A hundred years ago, autism had no name of its own. Children who would today be diagnosed with autism spectrum disorder were typically labeled “feebleminded,” “idiotic,” “imbecile,” or, if they came to the attention of a psychiatrist, schizophrenic. The word “autism” did exist by the 1920s, but it meant something entirely different: a symptom of adult schizophrenia, not a separate condition.
How the Word “Autism” Started
Swiss psychiatrist Eugen Bleuler coined the term “autism” in 1911, the same year he introduced the word “schizophrenia.” In his text Dementia Praecox or the Group of Schizophrenias, Bleuler used autism to describe a specific feature of schizophrenia: a withdrawal from reality so complete that it excluded other people. He defined it as “detachment from reality, together with the relative and absolute predominance of the inner life.” This was an adult psychiatric concept. It had nothing to do with children, developmental differences, or the condition we recognize today.
For roughly 80 years after Bleuler introduced it, autism remained formally tied to schizophrenia in psychiatric thinking. A child in the 1920s who didn’t speak, avoided eye contact, and lined up objects obsessively would not have been called autistic. That child would have received one of the era’s blunt, catch-all labels.
The Labels Applied to Autistic Children
In the early 20th century, children with significant developmental differences were sorted into a handful of categories that made little distinction between intellectual disability, mental illness, and what we now understand as autism. The most common terms were “feebleminded,” “idiot” (for those considered most severely affected), and “imbecile” (a middle category). These were not slurs at the time but formal clinical classifications, though they carried enormous stigma.
Some children whose behavior looked more unusual than delayed were diagnosed with childhood schizophrenia or classified under “dementia praecox,” the older term for schizophrenia that Bleuler had worked to replace. A child who rocked repetitively, resisted any change in routine, or seemed completely indifferent to the people around them could easily receive any of these labels. There was no framework for recognizing that these behaviors might represent a distinct neurological pattern rather than intellectual deficiency or psychosis.
What Happened to These Children
The consequences of these labels were severe. Families who had a child labeled feebleminded faced intense social pressure. The message from medical authorities was clear: institutionalize the child or keep them hidden. Families were stigmatized as “morally bad or genetically flawed.”
Institutionalization was booming. By 1905, public institutions averaged over 500 residents each, and by 1923, roughly 80 private institutions had sprung up under names like schools, farms, and hospitals. Overcrowding worsened steadily. People could spend entire days in a single room and often slept on the floor. The fences around these facilities, originally framed as protecting residents, now served to isolate them from society. Institution superintendents promoted the idea that people with disabilities were a “moral menace” capable of ruining the human species. Higher-functioning residents were sometimes paroled, but only after forced sterilization, including castration, vasectomy, or tubal ligation.
A child born in 1924 with traits we would now call autism faced this world. Whether they were nonverbal or highly verbal but socially unusual, they had no diagnosis that fit and no support designed for their actual needs.
Leo Kanner Separated Autism From Schizophrenia
The turning point came in 1943, when child psychiatrist Leo Kanner published a paper describing 11 children who shared a striking set of characteristics. He called the condition “early infantile autism,” borrowing Bleuler’s word but redefining it entirely. Kanner described these children as having an inability to relate to other people, extreme aloneness, preoccupation with objects, insistence on sameness in their environment, and unusual language patterns. Many of his 11 cases had previously been classified as feebleminded or schizophrenic.
Kanner drew careful distinctions from schizophrenia. The children he described had behaved this way from birth, not through a gradual deterioration. Their desire to maintain sameness and aloneness was a core feature, not a withdrawal from a reality they had previously engaged with. His paper gave clinicians, for the first time, a way to identify these children as something other than intellectually disabled or psychotic.
One year later, in 1944, Austrian pediatrician Hans Asperger independently described a related condition he called “autistic psychopathy.” His patients showed impaired social interaction, unusual language use, resistance to change, and intense focus on specific interests, but were generally higher functioning and less withdrawn than Kanner’s cases. Asperger’s work, published in German during wartime, remained largely unknown in the English-speaking world for decades.
The Slow Path to a Modern Diagnosis
Even after Kanner’s landmark paper, autism spent decades tangled up with schizophrenia in official diagnostic systems. It was not until 1980, when the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III) was published, that “infantile autism” appeared as its own diagnosis, formally separated from childhood schizophrenia. The DSM-III criteria explicitly stated that a child could not be diagnosed with infantile autism if they showed hallucinations, delusions, or the disordered thinking characteristic of schizophrenia. This was a direct line drawn between two conditions that had been conflated for most of the 20th century.
The diagnostic picture continued to shift. Asperger’s disorder was added to the DSM-IV in 1994. Then in 2013, the DSM-5 folded all previous subtypes into a single umbrella: autism spectrum disorder. Today’s criteria focus on two core areas. The first is persistent difficulty with social communication and interaction, covering everything from trouble with back-and-forth conversation to challenges reading body language and maintaining relationships. The second is restricted, repetitive patterns of behavior, such as repetitive movements, rigid routines, intense fixations, or unusual responses to sensory input like sounds, textures, or light. Clinicians rate severity across three levels based on how much support the person needs.
The distance between these criteria and the labels of a century ago is enormous. In the 1920s, a child who lined up objects for hours, never responded to their name, and screamed when furniture was rearranged would have been called feebleminded and possibly sent to an overcrowded institution. That same child today would receive a specific diagnosis, an understanding of their neurological differences, and access to support tailored to their needs. The behaviors were always there. What changed was the language, and with it, everything that followed.