The Rise and Fall of American Psychosomatic Medicine
Theodore M. Brown
New York Academy of Medicine
November 29, 2000
Because, I’ve discovered, my subject is something like a Rorschach, which stimulates people to project their own associations onto it, I think I need to begin with a little preface. So let me say at the outset what this paper is and what it is not. It is easier to say first what it is not. It is not an examination of philosophical ideas – dualism/holism/psychogenesis/somatogenesis – or of cultural attitudes and assumptions. Nor is it a discussion of physicians’ evaluations of their patients’ symptoms or of patients’ reactions to their physicians’ labeling behavior. Nor, for that matter, is it a discussion of a whole host of other interesting and important issues, which, at most, will enter my discussion passingly at various points. What this paper is, is a look at a particular field within American medicine and an attempt to understand how that field fit into the larger American medical context. It is about the dynamics of the rise and fall of psychosomatic medicine as both a part of and a reaction to the frequently changing shape of twentieth century American medicine.
Let me begin, then, by framing the rise and fall of American psychosomatic medicine as a field with two quotations, the first from March, 1988 at the forty-fifth Annual Meeting of the American Psychosomatic Society. On that occasion, Dr. Stanford Friedman, Chief of the Division of Adolescent Medicine at North Shore Hospital in Long Island, delivered a surprising and decidedly uncelebratory presidential address. Published a few months later in the fiftieth volume of Psychosomatic Medicine, the leading general American journal in the field, Dr. Friedman’s address included the following jarring remarks:
... When I joined the society in 1962, the society’s identity was clear. The organization represented the merger of mind and body, of behavior and disease, and of psychiatry and medicine. In particular, there was a scientific focus on the interactions of biologic and psychosocial factors in the etiology of disease. George Engel had developed the “multi-factoral” model ... [Others] were showing us the limitless scientific questions that could be pursued by a psychoendocrine orientation. ... Joining the society was an unambiguous commitment to an exciting, and scientific, approach to health and disease. [But, by contrast, I am struck by] the increased ambiguity of the present. To a great extent, the very term psychosomatic has lost meaning. ... [O]ur members, particularly our new members, do indeed now find it difficult, if not impossible, to view themselves as members of a clearly defined “group.” Compounding our identification problem is that, associated with our increasing scientific sophistication, has come subspecialization within what might have previously been seen as the exclusive domain of psychosomatic medicine. ... This society, in my opinion, is now especially vulnerable to “marginality.” ... [I]n an organization committed to an increasingly nebulous and ambiguous concept called “psychosomatic medicine,” who are we? ... [W]e are essentially invisible as a society to the general public and also to most of the medical community ... [W]e exist in partial vacuum and in isolation from the vast majority of our colleagues in medicine, psychiatry, and psychology ... [T]he average age of our membership is 50 years, or, put another way, about half our members are eligible to join the American Association of Retired Persons.
The full significance of Friedman’s painful assessment can best be appreciated by contrasting what he said in 1988 to what the editors of the journal had written in an upbeat “Introductory Statement” when Psychosomatic Medicine first launched into print about half a century before, in 1939:
... [The object of Psychosomatic Medicine] is to study in their interrelation the psychological and physiological aspects of all normal and abnormal bodily functions and thus to integrate somatic therapy and psychotherapy. Psychosomatic Medicine is not restricted to any specific field of pathology ... [I]t designates a method of approach to the problems of etiology and therapy rather than a delimitation of the area. ... The medical profession has awakened to the necessity of studying systematically ... the therapeutic utilization of the psychic component in the disease process, and in the emotional relationship between physician and patient. This intensive interest and research in medical psychology is a symptom of a new orientation toward the problem of disease, in fact, the sign of the beginning of a new era in etiological and therapeutic thought ... [P]sychosomatic medicine is both a special field and an integral part of every medical specialty. It is an essential approach to every medical specialty in so far as it is fundamental in all diagnosis and treatment.
In considering these two framing quotations, it is important to note that neither the editorial board in 1939 nor Friedman in 1988 were lone or maverick voices. On the contrary, there is considerable reason to believe that each statement spoke authentically for a particular generation of American psychosomaticists. That being the case, we have evidence here for a dramatic change within one of the signature fields of American medicine, and the crucial question for the medical historian is, then: Why, why did American psychosomatic medicine in the course of fifty years pass from great, optimistic hope to dispirited and depressing acknowledgment of its own marginality? How did it change from a field with a bold scientific as well as proselytizing mission into one, seemingly, with no clear mission at all? These are the questions I hope to answer tonight.
To begin the process of seeking answers, we have to consider what the circumstances were in the 1930s that gave rise to the field in the first place. Let’s begin by considering psychiatry, which in the years of psychosomatic medicine’s birth was pushing hard for recognition in America’s medical mainstream.
American psychiatrists had actually begun their campaign for increased professional status in the early years of the twentieth century. Until that time, they had largely been isolated from the rest of medicine, functioning primarily as “alienists” in psychiatric asylums and state hospitals. They pushed to be included on medical faculties and the staffs of general hospitals, urging more teaching of psychiatry to medical students and greater attention to the patient as an integrated “psychobiological” individual. Adolf Meyer was one of the great successes, who got a new position at Johns Hopkins and used it to promote the cause of psychiatry nationwide by calling for “subordination of the medically useless contrast of mental and physical.”
