Diagnosing in the Dark
The continuing relevance of Thomas Szasz's assault on psychiatric pretensions
Half a century after Thomas Szasz first declared "there is no such thing as 'mental illness,'" his radical critique of psychiatry is widely viewed as outmoded and simplistic at best, cruelly dogmatic at worst. "The opinion of official American psychiatry," Szasz writes in the preface to the 50th anniversary edition of The Myth of Mental Illness, "contains the imprimatur of the federal and state governments. There is no legally valid nonmedical approach to 'mental illness,' just as there is no such approach to measles or melanoma.…Debate about what counts as mental illness has been replaced by legislation about the medicalization and demedicalization of behavior."
Yet psychiatry's lack of scientific rigor is so obvious today that the profession's leading lights openly complain about it. In a January Wired article about the ongoing revision of the American Psychiatric Association'sDiagnostic and Statistical Manual of Mental Disorders (DSM), Gary Greenberg, a psychotherapist and journalist, recounts an interview with Allen Frances, lead editor of the manual's current (fourth) edition. "There is no definition of a mental disorder," Frances tells him. "It's bullshit. I mean, you just can't define it."
Since mental disorders officially exist in the United States only if they are listed in the DSM, which is the bible for mental health professionals and the key to insurance coverage, this is a pretty significant concession. It reinforces Szasz's point that psychiatrists invent mental illnesses by voting on whether to recognize them. "Old diseases such as homosexuality and hysteria disappear," he writes, "while new diseases such as gambling and smoking appear, as if to replace them."
The perils of this approach are evident in Greenberg's eloquently honest book Manufacturing Depression: The Secret History of a Modern Disease, which questions psychiatrists' authority to medicalize our moods even as it sympathizes with the suffering of depressed people and describes the author's own bouts of melancholy. Although his book has a Szaszian title (recalling the heretical psychiatrist's 1970 book The Manufacture of Madness), Greenberg mentions Szasz only once in passing. Jonathan Metzl, a professor of psychiatry and women's studies at the University of Michigan, has a bit more to say about Szasz in The Protest Psychosis: How Schizophrenia Became a Black Disease. Metzl implicitly criticizes Szasz and other opponents of forced treatment for inspiring the deinstitutionalization that began in the 1960s, which he says often left former mental patients with "nowhere to go and no one to turn to for help." Yet by tracking the shifting, politically driven definition of schizophrenia, commonly viewed as the mental disorder most clearly established as a disease, Metzl's eye-opening book casts doubt on psychiatry's status as a field of medicine, let alone one with a strong enough basis to justify coercively treating unwilling patients. Together he and Greenberg show that Szasz's objections to psychiatry's role in stripping people of their freedom and relieving them of their responsibility, no matter how often they are dismissed as quaint or simpleminded, remain logically and morally compelling.
As Greenberg makes clear in his Wired article and his book, mental disorders are defined by patterns of behavior, without regard to what causes them. By listing these criteria in the DSM, psychiatrists have achieved a high degree of diagnostic agreement, but they simply assume that people who are given the same label have the same underlying problem. In Manufacturing Depression, Greenberg quotes Thomas Insel, director of the National Institute of Mental Health, who told psychiatrists at the American Psychiatric Association's 2005 convention that the DSM "has 100 percent reliability and zero percent validity."
In Szasz's view, this lack of validity is unavoidable, because once a particular pattern of behavior can be confidently ascribed to a physical defect, such as the brain damage caused by advanced syphilis or Alzheimer's disease, it is no longer considered a psychiatric issue. "Contemporary 'biological' psychiatrists tacitly recognized that mental illnesses are not, and cannot be, brain diseases," he writes in his preface. "Once a putative disease becomes a proven disease, it ceases to be classified as a mental disorder and is reclassified as bodily disease." If every disorder in the DSM had a clear neurological cause, Szasz says, psychiatry would be indistinguishable from neurology.
Greenberg takes a less categorical stance, skeptical of the DSM enterprise but not quite prepared to give it up entirely. "A new manual based entirely on neuroscience—with biomarkers for every diagnosis, grave or mild—seems decades away, and perhaps impossible to achieve at all," he writes in Wired. "To account for mental suffering entirely through neuroscience is probably tantamount to explaining the brain in toto, a task to which our scientific tools may never be matched. What the battle over DSM-5 should make clear to all of us…is that psychiatric diagnosis will probably always be laden with uncertainty, that the labels doctors give us for our suffering will forever be at least as much the product of negotiations around a conference table as investigations at a lab bench."
