Encouraged by bacteriological discoveries, leaders of early-20th-century medicine hoped to reduce the causes of all disease to microbes. Clinicians observed that certain strains of bacteria appeared to infect particular targets, such as tooth sockets, sinuses, tonsils, the lungs, and the colon, resulting in specific diseases such as bacterial inflammation of the lining of the heart (endocarditis), kidney disease (nephritis), appendicitis, and rheumatoid arthritis. These observations were made credible in 1914 by Mayo Clinic physician E.C. Rosenow, who reported that he had isolated streptococcal bacteria from patients with some of these diseases and that the bacteria, when “injected intravenously in animals, tended to provoke inflammation in the same organs that had been affected in the patients from whom they had been obtained.” 

Rosenow’s discoveries lent credence to the claims of Frank Billings, professor of medicine at the University of Chicago and Rush medical schools, who had asserted in a 1912 article in the Archives of Internal Medicine that “focal infections” were responsible for conditions as varied as arthritis and nephritis. Billings, who served as president of both the American Medical Association and the Association of American Physicians, became the chief exponent of the focal infection theory. Relying on Rosenow’s data, Billings presented a series of lectures at Stanford University Medical School in 1915 in which he asserted that removal of the infected organ site eliminated the disease. Given Billings’s credibility and stature, many American (and not a few European) physicians adopted his view, performing a variety of minor and major surgical procedures—from tooth extractions to tonsillectomies to hysterectomies—to counter infections. Not until the mid-1930s was focal infection finally discredited, when researchers proved unable to repeat Rosenow’s laboratory experiments, and the introduction of antibiotics made these surgical procedures irrelevant even to their remaining advocates. 


Soon after Billings promulgated his theory, some psychiatrists began to make a connection between focal infection and intractable psychiatric conditions. One of the most influential of these psychiatrists, Henry Cotton, medical director of the New Jersey State Hospital in Trenton, is the subject of Andrew Scull’s new book, Madhouse: A Tragic Tale of Megalomania and Modern Medicine. Cotton argued that infected sites produced bacterial toxins that migrated to the brain, causing an array of mental disorders. Beginning in 1918, Cotton performed surgical procedures, ranging from removal of previously filled teeth, tonsils, and sinuses to major interventions such as colon resections and hysterectomies, on mentally ill patients. In one year alone (1919 to 1920), Cotton performed 200 such procedures; by 1922, he had operated on 1,720 mental patients, claiming improvement in 77 percent. 

The details are what enable Scull to place Cotton’s disastrous treatments in a much wider context, one that exposes how practitioners were seduced by their desires for scientific recognition, even when their own data and experience should have alerted them that they had placed their patients’ lives in mortal danger.

Elliot Valenstein’s widely read Great and Desperate Cures (1986) outlined Cotton’s psychiatric experiments, and Scull has now filled in all the missing details. Some readers may find these details overwhelming, and Scull’s discussions of the personal lives of his principal actors are at times distracting, but his goal includes making the complete record available. The details are what enable Scull to place Cotton’s disastrous treatments in a much wider context, one that exposes how practitioners were seduced by their desire for scientific recognition, even when their own data and experience should have alerted them that they had placed their patients’ lives in mortal danger. 

Despite overwhelming evidence that his interventions were killing many of his patients, and maiming many more, Cotton persisted, encouraged by like-minded psychiatrists in North America and Britain. Even when the slaughter became obvious to them, Cotton was protected by others. Most influential among these protectors was the distinguished Johns Hopkins psychiatrist, Adolf Meyer, under whom Cotton had served early in his career at the prestigious Worcester State Hospital. In 1906, at Meyer’s encouragement, Cotton traveled to Munich, where he spent a year learning the latest in brain pathology and histology from Emil Kraepelin, Franz Nissl, and Alois Alzheimer. For the rest of Cotton’s career, Meyer would serve as mentor and protector. 