Slide 1 – Adolf Meyer portrait
Despite Meyer’s bracing rhetoric, the movement of psychiatry into the medical mainstream remained, in fact, an uphill battle well into the twenties, and beyond. Successes in treating “cardiac neuroses”and other suspicious conditions during World War I and in the postwar period increased psychiatry’s legitimacy, and in June, 1924 Macfie Campbell, a former trainee and junior associate of Meyer, now Professor of Psychiatry at Harvard, gave the prestigious Shattuck Lecture at the Massachusetts Medical Society. Campbell used the occasion to promote a wider role for psychiatry in the practice of internal medicine.
Slide 2 – Campbell’s 1924 Lecture
He argued that, despite often naive and myopic assessments that considered disease processes “localized” in particular organs or systems, “we may have to remember that organs and systems are not isolated units but somehow or other contain within themselves the influence of the organism as a whole.” One of the largest set of influences of the “organism as a whole” derive from the individual’s failures to establish “harmony with the environment” in personal, family, or work situations, and these failures lead to somatic consequences. In very much the same spirit, in 1926 nationally prominent psychiatrist William Alanson White argued that since “mind and body ... [are but] two aspects of the organism, like two faces of a crystal ... every disease has a psychic component, and further ... this component has a history as long and as important ... as has the somatic component.”
In the later twenties and in the thirties one of the major vehicles for the advancement of psychiatry was the National Committee for Mental Hygiene, whose Division of Psychiatric Education was directed by Dr. Franklin Ebaugh of Colorado, another one of Meyer’s former trainees. Ebaugh promoted psychiatry’s ambitious agenda in a 1932 article in JAMA.
Slide 3 – Ebaugh’s JAMA paper
Ebaugh wanted psychiatry to “soak through” the entire medical school curriculum. In this way, the general physician would become more aware of precipitating life factors that are important in the functioning of all patients. They would become interested in knowing the total life situation surrounding the sick individual who, after all, “has a mind as well as a body ... [which] is just as important a part of the clinical study of a patient as the understanding of physical problems, and the two cannot be separated.”
These developments in psychiatry help explain trends that were obvious in American Internal Medicine in the same period, the most important of which was an interest in psychogenic connections that was evident already in the 1890s through the 1910s but that took a sharp upward, and possibly quite reactive, turn in the 1920s. Of course, relationships between mind and body, emotions and their physical expression, affective states and somatic disorders had been observed and discussed by physicians since antiquity, in a prominent way at least since Galen in the second century A.D. Galen was celebrated for his differential diagnoses, especially for those which distinguished between illnesses traceable to somatic causes and those which seemed to mimic them but were actually traceable to emotional causes instead. In one famous case he treated a young woman who seemed to exhibit the signs of physical illness but who, upon closer examination, revealed no organic pathology. After eliminating any possible humoral explanation, Galen identified the real, emotional, cause of her somatic symptoms.
Galen likewise contributed an important new interest in the balance not only of the humors but of what he called the “non-naturals,” among which he included the “passions or perturbations of the soul.” According to this doctrine, it was important for physicians to help patients keep their emotions in balance, for the sake of their bodies as well for their mental equanimity. The influence of strong emotions on physical health and illness thus became a central tenet of medical belief and grew progressively stronger in the medieval period. Moses Maimonides expressed the point this way in the twelfth century: “It is known ... that passions of the psyche produce changes in the body that are great, evident and manifest to all. On this account ... the movements of the psyche ... should be kept in balance ... and no other regimen should be given precedence.”
Ideas about the “balance of the passions” were popular in the Renaissance and early modern periods. One famous seventeenth century work showing how influential these ideas had become was Robert Burton’s The Anatomy of Melancholy which included the following observations about the possibly disastrous role of unchecked emotions: “the mind most effectually works upon the body, producing by his passions and perturbations miraculous alterations ... cruel diseases and sometimes death itself.” These ideas also showed up in medical textbooks by Friedrich Hoffmann and Herman Boerhave in the late seventeenth and early eighteenth century, respectively, and in treatises on mind and body by such leading medical figures of the mid-eighteenth century as Jerome Gaub and Robert Whytt. William Falconer weighed in with his Dissertation on the Influence of the Passions Upon the Disorders of the Body in 1788.
“Psychosomatic” relationships were thus commonplaces for a very long time and were part of the very fabric of traditional medicine. But they were significantly displaced in the nineteenth century when the rise of, first, pathoanatomy and, then, cellular pathology, bacteriology and biochemistry partially eclipsed traditional theory. And yet, as late as the 1880s and despite strong somatizing and localizing tendencies, we can still find substantial evidence for psychogenic sensitivity within the central core of mainstream medical practice and belief.
Consider, for example, Austin Flint’s Principles and Practice of Medicine, 5th edition, 1881.
Slide 4 – Flint title page
Flint was a major figure in American medicine from the 1850s to the 1880s, a leader of American pathoanatomy and early bacteriology, and a president of the AMA. He was also the author of the most important nineteenth century American medical textbook before Osler’s. What’s particularly notable about this textbook from my point of view is that, despite its heavy emphasis on pathoanatomy, physiology, biochemistry and bacteriology, Flint left plenty of room for psychogenic concerns, which he mainly included in his discussion of the “neuroses,” i.e. functional disturbances of the nervous system which were also extensively explored in the nineteenth century. Physicians like Flint turned to the nervous system to find a physiological connection between emotions and disease because in the nervous system they could work out a compromise of sorts between traditional ideas linking emotions and disease and the new, nineteenth century desire to extend the reach of localistic pathology. Since the nervous system was enormously complex and its workings subtle and elusive, it could be the locus of “functional” disorders, which were characterized by disrupted activity but where no inflammation or “appreciable morbid change in the nervous structure” could be found. In Austin Flint’s popular Treatise on the Principles and Practice of Medicine, functional disorders of the nervous system and the emotional causes that precipitated them became a major topic of clinical study.
