Psychiatry’s Legitimacy Crisis

By Andrew ScullAugust 8, 2012

All We Have to Fear: Psychiatry's Transformation of Natural Anxieties into Mental Disorders by Jerome C. Wakefield and Allan V. Horwitz

ABOUT 40 YEARS AGO, American psychiatry faced an escalating crisis of legitimacy. All sorts of evidence suggested that, when confronted with a particular patient, psychiatrists could not reliably agree as to what, if anything, was wrong. To be sure, the diagnostic process in all areas of medicine is far more murky and prone to error than we like to think, but in psychiatry the situation was — and indeed still is — a great deal more fraught, and the murkiness more visible. It didn’t help that psychiatry’s most prominent members purported to treat illness with talk therapy and stressed the central importance of early childhood sexuality for adult psychopathology. In this already less-than-tidy context, the basic uncertainty regarding how to diagnose what was wrong with a patient was potentially explosively destabilizing.

The modern psychopharmacological revolution began in 1954 with the introduction of Thorazine, hailed as the first “anti-psychotic.” It was followed in short order by so-called “minor tranquilizers:” Miltown, and then drugs like Valium and Librium. The Rolling Stones famously sang of “mother’s little helper,” which enabled the bored housewife to get through to her “busy dying day.” Mother’s helper had a huge potential market. Drug companies, however, were faced with a problem. As each company sought its own magic potion, it encountered a roadblock of sorts: its psychiatric consultants were unable to deliver homogeneous populations of test subjects suffering from the same diagnosed illness in the same way. Without breaking the amorphous catchall of “mental disturbance” into defensible sub-sets, the drug companies could not develop the data they needed to acquire licenses to market the new drugs.

In a Cold War context, much was being made about the way the Soviets were stretching the boundaries of mental illness to label dissidents as mad in order to incarcerate and forcibly medicate them. But Western critics also began to look askance at their own shrinks and to allege that the psychiatric emperor had no clothes. A renegade psychiatrist called Thomas Szasz published a best-selling broadside called The Myth of Mental Illness, suggesting that psychiatrists were pernicious agents of social control who locked up inconvenient people on behalf of a society anxious to be rid of them, invoking an illness label that had the same ontological status as the label “witch” employed some centuries before. Illness, he truculently insisted, was a purely biological thing, a demonstrable part of the natural world. Mental illness was a misplaced metaphor, a socially constructed way of permitting an ever-wider selection of behaviors to be forcibly controlled under the guise of helping people.

The problem was exacerbated when some psychiatrists sought to examine the diagnostic process. Their findings dramatically reinforced the growing suspicion that their profession’s claims to expertise were spurious. Prominent figures like Aaron Beck, Robert Spitzer, MG. Sandifer and Benjamin Pasamanick published systematic data that dramatized just how tenuous agreement was among psychiatrists, even the most prominent ones, regarding the nature of psychiatric pathology; consensus barely exceeded 50 percent whether the subjects were patients in state hospitals or out-patient settings. And in 1972, a systematic study of diagnostic practices in Britain and the United States found massive differences: New York psychiatrists diagnosed nearly 62 percent of their patients as schizophrenic, while in London only 34 percent received this diagnosis. And, while less than five percent of the New York patients were diagnosed with depressive psychoses, the comparable figure in London was 24 percent. Further examination of the patients suggested that these differences were byproducts of the preferences and prejudices of each group of psychiatrists, and yet they resulted in consequential differences in treatment.

Nor was this chaotic situation hidden from a larger public. In the legal profession, the civil rights movement of the 1960s led to the emergence of public interest law. A number of these attorneys broadened their focus from race to include other stigmatized and disadvantaged populations. By the early seventies, this led to the creation of a mental health bar, two of whose prominent practitioners seized on the results reported in these studies. They intimated that psychiatrists should no longer be credited with the status of “expert witnesses,” since their judgments amounted to “flipping coins in the courtroom,” as they put it. Shortly thereafter, a cleverly designed study by a Stanford social psychologist, David Rosenhan, appearing in the august pages of Science, poured gasoline on the flames. Rosenhan had eight pseudo-patients (including himself) show up at a dozen psychiatric hospitals complaining they were hearing voices and uttering the words “empty,” “hollow,” or “thud.” The so-called patients otherwise presented their normal selves. Seven received the diagnosis of schizophrenia, the eighth was labeled manic-depressive, and all were hospitalized for terms as long as 52 days. The article garnered massive media coverage, made Rosenhan a star and made of psychiatry a hapless buffoon.

