As is widely recognized, the asylum era in the United States was largely the product of the last two-thirds of the nineteenth century (Rothman Reference Rothman1971; Grob Reference Grob1973). Where the mentally ill had previously been for the most part left at large, the responsibility of their families, now segregating them into specialized institutions became the preferred policy. In this respect the United States was part of a far larger move towards isolating the mentally ill within specialized institutions, a movement that had its counterparts in Britain, in France, in Germany, indeed all across Europe, and from which Americans freely borrowed (Scull Reference Scull and Scull1981). Elsewhere, I have argued that the widespread adoption of the asylum had its roots in the rise of a market-based society and in the growing power of the state (Scull Reference Scull1977; idem 1993). Asylums had been created in an atmosphere of extreme optimism about the possibilities of curing the mentally ill, an optimism that proved spectacularly misplaced. By the last third of the nineteenth century, the utopianism that had attended their foundation had been replaced by a deepening pessimism, and a growing belief that the mentally ill were a hereditarily defective, biologically degenerate lot, best dealt with by isolating them and preventing them from reproducing their own kind.

The era that witnessed the adoption of the asylum as the preferred solution to the problem of grave mental illness also saw the creation of a steadily more organized and self-conscious group laying claim to expertise in the management and treatment of insanity. At first known in Anglo-American circles as alienists, medical psychologists, or asylum superintendents, only towards the close of the nineteenth century did they begin to call themselves (and be called) psychiatrists, and by the early twentieth century that was the term the profession increasingly preferred. In what follows, I shall examine in a necessarily somewhat schematic and abstract fashion certain key features about the evolution of that profession in the United States, and to suggest both the ideological and the institutional changes that have characterized its development. I shall also examine how, beginning in the middle decades of the twentieth century, clinical psychology emerged as a complementary and to some degree a rival profession seeking to control at least part of the market for the treatment of the mentally ill.

The history of American psychiatry as a profession from the mid-nineteenth through the end of the twentieth century is marked by three major shifts in its intellectual orientation – dramatic breaks with past ideologies which have been associated with equally profound changes in the nature of psychiatric practice. These resonate in important ways with changes in the mind and brain sciences in general – psychiatry, neurology, and physiology – in ways that are explored elsewhere in this topical issue. And indeed, at least some of the changes in psychiatry reflect clashes between its practitioners and those in neighboring disciplines, most especially psychology.Footnote ¹

I shall argue that these three eras have been marked by broader alterations in the meanings ascribed to mental illnesses and to what appear to be unambiguously physical nerve diseases in the larger culture, and in social practices. I speak here of the dominant tendencies that can be observed in the epochs of modernity and post-modernity, because there have always been undercurrents of dissent – vocal or otherwise – against the hegemonic ideas of a particular era. Still, the existence of radically different organizing ideologies, and the rapidity with which change has overtaken the profession are remarkable features of the psychiatric universe. There is every sign, moreover, as we move into a new millennium, that these wrenching changes may not be at an end. Despite frantic efforts to shore up the epistemological framework that has dominated psychiatry since 1980, a new sense of crisis is enveloping the field, and only the absence of a new organizing paradigm seems to stand in the way of the collapse of the current ruling ideology. In substantial measure, these recurrent crises (and temporary settlements that in turn give rise to new anomalies) reflect the underlying problem that the model of specific disease entities tied to a putative biological pathology has become hegemonic and well-accepted in medicine since the last third of the nineteenth century, but fits poorly when applied to the social, emotional, and behavioral troubles with which the psy complex concerns itself. Madness remains an epistemological mystery, an amorphous and ill-defined territory constantly contested and fought over. Proclamations that it is rooted in heredity, in psychodynamic conflicts with roots in childhood, or is the epiphenomenal manifestation of faulty brain biochemistry have been put forth by true believers as “scientific truth.” In reality, they are hypotheses about conditions whose defining characteristics remain frustrating vague and inchoate. We possess neither “a repertoire of tightly bounded, seemingly objective, and generally agreed-upon diagnostic categories” nor any ability to root mental illnesses “firmly on biopathological mechanisms” (Rosenberg Reference Rosenberg2006, 411). Pretense and premature certainty have abounded over the past century and more in psychiatry. Speculative somaticization has been replaced by speculative psychologization, and then back we have lurched to the soma. Certainty, or even a modicum of intellectual security, remains an elusive Holy Grail for psychiatry. In reality, its explanations for the behavioral and emotional pathologies with which it wrestles, even its attempts to create stable categories of disease entities as objects of research and intervention, remain transparently unsatisfactory for all but the infatuated. Contestation and controversy, trouble and turmoil are with good reason fixtures for psychiatry and neighboring professions: realities they may lament, but that they cannot long escape. The more the rest of medicine can call upon physical signs, laboratory findings, and imaging results to shore up its diagnoses and prescriptions, the less secure, the more marginal, the position of those whose clinical categories and objects of attention lack these epistemological moorings (ibid., 420–421).

What, then, are the defining characteristics of the three eras I have spoken of? When did their respective periods of dominance manifest themselves? What are some of the fundamental forces that seem to underlie the shift from one era to another? And why does a profession that proclaims its practices are founded upon “science” (implicitly portrayed as a progressive and cumulative process), instead seem to lurch periodically towards radically incompatible ends?

Ordinarily, one has difficulty attaching precise dates to major transformations like the ones discussed here, and to be sure, social phenomena like these have no neat and precise chronological boundaries. They exist in embryonic form before briefly achieving hegemonic status, and they generally enjoy a half-life that extends into the succeeding era. But bearing that caveat in mind, I think one can indeed treat each of these successive historical moments as having an identifiable beginning and (in all but the last case) end that is remarkably definite, to which one can attach a chronology that is anything but arbitrary or purely conventional. Those transitional points reflect and are prompted by larger social and political events that remain almost wholly outside professional influence or control. But the nature of the professional response is neither fixed nor given, and shapes the outcomes in important respects, in ways I shall seek to analyze later in this paper.

There is first, the period where institutional psychiatry predominates, and a veritable mania for experimenting upon the bodies of the mentally ill sweeps through the profession, one that lasts from the turn of the century through 1945; a second period, from 1945 to 1980, where psychoanalysis, dominates the psychiatric profession. By 1960, all but two departments of psychiatry in the United States were led by psychoanalysts or psychoanalytic fellow-travelers. The best neophyte psychiatrists opt almost universally to obtain training in psychoanalysis, leaving what they derisively refer to as “directive-organic” psychiatry to the less qualified, condemned to practice in overcrowded state hospitals. Psychoanalytic ideas dominate popular culture, from Hollywood to best-seller lists, and through the work of Benjamin Spock, become central to child-rearing practices. Madness is declared to be rooted in meaning, and psychiatry as a profession begins to desert the old institutions in droves. This is an era marked, in its later stages, by sharp declines in the in-patient census of state mental hospitals for the first time since they emerged on the scene in the nineteenth century, and the relocation of most of the seriously mentally ill into the “community,” whatever that term came to mean (the practitioners of community psychiatry left it deliberately ambiguous); and finally, there is the so-called neo-Kraeplinian era, when psychiatry is re-biologized and shifts drastically away from psychodynamic accounts of mental illness and psychotherapeutic approaches to its treatment, and embraces the twin gods of psychopharmacology and an ever-expanding universe of psychiatric disorders defined purely conventionally and descriptively. Madness then as now is proclaimed to be rooted only and always in the brain and its defects, genetic or biochemical. Within such an intellectual universe, the symptoms of mental illness are purely epiphenomenal, of no larger significance or interest – mere noise that testifies to the biological defects that are solely worthy of professional attention.