But what happened after the 1880s? As I see it, the leading edge of American academic medicine – what soon became known as “Internal Medicine” split into two lines of development. One can be called the “Osler line” and the other the “Barker line.” The “Osler line” is the one almost everyone knows about. Led by William Osler, a group of elite physicians followed the German example and defined a non-surgical field of “innere” practice. In the late nineteenth and early twentieth century, the field gained momentum as increasing numbers of talented young men became full-time clinical investigators and developed careers as institutionally-based practitioners. In 1908 the research-oriented Archives of Internal Medicine initiated publication, as an important marker event. In the teens and twenties the further growth of clinical investigation led to such important breakthroughs as the elucidation of the hormones and their role in the body, the development of insulin therapy for diabetes, and the discovery of liver therapy for pernicious anemia.
The main point is that the “Osler line” moved in a reductionist, biological direction. Medicine was an experimental science and the human organism – as Abraham Flexner put it in his famous Flexner Report of 1910 – was not essentially different from a frog.
Slide 5 – Osler looking down on patient
But there was also an often forgotten “Barker line,” led by Lewellys F. Barker, who was Osler’s successor as Professor of Medicine at Johns Hopkins. Unlike Osler, Barker was deeply interested in mental and nervous diseases, in psychotherapy for certain ostensibly somatic conditions, and, later, even in psychoanalysis.
Slide 6 – Lewellys Barker portrait
When Barker took over clinical and teaching responsibilities from Osler, he found one major area of deficiency. This is how he remembered the situation in his autobiography:
Admirable as Dr. Osler’s organization and conduct of the clinic had been, there was one field that had been inadequately cultivated, namely that of the functional nervous disorders ... When patients manifested symptoms of these disorders they were often rather lightly referred to as being “neurotics” and received too little attention from either staff or students. ... Dr. Osler had been trained in the pathological-anatomical ... school, and he was more particularly interested in organic rather than in functional disorders of the nervous system. ... It soon dawned on me that there was here at least one opportunity for improving the work of the clinic ... and I began an intensive study of the diagnosis and treatment of these functional nervous disorders.
Thus the “Barker line” in the history of late nineteenth and early twentieth century Internal Medicine. Its existence explains the popularity of various “psychotherapies” in the early years of this century. It also explains the enthusiasm with which some American, Barker-style clinicians turned to Sigmund Freud and his new psychoanalytic psychotherapy, even before World War I. (More on psychoanalysis in a moment.) During World War I, as you might expect, interest in Freud’s theories and methods grew dramatically.
To take one suggestive example, consider a paper published in 1919 by Dr. Alfred Cohn. Cohn was director of the Hospital of the Rockefeller Institute for Medical Research in New York City.
Slide 7 – Cohn paper
Cohn concludes his paper as follows:
In the Expeditionary Forces [during World War I] we were led, after reflection, to adopt the view that the Irritable Heart of Soldiers was the expression of a neurosis ... no matter what the predisposing cause, whether it be infectious disease, malfunctioning glands of internal secretion or gas-poisoning, the disorder ... depend[s] on anxiety and fear ... it is removed by the disappearance of the exciting cause and ... it is cured by measures designed to influence the neurotic state.
Dr. Cohn adds an important trailer:
For civil practice this experience has a lesson. There are many unfortunates who fall into this group of patients. It will not be seriously contended that the practice of medicine has adequately studied or solved the problems connected with this phase of disease.
What’s most notable about Cohn’s paper is that it is merely one of many with a similar orientation published in the medical literature of the period. It would thus appear that by 1919 many American physicians had joined the “Barker line” in Internal Medicine, that is, they had become interested in the role of emotion in the etiology and treatment of physical illness. The intensifying buzz from psychiatrists, plus experience with soldiers during World War I and with veterans in the immediate postwar period, served to underscore the role of emotional factors. The full extent of psychogenic disorder, already suspected by some before the war, was brought vividly home, and its occurrence was generalized to the entire civilian population.
What happened after 1919? The simple answer is that interest in the relationship between emotions and disease markedly increased in the twenties and thirties. The journal literature was crammed with articles such as these: “The Ophthalmologist and the Psychoneuroses” (JAMA, 1924); “Psychic and Emotional Factors in General Diagnosis and Treatment” (JAMA, 1927); “Psychogenic Fever” (New England Journal of Medicine, 1930); “The Psychogenic Origin of Organic Diseases” (New England Journal of Medicine, 1935). The journal literature also included strong advocacy statements, such as Charles Emerson’s repeated urging that his fellow Internitst – even those in the august Association of American Physicians – join him in venturing into the “Emotional Zone of Disease.”
Slide 8 – 1927 JAMA
The marked trend toward a more pronounced psychogenic orientation was evident even in the later editions of Osler’s Principles and Practice of Medicine. The tenth edition, published in 1925 and edited by Thomas McCrae, considerably widened the concept of hysteria – the prime “neurosis” – that Osler had somewhat reluctantly included in earlier editions. McCrae also pointedly added: “In periods of great stress, as in the recent war, it becomes a widespread and serious disorder.” In the thirteenth edition, published in 1935 and edited by Henry Christian, Christian interpolated this far-reaching remark:
Slide 9 – 1935/1938 edition of Osler’s textbook
It is to be remembered that with organic disease many, or possibly all, symptoms may be on a functional basis and that much benefit comes to the patient from treating the functional disturbances in addition to treating those derived from the organic lesion; always to treat the patient as a whole is a good rule to follow.