Cotton’s interventions were not uniformly praised. During the early 1920s, Cotton’s claims were interrogated and disputed at national psychiatric meetings by reputable psychiatrists, including Billings, who wrote that he was “unable to fully evaluate” Cotton’s work, warning of danger that focal infection could be “utilized with poor judgment and discretion by individual medical practitioners.” At the 1922 meeting of the American Psychiatric Association in Québec City, Canada, psychiatrists George Kirby (who had studied under Meyer) and Clarence Cheney, along with bacteriologist Nicholas Kopeloff, delivered a stinging indictment of Cotton’s surgical interventions. Based on their clinical research, they could find “no evidence on which to base a conclusion that removal of focal infection has itself brought about recovery” of any of Cotton’s patients. Moreover, Cotton’s bacteriological work was suspect. There was no “evidence to indicate anything more than the most gross relationship between particular species of bacteria and foci which have been considered and little to suggest that the bacteria found are in any way causally related to psychoses.” 

The criticisms of Cotton’s claims reached the Trenton State Hospital’s governing board, and one of its two physicians, Joseph Raycroft of Princeton University, asked Meyer to conduct an independent evaluation of Cotton’s procedures. In the fall of 1924, Meyer selected Phyllis Greenacre, a brilliant young psychiatrist at Hopkins, who later would become a leading American psychoanalyst, to undertake the investigation. Greenacre combed the Trenton State Hospital records and found that “there were elementary mistakes in the very first tables. As she looked more closely, she found several instances in which multiple admissions of the same patient were counted as though they were new cases each time...her suspicions were further aroused when she saw that virtually all the discharges were classified as cures.” 

The records and record keeping, according to Scull, were “so sloppy and unprofessional” that Greenacre “was appalled.” Greenacre’s interviews and detailed examinations of case reports confirmed her darkest suspicions. Typical of these cases was a young woman who had been admitted to Cotton’s care in 1922 with a diagnosis of recurrent manic depression. Within two days of admission the woman’s tonsils were removed, and nine days later “she underwent ‘the usual total colectomy,’ following which ‘she ran a fever of 100-102 for a month or six weeks, without the cause being discovered.’ ” She also had 16 teeth extracted and “was injected with vaccines derived from ‘streptococcus mitis and stock colon streptococcus.’ ” Discharged to her relatives when she “showed some more interest in her surroundings,” the woman, still depressed, was readmitted eight months later. When staff members reported that the patient was concealing stomach pains, Cotton ordered a laparotomy and colostomy. The woman died eight days later from post-operative peritonitis. 

After reading Greenacre’s report, Meyer wrote that her “investigation” of “Dr. Cotton discloses a rather sad harvest. His claims and statistics are preposterously out of accord with the facts.” Raycroft and the Trenton State Board received copies of Greenacre’s report, but, writes Scull, “remarkable as it seemed, they obviously had not fully grasped their import.” Meanwhile, in 1925, a New Jersey legislative committee began a series of hearings on financial waste in state government hospitals. In the midst of that investigation, former patients, their families, and disaffected hospital staff members at Trenton State Hospital testified about “patients being beaten, kicked, and dragged screaming into the operating room, of trolleys filled with body parts and not a few corpses streaming in the opposite direction.” As the stories multiplied, Cotton, now on the defensive, became unhinged and soon was confined to bed. 

Although the legislative investigating committee did learn of Greenacre’s investigation, Meyer kept them from viewing the actual report. Greenacre ultimately was called to testify but was asked only whether she had personally witnessed “cruel treatment of patients” and, in deference to Meyer’s instructions, volunteered nothing else. The investigation ended without condemnation of Cotton’s treatment of patients. Instead, the medical director was commended for his cost-cutting measures, and his accusers were dismissed as “disgruntled former employees and insane witnesses.” 

As a result of the suppression of Greenacre’s report, Cotton was able to continue to remove his patients’ teeth, resection their colons, and perform hysterectomies and other unneeded surgeries for another half decade until his not entirely voluntary retirement from Trenton State Hospital in 1930. 

Having regained his sanity, Cotton was free to return to “curing” his patients. As a result of the suppression of Greenacre’s report, Cotton was able to continue to remove his patients’ teeth, resection their colons, and perform hysterectomies and other unneeded surgeries for another half decade until his not entirely voluntary retirement from Trenton State Hospital in 1930. Cotton continued to maintain his private practice until his death three years later. 