In 1844, 13 American alienists met in Philadelphia to found a professional organization of asylum doctors. Fifty years later, their intellectual and institutional descendants met to celebrate the fiftieth anniversary of what was then called the American Medico-Psychological Association (renamed the American Psychiatric Association just over a quarter century later). Psychiatry was still pre-eminently and almost exclusively an institutional specialty. In the preceding quarter century, following the carnage of the American Civil War, another group of mental specialists had emerged, aiming to treat traumatic and other kinds of injuries and diseases of the brain and the nervous system. Neurologists had asserted a generalized claim to expertise in the diagnosis and treatment of nervous and mental disorders, and had had some success in attracting ambulatory patients with less serious mental diseases like hysteria and (a disorder one of their number had invented) neurasthenia, or weakness of the nerves. The neurologists had also been vocally critical of the scientific credentials and therapeutic skills of psychiatrists. Perhaps their most prominent and distinguished practitioner, Silas Weir Mitchell had been invited to address the assembled gathering of the nation's psychiatrists, held once more in Philadelphia.

Mitchell had initially declined the invitation, but when his hosts were persistent, he agreed to speak. In his opening remarks, he noted that while medicine and surgery had made extraordinary progress over the past half-century, psychiatry had been stagnant. “Your hospitals are not our hospitals; your ways are not our ways.” Psychiatrists lived in isolation, cutting themselves off from healthy criticism and scientific progress, and had become “almost a sect apart” and the consequences for profession and patient alike were “evil.” The very title to which the more senior members of his audience clung, “medical superintendent,” was “absurd.” Many listening to him had won their jobs through political patronage, still another “grave evil.”

The invective rained down on his audience for an hour and more. Mitchell denounced the “senile characteristics” of asylum boards; the repetition of “old stupidities in brick and stone”; the “neat little comedy” of outside inspections known about in advance and thus incapable of getting at the truth. But this was the least of the difficulty. The medical superintendent was the “monarch” of all he surveyed, but he was an emperor with no scientific clothes. “Where, we ask, are your annual reports of scientific study, of the psychology and pathology of your patients. . . . We commonly get as your contributions to science, odd little statements, reports of a case or two, a few useless pages of isolated post-mortem records, and these are sandwiched among incomprehensible statistics and farm balance sheets.” Asylum case records put on display an appalling state of affairs, an “amazing lack of complete physical study of the insane, . . . the failure to see obvious lesions,” a total ignorance of the diagnostic technologies indispensable to the practice of modern medicine, problems as visible in the most prominent and best-endowed asylums for the rich, as in the meanest, most over-crowded state hospital. The upshot of such conditions was what he himself had observed in the wards of an asylum in his home city of Philadelphia, where one saw “the insane, who have lost even the memory of hope, sit in rows, too dull to know despair, watched by attendants.” Nor were they the only victims of prolonged confinement: their captors, the psychiatrists, had fallen into the same sort of paralysis. “The cloistral lives you lead give rise, we [neurologists] think, to certain mental peculiarities . . . you [too] are cursed by that slow atrophy of the energizing faculties that is the very malaria of asylum life.” Indeed, Mitchell concluded, “I cannot see how, with the lives you lead, it is possible for you to retain the wholesome balance of mental and moral faculties” (Mitchell Reference Mitchell1894).

Psychiatrists thus stood accused, and not for the first or the last time, of being nearly as mentally unstable as those they locked up on behalf of society. Many simply shrugged, and safe in their bureaucratic niches, waited for the storm to pass. Society was by now accustomed to shutting up the mentally ill, and whether in the United States or in Europe, there was little inclination to leave the lunatic at large in the community. There was a place – a constricted and low status place to be sure – for psychiatrists willing to do little more than cast a medical gloss over homes for the mad. But among the more ambitious, such passivity was intolerable, and the half century following Weir Mitchell's 1894 address saw them embark on a steadily more extravagant series of therapeutic experiments on their captive audience, in a desperate attempt to reclaim their identity through scientific work as medical men and as professionals who treated and cured their patients (Scull Reference Scull2006, 129–149). Their actions were made possible by the status of those confined in their institutions as people who were to all intents and purposes, morally and legally dead: deprived of their selfhood, presumed to be lacking the capacity to make judgments about their own treatment, cut off from the larger world, and thus incapable of resistance.

The period between 1894 and 1945 marked in some ways a continuation of an earlier emphasis on the total segregation of the mentally ill in institutions. What was novel about this period, however, was the emergence of a whole host of treatments directed at what the more vocal psychiatrists insisted was a disease like any other: a disease, that is, rooted in the body, and thus to be attacked by somatic means. To be sure, among the therapeutically inclined, there was a small group of dissenters, operating for the most part outside the walls of asylums (or as they now came to be called, mental hospitals), and for the most part treating milder, less disturbed cases on an ambulatory basis. I shall return to a consideration of these dissenters who proclaimed madness had meaning and must be treated at the level of meaning by psychotherapeutics or “talk therapy.” But by their hundreds of thousands it was the institutionalized patients who provided the bulk of psychiatric practice and occupied the time and efforts of the overwhelming majority of the profession, including, in these years, its most prominent figures.

American mental hospitals had confined perhaps a hundred thousand patients at the turn of the century. They contained more than 400,000 patients by 1940, and a half million by 1950 – mental hospital populations had quadrupled during a period when the general population had merely doubled. And for these patients, the routines of “moral” or milieu therapy that had been on offer in the nineteenth century (often on little more than a token basis) were now increasingly supplemented if not replaced by a gathering array of somatic interventions.

First came operations to sterilize: tying the tubes for women, vasectomies for the men. Originally promoted by lay enthusiasts for eugenics (Brunius Reference Brunius2006) – restricting the breeding of the unfit – they were transformed in the hands of psychiatrists into something that purported to be of more direct therapeutic benefit to mental patients. It was a reworking of the meanings of the procedure as Joel Braslow has shown using the clinical records of California State Hospitals (the institutions where half of all involuntary sterilizations in the United States were performed), that was prompted by the discomfort many felt at the prospect of employing a “treatment” whose benefits seemingly benefitted the community, not the patients under their care (Braslow Reference Braslow1997, 54–70).

But sterilization was pursued enthusiastically in only a handful of states, California and Virginia most notable among them. The operations always encountered religious and ethical objections, and in many jurisdictions this opposition proved decisive.Footnote ² The other therapeutic interventions, however, often spread much more widely, and became central to psychiatry's efforts to re-emphasize its medical identity by insisting on the somatic roots of the disorders it treated.

The early twentieth century was heir to the triumphs of Pasteur and Koch. Though the two men had in fact quite distinct approaches to the etiology of disease, by the turn of the century those distinctions had been obscured. In truth, the image of immunology that enjoyed growing support was closer in many respects to Koch's belief in a one agent, one cause, one illness conception of the origins of disease, but regardless, it is safe to say that the bacteriological model of disease for a time swept all before it. Emil Kraepelin (1856–1926), the most prominent German psychiatrist of the fin-de-siècle period, attempted to produce some order from the chaos that was madness via a new nosology that purported to be derived inductively from thousands of cases confined in German asylums, introducing the basic division between dementia praecox and manic depressive psychosis, a fundamental organizing tactic that would eventually have a major impact on the world stage (Engstrom Reference Engstrom2003, 140–146). Little known outside the ranks of his professional colleagues but famous and highly influential among them, his Compendium de Psychiatrie went through eight editions in his lifetime, growing in size and influence with each successive edition. Adopting a view that had widespread support among his psychiatric colleagues in Germany,Footnote ³ Kraepelin repeatedly voiced the opinion that dementia praecox had a toxic origin (Kraepelin Reference Kraepelin1896, 36–37, 439; idem 1913, 931; Noll Reference Noll2011, 49–73), and this speculation found prominent supporters elsewhere. Ira Van Gieson, for example, who had been appointed the first director of the New York Pathological Institute, the research arm of the vast New York State mental hospital system (an institute explicitly set up in the aftermath of Weir Mitchell's address to counter the complaint about the scientific backwardness of psychiatry), was equally convinced that infection caused insanity (Van Gieson Reference Van Gieson1896; idem 1897–1898).Though he was soon maneuvered from office, hurt by a premature announcement that he was on the brink of discovering the “germ of insanity” and even more by his visible contempt for the psychiatrists his Institute was supposed to be serving, the basic hypothesis he was pursuing was one that drew considerable support (Dowbiggin Reference Dowbiggin1997, 51–54; Grob Reference Grob1983, 127–129).