Other important signs of the times were the “Convocational Oration” at the 1936 meeting of the American College of Physicians by Walter Cannon, America’s leading physiologist, on “The Role of Emotion in Disease.” In one sense Cannon’s oration capped a trend toward the study of emotion by physiologists (about which Otniel Dror has recently written), and in another it marked the scientific legitimation of the growing interest in psychogenesis by non-psychiatric clinicians.
Slide 10 – 1936 Cannon paper
And there were related publications by leading, biomedically-oriented clinical investigators – Harvard’s brilliant Soma Weiss is a good example – that explored such topics as “The Interaction Between Emotional States and the Cardiovascular System in Health and Disease,” the title of one of Weiss’ papers in the thirties. It was almost as if Soma Weiss were considering adding a prefix and hyphenating his name to Psyche-Soma Weiss.
Related to the psychogenic trend in Internal Medicine was another, closely connected one in medical education. What I refer to here was a systematic reform campaign by leading medical educators aimed at putting humanism, psychological awareness, and psychogenic presumption back into the medical curriculum from which, they claimed, they had recently been extruded. The reform campaign began in the teens and gathered momentum in the twenties and thirties.
For example, soon after being appointed professor of medicine at Harvard Medical School in 1912, largely on the strength of his biomedical research reputation, David Edsall drew the Social Service Department of the Massachusetts General Hospital into the teaching of fourth-year medical students. When Edsall became Harvard dean a few years later, the “social conferences” became the teaching responsibility of several other leading Harvard faculty. In 1925, Francis W. Peabody, soon to be famous for his broadly humanistic and psychogenically sensitive “The Care of the Patient” (1927), reported on his version of this teaching innovation. Once students were exposed to the social aspects of a case, he argued, “it would be a dull student indeed who could question a patient for even five minutes regarding his financial status, the character of his dwelling, the social group with which he is associated ... without becoming profoundly impressed with the various ways in which all of these factors may bear on the cause, course, and cure of the disease.” Peabody was here echoing concerns already expressed by medical school deans Charles P. Emerson (Indiana) -- that’s the same Charles Emerson we’ve already encountered – and Irving S. Cutter (Nebraska) a few years earlier.
The most interesting, ambitious, and well advertised of all the American efforts to engineer the “patient as a whole” back into the curriculum was that undertaken in the thirties by Dr. George Canby Robinson at Johns Hopkins. Robinson was a respected and highly visible scientific investigator, editor, and medical school dean. He had also been the first editor of the Journal of Clinical Investigation from 1924 to 1930 and president of the Association of American Physicians in 1932-1933. In 1936, he launched a research project at Johns Hopkins Hospital to explore the social and psychological factors that were present but not adequately addressed by medical staff in their dealings with hospital patients. This research project soon grew into a curricular initiative to create a holistic presence in the heart of America’s most unrepentantly reductionist biomedical institution. Robinson’s research and educational work was followed closely, reported often, and promoted vigorously by those who were aware that he was the first major American internist to devote himself full-time to these educational matters. Robinson’s efforts led to a 423-page monograph published in 1939 as The Patient as a Person.
Slide 11 – Patient as a Person title page
Although Robinson gave no acknowledgment to it, another very important trend in American medicine in the first three decades of the twentieth century was the growing popularity of psychoanalysis. That popularity went back to the pre-World War I period and Freud’s visit to the United States in 1909.
Slide 12 – Freud’s Clark University Lectures (1909)
By 1913, psychoanalysis had already achieved a foothold and a new journal, The Psychoanalytic Review, was launched. Then with World War I the pace quickened, as I have already suggested. In the twenties psychoanalytic ideas and methods sharply increased in popularity in both avant-garde cultural and more conservative medical circles. Around 1930, the already increased popularity spiked further with the entry of emigre analysts, who fled Europe and the Nazis. By 1933, psychoanalysts were invited to form a special section of the American Psychiatric Association, a significant institutional recognition.
Without pursuing the theoretical nuances and variations, it will suffice to say that psychoanalysis had an important psychogenic component from the start. In his 1909 Clark lectures, for example, Freud claimed that the somatic symptoms of hysteria are the expression of the patient’s secret repressed wishes. It was the process of “conversion” which translated a “purely psychical excitation” into physical terms. If constitutional factors allowed “somatic compliance” to occur, symptoms developed which “signified” several meanings simultaneously: the displacement of sensation; the copying of illness behavior; the search for “secondary gain”; and the “symbolization” of a “fantasy with a sexual content.”
Early psychoanalytic work focused on conversion reactions and symbolic representations connected only with the psychomotor or perceptual systems. By the 1920s, Freud and his followers moved beyond their earlier work to theoretical speculations about “vegetative” and organic diseases very broadly defined. Led by Freud himself and by such theorists as Sandor Ferenczi and George Groddeck, the psychoanalytic community, with Freud’s full blessing, extended its theories of “displacement reactions,” “erotization,” and “symbolic meaning” to a wide range of physical conditions not previously considered. Groddeck, for example, claimed that organic symptoms in any physical disorder can be understood as the symbolic play of the id on the plastic body through primary process, just as the imagery of dreams can be understood as the “work” of the unconscious upon the day residues and other visual tracings. Freud praised this work extravagantly, calling it the “medicine of the future.”