As Scull points out, “Meyer did more than cover up the parade of death and debility” that was documented in Greenacre’s devastating report. He also suppressed a follow-up and equally damning investigation by Solomon Katzenelbogen, who joined Phipps Psychiatric Clinic at Johns Hopkins in 1928 and was selected to head its Internal Medicine Laboratory. Like Greenacre before him, Katzenelbogen was astounded by “the very unsatisfactory examination of the mental status” of Cotton’s patients and called into question “the statistics referring to the recovery rates of certain psychotic types.” Moreover, Katzenelbogen found that the pathological testing “was slipshod and badly performed” and that the medical staff was not “sufficiently trained in pathophysiology.” According to Scull, Katzenelbogen “observed that when even careless work did not suffice to demonstrate pathology, Cotton and his staff simply manipulated the accepted boundaries of what constituted an abnormal reading.”

Cotton died of a heart attack in 1933. Scull writes: 

Knowing that what had resulted in Trenton was a piling up of the edentulous, the eviscerated, and the extinguished, Meyer chose the most prominent of professional platforms, The American Journal of Psychiatry, to praise the whole enterprise as “a most remarkable achievement of the pioneer spirit,” to laud Cotton as “one of the most stimulating figures of our generation,” to call for his work “to be carried on” through “prolonged observation and comparison”; and to lament that this therapeutic approach “now will have to be carried on without the leading and active spirit of the sincere and convinced protagonist.” 


Scull insists that Cotton’s story is no anomaly, writing that: 

...dismissing Cotton’s surgical assaults as an aberration is not the lesson we should draw from these events. On the contrary, the long-suppressed story demonstrates the extraordinary vulnerability of the mentally ill to victimization and the hollowness of professionals’ claims to police themselves. Morally, socially, and physically removed from the ranks of humankind, locked in institutions impervious to the gaze of outsiders, deprived of their status as moral actors, and presumed by virtue of their mental state to lack the capacity to make informed choices for themselves, patients were helpless to resist the interventions of those who controlled their very existence. 

Scull is correct in asserting that interventions practiced at Trenton State were not unique. I found that, in the 1920s and early 1930s, children at New York Hospital-Cornell Medical Center with a variety of diagnosed movement disorders, including convulsive tics, habit spasms, and chorea (many of whom today would be diagnosed with Tourette syndrome), routinely received tonsillectomies and often removal of their sinuses. Even though these surgeries failed to improve patients’ tics and vocalizations and even though children died of complications from the operations, physicians at the time considered these surgeries the first line of intervention. These procedures were not confined to the Northeast. On the West Coast, University of Oregon Medical School psychiatrist Laurence Selling claimed in the late 1920s that he had alleviated tics by removal of some or all of his patients’ tonsils and infected sinuses. The procedure moved in stages. Thus, when a 14-year-old boy failed to improve after removal of his tonsils and adenoids, Selling removed his antrum sinuses. When this also failed to halt the tics, Selling removed both ethmoid sinuses and claimed a cure. Selling never published follow-up studies, and the duration of “remissions” of Selling’s patients is unknown. 

Why did so many psychiatrists not only fail to see the flaws in Cotton’s claims but also, like Selling, to conduct similar interventions? Clearly, unlike Cotton, they were not all monomaniacs. The answer, Scull believes, lies in their uncritical attraction to reductionist science. Moreover, the atmosphere that enabled Cotton and the others to continue their interventions, despite the overwhelming evidence available that the results were more debilitating than the putative disease they were intended to cure, argues Scull, remains much the same in current psychiatry’s attempt to reduce mental illness to its biological substrates: 

Central, after all, to the medical identity of psychiatry for most of its history has been its metaphysical embracing of the body. From the humoral accounts of the origins of mania and melancholy that underpinned the medical claims to comprehend and treat mental disorder in the eighteenth century, to the declaration by the paymaster of American academic psychiatry, the National Institute of Mental Health, that the 1990s marked “the decade of the brain,” alienists have repeatedly tried to account for mental disorder in somatic terms. 

