Mens turbataA history of the troubled mind

Essay · 1969–2023

"On being sane in insane places": Rosenhan's study and its challengers

In 1973 David Rosenhan reported that eight sane people were admitted to psychiatric hospitals by claiming to hear voices. The study helped reshape American psychiatry; later research suggests it was not done as described.

On 19 January 1973 the journal Science published a nine-page paper by a Stanford professor of psychology and law, David Rosenhan. Its title, "On being sane in insane places", promised an answer to an old question, and the paper gave one: eight sane people had presented themselves at twelve psychiatric hospitals, claiming to hear voices, and every one had been admitted. "We cannot distinguish the sane from the insane in psychiatric hospitals," Rosenhan concluded1,4. Few papers in the social sciences have been read so widely or done so much.

Its fame did not come from discovering something new. Psychiatry's own researchers had shown for more than a decade that diagnosis was unreliable. What Rosenhan supplied was a story the public could not ignore, published in a journal it trusted. That is why the study mattered, and also why the evidence now assembled against it matters: the paper's influence ran far ahead of its data, and the data appear not to have been what the paper said.

Twelve admissions

According to the paper, the eight "pseudopatients" were a psychology graduate student in his twenties and seven older people: three psychologists, a paediatrician, a psychiatrist, a painter and a housewife; Rosenhan himself was one of them. Each telephoned a hospital for an appointment and said that he or she had been hearing voices, unclear but apparently saying "empty", "hollow" and "thud". Once admitted, they stopped reporting any symptoms and behaved normally1,7. The twelve hospitals, in five states on the east and west coasts, ranged from neglected rural state hospitals to a private hospital linked to a university4.

All were admitted. Those in public hospitals were diagnosed with schizophrenia (the successor of Kraepelin's dementia praecox), the patient in the private hospital with manic-depressive psychosis; they stayed from 7 to 52 days, 19 on average, and were discharged, the paper said, with schizophrenia "in remission"4,6. Staff read ordinary behaviour, such as note-taking, as symptoms, while some of the other patients guessed that the newcomers were journalists or researchers7. A ninth volunteer, Rosenhan wrote in a footnote, had been excluded for falsifying parts of his personal history4.

The paper's second experiment was, if anything, more damaging. Staff at a teaching hospital who doubted the findings were told that pseudopatients would try to gain admission over the next three months. Of 193 patients admitted, 41 were judged with high confidence by at least one staff member to be impostors, and 23 were suspected by at least one psychiatrist. Rosenhan had sent nobody4,6.

Why it landed

The claim had a long ancestry. In 1887 Nellie Bly had feigned madness, been committed to the women's asylum on Blackwell's Island, and written that once inside, the more sanely she behaved the madder she was thought; Elizabeth Packard had made a related charge about the asylum's power to define sanity a generation earlier (see The patients who wrote back). By 1973 the charge had academic support. In 1959 Benjamin Pasamanick and his colleagues warned that psychiatric diagnosis was too unreliable to be trusted in research or treatment; a review by Aaron Beck in 1962 found that psychiatrists agreed on specific diagnoses of functional disorders at best only about half the time; and a British–American study published in 1972 found that New York psychiatrists diagnosed schizophrenia in nearly 62 per cent of their patients and London psychiatrists in 34 per cent, differences that re-examination traced to the psychiatrists rather than the patients4. Erving Goffman had described the mental hospital as a damaging "total institution", Thomas Szasz had called mental illness a myth, and R. D. Laing had suggested that the madness was society's (see The rise and fall of the asylum)4.

Those critiques were read mainly by specialists. Rosenhan's paper was reported by at least 70 newspapers; Science gave nine pages of a later issue to protesting letters and his reply; and within weeks, on 1 February 1973, the trustees of the American Psychiatric Association held an emergency meeting to discuss the criticism of the profession4. American psychiatry in these years was dominated by psychoanalysis, whose practitioners regarded diagnostic labels as of little use, and its manual, DSM-II (1968), ran to 134 pages4.

Spitzer's reply and DSM-III

The most sustained answer came from the Columbia psychiatrist Robert Spitzer. In 1975 he argued that the study confused deception with diagnosis2. He quoted the psychiatrist Seymour Kety: a man who secretly drank blood and then came to an emergency department vomiting it would be diagnosed with a bleeding ulcer, and that would show nothing about whether medicine could diagnose ulcers2,7. On this view a hospital that believes a patient who reports hallucinations is doing its job, and psychiatrists should not be expected to assume that patients are pretending7.

Spitzer asked Rosenhan, in private letters, for the pseudopatients' admission records, with identities removed, so that he could see whether they had really reported only voices. Rosenhan declined, saying that he had to protect his sources and that other researchers must wait until he had finished a book on the study4. He had signed a contract with Doubleday in 1974; the book was never finished, and the publisher later sued to recover the advance4,5.