American physicians participating in the psychoanalytic movement traveled in the same direction as Freud and his immediate circle. In the teens and early twenties they wrote about hysterical conversion in the mode of the Clark Lectures, but in the later twenties and early thirties they joined the Groddeck bandwagon. This shift can be seen clearly in the work of Smith Ely Jelliffe, one of America’s leading early analysts.
Slide 13 – Smith Ely Jelliffe portrait
In the teens Jelliffe published papers on classic hysterical conversion; in the twenties and thirties he broadened his approach to include the study of “psychopathology and organic disease” in more general, Groddeck-like terms.
By the thirties, Jelliffe was joined by important emigre analysts such as Felix Deutsch and Franz Alexander. Deutsch moved in the Groddeck-direction, whereas Alexander tried to remain a Clark Lectures-style Freudian. Alexander was particularly popular with American audiences.
Slide 14 – Franz Alexander portrait
Alexander was invited to discuss his psychoanalytically-based ideas on psychogenic factors in organic disease before various important medical audiences. He was Harvey Lecturer at the New York Academy of Medicine and a featured speaker at the Central Neuropsychiatric Association meeting. The latter talk was published in JAMA in 1933.
Slide 15 – F. Alexander in JAMA (1933)
Psychoanalysis had clearly reached a stage of some acceptability, even within mainstream American medicine. And together with the ascent of psychiatry, the growing interest in psychogenic connections within Internal Medicine, and the attempt to reform medical education in the holistic direction, psychoanalysis helped push American medicine almost to the brink of a new field – “psychosomatic medicine.” But something was still missing, another element that when dropped like a seed crystal into a supersaturated solution would suddenly and dramatically precipitate the field. That missing element was organized American corporate philanthropy, which in the 1930s took a great interest in psychosomatic medicine.
Consider, for example, the Josiah Macy Foundation’s role in sponsoring, publishing, and disseminating Flanders Dunbar’s critically important and field-shaping monograph, Emotions and Bodily Changes.
Slide 16 – Dunbar’s 1935 book
This meticulous book included a bibliography of 2,251 entries which were organized and summarized in a text of 432 pages. Devoted to a “Survey of Literature on Psychosomatic Relationships,” Dunbar’s book not only consolidated an emerging field but gave it a name, the term “psychosomatic” occurring regularly in medical literature after 1935 but very rarely before. And all this was made possible by American corporate philanthropy.
Consider also George Canby Robinson’s The Patient as a Person, sponsored , published and disseminated by the Commonwealth Fund. This book legitimated and broadened the appeal of psychosomatic medicine by linking it to Robinson’s reputation as a clinical investigator, the Johns Hopkins Hospital, and the general improvement of American medical education.
Even more important than the Josiah Macy Foundation and the Commonwealth Fund was the Rockefeller Foundation, whose Medical Sciences Division was very ably directed in the thirties and forties by Dr. Alan Gregg.
Slide 17 – Portrait of Alan Gregg
I have become convinced that Gregg, with the comparatively vast resources and influence he commanded at the Rockefeller Foundation, was critically important to the development of American psychosomatic medicine and was very likely a key person behind the scenes in other important philanthropic endeavors as well – at a time, it must be emphasized, when the American government contributed essentially no resources to medical education and research.
Let’s watch Gregg in action. In 1933 he launched a major and carefully crafted program in psychiatry and related disciplines. One of the fields covered in the Foundation’s broadly conceived program was psychosomatic research, defined in largely experimental terms. Gregg wanted to create an “elite corps” of psychosomatic investigators oriented towards physiology but also interested in psychoanalysis yet not dominated by it or blindly adherent to its claims. He worked by a process that, to substitute for a moment a biological metaphor for the chemical one I used earlier, can be aptly described as “artificial selection” (meant in a neutral and strictly Darwinian sense with no pejorative intent). He picked the varieties in the herd he wanted to nurture and replicate, selectively fed and protected them, and then allowed the preferred few to multiply under carefully controlled conditions until, he hoped, they took over the field.
For example, consider Roy Grinker, who in the 1920s was a neurologist moving toward a broadly interdisciplinary approach. Enter Alan Gregg and the Rockefeller Foundation. In 1932, Gregg arranged a Rockefeller fellowship to allow Grinker to go to Vienna for an analysis with Freud, and when Grinker returned – originally to the new psychiatry department at the University of Chicago just funded by the Rockefeller Foundation – he began to devote a portion of his research time to psychosomatic medicine. Grinker was soon one of the acknowledged leaders of the emerging field, all nicely arranged by Gregg and the Rockefeller Foundation.
Stanley Cobb was another Gregg protege. He became famous as Psychiatrist-In-Chief at the Massachusetts General Hospital but had begun his medical career as a neurologist. He made his transition to psychiatry and to his position at Mass General with large and often repeated Rockefeller grants that Gregg carefully arranged. Beginning in the thirties, Cobb added psychoanalysts to his service and became involved in psychosomatic research. One of the analysts he employed for a time was Felix Deutsch, an emigre who had been close to Freud and who was not far from the Groddeck/Jelliffe end of the psychoanalytic spectrum. With Rockefeller funds and Gregg’s encouragement, Cobb used Deutsch in his work but kept him under careful surveillance – this is explicit in his correspondence with Gregg – and submitted his work to the exacting standards of the physiology laboratory.