And soon there was a demonstration that a substantial fraction of the mentally ill in asylums were indeed there because they had been infected. As many as twenty to twenty-five per cent of male admissions (and a smaller fraction of the female population) were victims of general paralysis of the insane (GPI), a progressive disorder first diagnosed in France in the 1820s, and one of the few unambiguous successes nineteenth-century psychiatrists could point to (Grob Reference Grob1983, 188–189; Braslow Reference Braslow1997, 72–74). The increasing suspicions of many that GPI was the product of syphilitic infection was decisively confirmed in 1913 by Noguchi and Moore, who observed the syphilitic spirochetes in the brains of its victims.

As well as reinforcing the conviction of most psychiatrists that mental illness had a somatic origin, these findings led indirectly to further therapeutic experiments based on this hypothesis. In the United States, led by Henry Cotton (Scull Reference Scull2005) of the New Jersey State Mental Hospital at Trenton (who had trained, inter alia, under Kraepelin) – and also independently in Britain, led by Thomas Chivers Graves, head of the Birmingham mental hospitals (Scull Reference Scull, Freeman and Berrios1996) – it led to the claim that all mental illness was the product of hidden chronic infections, focal sepsis, which produced toxins that poisoned the brain and thus resulted in the symptoms of mental disturbance. The effect was to license a wide-ranging assault on the presumed sites where these infections lurked, a program of “surgical bacteriology” that began with the wholesale removal of teeth and tonsils, and then extended in the United States to the excision of stomachs, spleens, colons, and other “unnecessary” organs. Though Cotton and Graves were scarcely isolated figures and continued their surgical interventions for decades, and though the notion of focal sepsis attracted vocal support from such leading centers of advanced scientific medicine as the Johns Hopkins, the University of Chicago, and the Mayo Clinic, including Llewellys Barker, William Osler's successor at Hopkins, Frank Billings, former president of the American Medical Association and long-time dean at Chicago, and Charles Mayo himself, not to mention such leading lights in British medicine and surgery as Sir Berkeley Moynihan, President of the Royal College of Surgeons and Sir George Robertson and Chalmers Watson of Edinburgh, their vigorous proselytizing for this form of treatment never attracted more than a minority of their colleagues – perhaps fortunately some might now conclude, since the “beneficiaries” of their therapies were, it subsequently materialized, left maimed or dead.

By contrast, the other therapeutic intervention spawned by the discovery of the syphilitic origins of GPI enjoyed broad acceptance internationally, and won for its discoverer one of the only two Nobel Prizes awarded to a psychiatrist. In the closing months of World War I, the Viennese psychiatrist Julius Wagner von Jauregg experimentally induced malaria in patients suffering from GPI. Earlier experiments with using other febrile agents to treat other forms of insanity had failed, but on this occasion, Jauregg announced he had achieved dramatic success in a condition regarded as uniformly fatal, and one that led to a particularly nasty end. At war's end, despite Jauregg's finishing on the losing side (and even being prosecuted for war crimes for his role in treating shell shock with painful electrical shocks),Footnote ⁴ his new technique rapidly drew converts and spread internationally. Patients believed to be suffering from GPI were deliberately given malaria, either using colonies of malarial mosquitoes or (more commonly) drawing blood from one malarial patient and injecting it to induce the disease in GPI cases. The uncertain and intermittent course of GPI and the lack of any controlled trials of the procedure's efficacy – a procedure that did not become standard in medicine till after the Second World War – makes one wary looking backwards of the extravagant claims of success that attended its use. But psychiatrists hailed it as a breakthrough, and the Nobel committee agreed with them in 1927 (Whitrow Reference Whitrow1993; Braslow Reference Braslow1997, 71–94). More importantly, the advent of this therapy spurred the search for other somatic interventions. In the absence of any check on experimentation on the captive populations in mental hospitals, and with the desperation psychiatrists felt to do something, anything that might improve their patients (and their own standing in the medical profession), a host of approaches were tried.

Many of these have sunk into oblivion – experiments with barbiturate-induced comas, injections of horse serum into the spinal canals of mental patients to induce meningitis, chilling patients down to temperatures barely consistent with life, heating them in sweat boxes, having patients breathe pure carbon dioxide, injecting them with cyanide, and so forth (Scull Reference Scull1994) – but others became attached to extravagant claims of therapeutic success. Cures of 70 or 80 per cent of schizophrenics were alleged for treatments as diverse as creating comas by injecting the newly discovered insulin, or grand mal seizures by injecting metrazol (a procedure that produced a feeling that one was about to die, followed by such severe fits that frequently spines and hip sockets were fractured). Most indelibly of all, there were large scale experiments with severing portions of mental patients’ frontal lobes (a technique introduced by the Portuguese neurologist Egas Moniz and popularized by his American disciple Walter Freeman in the mid-1930s, and the occasion for another Nobel Prize awarded in 1949) (Valenstein Reference Valenstein1986; Pressman Reference Pressman1998; El Hai Reference El Hai2005); and electro-convulsive therapy (ECT) (Shorter and Healy Reference Shorter and Healy2007) – the only one of these innovations to still enjoy considerable professional approval today.

It goes without saying that subsequent research has demonstrated that none of these innovations produced the sorts of results their promoters claimed for them – a failure made manifest in any event by the continued rapid upsurge in the mental hospital censuses through the middle of the twentieth century. For psychiatrists, however, they provided the illusion that they were engaged in a therapeutic enterprise, and one that involved treating sick bodies, just like the people they liked to think of as their colleagues, the medical profession at large, rather than simply serving as custodians of holding pens. The therapies were accepted by most of the profession as much for what they were not – psychotherapy, especially in its sexualized Freudian form – as for what they actually accomplished. Desperate patients (or more often, their families) clung to the notion that there might indeed be a medical breakthrough that would rescue their loved ones. And science journalists, ever-susceptible to hype and looking for a good story, were happy to relay reports of the latest miracle cure. Mental illness, mainstream psychiatrists repeatedly intoned, was a biological disorder and thus necessitated treatment by biological means.

It was Hitler and the exigencies of total war that shook up this consensus, and for a quarter century prompted American psychiatry to move in a completely contrary direction. Simultaneously, too, the war (and its Cold War aftermath) helped to create a rival profession, clinical psychology, with enduring effects on the mental health sector. To be sure, psychologists led by Robert Yerkes had managed to make themselves useful to the military in World War I, and in its aftermath, further development of IQ tests and other psychometric instruments had proceeded apace. Alongside its usefulness to the military, psychology had enjoyed a growing stature in the industrial arena, persuading professional managers that it had instruments that could help with their tasks. And its psychometric technologies had also been mobilized in the new child guidance clinics that had begun to proliferate with the support of the well-endowed foundations that emerged in this era, and in the identification and control of the “feeble-minded” and juvenile delinquents (Horn Reference Horn1989; Zenderland Reference Zenderland1998; Napoli Reference Napoli1981). In the latter setting, in particular, psychologists operated in clinical settings, rousing anxieties and opposition from psychiatrists who resented any “trespassing” on what they insisted was an exclusively medical territory. There were pointed suggestions that any invocation of terms like “diagnosis,” “treatment,” and “clinical” on the part of psychologists would be viewed as an affront, for these were terms and tasks that were the preserve of the medically qualified. Intra-professional tensions and hostility simmered beneath the surface right up to the outbreak of World War II, mitigated only by the fact that the largest group of applied psychologists busied themselves with non-clinical matters, and by the continued dominance of the traditional mental hospital in the delivery of mental health services ensured the bureaucratic subordination of psychologists who had the temerity to express interest in clinical concerns (Dana Reference Dana1917; Dickenson et al. Reference Dickinson1933; Grob Reference Grob1983, 243–265; Farreras Reference Farreras2001, chaps. 1 and 2).

