statue in park
Source: (c) David Hellerstein
What do you do when the truth comes out, even decades after the fact?
In recent years, we have become accustomed to statues being torn down, whether imperialists or slaveowners, or former heroes unveiled as crooks. Cultural monuments are also being demolished these days—though often difficult to dislodge from the public squares of our minds.
In psychiatry, I argue that the top candidate for demolition is the late Stanford psychologist David Rosenhan’s 1973 paper, “On Being Sane in Insane Places,” which was published in the preeminent journal Science almost 50 years ago.
In “On Being Sane,” Rosenhan famously describes sending eight people without mental illnesses for admission to psychiatric hospitals. They purportedly were instructed to report that they were hearing voices saying “empty,” “hollow,” and “thud”—symptoms of a nonexistent diagnosis concocted by Rosenhan, “existential psychosis.” All eight pseudo-patients (including Rosenhan himself) were promptly admitted to psychiatric wards and diagnosed with paranoid schizophrenia.
Immediately following admission, they were instructed to stop simulating “any symptoms of abnormality.” Though they purportedly weeks spent in the hospital conversing with other patients and staff and writing down observations about their experiences, none were detected as being sane—except by fellow patients. All were prescribed heavy doses of medications, and released after an average of 19 days, diagnosed with paranoid schizophrenia in remission.
Rosenhan’s electric accusations led to front-page newspaper and magazine stories across the country and changed psychiatry forever. Part of its appeal is no doubt Rosenhan’s passion, which is evident even today. Eloquent and incandescently outraged, he describes the mentally ill as “society’s lepers,” and evokes an overwhelming sense of powerlessness of poor souls incarcerated in psychiatric settings, stripped of civil rights and their very humanity. Psychiatric hospitals are revealed as heartless bureaucracies where staff spend hours in their glass-enclosed offices, barely conversing with patients—except for occasional kindly nurses who dispense numbing medications.
Somehow, concluded Rosenhan, psychiatry’s fatal flaw lay in its flawed diagnoses: “We have known for a long time that diagnoses are often not useful or reliable, but we have nevertheless continued to use them. We now know that we cannot distinguish insanity from sanity … How many people, one wonders, are sane but not recognized as such in our psychiatric hospitals?”
There was ample truth to Rosenhan’s observations. Mid-20th century psychiatric hospitals were often vast neglectful warehouses, with limited evaluation and treatment. The 1970s diagnostic system was flawed: A patient would routinely be diagnosed with schizophrenia in Britain and with manic depression in the US His anti-authoritarian critiques resonated with the Vietnam War era zeitgeist, coalescing with Thomas Szasz, RD Laing’s antipsychiatry fervor, Nurse Ratchet in Ken Kesey’s One Flew Over the Cuckoo’s Nestand the Chicago 7 circus trial and traitorous Watergate shenanigans.
“Sane in Insane Places” (henceforth referred to as SIIP) had a profound impact on psychiatry. Its apparent radical honesty inspired a generation of psychiatrists to provide compassionate and respectful care and to fix a dysfunctional mental health system. It accelerated the process of deinstitutionalization—for better or worse, since many people with mental illnesses now are incarcerated in prisons and receive little psychiatric care or become homeless; Many have arguably worse lives than back ward patients of earlier decades.
More productively, SIIP fast-tracked the development of the psychiatric diagnostic manual DSM-IIIpublished in 1980. The DSM-III and its successor volumes (now up to edition 5-TR) have revolutionized psychiatry, with the development of a reliable (though not often biologically “valid”) diagnostic system, leading to significant advances in clinical treatment and outcomes improving for millions, though it is now being supplanted by new neuroscience-based models. Worst, the paper has lent legitimacy to anti-psychiatry movements, stigmatizing both patients and their caregivers.
Today, SIIP retains a primary place in our culture. Not only does it have 4,416 citations in Google Scholar but its conclusions are presented uncritically in most introductory psychology textbooks published today.
Here’s the problem, and here’s why retraction is necessary, even fifty years after publication: “Sane in Insane Places” is a fraud.
Three years ago, in her book The Great Pretender, journalist Susannah Cahalan exhaustively investigated Rosenhan’s work. She reviewed primary sources, including medical records and Rosenhan’s notes, and conducted countless interviews. She discovered innumerable errors, distortions, misrepresentations, possible data fabrication, and deliberate suppression of contradictory findings.
