Online, someone is wondering if Will Smith is on the spectrum. Jennifer Lawrence? Low-key autistic. So is Matthew McConaughey, though he may be a savant. Thomas Pynchon seems to know a lot about town planning: distinctly spectrum-y. Anna Wintour’s limited diet, love of indoor sunglasses, and exacting standards have Asperger’s vibes. Other things that have been declared autistic-adjacent on Twitter: having an Aquarius moon, dudes in high school who shout rap lyrics, Tulsi Gabbard supporters, the sudden urge to climb a tree, the Build-A-Bear Workshop website.

What, exactly, are all these people talking about? You get the sense that we’ve strayed quite far from the official diagnosis of a neurodevelopmental disability that’s linked to difficulties with social interaction and communication and restricted or repetitive behavior. The colloquial use of terms like “autistic” and “on the spectrum” conjures up a set of related images of the autistic person: socially awkward, or maybe obsessive about a niche topic or area of expertise, or failing to recognize social cues or make eye contact, or affectless, or inconsiderate towards others and even self-absorbed. This stereotype is omnipresent today, but has been on the rise since the early 2000s, when the media first stoked widespread panic about a supposed childhood autism epidemic. 

Sometimes, what gets called autism is indistinguishable from ordinary rudeness. (One telling YouTube video from last September explains “How to Tell if Someone’s Autistic or a Jerk (or both?).”) Calling a terrible boyfriend or a coworker with bad anger management skills “spectrum-y” simultaneously puts a name to and partially excuses inconsiderateness, for how can one be taken to task for being autistic? Autism may be a throwaway insult or justification, but its ubiquitous everyday usage also reflects something more substantial about what’s churning through our collective mind. What it means to call someone autistic in common speech has changed, and not only because the science has changed. The more you study its evolving public perception, the more you realize that definitions of, and explanations for, autism are in an important sense a mirror for the non-autistic, neurotypical world. “Autism” is — and has been — as much a reflection of our collective anxieties as it is of an individual disability.

Even in clinical settings, the term “autism” and its implications have never been straightforward. Defining the condition remains difficult 80 years after the diagnosis was first used. It still lacks a singular presentation, a clear set of causes, stable diagnostic criteria, and universally agreed-upon treatments. Autism looks different in different people, and its effects vary over the course of their lives. As a saying in the field goes: if you know one autistic person, you know one autistic person. 

According to the most recent Diagnostic and Statistical Manual of Mental Disorders (DSM-5), published by the American Psychiatric Association in 2013, in order to be diagnosed with autism spectrum disorder, an individual must display “persistent deficits” in three categories of social communication and interaction — “social-emotional reciprocity,” “nonverbal communicative behaviors used for social interaction,” and “developing, maintaining, and understanding relationships” — along with at least two forms of “restricted, repetitive patterns of behavior, interests, or activities.” Critics of this “deficit model” argue that it pathologizes difference and is limiting to scientific and philosophical understandings of autism, as well as harmful to autistic people. The Autism Self Advocacy Network offers a different definition that stresses value-neutral difference, rather than deficit. “We socialize differently. Some of us might not understand or follow social rules that non-autistic people made up,” the group’s website explains. “We might have a hard time controlling our body language or facial expressions, which can confuse non-autistic people or make it hard to socialize.”

Of course, autism is also famously understood as a spectrum, a concept introduced in 1979 by the English child psychiatrist Lorna Wing, whose own daughter was autistic. After her epidemiological research indicated that autism was more widespread than previously believed, Wing rejected rigid diagnostic categories in favor of a mutable association of abilities and needs. Today, you can take online tests to measure how many autistic traits you have or what your “Autism Spectrum Quotient” is. For autistic people, this idea is frustrating. Having some autistic traits is not the same as being autistic. Everyone isn’t a little bit autistic; rather, some are autistic, and others are not. The spectrum, as it is now generally understood by the scientific and autistic communities, is a way of capturing autism’s range and complexity within the subset of people who have it, not the span of all human experiences. 