Freud himself had visited the United States briefly in 1909. Given the important place psychoanalysis came to occupy in the United States between 1945 and 1980, there has been a temptation retrospectively to assign more significance to the lectures he delivered at Clark University than is conceivably warranted by their modest impact at the time. Freud was only one of nearly 30 speakers on that occasion – the twentieth anniversary of the university's foundation – and he was far from being viewed as the most eminent or important of them. The roster included two Nobel prize-winners, just for a start, and other speakers on psychological and psychiatric matters were paid more than Freud, and were more prominent than he (Shamdasani Reference Shamdasani and Burnham2012; Skues Reference Skues and Burnham2012). Mythology aside, however, Freud did succeed on this occasion in attracting a small and devoted following, a small bridgehead that had secured a following among some of the rich and chattering classes. His acolytes came partly from the ranks of psychiatry, and partly from the neighboring profession of neurology, but the majority of both groups of specialists remained hostile to the claims of psychoanalysis: neurologists preferring to concentrate on diagnostic and prognostic refinement in dealing with a host of mostly debilitating and usually fatal diseases that seemed more obviously organic in origin; and asylum psychiatrists dismissing talk therapy as at once useless when applied to their psychotic clientele, and actively threatening to their tenuous claims to an identity as medics.

The Nazi persecution of the Jews, not yet as deadly as it would soon become, led many analysts to flee – Freud and his daughter to London, many of the rest to New York and the eastern seaboard of the United States (Makari Reference Makari2008, 468–483) – but while this provided some augmentation to the numbers of psychoanalysts in America, particularly in New York, and created all sorts of complicated sectarian rivalries within the analytic community, by itself, it did little to change the balance of power within psychiatry, or the experience of most psychiatric patients.Footnote ⁵

The war changed all that. Citing the experience with shell shock in World War I, American psychiatrists and psychologists succeeded in persuading the Washington powers-that-be that they should avoid a reprise of that wasteful and morale-sapping experience by screening all military recruits for signs of incipient mental instability (Capshew Reference Capshew1986). Robert Yerkes, who had directed the military's mental testing program in World War I was particularly active in stressing the usefulness of psychometrics to this process (Yerkes Reference Yerkes1941; Dallenbach Reference Dallenbach1946). The examiners rejected as many as one and three quarters million recruits on such grounds, only for breakdowns to occur among American troops at more than double the rate of the earlier war (Shephard Reference Shephard2001, 327). Mechanized warfare and the maintenance of mental health turned out, not for the last time, to be incompatible with one another in large numbers of cases (Grob Reference Grob1990; Shephard Reference Shephard2001).

Painfully, the lesson that under sufficient stress even the most stable of us are prone to mental collapse was relearned, and with it doubt was cast on the portrait of mental illness as a purely biological event, and something to which “normal” people were not prone. Treatment, too, though it often took the form of tea (or coffee) and cigarettes and not too much sympathy, was generally what could be thought of as psychotherapeutic: if successful, it took the form of keeping the men close to the front lines, persuading them that they were just suffering from “combat exhaustion,” and manipulating their sense of guilt at deserting their buddies in the platoon (Jones Reference Jones, Menninger and Nemiah2000). And this bias in favor of psychodynamic explanations and interventions was reinforced from the very top, because the prior decision to appoint William Menninger from the family clinic in Topeka, Kansas, as Brigadier-General and chief army psychiatrist ensured that someone broadly embracing Freudian ideas headed what was a quintessentially top-down organization (Grob Reference Grob1994, 193–194).

The significance of the rise of military psychiatry was amplified by the dramatic rise it fueled in the number of those practicing on the mentally ill. In 1940, on the eve of the war, the American Psychiatric Association had only 2,295 members, virtually all of them employed in traditional mental hospitals where they pursued their isolated and stigmatized careers. To these, we may add the small number of psychoanalysts who pursued an office-based practice – a group mostly working in a handful of eastern-seaboard cities, Boston, New York, Philadelphia, and Washington D.C., and in Chicago (with a further odd outpost in Topeka, Kansas). Psychiatric casualties among the armed forces were so numerous (twenty-five per cent and more of those seeing combat exhibiting serious symptomatology) that the demand for physicians to treat them greatly exceeded supply (Grob Reference Grob1990). Doctors were drafted into the field, and given a few weeks of emergency training before being let loose on the troops; and when even that didn't suffice, psychologists – previously members of a profession eschewing practical work and clinging to an identity as laboratory scientists – also found themselves called into the breach (Herman Reference Herman1995, 84, 92). Both groups found the clinical experience engaging for the most part, and many of the two thousand and more army psychiatrists, or “trick cyclists” as the military command dismissively called them, and their psychology counterparts sought to continue to practice in the field after the war finally came to a close.

The numbers of mentally disabled veterans provided one incentive for the federal government to invade the mental health territory – traditionally a purely state responsibility in the United States – and the extensive mobilization for total war had decisively broken down many of the objections that might formerly have been raised about states’ rights. The size of the fiscal drain mental illness represented for state budgets – it was the largest single expenditure for many – was another. And soon enough, the outbreak of the Cold War prompted yet further efforts to enhance federal involvement in science and medicine – both in training new practitioners, and in funding basic research. Whereas before the war, there had been essentially no federal involvement in funding medical research or basic science, in its aftermath, with the formation of the National Science Foundation and the National Institutes of Health, the era of Big Medicine and Big Science had arrived. Expansion was the order of the day (Greenberg Reference Greenberg1999; Appel Reference Appel2000; Farreras, Hannaway, and Harden Reference Farreras, Hannaway and Harden2004).

Psychiatry in the United States, which had been enriched by funding from the Veterans’ Administration at war's end (Friedman Reference Friedman1990), soon also saw major funding from the newly established National Institute for Mental Health (Scully et al. Reference Scully, Robinowitz, Shore, Menninger and Nemiah2000). The training programs funded by Washington fueled an extraordinarily rapid expansion in the numbers of psychiatrists and in academic psychiatry while specialties such as neurology and neuropsychiatry declined. And the dramatic changes in the size of the profession were matched by an equally dramatic shift in its intellectual orientation, and in the locus of psychiatric practice. The dominance of psychoanalytically-oriented practitioners – intellectually, politically, and eventually numerically – was immediately signaled by the election of William Menninger as president of the American Psychiatric Association in 1948, triumphing over the candidate of the old guard, Charles Burlingame, superintendent of the luxurious private mental hospital, The Institute of Living (Plant Reference Plant2005).

Whereas virtually all psychiatrists practiced in mental hospitals in the 1930s, as early as 1947 a majority had moved to outpatient work, and by 1948 only 16 per cent practiced their trade in traditional state hospitals. Numbers had grown from 2,200 pre-war to more than double that in 1948, and more than 27,000 by 1976 (Hale Reference Hale1995, 246). Virtually every major medical school had placed a psycho-analyst or someone sympathetic to psychoanalysis at the head of its department of psychiatry, and virtually all the best students entering the profession undertook training analyses at the local Psychoanalytic Institute. The duds of the profession, and those who failed to complete a full residency, were left to colonize the increasingly stigmatized and neglected mental hospital sector where, alongside foreign-trained physicians who were otherwise ineligible to practice medicine in the United States, they practiced a form of psychiatry their superior brethren dismissively referred to as “directive-organic psychiatry” – i.e., a second rate type of practice that continued to use some of the somatic therapies introduced earlier (Hale Reference Hale1995, 246–248, 253–256; MacIver and Redlich Reference MacIver and Redlich1959).