Devastatingly, Cahalan obtained psychiatric records for Rosenhan’s own pseudo-patient admission to Haverford State Hospital in Pennsylvania. Rather than merely claiming to hear voices, however, Rosenhan told the admitting psychiatrist he had symptoms for months, felt persistently suicidal, and that he put copper pots next to his head to block hallucinations—far worse symptoms than a single voice saying “thud. ” While Rosenhan later claimed that he excluded his Haverford admission from “Sane in Insane Places,” Cahalan convincingly demonstrates that her patient narratives primarily describe Rosenhan’s own hospital experiences.
Furthermore, Rosenhan’s paper excluded one actual participant, a young male graduate student. Decades later, Cahalan tracks him down. Now a psychology professor in Minnesota, Harry Lando had remarkably positive experiences in the hospital, bonding with staff and fellow patients, going on outings, and playing folk songs. In fact, life-transforming positive experiences! Rather than conceding that one subject’s report diverged from others, and tempering his study’s harsh conclusions, Rosenhan summarily deleted him. Excluding such “outliers” in a scientific paper is unforgivable.
Even more damning, Cahalan concludes that there is no evidence that there even were eight pseudo-patients and is only able to document two cases. Both in the 1970s and today, no one has ever been able to find primary source material for other subjects. Then there is Rosenhan’s strange behavior post-publication. He never conducted a larger study to replicate his small report. Though he received a substantial advance and wrote over 100 manuscript pages, he never published a book on the subject, leading Cahalan to conclude it would have exposed him as a fraud. Rosenhan even had to return his publishing advance.
Why was the paper even published? Cahalan was unable to find correspondence from Science to explain how the peer review process and by what criteria it was accepted for publication—or whether reviewers requested primary documents such as hospital notes or medical records to bolster Rosenhan’s claims.
In short, Cahalan’s Great Pretender confirms what psychiatrist Robert Spitzer contended in a 1975 paper, “On Pseudoscience in Science, Logic in Remission, and Psychiatric Diagnosis”: Rosenhan’s paper is pseudoscience. It falls far short of meriting publication in any scientific journal by today’s standards, and even by the standards of the 1970s.
Which leads to this conclusion:
Approaching 50 years after publication, I argue that SIIP should be retracted, joining the ranks of other now-discredited studies. In 2018 the New England Journal of Medicine retracted an interviewer to 2013 paper purporting demonstrating the superiority of the Mediterranean diet, after criticizing John Carlisle, a British anesthesiologist, pointed out flaws in the study participants’ randomization of study concluding, which made it impossible to conclude that the diet “caused” a 30 percent drop in heart attacks and strokes. Instead, the corrected and weaker paper could only conclude the diet was “associated with” such outcomes. A host of other retractions have proliferated since, including influential studies of cancer and kidney disease. If retracted, SIIP would be by far the most highly cited of all retracted papers; the Mediterranean diet paper had only 1,895 citations before retraction, compared to SIIP’s 4,416
It is time for the Science To begin a formal review, as done with other disputed studies, appointing a council of experts in psychiatric treatments, social psychology, history, ethics, and other disciplines. They should review the paper’s methods, its strengths and weaknesses, as well as its path to publication at Scienceand then recommend a process of restorative justice.
One could ask: What harm does a nearly 50-year-old paper do staying where it is, tucked in the Science archives like a verdigris statue standing forlornly in a park?
On the contrary: It is not neglected. Remaining as it is, unchallenged, “Sane in Insane Places” perpetuates lies. How ironic that a paper taking psychiatry to task for hypocrisy and flawed methods should be itself a collection of falsehoods, obfuscations, selective reporting, and outright fabrication. By portraying psychiatry as a harmful discipline led by uncaring professionals, it stigmatizes not only caregivers and institutions but also patients, those seeking and desperately needing help. Remaining unchallenged, it continues to spread an undeniably false narrative. Omitting Harry Lando’s story of healing dashes any hope of redemption.
Should “Sane in Insane Places” be pulled off the Internet, blown to bits like statues of overthrown dictators?
The problem is this paper has undeniable historical importance.
Perhaps the best solution would be to move it into a special park, so to speak, with other discredited intellectual monuments, surrounded by documents providing a contemporary context—as a cautionary tale. Amended or annotated, SIIP could note where Rosenhan’s claims are supported by documentation, where they are contradicted, and where their veracity is unknown. It could introduce the narrative of Harry Lando, whose psychiatric hospitalization was so healing. Updated periodically, it could advance the cathartic and prophetic fervor of Rosenhan’s paper, while openly admitting its grave flaws.
It could therefore stand as a monument to the harms and benefits of well-intentioned bad science—both an inspiration and warning to future generations.