Viewpoint Expert opinions on trends and controversies in autism research.
Philip Giordano
Opinion / Viewpoint

Correcting the record: Leo Kanner and the broad autism phenotype

by ,  /  26 April 2016
The Experts:

James Harris

Professor, Johns Hopkins University

Joseph Piven

Professor, University of North Carolina at Chapel Hill

Leo Kanner’s ideas about autism have been widely misunderstood. Recent media accounts and books report that he blamed parents for causing autism. But Kanner did not blame parents. In fact, his astute observations of parents gave him deep sympathy for them, and led to his fundamental insights about their behavior that fueled important advances in the genetics of autism.

Kanner’s classic 1943 paper on ‘early infantile’ autism was the first systematic description of this unique neurodevelopmental condition1. His conclusion that autism is an innate disorder set the stage for later studies in identical twins that confirmed the condition’s genetic basis2. In his paper, Kanner also described the parents of these children and their behavioral traits. He wrote that, overall, the parents seemed perfectionistic and preoccupied with abstractions, rather than showing a genuine interest in people.

Kanner addressed the question this observation raises: Are parents in some way to blame for their child’s condition? On the contrary, he wrote: “The children’s aloneness from the very beginning of life” suggests that parents’ behavior is not a causative factor. He went on to state that scientists must assume “that these children have come into the world with the innate inability to form the usual, biologically provided affective contact with people1.”

Kanner published his paper at a time when support for eugenic genetic engineering as a way to improve society was prevalent among scientists, making the diagnosis of an innate condition a serious matter. This was a time in the United States when sterilization of people with intellectual disability was legal. Moreover, in 1942, Foster Kennedy, a prominent neurologist from New York, proposed a “mercy death” for children with severe intellectual disability3.

Kanner publicly opposed Kennedy and offered a spirited defense for those with severe, innate developmental disorders. He insisted that all lives matter and that “we should extend the democratic ideal” to these individuals because much could be learned by studying their unique qualities4.

In defense of mothers:

Despite Kanner’s many publications and public statements, his views on the causes of autism have been widely misinterpreted. His attempts to distance himself from eugenic views, along with his descriptions of parents’ behavior, led others to mistakenly believe that Kanner had changed his mind about autism being innate and instead assigned parents much of the blame for their child’s autism.

On the contrary, Kanner held tightly to his original proposal that autism was an innate condition, which was widely understood to mean it had a genetic basis. His behavioral observations of parents contributed to a breakthrough concept that is wholly consistent with genes being a key part of the autism story. Instead of parenting causing autism, Kanner’s idea — which has since been validated — was that autism (and its genetic roots) underlies some of the behavior in a subset of parents.

From the 1940s through the 1960s, the negative behavioral view (in this case, aimed at parents) had strong adherents, in part because it seemed to offer some hope. Psychoanalysis — and the notion that treating people for psychological trauma in early life has broad benefits — held sway among the clinical and scientific establishment. In the case of autism, if parents were in some way to blame for their child’s condition, the parents could be treated psychoanalytically.

Unlike Kanner, Bruno Bettelheim, who then directed the University of Chicago’s Sonia Shankman Orthogenic School for emotionally disturbed children, although he was an art historian by training, believed that autism wasn’t innate, but rather the result of maternal aloofness. The press and Bettelheim popularized the idea of a parent who displayed such emotional detachment as a ‘refrigerator mother.’ Research at this time had shown that an extreme lack of social stimulation and emotional deprivation could severely affect development, which lent itself to the blame-the-parent ‘theory’ of autism5.

But Kanner saw through this idea. His subsequent research on children with autism showed there was no evidence for parental “mistreatment, overt rejection, or abandonment” as a factor in a child’s having autism. In addition, Kanner had long fought against psychoanalytic theories that blame mothers. Even before he published his seminal description of autism, he expressed this view in his widely read 1941 book, “In Defense of Mothers: How to Bring Up Children in Spite of the More Zealous Psychologists6.”

Heavy hearts:

The book opens with a sympathetic letter to mothers who have come to him with “heavy hearts” and in despair, saying, “It’s all my fault,” believing themselves to be to blame for their child’s problems. Kanner instead made clear his view that mothers are inundated with conflicting advice on child-rearing. He advised them to close the “door to the din from pseudopschological markets” and encouraged them to regain their self-confidence and trust their own feelings in the face of a barrage of parent-blaming.