The interest the war had stimulated among some psychologists in clinical work, and the sheer size of the problems posed by mental illness, encouraged the new federal bureaucracy to expand its subsidies for training clinicians to include psychologists (Herman Reference Herman1995, 84, 92, 94). Psychometrics, which had a long pre-war history and enjoyed considerable legitimacy, attracted some of the available funding (Danziger Reference Danziger1990). So far as clinical psychology was concerned, although there were one or two brief and somewhat feeble attempts to organize the new specialty along psychoanalytic lines, most notably at the Menninger Clinic, the fierce insistence of the Menningers and of other analysts on subordinating clinical psychologists rankled (Friedman Reference Friedman1990, 16–17; Grob Reference Grob1991, 107–108), and soon enough those launching the new profession created an alternative model, one linked tightly to the newly emerging knowledge factories, research universities, and embracing the laboratory as the source of knowledge and legitimacy.

Backed by federal funding, a well-attended conference was organized in August and September 1949 in Boulder, Colorado, on how to organize training for clinical psychology. Consensus was soon reached among the participants on a plan that its leaders welded together: training in academic psychology, statistics, and research methodology combined with clinical experience (Baker and Benjamin Reference Baker and Benjamin2000; Farreras Reference Farreras, Pickren and Schneider2005). The plan was built upon an approach developed by a working group the American Psychological Association had set up in 1947 to develop a detailed scheme for training in clinical psychology. Its central component was a four-year university-based doctorate that incorporated both training in academic psychology and research methods, and an internship in a clinical setting (Committee on Training in Clinical Psychology 1947). Importantly, this approach avoided an entirely separate training program for applied and clinical psychologists, and gave the academic core of psychology an incentive to collaborate with the new plans. Since the experimentally and laboratory-based discipline had always marginalized psychoanalytic perspectives on the human psyche, it also helped to ensure that the therapeutic approaches that would characterize the new clinical psychology would likely rest on very different intellectual foundations. Instead, behaviorally-oriented, laboratory tested standardized therapies would become the foundation for new forms of practice, centered upon the limited goal of treating symptoms, and eschewing the ambitious (and hard to measure) aim of restructuring the entire personality. By insisting on strong theoretical and research components in the training of neophytes, those laying down the parameters of the new clinical psychology were also asserting that its practitioners would be “scientists,” not mere technicians. The statistical and experimental prowess built in to their training program (and disdained by their psychoanalytic rivals), would prove a significant advantage over time, not least in the competition for federal research dollars.Footnote ⁶

The flow of federal dollars induced the universities to embrace the new model, ensuring a rapid expansion of the whole psychological enterprise. Moreover, as government support of professional training was increasingly complemented by an exponentially increasing budget for research in an environment where the federal paymasters defined the knowledge “relevant” to the mental health problems of the nation extremely broadly and laxly, clinical psychologists were able to capture the lion's share of that money. The very nature of the psychoanalytic perspective on mental illness, and its practitioners’ focus on the peculiarities of the individual case made them strikingly unsuccessful in the grantsmanship game. Psychologists enjoyed an ever-increasing competitive advantage as they developed more elaborate research protocols and analytic techniques, and consolidated their membership on the relevant study sections. And they generated a series of short-term interventions (collectively many of them sheltering under the rubric of cognitive-behavioral therapy) which carried both the imprimatur of laboratory science and statistical confirmation of their success in suppressing symptoms (Brand and Sapir Reference Brand and Sapir1964).

In the army, and in hierarchical bureaucratic organizations like traditional mental hospitals, always headed by someone with medical qualifications, subordination of the social scientists could always be assured. But as the bulk of psychiatric practice moved outside the institutions, the competition between these two rival forms of psychotherapy – psychoanalysis and the very different version developed by clinical psychologists – grew more heated. To be sure, psychoanalysts had the power and status that an M.D. degree conferred, but psychoanalysis, terminal and interminable, was directed at an astonishing array of psychological issues, and questions about the efficacy of their lengthy interventions were queries the analysts were ill-equipped to answer with compelling data. Marginalized by their medical brethren who saw talk as an odd sort of approach to the treatment of disease, their forms of psychotherapy were in some respects poorly placed to cope with the competition from their psychologist colleagues, whose alternatives were more finite and more focused (hence attractive to insurance companies who increasingly were providing limited coverage for mental health problems),Footnote ⁷ and whose prices (for theirs was a female-dominated and thus, as labor market economists would confirm, inevitably lower-paid profession) were distinctly cheaper. Psychoanalysts had elected to keep tight control over who was allowed to join their guild, but their decision to separate their training institutes from the legitimacy conferred by association with a university was a further handicap. So even as psychoanalytic psychiatry expanded, it did so upon increasingly shaky foundations. It faced in clinical psychology a competitor for ambulatory and affluent patients – far more desirable than the indigent and grossly disturbed psychotic patients who had traditionally thronged the wards of mental hospitals – that constituted a growing threat.

And growing its rival most certainly was. Prior to World War II, not a single PhD trained American psychologist would have identified as a clinical psychologist: the set was essentially an empty one. By war's end, membership in the American Psychological Association had risen but slightly, from 2,739 in 1940 to 4,173 in 1945. But fifteen years later, it had risen to 18,215, and a National Science Foundation survey taken just four years after that suggested that more than two-thirds of psychologists with a doctorate were working in the mental health field. The clinical tail was wagging the academic dog – a problem that had plagued psychiatry in an earlier era. By 1990, there were 63,000 clinical psychologists in practice, a number that has since doubled again.

These were ominous developments for psychiatry, but the threat they posed has largely been alleviated by the third major shift in the profession's ideological foundations, one that saw essentially an abandonment of psychotherapeutics by psychiatrists within a quite remarkably short space of time. That changed orientation was, I shall suggest, not for the most part driven by concerns about competition from clinical psychology, though it did a great deal to nullify or at least sharply mitigate that concern. And ironically, it for the moment has proved so successful that it has forced some segments of organized clinical psychology to attempt to breach one of the most entrenched monopolies our society has granted to a single group of professionals, sole control over the right to authorize the use of prescription drugs. They, too, these psychologists contend, ought to be licensed to prescribe psychotropic medications.

Both directly and indirectly, innovations in drug treatment for mental illnesses were one of the crucial sources of changed perspectives on the psychiatric patient's treatment. But as the timing of the intellectual revolution demonstrates, there was nothing automatic about its arrival. The advent of the psycho-pharmaceutical revolution dates back to the approval by the Food and Drug Administration (FDA) of Dilantin for the control of epileptic seizures in 1953,Footnote ⁸ and of the first so-called anti-psychotic medication, Thorazine,Footnote ⁹ in 1954, just over a quarter century before psychoanalytic hegemony abruptly collapsed. Millions of patients were taking the new magic potions, soon joined by such “minor” tranquillizers as MiltownFootnote ¹⁰ and Valium,Footnote ¹¹ during the very decades when psychoanalysis enjoyed unparalleled dominance within the psychiatric profession. Drugs and talk therapy somehow co-existed for years. And then they didn’t.

The initial response of many psychoanalysts to the new compounds was to reject them, rather as they rejected the symptomatic treatments proffered by the clinical psychologists: they were both band-aids that treated surface symptomatology while leaving the deeper personality disorders that allegedly underlay mental distress untouched. Not having major clinical responsibilities for the institutionalized psychotic patients, this was a stance that endured for a time, but as the drugs made further inroads, an alternative tack became popular among many psychoanalysts: drugs were a useful adjunct, a way of reducing florid symptomatology so that the “real work” of analysis could proceed (Sawrer-Foner Reference Sawrer-Foner1960). It was a conceit that worked for a time, but it was inherently unstable.

