When I look back at home videos of my daughter Isabel, I see the signs of autism clearly. But at the time, in 1992, I couldn’t. Autism was still considered rare. In one video, when Isabel was 15 months old, she sits quietly, putting coins in a piggy bank. She doesn’t respond to her name or look at us. My wife and I marvel at her focus and precision and predict she will be a scientist.
In a widely anticipated news conference on Monday, President Trump declared that there was “nothing more important” in his presidency than reducing the prevalence of autism. He claimed that his administration would virtually eliminate the condition, which he called a “horrible crisis” and which a top federal health official suggested might be “entirely preventable.”
The administration’s project is built on the premise that an autism diagnosis is a terrible tragedy and that scientists and doctors have failed to prevent what Health Secretary Robert F. Kennedy Jr. has called an “epidemic.”
But science has not failed. One reason we have so many questions about autism today is that we’ve learned so much about it and how to address it. Thousands of highly trained researchers and clinicians have generated an extraordinary amount of information about autism’s genetics and neurobiology, developed reliable early detection methods, expanded special education and improved behavioral and medical therapies. To think otherwise reveals a deep and willful ignorance of the history of autism and its present-day complexity.
Isabel was 2½ when she was diagnosed with the unwieldy and now obsolete “Pervasive Developmental Disorder, Not Otherwise Specified,” or PDD-NOS, an old term for someone with autism who had relatively low support needs, or who did not meet every criterion in the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R).
Many years later, when I asked our doctor if he had considered diagnosing Isabel with autism, he confessed that he was scared the word would devastate us. Autism was a frightening diagnosis in the ’90s, suggesting one’s child had a bleak future. He also understood how parents can blame themselves for their child’s problems. My wife and I searched our memories for what we might have done wrong — a chemical we used while renovating our new house, even that roller coaster ride before we knew she was pregnant.
It’s easy to forget how recently autism became a common diagnosis and a household word. It’s easy to forget how little researchers and clinicians knew about developmental disorders, how we would marginalize people with the symptoms we now call autism, stigmatize them and their families and deny them an education. Without knowledge of that past, we risk making policy decisions that will stall scientific and social progress.
What we have learned from dozens of rigorous studies is that, in contrast to claims Mr. Kennedy has made, vaccines do not cause autism. We know that more than 100 genes are associated with autism, which interact not only with one another but also with individuals’ environment and experience throughout their lifetime. The neurogeneticist Stanley Nelson once said, “If you had 100 kids with autism, you could have 100 different genetic causes.”
Researchers increasingly believe the search for a cause of a singular “autism” is misguided. There are autisms, rather than just autism. The diversity in symptoms, the severity of symptoms and the range of biological contributors are simply too large to be reduced to one uniform disorder. The neurobiology is too complex for easy explanations.
The more we learn, the more we realize that scientific answers to the question “What causes autism?” or even “What is autism?” will take a long time to unravel, no matter how quickly a government official wants them.
These days, when students in my classes on the anthropology of mental health ask me about the history of autism, they often put it this way: “When did it become a thing?” I know they are asking me when autism became widely recognized, but I like to interpret the question as, “How did we start believing that autism is something real?”
Doctors developed the concept of autism to make sense of a pattern of traits that frequently co-occur but were not previously grouped together under a single term. Autism is a collection of signs and symptoms that has no single identifiable pathological process. Doctors did not adopt the term because they found something new, like a bacterium or a virus, that we can test for in a laboratory.
My students are often surprised to learn just how unstable a concept autism is. Derived from the Greek word autos, meaning “self,” the term autistic was often used in the early 20th century to describe adults with schizophrenia who seemed absorbed by their own internal worlds and private thoughts. By the 1940s, clinicians began to think that when a child had the symptoms of autism, it was merely evidence of childhood-onset schizophrenia, a condition we now know to be extremely rare.
It wasn’t until the 1960s and early 1970s, when child psychiatry was just beginning to become a bona fide specialization, that clinicians started to more regularly differentiate autism from schizophrenia. Doctors disagreed about whether a diagnosis of autism should be reserved for children with average or above average intelligence, or if it should also encompass those with an intellectual disability, as well as additional medical issues such as seizure disorders.
The families of people with autism also suffered from a particular stigma around this time. The influential writer Bruno Bettelheim convinced many that autism was a devastating sickness caused by a single environmental factor: highly educated, cold and distant “refrigerator mothers,” who he believed must have neglected their children.
Without legitimate evidence, Mr. Bettelheim blamed a generation of loving mothers and urged families to send their autistic children to grim residential institutions to remove them from their mothers’ “black milk.” Although the authority of psychoanalysis declined during the 1970s, Mr. Bettelheim’s influence still surfaced. In 1994, a leading autism expert in the United States told my wife that if we wanted our daughter to improve, she should quit her job and become a stay-at-home mom. He didn’t look at me, and never suggested I should quit my own job.