This “attribution of madness to disorders of the body,” according to Scull, “has, unsurprisingly, led to a recurrent fascination with somatic treatments for presumed underlying pathology.” 


Scull makes a persuasive case, but his antipathy to organic explanations of mental illness may restrict our understanding of why so many psychiatrists, even those who openly rejected Cotton’s claims, seemed willing to give his interventions more credibility than they deserved. The way Scull tells it, one would have to be a monomaniac, like Cotton, or a blind conspirator to give any credence to infectious causes of mental illnesses. Starting with this premise, it is difficult for Scull to explain why so many seemingly sane neurologists and psychiatrists continued to believe that a connection existed between infection and mental disorders, unless, as he seems to imply, the entire psychiatric enterprise was and continues to be built on lies, self-deception, greed, and a deep-seated need to dominate nonconformists. That psychiatrists and physicians are often arrogant cannot be disputed, but, unfortunately, this is not a trait limited to any one profession. Was there, after all, something compelling about the connection between infection and the behaviors labeled as mental illness, which Cotton took to its illogical conclusion? 

Early on, Scull remarks that fever therapies for neurosyphilis, a serious infection of the brain caused by syphilis, provided the soil for focal infection interventions on psychiatric patients. In 1904, Nissl and Alzheimer established the infectious origin of neurosyphilis, and, two years later, Wasserman developed a reliable serological test that identified active infection. But the problem of an effective intervention against the infection remained. Around this time, the Viennese psychiatrist Julius Wagner-Jauregg and his colleagues discovered that syphilitic patients who contracted febrile diseases during the early years of their syphilitic infection rarely developed neurosyphilis. By the late teens, Wagner-Jauregg began injecting syphilitic patients with a tertian type of malaria to cause a fever, and, by 1921, he reported that 21 percent of these patients were able to resume a normal life, cured of syphilis. Later in the decade, he reported a cure rate of close to 30 percent. In 1927, Wagner-Jauregg was awarded the Nobel Prize in medicine “for his discovery of the therapeutic value of malaria inoculation in the treatment of dementia paralytica.” 

Scull cites Wagner-Jauregg’s malarial therapy as emblematic of the absurdity of biological psychiatric theory and practice and as evidence of the milieu from which Cotton emerged: 

Julius Wagner-von Jauregg, Freud’s contemporary and a notorious anti-Semite who later would become an enthusiastic Nazi, had long argued that fevers might bring about a remission of psychiatric disorders. From the late 1880s onwards... he had experimented with a variety of febrile agents, searching for a reliable means of inducing the physiological response he sought. 

Scull writes that, unable to produce the results he desired, Wagner-Jauregg was undeterred and: 

...pressed forward, increasingly focusing his efforts on the paretic...In the last months of the war [World War I], von Jauregg was presented with a shell-shocked patient from the Italian front who was simultaneously suffering tertian malaria. Finally he had found a reliable source for producing sustained fevers of up to 106 degrees Fahrenheit... Immediately, he made use of the source to transfer the malaria to first one and then to a whole succession of paretics, launching a therapy that he claimed produced near-miraculous results in two-thirds of his patients, and one that, after the war, spread rapidly across Europe and North America. 

Scull’s implication, along with his selection of adjectives, suggests that this was an insane and dangerous intervention. But it worked. 

Historian and psychiatrist Edward Brown wrote (in response to an earlier article by Scull making a similar argument) that at the turn of the century syphilis and its resultant general paresis (paralysis) were as menacing and widespread as HIV/AIDS is today. Its victims experienced a slow and horrible death through final stages of dementia and insanity. By inducing leukocyte production, malarial therapy called on the body’s immune defenses to defeat the spirochetes, the bacteria that caused syphilis. The intervention, and the theory that drove it, was based on a robust understanding of both how bacterial infection could result in insanity and how that infection could be defeated. Scull implies that Wagner-Jauregg was inadvertently passing syphilis from patient to patient by using blood from infected syphilitics to get febrile responses in other patients, although the evidence for this claim seems weak. As Brown reminds us, “malarial treatment continued to be used into the early 1950s.” The most authoritative textbook, the 1946 Neurosyphilis by H. Houston Merrit, Raymond Adams, and Harry C. Solomon, insisted that malarial therapy was “the simplest and most effective method of treatment of paretic neurosyphilis.” Brown admits that “there is no question that it was a desperate treatment. Even so, there was reason to be proud of it. After a hundred years of hopelessness and despair, it offered hope for people afflicted with a devastating disease.” The subsequent discovery and refinement of penicillin was, of course, a much more effective weapon against neurosyphilis, but the biological reductionists were correct to try to understand the insanity of tertiary syphilis was a sequel to an infection. Indeed, this connection had long occupied Meyer and was the subject of Greenacre’s early research at Hopkins. 