Problems arose on a number of levels. One of the most significant was something the psychoanalysts were reluctant to take seriously because in their world-view it scarcely mattered, but that became increasingly hard to ignore in the late sixties and early seventies: the question of diagnosis. Focused as they were on what they saw as the precise psychodynamics of the individual case, analysts saw diagnostic labels as essentially an irrelevance, artificial creations that added nothing of substance to the understanding of the patient's problems or to the treatment process. Karl Menninger (Reference Menninger1963) had even taken this skepticism to its logical conclusion, calling for the abolition of diagnosis in psychiatry altogether, on the grounds that it was not just a charade, but actively harmed patients by labeling them. Among analysts, it probably didn't help matters that the dominant psychiatric nosology to that point mostly derived from the work of Emil Kraepelin, someone Freud had regarded with contempt and referred to scornfully as “the Great Pope of psychiatry” (quoted in Makari Reference Makari2008, 236). To be sure, there was an official guide to diagnosis, promulgated by the American Psychiatric Association, which had appeared in two editions, in 1952 and 1968.Footnote ¹² They were slight documents, barely over a hundred pages long, and containing around the same number of separate diagnoses, each with a purported psychodynamic etiology attached. Beyond a broad distinction between psychoses and neuroses, there was little attempt at precision or operationalizability, and the manuals were lightly regarded and seldom consulted. When efforts were made, as by the Philadelphia psychiatrist Aaron Beck (Beck Reference Beck1962; Beck, Ward, Mendelson, Mock, and Erbaugh Reference Beck, Ward, Mendelson, Mock and Erbaugh1962), to examine the reliability of professional diagnosis using this template, the results showed that psychiatrists regularly disagreed amongst themselves to a quite extraordinary extent. A systematic cross-national study comparing diagnostic practices and labels in Britain and the United States by John E. Cooper and associates appeared in 1972 and made the point much more forcefully. Schizophrenia was diagnosed far more frequently in the United States, as was manic-depressive psychosis in Britain, and compelling evidence demonstrated that this outcome was the artifact of two very different psychiatric cultures. American psychiatrists “saw” schizophrenia where their British counterparts were convinced the problem was manic-depressive illness.

What was one to make of diagnostic practices that produced such contradictory results? How could one have confidence in the scientific legitimacy of a profession so demonstrably unable to agree on what was wrong in an individual case or indeed whether anything at all was amiss? The threat to psychiatry's legitimacy was patent and soon became much worse. A Stanford psychologist, David Rosenhan, conceived of a novel way to dramatize psychiatry's diagnostic incapacity, sending a series of pseudo-patients to local mental hospitals with instructions to report limited auditory hallucinations, but otherwise to behave completely normally. All were diagnosed as either schizophrenic or manic depressive, at 12 different mental hospitals, and all were admitted and medicated. Despite behaving normally thereafter, it took between 7 and 52 days for them to be released, which only took place after they admitted being ill. Publishing his findings in Science and concluding that they were an indictment of psychiatry's failings (Rosenhan Reference Rosenhan1973), Rosenhan's study drew enormous media attention and, fierce rejoinders from some psychiatrists notwithstanding, contributed to public cynicism about the profession's scientific standing.

Soon criticisms of psychiatry's competence were not confined to the court of public opinion. The civil rights movement of the 1960s had helped bring into being a public interest bar interested in rights issues, and in the 1970s, some of these lawyers became active in mental health issues, suing for patients’ rights to treatment, then for the right to refuse treatment; against “institutional peonage” (the use of patients to perform hospital labor without pay, under the guise of “work therapy”; and for the right to challenge involuntary commitments for mental illness in court. The profession's diagnostic incompetence was regularly cited as a way to impeach psychiatrists’ claims to expertise, with one law review article comparing the reliability of their testimony to “flipping coins” (Ennis and Litwack Reference Ennis and Litwack1974).

Pressures to amend this unsatisfactory situation were emerging from another quarter as well – from the ranks of the pharmaceutical houses who were anxious to expand the market for drugs for psychiatric disorders, which had proved enormously lucrative. For bringing new drugs to market required data from controlled clinical trials to demonstrate “efficacy,” or at least some degree of improvement on a rating scale versus placebo. That in turn meant that the patients had to share significant characteristics that warranted lumping them together into a treatment and a control group, difficult to accomplish so long as the lack of reliability and reproducibility of diagnoses made such identifications uncertain. As it became apparent that patients responded variably to particular chemical compounds, drug companies also sought to isolate what the responders shared in common, since that would allow market segmentation and the development of separate lucrative markets for their wares. Absent stable diagnoses, it was difficult, often impossible, to replicate prior findings. And if the range of mental disturbances could be carved up, drug companies would have a whole series of targets to aim at when trying to demonstrate the value of new compounds, and psychiatrists themselves some sort of rationale for the prescriptions they wrote. Looming in the background, as well, was the knowledge that a new version of the so-called International Classification of Diseases (the ninth edition usually referred to as ICD-9) was being prepared under the auspices of the World Health Organization, prompting concerned bureaucrats at the American Psychiatric Association to worry about harmonizing American diagnostic practices with this document (Shorter Reference Shorter2009, 154). (The anxiety would turn out to be misplaced. It turned out that, with the power of the pharmaceutical industry behind them, the versions of the American Psychiatric Association's Diagnostic and Statistical Manual, from the third edition onwards, would become the hegemonic documents, relied upon in most of the world.)

The move to create more reliable diagnoses was thus over-determined. That did not mean that most psychoanalysts embraced it. Instead, they largely ignored the task force that the American Psychiatric Association assembled to address this issue. Only one of their number, John Frosch, joined the process, and soon marginalized, he stopped attending. The re-vamping of the diagnostic schema thus fell by default into the hands of biologically-inclined psychiatrists, led by Robert Spitzer of Columbia University. The upshot, ratified after a brief struggle in which Spitzer and his allies comprehensively out-maneuvered the analysts was a vastly expanded new edition of the Diagnostic and Statistical Manual. Gone were all the speculative psychodynamic etiologies for the comparatively small numbers of mental disorders that editions one and two had recognized. Gone was the distinction between neuroses and psychoses, as well as such illnesses as “hysteria” that had hitherto formed the bread and butter of psychoanalytic practice. Protests from the analysts led to some minor modifications to the scheme (parenthetical references to alternative labels preferred by the Freudian fraternity), but these were purely tactical concessions to quiet the critics, rapidly abandoned in subsequent editions of the manual. (Interestingly, the decision by mainstream psychiatry to abandon the word hysteria was not echoed elsewhere. The category persisted, for example in psychometrics, with technologies like the MMPI maintaining references to hysteria.)

Treatment of symptoms had been scorned by psychoanalysts as beside the point: these were just the visible signs of the underlying disorders of the personality that needed attention. Treating symptoms, they insisted, was like playing “whack a mole,” so long as the underlying defects of personality were left untouched. Now, on the contrary, they became the very markers of disease, the key to deciding what ailed the patient one was confronting. Unable to demonstrate convincing chains of causation for any major form of mental disorder, Spitzer and his team opted to abandon any pretense of doing so. Instead, they devised ways of maximizing inter-rater reliability, creating a “tick the boxes” approach to diagnosis that ensured that psychiatrists examining a particular case would agree on what was wrong. Facing a new patient, the practitioner would record the presence or absence of a given set of symptoms, and once a certain threshold had been reached (six out of ten symptoms in the case of schizophrenia), a diagnostic label was applied, with co-morbidity invoked to explain those cases where a patient qualified for more than one diagnosis.

Disputes about what belonged in the manual were resolved by committee vote, as was the arbitrary decision about where to situate cut-off points: i.e., how many of a laundry list of symptoms a patient had to have before he or she was declared to be suffering from a particular form of psychiatric illness. Questions of validity were simply set to one side. If diagnoses could be rendered mechanical and predictable, consistent and replicable, that would suffice (for detailed documentation of this point, see Kutchins and Kirk Reference Kutchins and Kirk1997). Latent in this approach was the possibility, soon realized, of a vast increase in the range of conditions that could be redefined as pathology and thus brought within the ambit of psychiatric intervention.

