The word “autism” is over a century old, coined in 1910 by the Swiss psychiatrist Eugen Bleuler to describe a feature of schizophrenia. But the condition we recognize today as autism spectrum disorder is a far younger concept, one that has been repeatedly redefined, expanded, contracted, and fought over since the 1940s. The distance between Bleuler’s original term and a modern autism diagnosis is not just a matter of medical refinement; it reflects dramatic shifts in how societies have understood childhood, disability, and the boundaries of “normal” human behavior.
The Word Before the Condition
Bleuler introduced “autism” from the Greek word “autos” (self) to describe a withdrawal into inner fantasy life that he considered a hallmark of schizophrenia. For roughly 80 years, that was what the word meant in psychiatry: a symptom of psychosis, not a standalone diagnosis.1PubMed. Autism in schizophrenia and its original link to self-disorder: returning a borrowed concept This matters because when early researchers later applied the same term to children, they were borrowing a concept already loaded with associations to schizophrenia. That baggage would haunt autism research for decades, entangling childhood autism with adult psychosis in ways that slowed its recognition as a distinct condition.
Through the 1920s and 1930s, “autism” floated through the work of child psychologists and psychoanalysts like Jean Piaget and Lauretta Bender, who used it loosely to describe hallucinations and unconscious fantasy life in infants.2PubMed Central. How autism became autism: The radical transformation of a central concept of child development in Britain No one yet conceived of it as a developmental condition with its own trajectory. The children who would later be recognized as autistic were, during this period, often institutionalized under labels like “mental defective” or “childhood schizophrenia,” categories that told clinicians almost nothing useful about what was actually going on.
The Forgotten Pioneer
The standard history of autism usually begins in 1943 with Leo Kanner’s landmark paper describing eleven children with what he called “early infantile autism.” But there is a significant earlier chapter. In 1925, a Soviet child psychiatrist named Grunya Efimovna Sukhareva published a paper describing six children whose presentation, in modern terms, would meet criteria for what we now call autism spectrum disorder. She documented their social difficulties, restricted interests, and unusual communication patterns with striking accuracy, nearly two decades before Kanner or Hans Asperger published their own observations.3PubMed Central. The Work of Grunya Efimovna Sukhareva in the Field of Autism Spectrum Disorder One Hundred Years After Her Original Description
Sukhareva’s work was published in Russian (and later in German), but it never gained the traction of Kanner’s or Asperger’s English-language descriptions. Her erasure from the mainstream history of autism has attracted growing attention in recent years. It raises uncomfortable questions about whose observations get credited in medical history and how language barriers and geopolitics can bury foundational work.
Kanner and Asperger in the 1940s
In 1943, working at Johns Hopkins University, Leo Kanner described a group of children who shared a distinctive pattern: extreme aloneness from the beginning of life, an obsessive desire for sameness, and a peculiar relationship with language. He called the pattern “early infantile autism,” borrowing Bleuler’s word but giving it an entirely new meaning. The following year, in Vienna, Hans Asperger independently described a group of children he called “autistic psychopaths,” who had intense special interests, unusual social behavior, and often strong verbal and intellectual abilities despite their social difficulties.
Asperger’s patients represented what a retrospective analysis found to be a relatively high-functioning subgroup: roughly two-thirds of the cases reviewed met ICD-10 criteria for Asperger syndrome, while about a quarter fulfilled criteria for autism.4PubMed Central. A retrospective analysis of the clinical case records of ‘autistic psychopaths’ diagnosed by Hans Asperger and his team at the University Children’s Hospital, Vienna Asperger’s descriptions would not reach the English-speaking world in a major way until Lorna Wing translated and championed his work in 1981, creating a long gap during which Kanner’s narrower definition dominated.
For decades, this meant “autism” referred almost exclusively to children with severe impairments: no speech, no apparent interest in other people, rigid and repetitive behaviors. Children who were verbal, intellectually capable, and socially interested but awkward were not considered autistic. They simply fell through the diagnostic cracks.
Asperger’s Troubling Legacy
The history of autism cannot be told honestly without addressing what researchers have uncovered about Hans Asperger’s relationship with the Nazi regime. A thorough investigation of wartime records found that Asperger accommodated himself to Nazi rule and was rewarded with career opportunities. He joined several organizations affiliated with the Nazi party, publicly endorsed race hygiene policies including forced sterilizations, and on multiple occasions actively cooperated with the child “euthanasia” program, which killed disabled children deemed unworthy of life.5PubMed Central. Hans Asperger, National Socialism, and “race hygiene” in Nazi-era Vienna
This revelation, published in 2018, shook the autism community. The American Psychiatric Association had already folded “Asperger’s disorder” into the broader autism spectrum diagnosis in 2013, but for many autistic people who had identified with the Asperger label for years, the historical findings added a painful dimension to an already contentious diagnostic change. Some researchers have since argued that Sukhareva, not Asperger, deserves recognition as the true early describer of what was long called Asperger syndrome.
