looks like another success story. Years of research and clinical observation have yielded catalogues of presumed mental dysfunction, culminating in the Diagnostic and Statistical Manual of Mental Disorders, or DSM. First produced by the American Psychiatric Association seven decades ago, and currently in its fifth edition, the DSM organizes conditions into families such as “anxiety disorders,” “sexual dysfunctions,” and “personality disorders.” Each diagnosis is described by clear criteria and accompanied by a menu of information, including prevalence, risk factors, and comorbidities. Although clinicians and researchers have understood the DSM to be a work in progress, many had faith that the manual’s categories would come to approximate natural kinds, exhibiting, as the Columbia psychiatry professor Jerrold Maxmen put it in 1985, “specific genetic patterns, characteristic responses to drugs, and similar biological features.”

More than any other document, the DSM guides how Americans, and, to a lesser extent, people worldwide, understand and deal with mental illness. It decrees psychiatric vocabulary, having codified terms like “attention-deficit/hyperactivity disorder” and “post-traumatic stress disorder.” It determines which conditions are taught in medical schools, which can be treated by F.D.A.-approved drugs, and which allow people to collect disability benefits and insurance reimbursements. Through its classification of mental illnesses, it establishes their prevalence in the population and indicates which ones public policy should target.

The DSM as we know it appeared in 1980, with the publication of the DSM-III. Whereas the first two editions featured broad classifications and a psychoanalytic perspective, the DSM-III favored more precise diagnostic criteria and a more scientific approach. Proponents hoped that research in genetics and neuroscience would corroborate the DSM’s groupings. Almost half a century later, however, the emerging picture is of overlapping conditions, of categories that blur rather than stand apart. No disorder has been tied to a specific gene or set of genes. Nearly all genetic vulnerabilities implicated in mental illness have been associated with many conditions. A review of more than five hundred fMRI studies of people engaged in specific tasks found that, although brain imaging can detect indicators of mental illness, it fails to distinguish between schizophrenia, bipolar disorder, major depression, and other conditions. The DSM’s approach to categorization increasingly looks arbitrary and anachronistic.

Steven Hyman, who directed the National Institute of Mental Health from 1996 to 2001, told the Times that he considered the manual an “absolute scientific nightmare.” In 2009, four leaders of the DSM-5 revision wrote about their hopes to “update our classification to recognize the most prominent syndromes that are actually present in nature.” The outcome didn’t live up to those aspirations. In April, 2013, weeks before the DSM-5’s slated release, Thomas Insel, then the director of the N.I.M.H., remarked, “The final product involves mostly modest alterations of the previous edition.” As a result, he announced, the institute “will be re-orienting its research away from DSM categories.”

In “DSM: A History of Psychiatry’s Bible” (2021), the medical sociologist Allan V. Horwitz presents reasons for the DSM-5’s botched revolution, including infighting among members of the working groups and the sidelining of clinicians during the revision process. But there’s a larger difficulty: revamping the DSM requires destroying kinds of people. As the philosopher Ian Hacking observed, labelling people is very different from labelling quarks or microbes. Quarks and microbes are indifferent to their labels; by contrast, human classifications change how “individuals experience themselves—and may even lead people to evolve their feelings and behavior in part because they are so classified.” Hacking’s best-known example is multiple personality disorder. Between 1972 and 1986, the number of cases of patients with multiple personalities exploded from the double digits to an estimated six thousand. Whatever one’s thoughts about the reality of M.P.D., he observed, everyone could agree that, in 1955, “this was not a way to be a person.” No such diagnosis existed. By 1986, though, multiple personality disorder was not only a recognized psychiatric label; it was also sanctioned by academics, popular books, talk shows, and, most important, the experiences of people with multiple personalities. Hacking referred to this process, in which naming creates the thing named—and in which the meaning of names can be affected, in turn, by the name bearers—as “dynamic nominalism.”

Three new books—Paige Layle’s “But Everyone Feels This Way: How an Autism Diagnosis Saved My Life,” Patric Gagne’s “Sociopath: A Memoir,” and Alexander Kriss’s “Borderline: The Biography of a Personality Disorder”—illustrate how psychiatric classification shapes the people it describes. It models social identities. It offers scripts for how to behave and explanations for one’s interior life. By promising to tell people who they really are, diagnosis produces personal stakes in the diagnostic system, fortifying it against upheaval.

Just as personality tests (see, I’m an introvert!), astrological signs (I’m a Libra!), and generational monikers (I’m Gen Z!) are used to aid self-understanding, so are psychiatric diagnoses. When Paige Layle was fifteen, a psychiatrist told her that she had autism spectrum disorder. She describes the rush of clarity she experienced when hearing the DSM-5 criteria: “I’m not crazy. I’m not making it up. I’m not manipulative or trying to fake anything. . . . There’s a reason why I’m the way that I am.”

Layle’s diagnosis—and her discovery that “so many of my questions have one clear answer”—comes halfway through her book. The chapters leading up to it recount the frustration and confusion of being inexplicably different. She lists her “traits” early on: picking at her skin, pulling out her hair, cutting tags from her clothes, dissociating at the sound of fireworks. By the first grade, she says, she could read almost as well as a sixth grader, and in later years, she consistently earned high marks, yet she failed to decipher implicit meanings of texts in English class. She also struggled with anxiety; she was, in her words, “the weird kid who cried all the time and was so stressed that she wanted to die every single day.”

In 2020, she posted a video on TikTok slamming a trend that used “autistic” to mean “dumb.” It went viral. With a bluntness easily mistaken for vanity, she told the camera, “Hi, there, I’m autistic. I’m also the smartest person I’ve ever met, O.K.?” In the years since, she has become one of the most popular autism-awareness activists on the platform, attracting more than two and a half million followers. Bubbly and socially perceptive, with expertly applied makeup and a way of looking into the camera that’s both intimate and intense, she defies common expectations of what an autistic person looks like—while also delivering a message about the diversity of autism’s expressions, and especially its different presentations in men and women.

Some organizations and style guides now urge what is called person-first language (“person with autism”) over diagnosis-first language (“autistic person”). These recommendations are “straight-up, basic ableism,” Layle writes in her book. “I am autistic. I am always autistic, in every situation, all of the time, 24-7. . . . I would be a completely different person if I wasn’t autistic, and the idea of who that person would be is something I could never figure out, because autism is such an integral part of my identity. To take it away would be to take all of me away, and I find it ableist to want to try.”