The search for easy explanations also persisted. In 1998, a British doctor, Andrew Wakefield, published an article in The Lancet describing 12 children who appeared to have normal development but quickly lost skills, including language skills, following administration of the measles, mumps and rubella vaccine. It was later revealed that Mr. Wakefield had misrepresented the findings: Many of the children’s symptoms predated the vaccinations, and others did not have autism at all.
The article was retracted and Mr. Wakefield lost his medical license, but the damage was done. The myth that vaccines cause autism has persisted, despite studies covering millions of children showing that there is no link. And now the president of the United States is echoing Mr. Wakefield’s discredited claims about children “lost” to autism because of vaccines.
In the last three decades, I’ve seen countless other attempts to discover specific causes of autism — including scientifically plausible topics, like de novo mutations in the sperm of older fathers or high levels of prenatal testosterone, and the absurd, such as television and pasteurized milk.
President Trump claimed on Monday that “since 2000, autism rates have surged by much more than 400 percent.” But the definition of autism has changed so much that comparing diagnostic rates from different eras is misleading.
When “infantile autism” was first introduced into the American Psychiatric Association’s diagnostic manual in 1980, it was defined so narrowly that few children qualified. But by the time Isabel was diagnosed, the concept had been broadened to “autistic disorder,” with much greater flexibility in symptoms and age of onset. Core traits like “pervasive lack of responsiveness to other people” were broadened to include looser, more subjective descriptions, such as preferring solitary play.
The manual’s next version, the 1994 DSM-IV, continued that expansion, adding new subtypes, including the short-lived Asperger’s disorder, which described autistic children without significant language delays and was often seen as less stigmatizing. It was a diagnosis clinicians often made for highly verbal children, sometimes to assure parents that their children’s challenges were not profound, much as Isabel’s first doctor chose the diagnosis of PDD-NOS in lieu of autism.
But even as researchers added subtypes, many worried that they could not reliably distinguish between different types of autism. By the time the fifth and current edition of the diagnostic manual was published in 2013, Asperger’s had been removed and autism was reconceptualized as a continuous spectrum. The decision was fiercely contested: Some people with Asperger’s feared a loss of access to services or even to the identity and community that had been built around the label.
Isabel now falls under the label autism spectrum disorder, a diagnosis so broad that it includes people who are nonspeaking and require intensive support, as well as people who work at the highest levels of academia, government and technology whose mild autistic features help them succeed professionally. Isabel certainly doesn’t care about diagnostic manuals, which say little about what shapes the experiences and identities of people with autism. To her, autism is what gives her perfect pitch, her ability to solve complex jigsaw puzzles even when they are picture-side down, and her quirky, sometimes belly-laugh-inducing sense of humor.
Some autistic self-advocates argue that for many people, autism should not be considered a disorder or a disability at all, but rather a different way of being human. Still other advocates draw attention to the fact that autism is often accompanied by severe challenges like seizures, intellectual disability and self-injurious behaviors. Concerned that such serious issues are getting lost in the public discourse that is celebrating the abilities of autistic people with low support needs, they are encouraging clinicians to name a new subtype profound autism.
The classification of illnesses is often based less on biological facts than scientific and social consensus at particular historical moments. It’s a dynamic process of lumping and splitting, of changing definitions and measures — whether one is describing autism, hypertension, Type 2 diabetes, obesity, or other conditions whose cutoff points keep shifting. It’s even conceivable that, years from now, scientists will eliminate the word autism altogether and replace it with language that describes behaviors, or biological or genetic pathways, shared by many different conditions, rather than a distinct syndrome.
Diagnostic categories rarely reflect clear boundaries between health and illness. No one knows for certain where on a continuum the adaptive anxiety that compels humans to look both ways before crossing the street becomes an anxiety disorder, where sadness becomes clinical depression, or social awkwardness becomes autism. Indeed, family and twin studies show that autism symptoms can be found throughout the general population, and especially among relatives of people with autism. Identifying the point at which something becomes a disorder is a clinical judgment about whether people’s symptoms impair their daily functioning and warrant treatment and services.
Sociocultural factors also play an important role in how we define and treat the cluster of symptoms we refer to as autism. Some Navajo Indians have viewed it as a state of becoming; in some areas of sub-Saharan Africa it is linked to spirit possession; some ultra-Orthodox Jews in Israel view autism positively as an ability to connect with God; until the turn of this century, the French psychiatric establishment labeled autism a psychotic disorder.
In contrast to Mr. Trump’s comment Monday that “there are certain groups of people that don’t take vaccines and don’t take any pills that have no autism,” people have had the symptoms long before the first vaccine was invented in the late 1700s, and certainly long before the word was coined in the 20th century. In past centuries, many people with crude labels like “insane” or “feebleminded” were confined to asylums because it was considered culturally appropriate to do so.
The president also falsely suggested that autism was virtually absent among the Amish and in Cuba. The Amish do have autism, though they may be diagnosed less often because they can be reluctant to seek specialized medical care, and there are multiple schools in Cuba specifically for children with autism.