To address the embarrassment, one of the profession’s internal critics, Robert Spitzer of Columbia University, persuaded the American Psychiatric Association to authorize the development of a new diagnostic manual. The document he and his Task Force produced, approved and published in slightly modified form in 1980 as the third edition of the Diagnostic and Statistical Manual of the American Psychiatric Association (DSM III for short) launched a revolution in American psychiatry whose effects are still felt today. Versions III R (revised), IV, and IV TR (text revision) and DSM 5 (to be released in 2013) have been produced with numbing regularity. The advent of DSM III and its descendants constitute the backdrop to the argument presented in the new book by Allan Horwitz and Jerome Wakefield, All We Have to Fear: Psychiatry’s Transformation of Natural Anxieties into Mental Disorders.           

Horwitz and Wakefield want to argue for the harmful impact of what is often called the neo-Kraepelinian revolution in psychiatry. Emil Kraepelin was the fin-de-siècle German psychiatrist who launched the fashion for descriptive psychopathology and first made the distinction between dementia praecox and manic-depressive illness. Horwitz and Wakefield suggest that the efforts of Kraepelin’s late-twentieth century successors to make psychiatric diagnoses more rigorous and predictable have instead enabled psychiatric pathology to get out of hand. They identify two problems: the psychiatric profession’s obsession with simplistic, symptom-based diagnoses, and the looseness of its criteria for defining mental states as pathology. All sorts of anxieties that are in reality part of the normal range of human emotion and experience have been transformed by professional sleight of hand into diseases. The upshot, they contend, is that whereas thirty years ago less than five percent of Americans were thought to suffer from an anxiety disorder, nowadays some widely cited epidemiological studies have decreed that as many as 50 percent of us do so.

Horwitz and Wakefield are scarcely the first scholars to suggest that rising rates of mental illness are a reflection of the widening and loosening of diagnostic schema. Three decades ago, the British psychiatrist Edward Hare and I engaged in a vigorous debate on this issue in the pages of the British Journal of Psychiatry. He argued that the growing number of lunatics in Victorian museums of madness were victims of a new viral disease, schizophrenia, and I countered that it was more probable that other factors were at work — namely, the amorphousness of nineteenth century definitions of madness, the decreasing willingness and ability of families to cope with difficult or impossible relations, and the eagerness of psychiatrists to enlarge their sphere of operations. Of more contemporary relevance, a range of commentators have noticed the explosive growth of depression as a diagnosis, to the point where it is now frequently termed ‘the common cold’ of psychiatry; the equally dramatic expansion in the number of children being diagnosed with ADHD; the appearance out of nowhere of juvenile bipolar disorder, which apparently became forty times as common between 1994 and 2004; the epidemic of autism, a formerly rare condition afflicting less than one in five hundred children in 1990, which has now mushroomed into a disease found in one in every ninety children. More than a few scholars have been tempted to attribute these seismic shifts not to any real alteration in the numbers of sufferers from these disorders, but to disease-mongering by the psychiatric profession and by Big Pharma, the multi-national pharmaceutical industry that obtains a huge fraction of its profits from the sale of drugs aimed at mental disorders of all sorts.

Among the most zealous critics of the expanding psychiatric empire have been two unlikely souls: Robert Spitzer, the principal architect of DSM III, and Allen Frances, who played a similarly large role in the construction of DSM IV. As the latest edition of that tome, the largest thus far and the most delayed, struggles to be born, those assembling it have been assaulted by Spitzer and Frances for creating a version built on hasty and unscientific foundations; they claim it pathologizes everyday features of normal human existence, and that, like its predecessors, it will create new epidemics of spurious psychiatric illness. Allen Frances, in particular, has taken to uttering frequent mea culpas, taking the blame for loosening the criteria for diagnosing autism in DSM IV, and thus, so he claims, sowing fear and mislabeling thousands and thousands of children.