Despite the obvious flaws that marked the enterprise from the outset, DSM III was a political and “technological” triumph. Its publication in 1980 marks the third and final shift in the orientation of psychiatry, and it is one that persists today, despite some increasing signs of strain (Paris Reference Paris2005; Frances Reference Frances2013; Spitzer Reference Spitzer2009; Spitzer, Williams and Endicott Reference Spitzer, Williams and Endicott2012). Its appearance led to a conceptualization of mental illnesses, by the profession and public alike, as specific, identifiably different diseases, each amenable to treatment with different drugs (Healy Reference Healy2002). Despite the absence of any plausible claim to link its labels to physical phenomena, the artificially created stability of its categories was invoked to render these entities “facts” about the world, features of nature itself, and to suggest that such stable entities must rest upon biological foundations. Neo-Kraepelinian psychiatry, as this new approach soon labeled itself, was biologically reductionist, a point made forcibly when President George H. W. Bush proclaimed the 1990s “the decade of the brain” (Presidential Proclamation 1990). With quite remarkable rapidity psychoanalysis was cast aside, defenestrated from its privileged position in departments of psychiatry where neuroscientists and psycho-pharmacologists replaced its acolytes.

The rapid triumph of the new approach was crucially aided, I suggest, by two major factors besides the fact that it provided a “solution” to a source of profound professional embarrassment. On the one hand, it proved immensely attractive to the ends of the medical insurance industry, for whom it provided stable guidelines to a territory it was beginning to cover, and one whose uncertain boundaries and therapeutic miasma had hitherto proved profoundly threatening. If the means for regularizing the assignment of illness could be linked to standardized patterns of treatment, then uncertainty could be replaced by predictability, and finite limits to insurance coverage set, an important end from the insurers’ point of view. Very quickly, therefore, reimbursement for psychiatric care came to be tied to use of the DSM's categories: no diagnosis, no payment. The DSM thus became a technology impossible to ignore and impossible not to validate, if only by employing its categories. Moreover, as the manual passed through subsequent revisions, always expanding in size and in the range of conditions now alleged to constitute medical conditions, another subtle shift occurred, one underlining the importance of the second factor that underlay the triumph of the neo-Kraepelinians. For where once the disease had led to the treatment, soon the treatment led to the disease (Rosenberg Reference Rosenberg2006). If only a subset of patients responded to a particular form of treatment, why they must be suffering from a separate kind of mental disorder, so a new “illness” emerged, one created in the marketing departments of Big Pharma (Healy Reference Healy2008, 198–244).Footnote ¹³

For the other force behind the stabilization of the psychiatric profession around DSM III and its successors were the intimate links that developed between specific diagnostic categories and specific forms of drug treatment. That patients (allegedly and sometimes in reality) responded to particular classes of drugs was cited as “proof” that they suffered from a particular form of illness. And that the drugs had effects on the course of mental illness was simultaneously “proof” that mental illness was rooted in the brain. Why else would the pills work, if they were not manufacturing normal brains? After the fact, therefore, accounts were constructed about why a particular class of drugs worked: phenothiazines combated schizophrenia because they affected dopamine levels in the brain: or Prozac and other Selective Serotonin Re-uptake InhibitorsFootnote ¹⁴ relieved depression because depression was rooted in a shortage of seratonin in the brain. These were scientific fairy-tales, ones swiftly discredited (Healy Reference Healy2002, 207–224; Lacasse and Leo Reference Lacasse and Leo2005). But they were retailed to the public by professionals and science journalists alike so plausibly and repeatedly that many embraced the portrait of mental illnesses as biologically rooted events, probably dependent upon disturbances in brain bio-chemistry.

Not least, for the families of the mentally ill, whom the psychoanalysts had implied bore primary responsibility for the disorders of their offspring (refrigerator mothers being particularly at fault), the neutral language of biology provided a welcome alternative, and newly formed pressure groups like NAMI (the National Alliance on Mental Illness) became fierce proponents of the new orthodoxy (their value recognized in very tangible form by the pharmaceutical industry, which provided them with large hidden financial subsidies). In 1999, for example, Mother Jones reported that 18 drug companies had given NAMI at total of $11.72 million between 1996 and mid-1999. Eli Lilly, which had contributed $2.87 million during that period, gave $1.1 million in 1999 alone. In addition, it “loaned” NAMI one of its executives, Jerry Radke, to provide the group with advice on “strategic planning” (Silverstein Reference Silverstein1999). Eight years later, the Wall Street Journal indicated that Lilly continued to fund NAMI to the tune of $544,500 in 2007, and that GlaxoSmithKline had given more than $12 million to similar groups in Europe (Johnson Reference Johnson2007; and, more generally, Healy Reference Healy2004, chap. 4).

The assertion that psychoses represent forms of brain disease persists, though its evidentiary basis is weak, and rests more on programmatic claims and plausible hunches, than on demonstrable evidence. Baldessarini's conclusion from almost a decade ago still summarizes the basic state of play:

The tight linkages formed between the new version of psychiatric knowledge and the pharmaceutical industry reflected the mutually beneficial outcomes this provided for both professionals and industry. Psychoanalysts had a terrible track record of attracting large external funding. Less than 5 per cent of the NIMH research budget came their way even during the period of their hegemony (Scull Reference Scull2010, 35). They were not trained as grantsmen, and they scorned the large-scale research enterprises that the federal bureaucrats preferred. But in the era of Big Science and Big Medicine, research universities became addicted to the flow of research funding – indeed their very existence depended upon these vast new flows of money, from Washington and elsewhere. Psychoanalysts were already suspect for their emphasis on sex and talk. But academic psychiatrists with psychoanalytic inclinations were doubly marginalized in the knowledge factories because they failed to bring in the coin of the realm: research dollars. By contrast, the new biologically-oriented, pharmacologically entranced type of psychiatrists soon became the darlings of medical school deans, for large sums flowed in to underwrite their work. And those sums materialized because drugs for psychiatric conditions were among the most profitable products manufactured by the global pharmaceutical industry.

Prozac, the first of the blockbuster SSRIs, reached the American market for the first time in December 1987. In its first full year on the market, it had sales of $125 million, rising to $350 million in 1989 – and this exponential growth pattern spread throughout the entire market for psychotropic medications. Between 1986 and 1991, world sales rose from $2 billion to $4.4 billion. By 2001, three of the top 10 drugs in the United States, by far the most lucrative drug market in the world, were SSRIs – Zoloft, Paxil, and Prozac. And sales continued to grow apace, though the “science” that underpinned their popularity was by now thoroughly discredited, the symptomatic relief they provided increasingly doubtful, and the side-effects of taking such pills a source of growing concern. Yet the anti-depressant market – largely a market for SSRIs – continued to boom, with sales rising from $5.1 billion in 1997 to $12.1 billion only five years later (Shorter Reference Shorter2009, 198–199).

More generally, for psychiatrists the neo-Kraepelinian synthesis provided a new grounding for psychiatric authority and a less time-consuming and more lucrative foundation for psychiatric practice. Fifteen minutes spent prescribing pills was far easier and such a pattern of practice was far more rewarding financially than an hour spent providing psychotherapy (Harris Reference Harris2011). The result has been a rapid abandonment of what had been the lynch-pin of out-patient psychiatric practice, and the ceding of that portion of the marketplace to the cheaper competition: clinical psychologists and psychiatric social workers. Psychiatric training has de-emphasized psychotherapeutics, and little remains of an earlier form of office-based psychiatric practice (Luhrmann Reference Luhrmann2000). Though this has opened up a large space for the psychologists to colonize, they have done so under the shadow of the enormous influence of the DSM, whose categories they have been forced to embrace as the price for being paid, though it is their competitors the psychiatrists who uniquely control this instrument. In parallel fashion, psychiatrists outside North America have also found themselves constrained to adopt its ways of categorizing the objects they treat, such has proved its hegemonic power, not least because of the tight linkages between its categories and the drugs the profession now depends upon. For psychologists and clinical social workers, moreover, the links to pharmacology highlight still another concern: to the extent hoi polloi have come to embrace the notion that mental illnesses are ultimately biological phenomena, the product of faults in the brain, and more specifically faults in its biochemistry, they have also absorbed the lesson that the weapons of choice in battling their depredations must accordingly be chemicals that purport to rectify such imbalances. Yet because the medically qualified monopolize the power to prescribe such potions, here is another way in which psychologists subtly find themselves relegated to a second-class, subordinate status in the psy complex.