The “Refrigerator Mother” Years
In the 1950s and 1960s, the dominant explanation for autism was psychodynamic: the problem was assumed to lie in the child’s emotional environment, specifically with the mother. Bruno Bettelheim, a psychologist at the University of Chicago, popularized the theory that emotionally cold or rejecting mothers caused their children to withdraw into autism. The term “refrigerator mother” became widely used, and parents, especially mothers, bore the blame for their children’s condition.6PubMed. From ‘Refrigerator Mothers’ to Empowered Advocates: The Evolution of the Autism Parent
This era caused enormous harm. Families were told their parenting had damaged their children. Mothers underwent psychoanalysis. Children were sometimes removed from their homes. The refrigerator mother theory had no empirical basis, but it aligned with the psychoanalytic framework that dominated mid-century psychiatry, and it persisted for roughly two decades before biological evidence began to dismantle it.
The Biological Turn
The first major blow to the psychodynamic model came from genetics. In 1977, a twin study examined 21 pairs of same-sex twins where at least one had autism, as diagnosed by the criteria of Kanner and Rutter. The results demonstrated the importance of hereditary influences in the cause of autism, showing that identical twins were far more likely to share the condition than fraternal twins.7PubMed Central. Genetic influences and infantile autism If autism were caused by bad parenting, twin type would not matter, since both twins share the same parents. The finding pointed clearly toward biology.
This study was a turning point. Through the late 1970s and 1980s, the closure of institutions for people labeled as “mentally defective” and the growth of speech therapy services also encouraged new frameworks for understanding autism that focused on cognition and language rather than unconscious conflict.2PubMed Central. How autism became autism: The radical transformation of a central concept of child development in Britain The shift was not just intellectual; it was practical. Once researchers stopped looking for what the mother did wrong and started looking at how the autistic brain develops differently, useful interventions became possible.
How the Official Diagnosis Evolved
Autism did not appear as a standalone diagnosis in any edition of the Diagnostic and Statistical Manual (DSM), psychiatry’s official classification system, until 1980. Before that, children who would now be diagnosed as autistic were typically classified under childhood schizophrenia. DSM-III marked the first time autism was recognized as a distinct condition, separated from psychotic disorders.8PubMed Central. The Diagnosis of Autism: From Kanner to DSM-III to DSM-5 and Beyond
Each subsequent revision broadened the boundaries. DSM-III-R in 1987 loosened the criteria. DSM-IV in 1994 introduced several related diagnoses, including Asperger’s disorder, pervasive developmental disorder not otherwise specified (PDD-NOS), and childhood disintegrative disorder, creating a cluster of conditions under a shared umbrella. Then DSM-5 in 2013 collapsed all of those subcategories into a single diagnosis: autism spectrum disorder, with severity levels replacing the old subtypes.
That consolidation remains contested. Since its first formal description in 1943, repeated attempts to set clear diagnostic boundaries have struggled with the sheer heterogeneity of autism. Excessive overlap between subcategories, frequent use of “not otherwise specified” labels, and diagnostic uncertainty in borderline cases all pointed to the limitations of trying to carve autism into neat categories. The spectrum model was, in part, an acknowledgment that the categorical approach was failing.
Why Autism Numbers Keep Climbing
Autism prevalence estimates have risen dramatically over the past few decades, from roughly 4 in 10,000 children in the 1960s to about 1 in 36 in recent U.S. estimates. This increase has fueled widespread anxiety about whether something in the modern environment is causing more autism. The evidence, however, points primarily to changes in how we count.
Research tracking educational classifications found that higher autism prevalence was significantly associated with corresponding declines in the prevalence of intellectual disability and learning disability diagnoses, suggesting that children who would previously have been classified differently were now being identified as autistic.9Pediatrics. The Contribution of Diagnostic Substitution to the Growing Administrative Prevalence of Autism in US Special Education In other words, much of the rise reflects diagnostic substitution: the same children receiving a different label, not new children developing a new condition.
A broader review of the evidence reached a similar conclusion: the majority, if not all, of the reported rise in incidence and prevalence is due to changes in diagnostic criteria, increasing awareness among parents and professionals, the development of specialist services, and the expansion of the autism concept from Kanner’s narrow definition to the wide spectrum used today.10PubMed. The epidemiology of autistic spectrum disorders: is the prevalence rising? None of this rules out some true increase, but the evidence makes it clear that the diagnostic and social changes are doing most of the heavy lifting.
Who Gets Missed
The history of autism diagnosis is also a history of who was left out. Because early descriptions were based almost entirely on boys, the diagnostic criteria that developed reflected male presentations of autism. Women and girls remain significantly underdiagnosed, in part because of the male-oriented structure of current criteria, higher rates of co-occurring psychiatric conditions that can mask autistic traits, and the tendency of girls and women to engage in social camouflaging, learning to mimic neurotypical social behavior to fit in.11PubMed Central. Underdiagnosed and Misunderstood: Clinical Challenges and Educational Needs of Healthcare Professionals in Identifying Autism Spectrum Disorder in Women A woman who has spent her life carefully studying and imitating social cues may not trigger the recognition patterns that clinicians were trained to look for.