Kanner went on to respond directly to accusations that parents were to blame. He was appalled at Bettelheim’s theory that so-called ‘cold mothers’ were the cause of autism. At the first meeting of the parent advocacy group the National Society for Autistic Children (now called the Autism Society) in 1969, he said, “From the very first publication until the last, I spoke of this condition in no uncertain terms as ‘innate.’ But because I described some of the characteristics of the parents as persons, I was misquoted often as having said that ‘it is all the parents’ fault.’” Then in his typical warm, wry, jovial manner, he announced to the parents in attendance: “Herewith I especially acquit you people as parents.” Many of the parents in the audience rose and tearfully cheered him. The society awarded Kanner an official citation for his contributions to children with autism and their families.

But despite Kanner’s open denial that he had ever blamed parents, the specter of the ‘refrigerator mother’ theory continues to haunt popular histories of his work. Some recent histories relate that Kanner eventually gave in to others’ views and came to agree with professionals who blamed the parents, just as they did. Critics often add that, after recanting much later, Kanner went back to his original proposal that autism was innate — but that by then, it was too late and the damage was done. This version of events, however compelling, is not accurate.

Telling traits:

We revisited Kanner’s historical contributions on autism, seeking to clarify the confusion about how Kanner’s descriptions of parents’ behavior have been interpreted. We conclude that Kanner, in describing the parental behavior he observed, was describing the features of the broad autism phenotype — a genetic predisposition for autism-related traits that are thought to be an expression of autism’s genetic causes. Kanner’s observations may in fact constitute the first descriptions of a broad autism phenotype, a concept that has been widely studied and validated.

In his evaluations of children with autism, Kanner recognized that some parents have characteristics that are milder but qualitatively similar to the defining features of autism. In particular, these parents were overly focused on detail and had limited interest in social interactions.

In 1956, Kanner and his collaborator at Johns Hopkins University in Baltimore, Leon Eisenberg, published a paper describing autism as a psychobiological disorder that both genetic background and the environment contribute to7. “If one considers the personalities of the parents who have been described as successfully autistic, the possibility suggests itself that they may represent milder manifestations and that the children show the full emergence of the latent structure,” they wrote.

In the nearly 75 years since Kanner’s first publication on autism, his recognition of the broad autism phenotype in a subset of parents set the stage for later family-genetic studies demonstrating that autism is a genetic disorder. Rigorous behavioral measures and brain imaging studies now support the existence of a broad autism phenotype among some parents of children with autism8,9,10.

Dynamic role:

In their 1956 publication, Kanner and Eisenberg acknowledged that although the psychological environment does not cause autism, it does affect child development. “It is difficult to escape that the emotional configuration of the home plays a dynamic role” in the development of children with autism, they wrote7. If parents had difficulty recognizing their children’s social cues, they could inadvertently adversely affect their child’s development — just as any parent could if they failed to find a way to emotionally engage with their child.

Child-rearing strategies, behavioral interventions and home environment make a difference in both typical and atypical neural development in children. Ample evidence exists for the idea that enriched social environments lead to better outcomes for children at risk of autism, and current early intervention research focuses on optimizing the dynamic engagement of the children with their parents11. This ongoing research may lead to more successful early behavioral interventions for children with autism, helping to improve their language skills and ability to attend to social cues.

As founding director of child psychiatry at John Hopkins University from 1930 to 1959, Kanner was very much a pioneer in providing clinical support to families of children with developmental disorders. In 1935, he wrote the first textbook of child psychiatry, one that defined child psychiatry as a medical discipline. His passion for social justice is a potent reminder of the importance of finding appropriate services for all children. To those who knew him,* Kanner was a wise and compassionate physician who cared about all children and supported their parents. His career continues to be an inspiration to us all.

James Harris is professor of psychiatry and behavioral sciences at Johns Hopkins University School of Medicine in Baltimore, where he is founding director of the Developmental Neuropsychiatry Clinic. Joseph Piven is Thomas E. Castelloe Distinguished Professor of Psychiatry and director of the Carolina Institute for Developmental Disabilities at the University of North Carolina at Chapel Hill.

* James Harris was Leo Kanner’s student during his residency training and was a successor to Kanner as director of the child and adolescent psychiatry division at Johns Hopkins University School of Medicine. Harris gave the memorial tribute to Kanner at the plenary session of the American Academy of Child Psychiatry in 1981, the year of Kanner’s death.