Meanwhile the APA had appointed Spitzer to lead the revision of its manual. DSM-III, published in 1980, ran to 494 pages and replaced loosely described categories with explicit lists of criteria intended to make psychiatrists in different places agree; it also introduced new categories, among them post-traumatic stress disorder (see War and the mind: from nostalgia to PTSD)4,8. The manual helped to end the dominance of psychoanalysis in American psychiatry and to establish the descriptive, Kraepelinian approach associated with the biological turn4.

The search for the pseudopatients

Rosenhan died on 6 February 2012, aged 82, remembered by colleagues as a generous teacher and a pioneer of psychology applied to law; one called his paper "a moral outcry" against wrongful diagnosis and its consequences6. The journalist Susannah Cahalan later began research for a book about the study, with access to his private papers, including notes, a diary, the unfinished book manuscript and his own hospital records. The Great Pretender was published in November 20193,4.

Cahalan could identify only one of the seven other pseudopatients reported in the paper, a former Stanford graduate student, and found the excluded ninth, the psychologist Harry Lando4,5. Her findings, and Scull's reading of the documents she shared with him, are these4,5:

  • Rosenhan's own admission, to Haverford State Hospital on 6 February 1969, was not as described. The hospital's notes record that he reported far more than three words: months of being unable to work, the belief that he could hear what other people were thinking and was sensitive to radio signals, and deep low mood; he was diagnosed with schizophrenia of a schizo-affective type4.
  • The graduate student recalled no long preparation, though the paper said the pseudopatients had been carefully trained, and his hospital chart contains no discharge diagnosis of schizophrenia "in remission"4.
  • Lando had published his own account in 1976. He spent 19 days on a well-staffed, unlocked ward and called his impressions "overwhelmingly positive", the opposite of the paper's picture; a detail from his stay nonetheless appears in the published paper, attributed to an unnamed pseudopatient4.
  • An earlier draft of the paper, which Rosenhan circulated to a colleague, included nine pseudopatients, but its figures for staff contact with patients were identical to the published figures for eight4.
  • No trace of the other pseudopatients could be found in Rosenhan's files except a list of pseudonyms, and a two-year search, including a private investigator, found none of them4.

What survives

Cahalan did not conclude that Rosenhan's questions were empty. She believed he had "exposed something real": the stigma attached to psychiatric diagnosis, the unequal treatment of mental and physical illness, and the limits of psychiatry's diagnostic tools5. Those problems were documented before 1973 by better methods, in the reliability studies that Rosenhan's paper overshadowed4.

The study's history is also a lesson in how evidence about psychiatry has been weighed. For half a century a paper with anonymous participants, unnamed hospitals and no available raw data was taught as settled fact, because it confirmed what many readers already believed about the asylum (The rise and fall of the asylum, deinstitutionalisation). The same scepticism that Rosenhan turned on psychiatric diagnosis, applied to his own paper, has turned out to be justified.

Key dates

  1. 6 February 1969 David Rosenhan, under a false name, is admitted to Haverford State Hospital in Pennsylvania, the first admission later reported in his study.
  2. 19 January 1973 Science publishes Rosenhan's 'On being sane in insane places'.
  3. October 1975 Robert Spitzer publishes his critique of the study in the Journal of Abnormal Psychology.
  4. 1980 The American Psychiatric Association publishes DSM-III, prepared by a task force led by Spitzer.
  5. 5 November 2019 Susannah Cahalan's The Great Pretender reports that Rosenhan's study cannot be reconciled with his own papers.
  6. February 2023 Andrew Scull, in History of Psychiatry, calls the study 'a spectacularly successful case of scientific fraud'.

See the full timeline

Sources

  1. Primary source David L. Rosenhan, On being sane in insane places, Science 179(4070): 250–258 (1973-01-19).
  2. Scholarship Robert L. Spitzer, On pseudoscience in science, logic in remission, and psychiatric diagnosis: a critique of Rosenhan's 'On being sane in insane places', Journal of Abnormal Psychology 84(5): 442–452 (1975-10).
  3. Scholarship Susannah Cahalan, The Great Pretender, Grand Central Publishing (Hachette Book Group) (2019-11-05).
  4. Scholarship Andrew Scull, Rosenhan revisited: successful scientific fraud, History of Psychiatry 34(2): 180–195 (2023).
  5. Scholarship Christie Aschwanden, An undercover journey into the heart of madness (review of The Great Pretender), Undark Magazine (2019-12-27).
  6. Scholarship Lee Ross, David Kavanagh, David L. Rosenhan (1929–2012) [obituary], American Psychologist 68(6): 469 (author's version, QUT ePrints) (2013).
  7. Wikipedia Rosenhan experiment, Wikipedia.
  8. Official History of PTSD in Veterans: Civil War to DSM-5, US Department of Veterans Affairs, National Center for PTSD.

Cross-checked with Wikidata Q402915 and the English Wikipedia article.

Pages that mention "On being sane in insane places": Rosenhan's study and its challengers