Age is another major factor. Without a clear definition of autism’s behavioral features existing in the past, individuals born before roughly 1980 often went entirely undiagnosed or were misdiagnosed with other conditions.12PubMed Central. Living with autism without knowing: receiving a diagnosis in later life A growing number of adults are now receiving autism diagnoses in midlife or later, sometimes in their 50s, 60s, or beyond. Research on adults diagnosed after age 35, all born before 1975, found that many had experienced significant negative experiences including trauma throughout their lives. For most, receiving an autism diagnosis had a profoundly positive impact on their sense of self, allowing them to understand their past and feel good about their identity for the first time.13PubMed. ‘A way to be me’: Autobiographical reflections of autistic adults diagnosed in mid-to-late adulthood
These late diagnoses are not niche cases. They represent an entire generation (or several) of people for whom the diagnostic infrastructure simply did not exist. The experience of growing up autistic without a framework to understand it, of being told you are anxious, depressed, difficult, or just odd, is a shared story among late-diagnosed adults that highlights how profoundly diagnostic history shapes individual lives.
Modern Diagnostic Tools and Their Limits
Today, the gold standard for autism assessment typically involves structured instruments like the Autism Diagnostic Observation Schedule (ADOS) and the Autism Diagnostic Interview-Revised (ADI-R), often used in combination. Research has examined whether both tools are truly necessary: one study found that a streamlined model using just 11 features, seven from the ADOS and four from the ADI-R, achieved 85% accuracy with 93% sensitivity and 78% specificity, performing comparably to using the full set of 65 features.14PubMed Central. Is the Combination of ADOS and ADI-R Necessary to Classify ASD? Rethinking the “Gold Standard” in Diagnosing ASD
These tools are a dramatic improvement over the informal clinical impressions that governed diagnosis for most of autism’s history, but they are not without problems. They were developed and validated largely on white male children, which helps explain why women, people of color, and adults are more likely to be missed or misidentified. Access is another barrier: a full diagnostic assessment is expensive, requires trained specialists, and may involve months-long waiting lists. In many parts of the world, the tools are simply unavailable.
The Neurodiversity Reframing
Perhaps the most significant recent development in autism’s history is not medical at all. Beginning in the late 1990s, autistic self-advocates began organizing around the concept of neurodiversity, which frames autism not as a disease to be cured but as a natural variation in human cognition. This represents a fundamental challenge to the traditional psychiatric perspective, which treats autism as a disorder requiring medical intervention. The neurodiversity movement instead emphasizes societal adaptation and inclusion.15PubMed. History of the Autism Diagnosis- How the Perspectives Have Changed
Autistic culture, as researchers have described it, can reframe personal difficulties as a shared political struggle, defying the stereotype of autism as a condition defined by social withdrawal.16Human Development. Come as You Are: Examining Autistic Identity Development and the Neurodiversity Movement through an Intersectional Lens This movement has real consequences for diagnostic practice. Identity-first language (“autistic person” rather than “person with autism”) has become preferred by many in the community. The expansion of the autism diagnosis itself, some argue, supports the neurodiversity paradigm by recognizing a wider range of human cognitive styles as autistic, which in turn fosters greater inclusion.15PubMed. History of the Autism Diagnosis- How the Perspectives Have Changed
Applied Behavior Analysis and Ongoing Controversy
No account of autism’s diagnostic history is complete without the treatments that followed from it, and none has generated more debate than Applied Behavior Analysis (ABA). ABA emerged in the mid-twentieth century and became the dominant intervention for autistic children in English-speaking countries, largely through the influence of psychologist Ole Ivar Lovaas. His early work in the 1960s used intensive behavioral techniques, including physical aversives like slapping and electric shocks, to modify autistic children’s behavior. Reception was initially positive despite concerns about efficacy and the use of punishment. His 1987 study, which reported dramatic improvements, was widely treated as validation of the approach, although both his methods and his understanding of autism had changed considerably since the 1960s.17History of the Human Sciences. ‘The line between intervention and abuse’ – autism and applied behaviour analysis
Since the early 1990s, ABA has faced growing criticism, especially from autistic adults who experienced it as children. The neurodiversity movement in particular has challenged the premise that autistic behaviors need to be extinguished rather than accommodated. Modern ABA practitioners generally disavow the use of aversives, and the field has evolved considerably. But the fundamental tension remains: is the goal of therapy to help an autistic person navigate the world on their own terms, or to make them appear less autistic? How that question gets answered depends heavily on who is doing the answering, and the history of autism shows that the people being diagnosed have rarely been the ones making that call.