Slide 18 – Stanley Cobb portrait
As a third example, let us consider Franz Alexander again. Alexander was the only psychoanalyst to receive large direct grants from the Rockefeller Foundation. He’s the exception who proves the rule of what Gregg was deliberately trying to accomplish. At one time in his career Alexander had been a physiologist and, although in the 1920s he became a popular training analyst in Berlin and a major contributor to psychoanalytic theory, he remained interested in physiology to some extent and, even more important, he stayed aloof from the Groddeck/Jelliffe - style of psychogenic speculation. With Gregg’s strong encouragement and the irresistible lure of Rockefeller money, Alexander moved strongly in the direction of physiologically-oriented psychosomatic studies.
Alexander proceeded along two complementary lines. First, he critiqued the inadequacies of Groddeck/Jelliffe psychoanalytic explanations of organic disturbances, claiming that these theories were often internally inconsistent from a logical point of view and simply implausible in the assertions they made about the direct somatic influence of “fantasies” or “wishes.” Second, Alexander offered an alternative explanatory model for those psychogenic pathologies -- like ulcer and asthma -- he explicitly differentiated from the conversion phenomena of original and extended Freudian theory. This alternative model rested on the assumption of a necessary and specific “chain of intermediary physiological processes ... interpolated between psychological stimulus and organic end-result.”
In 1935 Alexander secured his first grant from the Rockefeller Foundation – $100,000 for three years. Here was an unmistakable instance of carefully controlled artificial selection. In his negotiations, Gregg had withheld the Rockefeller grant until he had assurances that he would get what he wanted from Alexander, and he repeated the same process before awarding a renewal grant in 1938.
Thus, to turn back to the chemical metaphor again, the Rockefeller Foundation as Alan Gregg used its resources and influence was the seed crystal needed to precipitate psychosomatic medicine from the general “soup” of twenties and thirties American medicine. The ingredients had been present for a while and steadily increased in concentration, but now they came together in a substantial and visible way. And when, as the signal event, the new journal Psychosomatic Medicine was launched in 1939, Rockefeller influence was evident on almost every page. The majority of the studies undertaken in the early years – including several by Grinker – had been funded with Rockefeller money at Rockefeller-shaped institutions under the direction of investigators whose careers were molded by Rockefeller support. The journal’s editors included Rockefeller grantee Helen Flanders Dunbar, Rockefeller grantee Franz Alexander, and Rockefeller grantee Stanley Cobb.
Slide 19 – PSM title/editorial page
So, in 1939, an exciting new medical field had come into being in America, and everything seemed to be working to guarantee it a limitless future. No wonder its leaders felt that they were participating in a “new era in etiological and therapeutic thought” and that psychosomatic medicine was an “integral part of every medical specialty,” as I quoted them saying at the start of this talk. But what happened over the next fifty years? And why did everything turn so sour by 1988 that the field seemed moribund or entirely marginal even to its leaders, as I also made clear at the start? To answer these questions in the last part of my talk, I want to proceed pretty much as I have in the first, except that, in the interest of time, I will approach matters more schematically and telegraphically than I have thus far. I will also organize my overview by decade and highlight only what I regard as the most significant examples or turning points.
Let’s start with the 1940s. In this decade the field generally did quite well and was able to sustain a great deal of its original excitement as a major new clinical and research area. It continued the momentum of the thirties and carried along some of the same research funding. To this was added, in a major way, the opportunity afforded by World War II, particularly the opportunity provided by military-sponsored research. Roy Grinker’s work on the psychosomatic problems of aviators under combat conditions in North Africa and domestic research at Cincinnati on the physiology and psychology of altitude sickness, both sponsored by the Air Force, are just two examples of the many research projects that were funded and pursued to fruitful results. The forties also saw the recruitment of new investigative talent in Internal Medicine. George Engel’s “conversion” to psychosomatic research in this decade -- as he was convinced and cajoled out of his stubborn resistance to what he had previously regarded as the “laughable hogwash” of the psychiatrists and psychoanalysts -- is the one with which I am most familiar (because of Engel’s relationship to Rochester), but his was only one of several similar stories from the period.
Psychosomatic medicine was also quite successful educationally. There was sufficient interest, in fact, to inspire the first textbook on psychosomatic medicine for medical students.
Slide 20 – Weiss & English (1943)
As an indicator of its success, Weiss & English went through three editions in a very short time.
Even more dramatic was psychosomatic medicine’s success in popular culture. Flanders Dunbar and Franz Alexander each published “psychosomatic best sellers,” Dunbar’s even being a Book of the Month Club selection while Alexander enjoyed almost “cult” status for his combined psychoanalytic and physiological model.
Slide 21 – Dunbar’s Mind and Body (1948)
Slide 22 – Alexander’s Psychosomatic Medicine (1950)
Slide 23 – Alexander’s model, linking unconscious intrapsychic conflicts and
physiology in the genesis of peptic ulcer
Psychosomatic medicine was so much the “rage” around 1950 in the United States that popular magazines like Newsweek, Readers Digest, and Ladies Home Journal ran articles by the dozens and psychosomatic theories even found their way into the lyrics of a popular Broadway show – “Adelaide’s Lament” in Guys and Dolls (“A goil could develop a cough”).
In the 1950s, there were some signs of trouble, although they were quite subtle at first. Franz Alexander and others committed to psychoanalytic approaches were still quite powerful as forces in psychiatry and medicine. But the fifties also saw a shift to general systems theory on the part of older investigators like Grinker and younger ones like Engel, and growing focus on somato-psychic as well as psychosomatic pathways. A parallel shift was detectable, from unconscious intrapsychic conflicts retrieved by psychoanalytic methods to directly observed conscious and semi-conscious maladaptive behavior studied by clinical and experimental psychologists (who grew dramatically in number at mid-century). This shift to observed maladaptive behavior was likewise linked to the powerful advance of laboratory methods and the rise of models derived from organismic physiology and the social sciences. On the one side, physiologically-oriented investigators like neurologist Harold Wolff in New York and endocrinologist Hans Selye in Montreal emphasized stress and hyperexcitability, which left the maladapted organism stuck in the “on” position.