In addition, in my own fieldwork in remote regions such as the Ituri Forest in the Democratic Republic of Congo and the Kalahari Desert in Namibia — where some communities have little or no access to vaccines or modern medicine — I found many children and adults with autism and intellectual disabilities. In the Kalahari, one little boy who would almost certainly be diagnosed with autism in an American doctor’s office, is highly valued for his abilities. Despite being nonspeaking, and despite recently losing his hearing as a result of measles, the boy, his father told me, “is great herding goats. He always knows where they are in the day or night.”
In the United States, both race and income have long shaped who is recognized as autistic. Black children have historically been much more likely to receive a diagnosis of “conduct disorder” or “emotional disturbance” rather than autism — possibly because of a cultural and clinical bias that wrongly associated minority children with disruptive behavior.
This has often led to treatments and school placements different from those given to their white autistic counterparts. In the documentary “Refrigerator Mothers,” a mother named Dorothy Groomer recalled doctors at the University of Illinois telling her decades ago that her son could not have autism. “We did not fit their mold,” she says. “We did not fit the classic mold for autism. Which is white, upper middle class and very, very bright.”
Recent prevalence estimates for the Centers for Disease Control and Prevention suggest that racial gaps in autism rates have narrowed, a good sign that children once underserved are being identified and getting the services denied to them in the past.
Had Isabel been born in the 1960s, she might have received any number of diagnoses, because there was no standardized classification system for disabilities in American schools. A child with the symptoms of autism could be labeled a “slow learner,” “educable mentally retarded,” “emotionally blocked” or any other term a school system wanted to use.
It wasn’t until the passage of the 1975 Education for All Handicapped Children Act that the federal government guaranteed that all children with disabilities could go to school. Shortly after that, a uniform set of disability categories was established, but the list did not include autism until the early 1990s. It took many years before autism programs were established in schools nationwide.
The proportion of children receiving special education services in the United States has had only a small increase over the past couple of decades, but the labels have changed. Many children who once might have been classified as having an intellectual disability, emotional disturbance or specific learning disability are now classified as having autism.
Indeed, one study of special education enrollment data on 6.2 million schoolchildren between 2000 and 2010 found that the reduction of intellectual disability classifications alone accounted for 64 percent of the rise in autism classifications during that decade. Such reclassification helps explain how what may seem like an epidemic may actually be the result of greater awareness and changing definitions.
When Mr. Kennedy and Mr. Trump seek to simplify autism to fit their own cause-and-effect narrative, they risk muting the voices of the vast majority of autistic self-advocates, scientists and service providers who do not find the Trump administration’s project credible or even feasible. Mr. Kennedy and Mr. Trump also threaten to disrupt the trajectory of the research that has made the science of autism so vibrant and productive. What’s more, when they talk about people with autism as casualties of a medicine or a vaccine, they risk reinforcing the stereotype that people with autism are damaged, and that this damage could have been prevented.
Remember Mr. Bettelheim? He demonstrates how just one person can derail scientific progress. He had virtually no scientific training, yet he convinced many that poor mothering was the single environmental condition that could explain the causes of autism. Charismatic and persuasive, he named himself an authority on autism and hijacked clinical research for years. We lost time that could have been devoted to other areas of study and wasted precious resources.
Instead of pursuing an ill-conceived search to link vaccines and autism, we need wide-ranging research on how the genes associated with autism work, how autism intersects with co-occurring conditions like A.D.H.D., depression and anxiety, and how sex and gender play roles in both the onset of symptoms and a person’s life experience.
There is an especially pressing need to study how people with autism change over their life span, and what treatments and services help them thrive. Autism research has long focused on children, but people’s ability to change doesn’t end when they turn 21. Isabel is today one of the happiest and busiest people I know, but many of the strides she made to get there happened as an adult.
The history of autism also suggests that we need to have realistic expectations about the pace of ethical science. Research in genetics and neuroscience has produced a staggering number of discoveries, but most have not yet been translated into novel ways to treat neurological and mental disorders, even disorders that are already well understood genetically (such as Huntington’s disease). Good science takes years, not months.
What matters most to Isabel isn’t finding someone to blame, since she likes herself the way she is. What matters to her is continuing to build a meaningful life with the social supports and opportunities that were unavailable to people with autism in the past.
We’ve come too far to go back to a time when autism was a stigmatized condition defined solely in terms of deficits; when mothers were made to feel guilty; when the pressures for conformity so outweighed the desire for diversity that people with autism had few chances to get an education or a job; when one person with power could dominate the discourse on autism; and when the people whose well-being was at stake were excluded from shaping the narratives and policies that can alter their lives and happiness.
Roy Richard Grinker is a cultural anthropologist who teaches about mental health at George Washington University. His books include “Nobody’s Normal: How Culture Created the Stigma of Mental Illness.”
The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: letters@nytimes.com.
Follow the New York Times Opinion section on Facebook, Instagram, TikTok, Bluesky, WhatsApp and Threads.