In place of what Philip Rieff once called “the triumph of the therapeutic” (Rieff Reference Rieff1987), we thus seem to be living in an epoch in which wholly new values and forms of knowledge have come to the fore. In the aftermath of the psychopharmacological revolution and the immense wealth it spawned, modern laboratory science has helped to hide the banality of the intellectual foundations on which this whole complex edifice rests, a set of political deals and compromises hatched in committees and presented to the outside world as “wie die Dinge sind,” how things are. But that foundation is not just banal, it is a ramshackle Rube Goldberg apparatus,Footnote ¹⁵ one threatened with collapse from two directions. On the one hand, the drugs revolution has been vastly oversold. The problem is that not only are anti-psychotics and anti-depressants not psychiatric penicillin, but that the symptomatic relief that these drugs provide to some (very far from all) of those to whom they are prescribed often comes at a very heavy price indeed. It took psychiatry two decades and more (Crane Reference Crane1973) to acknowledge the iatrogenic price schizophrenics paid in return for whatever benefits flowed from taking phenothiazines like Thorazine (and for many psychotics the “benefits” were wholly illusory). And despite decades of research, attempts to ward off or treat such stigmatizing and distressing iatrogenic disorders as tardive dyskinesia and motor control problems have proved fruitless. That, plus the desire of drug companies to reap new monopoly profits from a new generation of drugs, so-called atypical anti-psychotics like Risperdal, Clozaril, and Zyprexa,Footnote ¹⁶ has prompted shifts in prescribing practices and some associated changes in the typical side effects patients experience, towards heightened risks of metabolic disorders, rapid weight gain, diabetes, heart disease, and premature death (which is not to say that neurological complications disappear when the new drugs are substituted for first generation drugs). As a recent commentary in The Lancet put it: “As a group they are no more efficacious, do not improve specific symptoms, have no clearly different side effect profiles than the first-generation anti-psychotics, and are less cost effective. The spurious invention of the atypicals can now be regarded as invention only, cleverly manipulated by the drug industry for marketing purposes and only now being exposed” (Tyrer and Kendall Reference Tyrer and Kendall2009, 4).

Similarly, SSRIs like Prozac, initially hailed as making depressed patients “better than well,” are now facing mounting questions about their efficacy and the range and seriousness of the side effects associated with their use. Their supposed mechanism of action has long been exploded (drugs that lower levels of seratonin in the brain have been shown to be as efficacious as those that raise it, for instance). And over the past two decades, very serious questions have been raised about the “science” on which the approval of these drugs was based. Drug companies now “own” the data on the clinical trials, and there is mounting evidence that they suppress the trials that produce results they don't like, and manipulate the data that do see the light of day in disreputable fashion (DeAngelis and Fontanarosa Reference DeAngelis and Fontanarosa2008). Articles in major and minor peer reviewed journals are often ghost written by Big Pharma employees, while the names of major academics appear on the relevant papers (Healy Reference Healy2012, 104–113). Evidence-based medicine turns out to rest on evidence not worthy of the name. And once a drug has secured regulatory approval as useful for one type of illness, the pharmaceutical industry has directly and indirectly conspired to expand its markets by encouraging so-called “off-label” uses – prescribing the drug as useful for other disorders when no solid scientific evidence exists that this is the case (Landefeld and Steinman Reference Landerfeld and Steinman2009). As class action law-suits have led to the release of documents that demonstrate just how routine these practices have become, and as some of the newly released data have shown that, perversely, so-called anti-depressants may heighten the incidence of suicidal behavior, so questions have arisen, not just about the rapacity of the industry, but also about the ethics of a profession that has collaborated in these behaviors (Healy Reference Healy2012).

If the links to psychopharmacology are now proving a double-edged sword, the construction of the nosological Bible is also coming under increased criticism. The very arbitrariness of the foundations on which it was built put no obvious obstacles in the path of revising it, and pressures to do so came from many quarters. One significant sort of pressure is one familiar to students of psychiatric history: the tendency of the profession to widen the range of behaviors that fall within its ambit. Various forces drive this process. Just to cite a few: demands from potential patients or their families, who seek relief from problems and see medical labels as the route to resources (something that plagues those professionals who subsequently attempt to reign in diagnostic imperialism because that threatens to curtail resources flowing in particular directions); the attractions of expanding professional markets, and perhaps, by including more marginal cases, finding more readily treatable objects; and the constant efforts by the drug industry to pathologize behaviors and expand demand for lucrative new products (Horwitz and Wakefield Reference Horwitz Allan and Wakefield2012).

Every few years, therefore, we have seen revisions to the DSM, each presented as “fixing” the flaws of the previous edition and incorporating new layers of understanding. The upshot is a manual that has expanded like the Yellow Pages on steroids, from the 104 pages of DSM I to the 992 pages of DSM IV TR (text revision). DSM III, published in 1980, became DSM IIIR (revised) in 1987, which in turn begat DSM IV in 1994, and DSM IV TR (Text Revision) in 2000. Not only has the range and variety of “illnesses” one can potentially suffer from grown dramatically with each version, but the definitions of particular disorders has shifted dramatically over time, and often in ways that have produced soaring increases in the numbers of those “suffering” from a particular condition. Following changes to the definition of juvenile bipolar disorder in 1994, the number of patients with this diagnosis grew forty-fold in a decade (Blader and Carlson Reference Blader and Carlson2006; Harpaz-Rotem and Rosenheck Reference Harpaz-Rotem and Rosenheck2004; Moreno et al. Reference Moreno2007). An autism epidemic broke out, and a previously rare condition, seen in only one in every 500 children in 1994 was found in one of every 90 children only 10 years later. Hyperactivity as a diagnostic category has similarly seen an explosive growth in numbers, and an associated sky-rocketing in the prescribing of drugs that purport to control it. Among adults, one in every 76 Americans qualified for welfare payments based upon mental disability by 2007. After a certain point, such data begin to strain credulity.

If psychiatrists’ inability to agree among themselves on a diagnosis threatened to make them a laughing-stock in the 1970s, the relabeling of a host of ordinary life events as psychiatric pathology now seems to promise more of the same: oppositional defiant disorder, school phobia, narcissistic and borderline personality disorder seem poised to be joined by such things as pathological gambling, binge eating disorder, hypersexuality disorder, temper dysregulation disorder, mixed anxiety depressive disorder, minor neuro-cognitive disorder, and attenuated psychotic symptoms disorder. That these diagnoses provide lucrative new markets for psychopharmacology's products raises questions in many minds about whether commercial ends are illegitimately driving the expansion of psychiatric means – a concern that is scarcely allayed when one recalls that great majority of the members of the DSM Task Force are recipients of drug company largesse.

Relying solely on symptoms and behavior to construct its illnesses, and on organizational fiat to impose its negotiated categories on both the profession and its public, psychiatry is now facing a revolt from within its own ranks. The general editors of both the third and the fourth editions of the manual (Robert Spitzer Reference Spitzer2009 and Allen Frances Reference Frances2009 respectively) have been campaigning loudly against the proposed revisions, accusing their authors of producing a text based on hasty and unscientific foundations, of pathologizing normal features of human existence, and of threatening to create new epidemics of spurious psychiatric diseases. Increasingly, the vast enterprise that is descriptive psychiatry seems only to survive because of the lack of a plausible rival. The frailty of the foundations of much that passes for the science and medicine of mind and brains has become ever more exposed once again, and the many professions involved once again confront, as they did so often in the past, profound questions about their legitimacy (Rosenberg Reference Rosenberg and Rosenberg1992).