Before focusing on Horwitz and Wakefield’s contribution to this debate, it is worth acknowledging that Spitzer and Frances’s claims have proven to be highly controversial. Not unexpectedly, given the huge revenue the American Psychiatric Association rakes in from each edition of its manual, and the centrality of that book’s place to psychiatry’s claims to be a science, the oligarchs who run its operations have been swift to condemn the renegades. The oligarchs have launched a series of ad hominem attacks on the renegades’ motives and on the nature of their criticisms. Interestingly, equally fierce if not fiercer reactions have been manifested from an entirely different source: the relatives of those who have been diagnosed with ailments whose boundaries Spitzer and Frances want to shrink. Particularly vocal in online discussions have been the parents of children diagnosed with autism, for whom the loss of the label will mean being deprived of social services and support that is conditional on retaining that status. At times, the vituperation that has rained down on Frances’s head has been extraordinary — and indeed it’s hard not to form a mental image of families all across the country sticking pins into a Frances voodoo doll. Whatever other lessons are derived from this state of affairs, one point should be obvious: It is not just professional imperialism on the part of psychiatrists, nor the greedy machinations of Big Pharma, that explains the burgeoning problem of mental disorder in early twenty-first century America. And a burgeoning problem it is. To cite just one statistic[EM1] , one in every 76 Americans in 2007 qualified for welfare payments based on mental disability. As we examine Horwitz and Wakefield’s work on anxiety disorders, it is therefore important to bear in mind that theirs is just one piece of a larger puzzle. Indeed, the same authors have already examined another example of this phenomenon, the medicalization of sadness, and its transformation into pathology.

Horwitz and Wakefield rightly place the DSM in its various post-1980 incarnations at the center of their explanation of how we are to account for the massive growth in the numbers of people diagnosed with pathological anxiety. DSM III “solved” the legitimacy crisis that psychiatry faced in the late 1970s. As long as one employed its methods and categories, high levels of agreement among psychiatrists confronting the same case were all but assured. In that sense, psychiatric diagnosis became, as statisticians would put it, more reliable. How was that feat accomplished? By rendering the diagnostic process mechanical, employing a tick-the-boxes approach to deciding whether or not someone had a mental disorder, and if so, what disorder it was. Display any six out of ten symptoms, and voilà, a schizophrenic. Tick another set of boxes and you had General Anxiety Disorder (GAD), and so forth. A given patient might potentially have several “illnesses” at once, a problem alleviated by setting up a hierarchy of psychiatric diseases and awarding patients the most serious of them, or by creating a category called “co-morbidity” and thereby accepting the presence of multiple illnesses. The overlap in symptomatology between two schizophrenics with the “same” disease might be as few as two out of ten symptoms.

Why is psychiatry forced to rely on a grab bag of symptoms to make its diagnoses? Because, fundamentally, it has nothing else to offer. The cause of the overwhelming majority of psychiatric disorders remains as obscure as ever. Periodic weightless claims, endorsed by credulous science journalists, that schizophrenia is triggered by a newly discovered gene or by a dopamine deficiency in the brain, or that people suffering from depression have a shortage of serotonin, which can be reversed by taking a Selective Seratonin Reuptake Inhibitor (SSRI) such as Prozac to immerse their synapses in a serotonin bath, are so much biobabble ­­­— scientific nonsense that has proved good marketing copy for Big Pharma but is otherwise worthless.

This reliance on symptoms, and on the simplistic approach of counting symptoms to make a diagnosis, creates a bogus confidence in psychiatric science. Such categories have an element of the arbitrary about them. When Robert Spitzer and his associates created DSM III, they liked to call themselves DOPs (data-oriented persons). In fact, DSM’s categories were assembled through political horse-trading and internal votes and compromise. The document they produced paid little heed to the question of validity, or to whether the new system of categorizing mental disorders corresponded to real diseases out there. And subsequent revisions have hewed to the same approach. With the single exception of Post Traumatic Stress Disorder (PTSD), which, as its name implies, is a diagnosis having its origins in trauma of an extreme sort, the various categories in the DSM, including the anxiety disorders that preoccupy Horwitz and Wakefield, are purely symptom-based. (The construction of the PTSD diagnosis, incidentally, as the authors show, was every bit as political as the creation of the other DSM categories.) Because so much depends on the wording that describes the symptoms to be looked for and on how many symptoms one needs to display to warrant a particular diagnosis (why do six symptoms make a schizophrenic, not five, or seven?), small shifts in terminology can have huge real-world effects. The problem is magnified in studies of the epidemiology of psychiatric disorders. As Horwitz and Wakefield point out, to make studies of this sort cheaper and allow those producing them to employ laypeople to administer the necessary instruments, the diagnostic process is simplified even further in these settings. They write that psychiatric epidemiologists make “no attempt to establish the context in which worries arise, endure, and disappear so as to separate contextually appropriate anxiety from disordered anxiety conditions [and thus they] can uncover as much seeming psycho-pathology as they desire.”