  1. Kanner L. Acta Paedopsychiatr. 35, 100-136 (1968) PubMed
  2. Folstein S. and M. Rutter J. Child Psychol. Psychiatry 18, 297-321 (1977) PubMed
  3. Kennedy F. Am. J. Psychiatry 99, 13-16 (1942) Full text
  4. Kanner L. Am. J. Psychiatry 99, 17-22 (1942) Full text
  5. Spitz R.A. Psychoanal. Study Child 1, 53-74 (1945) PubMed
  6. Kanner L. (1941). In defense of mothers: How to bring up children in spite of the more zealous psychologists. Springfield, IL: Charles C. Thomas.
  7. Kanner L. and L. Eisenberg Psychiatr. Res. Rep. Am. Psychiatr. Assoc. 55-65 (1957) PubMed
  8. Piven J. et al. Am. J. Psychiatry 154, 185-190 (1997) PubMed
  9. Losh M. et al. Arch. Gen. Psychiatry 66, 518-26 (2009) PubMed
  10. Yucel GH. et al. Cereb. Cortex 25, 4653-4666 (2015) PubMed
  11. Green J. et al. Lancet Psychiatry in press

11 responses to “Correcting the record: Leo Kanner and the broad autism phenotype”

  1. Planet Autism says:

    “Instead of parenting causing autism, Kanner’s idea — which has since
    been validated — was that autism (and its genetic roots) underlies some
    of the behavior in a subset of parents.”

    Here it sounds like you’re saying the parents were autistic themselves. Yet further down you say “Kanner recognized that some parents have characteristics that are milder
    but qualitatively similar to the defining features of autism.” and “We conclude that Kanner, in describing the parental behavior he observed, was describing the features of the broad autism phenotype — a genetic predisposition for autism-related traits that are thought to be an expression of autism’s genetic causes.”

    I don’t actually believe in the BAP, I believe you are either autistic or you are not. There are brain differences. Some autistic parents may have had a fortuitous start, conducive environment or a lucky break which enabled them to blend in, mimic, use cognition to calculate how to muddle through. This does not mean they were not autistic. There is a great blog post here: which wrote everything I believed inside but couldn’t quite explain.

    There is also a brilliant paper here “Invisible at the End of the Spectrum: Shadows, Residues, ‘BAP’, and the Female Asperger’s Experience” (

    It’s very frustrating when professionals don’t understand the nature of autism and talk about autistic traits as if you can just have a few and still be neurotypical. It’s responsible for the “not enough traits for a diagnosis” scenario ( which means autistics are not diagnosed and condemned to lack of support and understanding, with all the negative effects that has on outcomes.

    What I think happens, is that professionals often don’t understand masking and mimicking and they misjudge based on that. Imagine of any of Kanner’s parents were Temple Grandin, Nicola Tesla, Einstein etc. – he would have been saying they had some autistic like traits and were BAP but would have deemed them not autistic because they were managing, were successful and not locked away in a home for the retarded as they used to do!

    • Science is not about what you “believe in”. Science is about formulating and testing hypotheses, and the BAP hypothesis is currently supported by solid data. Can you point to some actual SCIENCE that discredits BAP and not mere opinion pieces? And don’t pull the autistic card on me–I’m also autistic.

      • Planet Autism says:

        I don’t need to ‘pull the autistic card’, that would have absolutely nothing to do with this. My belief is a science because it is generated from my brain. My brain is explained by biology. Science also has spectacularly failed in the past. Science is evolving all the time. There are current science claims that will be proven to be wrong. We don’t need to talk about science with this. Anyone that relies solely on science is either too rigid (common in autistics) or is a scientist being precious about their field. Science is not only about the findings, it’s about the correct interpretation of those findings. As people that like to rely solely on scientific claims usually like to say ‘correlation does not equal causation’. Don’t get me started on scientific fraud either. Just out of interest, what do you see as the defining scientific proof that BAP exists?

  2. Seth Bittker says:

    I think Kanner was a brilliant observer and a compassionate man, and his work in identifying autism and describing it was very important. However, it seems to me that the authors of the above article are engaging in revisionist history and leaving out some of Kanner’s statements between 1949 to 1960 to cleanse Kanner from any association with theories blaming parents for their child’s autism.

    The evidence shows that it was in fact Leo Kanner who first postulated the “refrigerator mother” theory of autism which would do so much damage. In a paper from 1949, Kanner wrote that it may be that autism is due to a “genuine lack of maternal warmth” and fathers of those with autism rarely played with their children. He noted that children with autism were exposed from “the beginning to parental coldness, obsessiveness, and a mechanical type of attention to material needs only…. They were left neatly in refrigerators which did not defrost. Their withdrawal seems to be an act of turning away from such a situation to seek comfort in solitude.” See: In an interview from 1960, Kanner said of parents of children with autism, that they “just happening to defrost enough to produce a child.” See “The child is father”. TIME. 1960-07-25. These are condemnatory statements about parents of children with autism, and the first one directly ties the coldness of parents to behavior of the child.