Scull rejects all claims of an organic cause of mental illness in favor of a psychogenic one. The tale of Cotton’s outrageous interventions is used to imply that any claim of an infection as the source of a psychiatric disorder is, ipso facto, ridiculous. But the example of neurosyphilis is not unique. Bacterial antibodies had long been suspected and are now understood as the cause of Sydenham’s chorea, and a similar mechanism is probably part of the cascade of many obsessive-compulsive disorders and serves as the environmental trigger for some people with Tourette syndrome. Disorders such as multiple sclerosis, which were, until the 1930s, often diagnosed and treated as psychogenic now appear to be virus mediated, and it would be malpractice to treat patients as if the cause and progression of their condition were mental or emotional, rather than physical. Other disorders, which indisputably are organic, but which in the past were seen as psychogenic and treated by psychiatry, include Huntington’s disease, amyotrophic lateral sclerosis (Lou Gehrig’s disease), and Tourette syndrome. The evidence that most schizophrenias have an organic substrate seems difficult to ignore. 

At one point, Scull derides as absurd Cotton’s belief that infection could be spread by grandparents kissing babies. Actually, in this case, Cotton was probably correct, at least about bacterial transmission. Research suggests that the gum disease, gingivitis, is most often and increasingly passed to infants from infected adults kissing babies. Infected teeth also can be part of a cascade for acquired heart disease. Having made this connection does not, however, authorize a prohibition of baby kissing or removal of the teeth of grandparents, but rather a more systematic investigation of the extent and risk of possible transmission. 

Likewise, Scull has exposed the mindless translation of infectious and organic hypotheses from theory and research to medical intervention. What separates malarial fever interventions from Cotton’s promiscuous interventions is the doctrine of lesser harms, elucidated wonderfully in Lesser Harms: The Morality of Risk in Medical Research, a recent book by another medical sociologist, Sydney Halpern. From that perspective, malarial fever therapy for neurosyphilitics was justifiable because, lacking any other effective intervention, the procedure could save some, if not all, surely doomed lives. Once penicillin was available, fever therapies would be unethical, because they would subject syphilitic patients to a greater harm. The interventions that Cotton performed could never meet the test of lesser harms. The problem is how to know the difference and what standards to apply. How we define a “cure” is complicated by what is classified as illness and who is authorized to make that classification. Scull is right to challenge researchers and physicians to examine these questions; but, it would be wrong to conclude that, because of classification uncertainty, there is no neuropsychiatric illness or that the causes of psychic distress can never be infective or organic. 

Unfortunately, much of what authorized Cotton and misled his defenders continues to inform psychiatric practice. The gap between the tentative nature of current neuroscientific and molecular research and its too eager translation into therapeutics continues; it may have gotten worse. The promiscuous re-prescription (by both general practitioners and psychiatrists) of mood-altering and stimulant pharmaceuticals is epidemic and probably dangerous. Given the influence of the pharmaceutical industry in clinical research, it is increasingly difficult for skeptical voices to be heard. Nevertheless, it would be sad if the only response left to critics would be Luddite—if the only choice the distressed had was between over-eager and partially-informed interventionists and nihilistic critics. What is labeled as mental illness represents real distress. Most often that distress reflects an interaction of organic substrates with existential conflict. Identifying this interaction requires more than a theory; treating it should demand more than power. What is essential is a deeper and continually skeptical investigation of scientific claims, not a counter theory informed by only past failures and excesses, but rather a robust interdisciplinary approach that recognizes past mistakes and teaches us appropriate lessons.  