By contrast, at least initially, psychiatrists were expected to exercise some independent clinical judgment when reaching their professional judgments. Being anxious and fearful is, under some circumstances, a natural and healthy human response to the world. How are we to distinguish between healthy or normal fears — perhaps even fears that are exaggerated but had their origins in an earlier period of our evolutionary history — and pathological forms of anxiety? Allow too much room for clinical judgment and the goal of standardizing psychiatric diagnosis goes away. Eliminate it and the anxieties that people naturally feel when they’ve survived a bad marriage, recovered from a serious disease, or lived through a war or a disaster like Katrina, are all-too-readily relabeled as illness. DSM attempted to cope with this problem by insisting that the anxiety had to be “excessive” and “prolonged,” six months in duration or longer, and to be perceived as “abnormal” or disabling by those subject to these emotions. These are inadequate and fallible correctives, but they did something to make it less likely that normal people would be called “mentally ill.” As the manual went through successive editions, however, and as its categories were simplified to make the job of epidemiologists easier and cheaper, the effect, as Horwitz and Wakefield argue, was steadily to enlarge the numbers of ordinary people drawn into the ranks of the mentally unstable, often to a spectacular degree. And because of the seemingly scientific basis of the labels, the consistency with which cases were diagnosed, and the translation of human judgment by means of this verbal alchemy into statistics, the multiplication of the anxious and nervous (as with other psychiatric categories) has proceeded in relentless fashion.

Through detailed analyses of the underlying terminological changes and their effects, Horwitz and Wakefield show how “social phobia” multiplied six-fold in the course of a decade. They document a similar pattern with PTSD, with Social Anxiety Disorder (SAD), and a whole variety of other anxiety disorders. Less satisfactorily, they make some attempt to link these developments to issues of professional imperialism, the financial interests of Big Pharma, and even the demands of patients and more especially of patients’ families, for whom a particular diagnosis may be the sine qua non of obtaining access to insurance payments and other forms of social support. Two other critics of the DSM, Kutchins and Kirk, have suggested that the looseness of its categories means that “the prevalence rates in the United States will rise and fall as erratically as the stock market.” To this comment, Horwitz and Wakefield add a rueful and all-too-accurate coda: “Kutchins and Kirk are only half correct. Prevalence rates in recent epidemiological studies go in only one direction: upward.”

All We Have to Fear is nonetheless a curate’s egg of a book. There are good bits and bad bits. Horwitz and Wakefield manage to make a strong case for the prosecution: Psychiatry has indeed lost its way and seems increasingly unable to resist pathologizing ordinary life. But before the reader gets to that case, he or she will have to plow through the seemingly endless and tedious pages of evolutionary psychology that make up the key sections of the book’s first three chapters. Here one finds claims about genetic endowments that were built into human beings at the time of cave-men and hunter-gathers, and persist as part of our mental constitution. These inheritances from the past are invoked to explain our contemporary fears and anxieties, even ones of quite specific sorts. The alleged features of normal human nature and the supposed hold our genes have over our behavior are as speculative as most neuro-maniacal accounts of modern man. More importantly, they are unnecessary, and get in the way of an argument that depends on no more than the self-evident proposition that all of us experience fears and anxieties, which are intensified in certain social situations and by large-scale trauma, but which cannot be termed “mental illnesses.”

Even setting that objection aside, the remainder of the book is heavy-going. Much of the discussion is wandering and repetitive. The same arguments are mobilized again and again, moving across only slightly varied terrain. What could have been a long article thus becomes a book of sorts — one that many readers will have trouble finishing. This is too bad, because contemporary psychiatry is on the brink of one of those periodic crises of legitimacy that have been so notable a feature of the profession’s history over the past couple of centuries; the story Horwitz and Wakefield recount helps us to understand one of the reasons why renewed turmoil threatens to engulf the psychiatric enterprise.


Recommended Reads:

LARB Contributor

Andrew Scull taught at the University of Pennsylvania and Princeton prior to becoming a professor at UCSD. His articles have appeared in leading journals in a variety of disciplines, including British Journal of PsychiatryPsychological MedicineLancetEuropean Journal of SociologyMedical HistoryVictorian Studies, and Stanford Law Review. He has held fellowships from (among others) the Guggenheim Foundation, the American Council of Learned Societies, and the Davis Center for Historical Studies, and in 1992-1993 was the president of the Society for the Social History of Medicine. His most recent books are Madness in Civilization: From the Bible to Freud, and from the Madhouse to Modern Medicine, and Psychiatry and Its Discontents. In the spring of 2022, Belknap Press/Harvard University Press will publish his new history of American psychiatry, Desperate Remedies: Psychiatry’s Turbulent Quest to Cure Mental Illness.


LARB Staff Recommendations

Did you know LARB is a reader-supported nonprofit?

LARB publishes daily without a paywall as part of our mission to make rigorous, incisive, and engaging writing on every aspect of literature, culture, and the arts freely accessible to the public. Help us continue this work with your tax-deductible donation today!