    Admittedly Kanner was a compassionate man, his hypothesis seemed reasonable given the data available at the time, and it seems his position evolved to one of greater sympathy for the parents of those with autism. At the time there was no knowledge of autism biochemistry or autism genetics. In addition given that Kanner seems to have been recognizing the “broader autism phenotype” in some parents as the authors of the above article have noted, it is easy to see how Kanner could make the mistake of jumping to the conclusion that parental behavior was a cause of their child’s autism. So to be clear, I do not blame Kanner for developing his “refrigerator mother” theory, but that does not mean we should forget that it was he who originated it and at least for a time he seemed to believe it.

    • Planet Autism says:

      I thought I’d read somewhere about those comments of his, it was like stepping into an alternate reality reading the above article – I thought I must be going mad to have imagined it! That’ll teach me to check out my doubts next time.

  3. Steve Silberman says:

    Let me say first off that I completely agree with Piven and Harris that what Leo Kanner was seeing in parents was the broad autism phenotype. I believe that’s a very important insight, and one that I came to myself after a very thorough review of Kanner’s publications over the course of more than six years. I also agree that Kanner was a compassionate and deeply caring man and an acute clinical observer; he was also a highly lucid and compelling writer. But I also agree with commenter Seth Bittker below that what Piven and Harris are doing here is not simply “correcting the record,” as the headline insists, but engaging in historical revisionism.

    The fullest explanation of my thoughts on Kanner — with exhaustive citations — can be found in my book “NeuroTribes: The Legacy of Autism and the Future of Neurodiversity.” I assume that Piven and Harris had my book in mind — as well as John Donvan and Caren Zucker’s “In a Different Key” — when they refer to “popular histories” of his work that allegedly “misunderstand” Kanner’s position on parents. I would suggest that any understanding of Kanner that attempts to render his public position on this crucial issue as single-minded and unchanging is inherently flawed, as he changed his position strategically over the years, depending on which audience he was addressing and the prevailing winds in psychiatry at the time. That’s what makes the record of his statements so perilously subject to cherry-picking.

    As I explain in “NeuroTribes,” the tension between his various positions on parents was present from the start, in his 1943 paper. His suggestion that autism might be “innate” was indeed highly significant (and paralleled Hans Asperger’s equally prescient insight that the causes of autism are “polygenetic”); but Kanner left that conclusion more open-ended than Piven and Harris suggest. He wrote:

    “One other fact stands out prominently. In the whole group, there are very few really warmhearted fathers and mothers. for the most part, the parents, grandparents, and collaterals are persons strongly preoccupied with abstractions of a scientific, literary, or artistic nature, and limited in genuine interest in people. Even some of the happiest marriages are rather cold and formal affairs. Three of the marriages were dismal failures. The question arises whether or to what extent this fact has contributed to the condition of the children. The children’s aloneness from the beginning of life makes it difficult to attribute the whole picture exclusively to the type of the early parental relations with our patients.”

    Note, Kanner was quite clear about his mixed feelings on the subject: “The question arises,” he writes, before suggesting that “the whole picture” of autism might not be “exclusively” caused by disturbed relations with parents. As I explain in “NeuroTribes” at length, in these statements, Kanner was being a good student of his mentor, Adolf Meyer, who taught his disciples to reject any single explanation for complex disturbances of behavior, but to look for both biological and psychological contributing factors.

    By 1948, when Kanner became the subject of an influential article in Time called “Frosted Children,” his opinion had become less nuanced. The entire article is a condemnation of parents, beginning with the line, “Leo Kanner used to stand up for parents,” and going on to these fateful lines under the heading “Cold Perfectionists”:

    “But there was something wrong with all of [these parents]. They showed a ‘mechanization of human relationships,’ described themselves and their spouses as undemonstrative. There was, Dr. Kanner found, ‘no glamor of romance in premarital courtship, no impetuousness in postnuptial mating.’ He saw only one mother hug her child warmly and bring her face close to his; many of the busy fathers hardly knew their children. The parents wanted to do the right thing by them; but their idea of the right thing was ‘the mechanized service of the kind which is rendered by an over-conscientious gasoline station attendant. The children, says Dr. Kanner, were ‘kept neatly in a refrigerator which didn’t defrost.’
    Were the cold parents freezing their children into schizophrenia? Dr. Kanner did not say yes or no; but he has found no case of infantile autism among children of “unsophisticated” parents. Said he of his pathetic patients: ‘Their withdrawal seems to be an act of turning away . . . to seek comfort in solitude.'”