Part of the confusion around what autism is may stem from the varying ways that it has historically been diagnosed, and particularly the historical overlap between so-called “high-functioning” autism and Asperger’s syndrome, which for a long time had indistinguishable criteria. Asperger’s — which shows up as intense interests in specific subjects, often related to science and math, and social awkwardness — is named after the Austrian psychiatrist Hans Asperger. In the 1930s and early 1940s, Asperger and the child psychiatrist Leo Kanner were working simultaneously, the former in Vienna and the latter in Baltimore. Both researchers independently borrowed the term “autism” from the Swiss psychiatrist Eugen Bleuler, who had used it to refer to “a definite withdrawal from the external world” in his schizophrenic patients.

While both Kanner and Asperger characterized their autistic patients as socially isolated, there were marked differences in the case studies in their initial papers, published in 1943 and 1944, respectively. Kanner described eleven children with “an inability to relate themselves in the ordinary way to people and situations,” emphasizing their “basic desire for aloneness and sameness.” He saw his patients as intelligent but experiencing unusual difficulties, including with language: he writes, for example, of a child who would recite poetry but would not readily ask or answer questions. Some of Asperger’s patients, by contrast, seemed comfortable speaking at length on their subjects of special interest — he called them “little professors.” While Kanner believed autism to be a rare and narrowly defined condition, Asperger thought that there were many ways autism could present, and that some autistic adults could achieve high levels of professional success and social integration. But few people knew about Asperger’s more expansive understanding of autism: his contributions were primarily limited to German-language scientific literature until the 1970s, while Kanner’s work was distributed internationally. (For a long time, Asperger was held up as a kind of hero, a forerunner of the movement to recognize and accommodate neurodiversity. Then, in 2018, it came to light that he had in fact believed his less high-functioning patients were “ineducable” and collaborated with the Nazis to have them transferred to the killing center at Spiegelgrund.) 

It was Lorna Wing who popularized Asperger’s work in the English-speaking world in a 1981 journal article, using his findings to bolster her new theory of an autism “continuum,” or spectrum. The spectrumification of autism, and the introduction of Asperger’s syndrome (as Wing called it), made diagnoses increasingly common. Until the 1980s, autism was thought to occur in two to four children per 10,000; in 1985, that number jumped to one in 2,500, per the Centers for Disease Control and Prevention. 

Over the decades, the rate of autism diagnoses continued to climb. By 1995, the CDC’s estimate was one in 500; in 2000, one in 150; and as of 2021, one in 44. Meanwhile, the changing official relationship between autism and Asperger’s reflected ongoing confusion about the nature of both conditions. “Autism,” once categorized under childhood schizophrenia, first appeared as its own diagnosis in the 1980 DSM-III, and the 1987 edition broadened the criteria for diagnosis. Asperger’s syndrome was added to the 1994 DSM-IV alongside autism, but it was difficult for diagnosticians to differentiate clearly between the two: after not even two decades on the books, Asperger’s was removed from the 2013 DSM-5. (The diagnosis remained in the International Classification of Diseases until this year.) 

In the 1950s and 1960s, the dominant explanation for autism was the “refrigerator mother” theory, which held that emotionally neglectful caregivers caused their children to develop autism. The term came from Kanner, who in a 1948 Time magazine article used it to describe parents who keep their children “neatly in a refrigerator which didn’t defrost.” But it was popularized by the public intellectual Bruno Bettelheim, who talked it up everywhere from The Dick Cavett Show to The New York Times. Bettelheim had no formal psychiatric or medical training, but he was from Vienna and willing to omit enough details (and falsify enough credentials) to fit the image of a Freudian expert. A Holocaust survivor who had launched his career with an article on the psychology of Nazi concentration camps, Bettelheim compared autistic children to his fellow prisoners in Dachau and Buchenwald in his 1967 bestseller The Empty Fortress. 

By the 1980s, autistic people were no longer seen as the results of bad parenting or genocidal dehumanization. Instead, they were viewed as inherently inhuman. Psychologists had described autistic children as “unable to achieve empathy” as early as 1962, but it was the English psychologist Simon Baron-Cohen — a cousin of Sacha’s — who popularized the understanding of autism as an “empathy deficit.” In a 1985 study, Baron-Cohen, along with Alan Leslie and Uta Frith, argued that to be autistic is to lack a theory of mind, or the recognition that other people have their own internal worlds. A theory of mind is what allows us to reflect on others’ mental states, and therefore to empathize. It is, according to Baron-Cohen, “one of the quintessential abilities that makes us human.” 