Slide 24 – “links between stress & disease” from Simmons & Wolff (1954)
Slide 25 – Hans Selye portrait
Slide 26 – title page of Stress (1950) by Selye
On the other side, George Engel and his colleagues in Rochester emphasized losses, deprivations and adaptive hypoexcitability that caused the organism to enter a state of “conservation-withdrawal” and to become generally “shut-down.” Their most dramatic work derived from naturalistic experiments on an infant, “Monica,” who was fortuitously admitted to Rochester’s Strong Memorial Hospital. Monica had a gastric fistula, which meant that her stomach secretions could be easily monitored and closely correlated with her observed affective states. Joyful reunions with trusted experimenters and care-givers were associated with copious secretions, but when Monica became emotionally disengaged or withdrew – as in the presence of a stranger – her gastric secretion ceased entirely and even became unresponsive to histamine. It wasn’t a very long step to the assumption that her immune system would also function with reduced capacity in these circumstances.
Slide 27 – Monica affects
While all this was going on, Internal Medicine in the fifties moved in a biochemical and generally reductionist direction, separated itself from organismic physiology, and disconnected from psychiatry. Psychiatry headed in several different directions at once – to new versions of psychoanalytic, behaviorist and group-psychological therapies and to psychotropic medication, all at the same time – and defined psychosomatic medicine more and more as a subspecialty area, both in terms of research (increasingly, actually done by Ph.D. psychologists) and as a clinical activity organized in “liaison” programs. Internists and other clinicians interested in the practice dimensions of psychosomatic medicine tended to withdraw from the research-oriented American Psychosomatic Society and formed their own psychosomatic Academy.
One of the major reasons all this happened was because there was a general growth and multiplication of fields and subdisciplines derived from rapidly expanding government funding. The NIH budget alone was $400,000,000 per annum by 1960, and there was also the NIMH, community mental health programs, the Veterans Administration System, etc. Private health insurance also dramatically expanded and increasingly underwrote psychiatrically-related programs. Private philanthropy – of the Rockefeller Foundation and others – became much less important as a funder of medical research and increasingly played a role of greater relative influence in medical education. Even there, its funds were swamped by the infusion of federal resources. But federal resources, it is important to note, flowed from no single, central source and were disbursed according to no single, organizing vision. We were a far cry from the day when Alan Gregg controlled the disbursement of Rockefeller funds in an exercise of artificial selection. Increasingly, research fields were left to their own momentum and thus tended to proliferation, subdivision, fractionation, and fragmentation.
At the beginning of the next decade, E.D. Wittkower tried to capture the hard truth in his 1960 presidential address to the American Psychosomatic Society.
Slide 28 – Wittkower title page
... [there] is a decline in participation in psychosomatic research of medical specialists other than psychiatrists ... of late the interest in our field of physicians and surgeons has greatly subsided; the number of their articles has dwindled to a mere trickle during the last few years. ... There is a growing cleavage noticeable in psychosomatic publications between those with more and more psychiatry and less and less physiology, and those with more and more physiology and less and less psychiatry. ... the decline in interest in psychosomatic medicine on the part of physicians is due to an abandonment of the originally intended concerted effort, resulting in a fragmentation into particularistic interests.
As Wittkower feared, fragmentation into particularistic interests continued in the sixties as the various rifts deepened and as competing groups developed differing subdisciplinary emphases and rival research programs. Those most interested in clinical applications also tended to divide into separate “schools.” “Stress” became more and more popular, even faddish, and separate journals developed, which, of course, fragmented the field further. Meanwhile, both research studies and clinical interventions tended to ignore psychoanalysis and emphasized bioconditioning, sleep deprivation, sensory overload, and biobehavioral management, instead. On the other side, Engel and the “Rochester school” pushed the conservation-withdrawal model, continued to promote psychoanalysis although now with an emphasis on “object relations” and “organismic” theory, and moved in the direction of studying the “final common pathway to disease” by concentrating on the pre-onset circumstances to a wide variety of conditions, pre-onset circumstances that were marked by common affects of “helplessness and hopelessness” and the related “giving-up – given-up” complex. All this, as I see it, was really a sophisticated and clinically rich extension of the Monica studies and notions of organismic “shut down” during “conservation-withdrawal.”
Slide 29 – Engel portrait
Slide 30 – Engel’s Menninger Lecture paper (1968)
There’s a subtle but important dimension to Engel’s success that also needs to be noted. As his approach gained recognition in Internal Medicine, it became less recognizably psychoanalytic even though Engel readily acknowledged that he had drawn much of his inspiration from psychoanalytic theory. It is hard to ignore the possibility that he was adapting to his circumstances and making certain tradeoffs, because it is now generally acknowledged that psychoanalysis experienced a dramatic drop-off in popularity during the sixties. This is one of the major themes in Nathan G. Hale’s recent and long-awaited book.
Slide 31 – Hale title page
The evidence is also clear in the medical literature of the period. As just one example of many that could be presented, consider Gatfield and Guze’s 1962 paper on the diagnosis of hysteria published in the journal, Diseases of the Nervous System. Their principal point – typical at the time – was that the hysteria diagnosis had been very badly overextended and often dangerously applied to cases that on follow-up proved to be clear instances of organic disease.