From Madhouse: A Tragic Tale of Megalomania and Modern Medicine by Andrew Scull. © 2005 by Andrew Scull. Reprinted with permission of Yale University Press.

Henry Cotton’s experiments on his Trenton patients were not an isolated and transient phenomenon. The notion of focal sepsis upon which he seized to explain psychosis and promote his campaign for surgical bacteriology was embraced by some of the best medical minds of his era, and its significance and practical application extended far beyond the marginal realm of psychiatry. In general medicine over several decades, millions of tonsils were sacrificed on this particular altar, and major figures in physic and surgery accepted the basic idea that focal sepsis could cause chronic disease as plausible and promising. 

And what many found plausible was not just this general idea, but its specific applications to psychiatry. On at least two occasions, as we have seen, the best and the brightest— or rather, the most prestigious and most prominent, at least in British medicine—flocked to sing Henry Cotton’s praises and embrace his theories and therapeutics as a signal contribution to twentieth-century medicine. They did so at length and in very public settings, and with scarcely a whisper of dissent, even when the object of their hosannas acknowledged that his most drastic interventions killed almost a third of those he treated. Visitors from three continents descended on Trenton. All pronounced themselves profoundly impressed by what they had seen. In Britain, T.C. Graves, who had independently happened upon and embraced the focal sepsis hypothesis, actively and aggressively treated all the mental patients hospitalized in the Birmingham area along Cotton’s lines. He presided over all the mental hospitals in England’s second city till well after the end of the Second World War, and continued assaulting sepsis during the four years he occupied the post of president of the Royal Medico-Psychological Association in the 1940s (the longest period anyone has ever held this position). 

And Cotton was by no means a prophet without honor in the United States. A substantial number of his fellow asylum superintendents visited Trenton in the twenties and lamented that the parsimony of their local state legislators kept them, too, from embarking on so ambitious, so laudable, so desirable an antiseptic program. Respectable figures like Hubert Work, the president of the American Medical Association, and Stewart Paton, author of perhaps the most influential American textbook of psychiatry in the early twentieth century, blessed Cotton’s endeavors, as did medical men whose expertise was of more dubious provenance but whose public influence was large: the nation’s most famous medical agony aunt, Senator Royal Copeland, and the entrepreneur who had built the country’s most lucrative funny farm for the fashionably nervous and the prostrated neurasthenic, John Harvey Kellogg of the Battle Creek Sanitarium. Cotton’s assistant physicians at Trenton included at least two Hopkins-trained psychiatrists who went on to very prominent careers in North American psychiatry—Clarence Farrar, editor of the American Journal of Psychiatry from 1931 to 1965, and Franklin Ebaugh, director of the Division of Psychiatric Education at the National Committee for Mental Hygiene and chair of the Department of Psychiatry at the University of Colorado. Neither of them is known to have objected to what went on, either then or later. Indeed, during his time at the Trenton hospital, Ebaugh gave every sign of being a willing and enthusiastic participant in the search for sepsis. Right down to Cotton’s premature demise, for a period of more than a decade and a half, the moneyed classes flocked to the doubtful charms of New Jersey’s capital city, bringing loved ones whose teeth, tonsils, and colons were willingly sacrificed in a frantic search for sanity—a practical (and profitable) endorsement of Cotton’s theoretical claims.

About Cerebrum

Bill Glovin, editor
Carolyn Asbury, Ph.D., consultant

Scientific Advisory Board
Joseph T. Coyle, M.D., Harvard Medical School
Kay Redfield Jamison, Ph.D., The Johns Hopkins University School of Medicine
Pierre J. Magistretti, M.D., Ph.D., University of Lausanne Medical School and Hospital
Helen Mayberg, M.D., Icahn School of Medicine at Mount Sinai 
Bruce S. McEwen, Ph.D., The Rockefeller University
Donald Price, M.D., The Johns Hopkins University School of Medicine
Charles Zorumski, M.D., Washington University School of Medicine

Do you have a comment or question about something you've read in CerebrumContact Cerebrum Now.