Though Baron-Cohen has since specified that autistic people only lack cognitive empathy (recognizing what others are thinking or feeling) but display affective empathy (responding emotionally to what others are thinking or feeling), the idea that autistics lack all empathy persisted decades later. “It’s as if they do not understand or are missing a core aspect of what it is to be human; to be and do like others and absorb their values,” psychologist Bryna Siegel, then director of the Autism Clinic, University of California, San Francisco, told USA Today in 2002. In his 2002 book, The Blank Slate: The Modern Denial of Human Nature, Harvard psychologist Steven Pinker compared autistic people to “robots and chimpanzees,” categorizing all three as groups that lack a theory of mind and are therefore incapable of cultivating culture, a fundamental human ability.

In the early 2000s, Baron-Cohen decided that autism was actually the manifestation of the “extreme male brain.” While male brains, Baron-Cohen argued, are adept at systematizing, female brains are oriented towards empathetic thought. Autism, then, can be seen as an overdevelopment of the former at the expense of the latter. This was not an entirely new idea; something like it was first proposed by Asperger himself, who wrote in 1944 that “the autistic personality is an extreme variant of male intelligence.” But at the time, gender essentialism was particularly in vogue. Men Are from Mars, Women Are from Venus, from 1992, was a smash hit self-help guide. As rhetoric scholar Jordynn Jack argues, Baron-Cohen likely borrowed its structure, tone, and style for his 2003 book, The Essential Difference: Male and Female Brains and the Truth about Autism. 

From the image of an autistic person as a young man incapable of empathy, it was not a far leap, when mass shootings entered the culture, to what a 2015 New York Times op-ed denounced as the “myth of the autistic shooter.” Ever since Columbine, mass shootings have inspired speculation about whether the antisocial young perpetrator was autistic. After a mass shooting at Umpqua Community College in 2015, it came to light that the killer’s mother had alleged in an old Yahoo post that he had Asperger’s, and a Facebook page called “Families Against Autistic Shooters” materialized. “What do all shooters over the last few years have in common?” its creator sermonized. “A lack of empathy and compassion due to autism!”

There is no scientific evidence that autism causes a predisposition to extreme violence — in fact, autistic people are more likely to be the victims, rather than the perpetrators, of violence. But that has not stopped autism from being cited in an increasing number of high-profile criminal defenses. For instance, following his arrest for participating in the January 6 Capitol riot, the lawyer for Jacob Chansley, the “QAnon shaman,” suggested to the media that Chansley’s participation in the riot was caused by his Asperger’s syndrome. (The lawyer also described many of the Capitol rioters as “on the goddamn spectrum.”) Pedophile financier Jeffrey Epstein likewise suggested to a prison psychologist that he was possibly autistic in order to get better treatment while he awaited trial on federal sex trafficking charges. He was disturbed (due to neurological differences from the other prisoners, he claimed) by the noise of the running toilet in his cell. 

As autistic teens grew scarier in the public imagination, parents became increasingly afraid of having autistic children. Some began to warn that autism was the result of an adverse reaction to early childhood vaccines — a conspiracy theory that betrays the assumption that autism is a fate worse than potentially deadly infectious diseases. In the early twenty-first century, autism was cemented in the public imagination as the paradigmatic childhood developmental disability — overtaking, for instance, Down syndrome — thanks to a variety of factors including Dustin Hoffman’s portrayal of Tom Cruise’s autistic brother in Rain Man (the highest-grossing film of 1988), a rash of memoirs by autistic people and their parents, Jenny McCarthy’s 2007 assertion on Oprah that a vaccine had caused her son’s autism, and, notably, a massive public awareness campaign by a generic-seeming nonprofit called Autism Speaks. 

Autism Speaks was founded by NBC Universal’s then-CEO Bob Wright and his wife, Suzanne, in 2005, a year after their grandson was diagnosed with autism. An initial $25 million donation by Home Depot cofounder Bernie Marcus quickly established the charity as the most prominent autism advocacy organization in the U.S. A now-infamous 2009 ad titled “I Am Autism” depicted the condition as an invisible monster lurking within a family, with a voiceover from the perspective of autism itself. “I work faster than pediatric AIDS, cancer, and diabetes combined,” the voice threatened. “You have no cure for me. Your scientists don’t have the resources, and I relish their desperation. Your neighbors are happier to pretend that I don’t exist — of course, until it’s their child.” In 2013, Suzanne Wright published a post on the organization’s blog comparing autistic children to missing and gravely ill children, describing their families as living “moment-to-moment,” “in despair,” and “in fear of the future.”