Slide 32 – Gatfield & Guze table (esp. Cases 16, 17, 18)
Other cases were simply less serious than originally thought and often tended to resolve themselves. And hysteria itself was very likely genetically grounded and biologically based, no longer the premier symbolic and psychodynamic condition celebrated in Freudian theory.
With psychoanalysis gone as a central inspiration, the psychosomatic field tended to fragment even more rapidly and bitterly, while reductionist approaches proliferated in Internal Medicine for diseases once thought to be “classic” psychosomatic conditions – like asthma, ulcerative colitis, peptic ulcer. The psychosomaticists who remained active in clinical practice tended to shift to “somato-psychic adjustment,” that is, to the management of patients’ psychological reactions to illnesses now presumed to be largely somatic.
The seventies were overshadowed by the “molecular revolution,” which affected all aspects of medicine. Internal Medicine continued further down the path it had already begun to follow in the sixties, and psychiatry, more dramatically, suddenly entered the era of “neuroscience.” Solomon Snyder’s work on neurotransmitters was just one example of what marked the decade.
Slide 33 – Solomon Snyder’s neurotransmitter diagram
Now that it was getting “molecularized,” psychiatry no longer had to struggle to get medicine’s attention; it merely needed to “fit in.” Those psychosomaticists who were up with the times and current in research – Herbert Weiner is a good example – tried to import the latest findings of psychoendocrinology and neurochemistry into psychosomatic research, often using animal models as their principal focus but then generalizing to human disease.
Slide 34 – Weiner, Psychobiology title page
Older-generation psychosomaticists – like Engel, Wittkower and Z.J. Lipowski – tried to cope with the changing times by constructing general theories that attempted to encompass everything. But while they achieved some recognition and acknowledgment (and in Engel’s case added to the vocabulary with his newly minted “biopsychosocial” alternative), the theorists were generally given lip service but in fact were marginalized or ignored, brought out only on ceremonial occasions, or consigned to the role that Alvan Feinstein brilliantly and sardonically labeled “clinical exhortation.”
Slide 35 – Engel’s 1977 Science paper
Slide 36 – Engel’s 1980 “Clinical Application” paper
In the eighties, there were two notable developments. One was the emergence of the exciting new field often called “Psychoneuroimmunology” (there are many variants), which psychosomaticists saw as the vindication of their long quest. Basic scientists were now able to demonstrate direct micro-anatomical connections between the nervous and immune systems; show that anatomical lesions in or the electrical stimulation of parts of the brain influence antibody production in the spleen and lymph nodes; and identify receptor sites for neuroendocrine hormones and neurotransmitters on cells of the immune system. This rigorously demonstrated “cross-talk” between the immune and neuroendocrine systems has been taken by Engel and others to mean legitimation for the Biopsychosocial Model and, behind that, for the psychosomatic approach in general. Psychoneuroimmunologists and molecular neuroscientists themselves are not so sure and are much less willing to generalize or philosophize about what their discoveries mean. They are, in any case, very far removed from older generation psychosomatic researchers and the American Psychosomatic Society, which goes on in its now familiar ways, as Friedman’s remarks in 1988 quoted at the outset of this paper make clear. Psychoneuroimmunology has become its own specialty area or, actually, several competing specialty areas.
Slide 37 – Ader, Psychoneuroimmunology
Finally, the second development of the eighties was the growing focus on “somatization” as the clinical dimension of psychosomatic medicine. “Somatization” in various forms appeared in the Diagnostic and Statistical Manual of the American Psychiatric Association: the DSM III of 1980, DSM IIIR of 1987, and the DSM IV of the 1990s.
Slide 38 – DSM IV: “Somatization Disorder”
Psychosomatically-inclined clinicians – for example, Engel’s former fellows now in primary care internal medicine and family medicine – often think and write in terms of “somatization,” usually using the term far less rigidly and more fluently than the DSM with its strict categories and criteria would seem to allow. But in their hands, “somatization” is a deliberately and aggressively non-theoretical construct. It captures one aspect of patient behavior and must be studied in clinical contexts, using clinical methods. It is, decidedly, not a subject for rigorous experimental examination or abstract discussion, and it is notable, therefore, that clinicians most interested in somatization have little inclination to present their findings to the American Psychosomatic Society or even to attend its meetings. No more than Psychoneuroimmunology does “Somatization” seeem capable of reviving the now largely moribund field of American psychosomatic medicine.
American psychosomatic medicine as a research field with a clear focus, optimistic outlook, and strong sense of clinical mission is gone! Times have changed, and the circumstances which gave rise to it and the forces that sustained it no longer exist. But could things have turned out differently? Perhaps, if the flaws identified in early formulations of the field – e.g. Alexander’s – had been acknowledged as the result of well-intended but too hasty efforts and were then amended in friendly rather than in competitive or hostile fashion. Perhaps, too, if the centrality of studying affective states pre symptomatic onset with “clinical-anamnestic methods” was accepted, and integrated with laboratory work, as in Engel’s “Monica” studies. But “clinical-anamnestic methods” may have been abandoned by the field too quickly as psychoanalysis fell out of favor and as the push to the laboratory and psychometric testing became so insistent because of the pressure of grants, career-building, and institutional requirements. Could it have been different? Could American medicine of which psychosomatic medicine was an integral part have developed in alternative fashion? It may not be fair to answer a question with a question, but that’s where I think we are right now.
Copyright: The Author
Department of History
University of Rochester
Rochester, N. Y.