    Granted, Kanner’s description could be seen as a rather unkind description of the Broad Autism Phenotype. But that’s much easier to do in retrospect *now that we know such a thing exists.* In its historical context, Kanner’s description was much more apt to be interpreted by his colleagues and the general public as a description of wilful parental inadequacy — as indeed it was. In subsequent papers, Kanner would cast aspersions on mothers who insisted on going to college rather than being stay-at-home caregivers. This was not framed as some inevitable result of their genetic makeup, but as a failure of parenting that could be corrected with expert advice from a child psychiatrist.

    I could go on at length; but that’s what my book is for. Perhaps the clearest and most unambiguous refutation of Kanner’s “open denial” of responsibility for the refrigerator parenting theory, as Piven and Harris put it, was in the memorial statement written after his death by his chief disciple Leon Eisenberg and autism pioneers Stella Chess and Eric Schopler, published in the Journal of Autism and Developmental Disorders, Vol. 11, No. 3, 1981. “[Kanner’s] caring objectivity gave him the strength to remain faithful to empirical evidence, and to change his working assumptions about the autism syndrome as new data were developed,” they wrote. “It came as no surprise to those of us who knew him when the man credited with the term refrigerator mother explained to the members of the National Society for Autistic Children, at their annual meeting in 1971, that the blame for their child’s autism implied by this term was now established as inappropriate and incorrect.”

    What did they mean by saying that Kanner had “change(d) his working assumptions”? Piven and Harris would have us believe that he never did. But Eisenberg, Schopler, and Chess — who knew Kanner’s history intimately — knew better.

    • Planet Autism says:

      Well, I will have to maintain my stance on BAP until someone shows me some concrete proof that it exists.

      The parent blame has gone nowhere, it is alive and well in the professional world. There are more than Kanner to blame for that. Families are being torn apart by the state due to misrepresentation, discrimination and misunderstanding of autistic parenting. It’s a scandal of epic proportions.

      I do also suspect that some of what he witnessed was at least in part due to cultural norms of the age. As an autistic parent to two autistic children, we are all extremely loving with one another so the affection deficit is a total myth. As the expression about meeting one person with autism goes…

      We have our gifts and talents, but we are all disabled by our condition too, by the way…it is in no way ‘just a difference’.

      • Steve Silberman says:

        > we are all disabled by our condition too, by the way…it is in no way ‘just a difference’.

        I’m not sure if that comment was directed to me, but I emphasize the point that autism should be considered a disability, and accommodated as such, in nearly every talk I give, along with talking about its potentially potentially positive aspects when provided with adequate support,

        • Planet Autism says:

          I’m pleased to hear that. Because it’s not only the social model of disability that accounts for our difficulties and challenges. Some in the autistic world are even very strident about it being seen as just a difference and are furious at the notion that it’s epigenetic and that this therefore constitutes an insult from something in the environment in those with a particular genetic make-up. Temple Grandin talks about autistics not moping around and getting on with it to succeed, but she has come from a background of early diagnosis and all the right supports to achieve what she has. Most are not so lucky. Many are diagnosed very late, misdiagnosed, even when diagnosed denied the support they should have had. Autism awareness in it’s full meaning is virtually zero and those that should adhere to equality laws to give autistics a shot, don’t. So telling people it’s a gift and a difference to change attitudes, won’t do the trick. In fact it does the opposite, it makes things a lot harder.

  4. Florencia Ardón says:

    Just adding–if you truly believe autism is genetic, please don’t speak about people being ‘at risk’ for autism. One is either autistic or not. One is not ‘turned’ autistic by an external condition or situation.

    • Or one could be scientific about it and admit that non-syndromic autism (the majority of cases are non-syndromic) is NOT genetic. It is, instead, a sporadic condition with multiple genetic risk factors. The idea that autism is genetic has been left behind by most professionals in the field. Instead, the emerging consensus is that multiple genetic variations can play a role in elevated or reduced possibility of developing autism, but environmental influence also plays a non-insignificant role.

Leave a Reply

Your email address will not be published. Required fields are marked *