But the associations with autism that have come to the fore in recent years haven’t all been negative. A third autism archetype has emerged alongside the school shooter and the severely impaired child: that of the socially awkward tech savant. In his 2001 article “The Geek Syndrome,” journalist Steve Silberman theorized that the concentration of math and science types in Silicon Valley was producing a baby boom of autistic kids, and over the following decades geekiness (genetic or otherwise) went from stigmatizing to aspirational. Elon Musk recently revealed himself to have been diagnosed with Asperger’s in a Saturday Night Live monologue. “To anyone I’ve offended,” he joked, “I just want to say, I reinvented electric cars and I am sending people to Mars in a rocket ship. Did you think I was also going to be a chill normal dude?” Musk has been called an “Aspie supremacist.” He isn’t unaware of empathy or social cues — he’s just too busy making money and flying to space to bother learning them.

Fears that your children won’t turn out well, that the young loner you know might be dangerous, that there’s something off in relationships between men and women, that technology is changing the world too quickly —  these are not unique to the discourse around autism. Stereotypes about autism correlate entirely to major social phenomena of the past twenty years: helicopter parenting, mass shootings, a crisis of masculinity, the rise of Silicon Valley. Look back further, and you see the same thing: in the mid-century, Americans were afraid of working mothers and the breakdown of the nuclear family, and haunted by the recent history of the Holocaust; both anxieties were reflected in the prevailing understandings of autism at the time. A few decades later, concern about broad social cohesion — an extension of our ability to empathize with, and relate to, one another — was on the rise. When you look at it closely, the timeline of changing autism diagnoses is also a timeline of the anxieties haunting the popular imagination. And if autism diagnoses are some kind of prism through which to see society more clearly, autistic people are its unfortunate objects — refracted and distorted but never truly seen.

Ever since Kanner framed his autistic patients’ parents as cold, highly intelligent, and upper-middle class, working class and non-white children have been under-diagnosed with autism, in part because the diagnostic criteria were not written with them in mind. The same goes for autistic girls, women, and non-binary people, who do exist but, due to the association between autism and maleness, are routinely underserved by the medical community and denied necessary services.

Baron-Cohen, who was knighted in 2021 for his autism research, is widely disparaged within the autism community, and many of his ideas are rejected as offensive: in a 2011 “Reply to Simon Baron-Cohen,” the autistic writer Rachel Cohen-Rottenberg observed, “Autistic people face lives of substandard care, isolation, and abuse because we are considered to have been born without a core component of humanity.” Some efforts to advocate for families with autistic children have been equally harmful. “I have tried to help Autism Speaks staffers understand how destructive its messages have been to the psyches of autistic people,” wrote John Elder Robison, the sole autistic person on the board of Autism Speaks, upon resigning. “We are not problems for our parents or society, or genes to be eliminated.” And rather than promoting a greater degree of acceptance for all autistic people, tech culture has produced the kind of Asperger’s supremacy exemplified by Musk. Autistic writer and advocate Lydia X.Z. Brown has criticized the “shiny Aspie,” or someone who is vocally proud of having Asperger’s syndrome but works at distancing themselves from people with more noticeable forms of autism and “appearing as non-disabled as possible as much as possible.” In response to the colloquial use of the language of autism, the autistic community has tried to correct widespread assumptions using the label #actuallyautistic on TikTok and Twitter, but these efforts seem to have had little effect on the casual autism speculation that proliferates online, even as flippant remarks about other conditions are now viewed as off-limits. 

“If I know that I do not understand people and I devote all this energy and effort to figuring them out,” asked autistic activist Jim Sinclair in 1989, “do I have more or less empathy than people who not only do not understand me, but who do not even notice that they do not understand me?” The autistic scholar Damian Milton has called this the “double empathy problem,” which occurs when an autistic person and a non-autistic person cannot recognize each other’s thoughts and feelings — in other words, autistic people are not the only ones who have trouble bridging the divide. But even as autistic people have worked to articulate their own ways of being and help us better understand one another, neurotypical scientists, writers, and activists have spent the past half-century projecting their fears onto autistic people. If the history of ideas about autism tells us anything, maybe it’s that neurotypicals are the ones who have trouble with empathy.