The example that is the focus of Greenberg's book, depression, plainly illustrates the arbitrariness of the DSM's diagnostic criteria. Since everybody gets the blues, psychiatrists need to distinguish between normal sadness and pathological sadness, if only to preserve their own credibility as doctors treating illness. But such line drawing is unavoidably subjective. As Greenberg notes, the official definition of "major depression" excludes people who have experienced the death of a loved one within the previous two months. The American Psychiatric Association (APA) has decreed that 60 days of mourning is appropriate, while 61 is not. Up to the two-month line, you are experiencing normal grief; after that, you are sick.
Not only is the cutoff arbitrary, but so is the decision to count only death as a legitimate excuse for "a period of at least 2 weeks during which there is either depressed mood or a loss of interest or pleasure in nearly all activities." As Greenberg observes, "It's not clear why bereavement is the only exempt condition, why, for instance, misfortunes like betrayal by a lover or severe financial loss or political upheaval or serious illness—or for that matter a noncatastrophe, the slow accretion of life's difficulties or a loss of faith in one's government or simply existential despair kindled by an awareness of mortality—do not also spare people from the rolls of the diseased."
Greenberg does not think there is anything necessarily adaptive, redemptive, or ennobling about depression. He recalls his own struggles with it, including "the time I found myself on the floor watching dust specks float through sunbeams for hours (because they happened to be in my line of sight, because looking at anything else or closing my eyes and staring at my own black insides would just take too much effort), racked by some unspecifiable pain, like my whole being was a phantom limb, and thinking about the lady in the Life-Fone pendant ad, the one who has fallen and can't get up." He describes clients whose unremitting self-reproach has sucked every bit of pleasure from their lives, who are immobilized by dread and hopelessness.
Yet Greenberg rebels at Prozac apostle Peter Kramer's confident assertion that "depression is neither more nor less than illness." He argues that "the medical industry…has acquired far too much power over our inner lives—the power to name our pain and then sell us the cure one pill at a time." While Greenberg details how pharmaceutical companies have profited from treating depression as "a widespread chronic disease," he does not claim they foisted this concept on us. The idea appeals to Americans, he suggests, because it gives them permission to take mood-altering substances without running afoul of the principle that Harvard psychiatrist Gerald Klerman called "pharmacological Calvinism"—the belief that "if a drug makes you feel good, it must be morally bad."
Greenberg agrees that drugs, including psychotherapeutic catalysts such as MDMA (Ecstasy) as well selective serotonin reuptake inhibitors (SSRIs) such as Prozac, can help depressed people feel better. And although he bemoans "the medical industry's invention of a disease out of our daily troubles and aspirations," he concedes that pills might be the best choice for some people in some situations. But he emphasizes the crucial role that the placebo effect seems to play in the impact of SSRIs. In clinical trials, drugs like Prozac perform only slightly better than placebos, so slightly that the difference is "not clinically significant," according to a 2002 review of the evidence by Irving Kirsch, a professor of psychology at the University of Hull, who elaborates on his findings in The Emperor's New Drugs (Basic Books). The difference is so small that it may be partly or entirely due to expectations primed by the drug's side effects. These results (along with Greenberg's own experience as an experimental subject, which he describes) suggest the power of hope, kindled by the rituals of self-improvement, as an antidote to depression.
Many depressed people, of course, report dramatic results from taking SSRIs, and there may be more to it than the placebo effect. Perhaps the clinical trials lump together too many different kinds of depressed people and therefore fail to show how effective these drugs can be. But does that mean some of these subjects, diagnosed based on the APA's official criteria, are not "really" depressed, or does it mean the DSM mistakenly lumps together disparate conditions with different causes based on superficial resemblances? The weak experimental evidence in favor of SSRIs and the mixed real-world experiences of people who take them are a standing rebuke to the medical model of depression, which says it is a disease caused by an imbalance of brain chemicals, an imbalance that can be rectified with pills. "Depression is nothing more or less than its symptoms," Greenberg writes. "For all the scientific language and scholarly discourse, for all the doctors' claims that they've found the wellsprings of demoralization, there's still no actual biochemical glitch that lies behind the symptoms."
It's not hard to see how happy pills fit the medical model of depression. But Greenberg also portrays cognitive therapy, which aims to root out self-defeating habits of thought, as reinforcing the idea that depression is a disease. That's an odd judgment, since cognitive therapists approach depression as a learned condition that can be unlearned, which fits more with Szasz's view of mental illnesses as "problems in living" than it does with the conventional psychiatric perspective. Greenberg's distaste for cognitive therapy's narrow focus on "distorted cognitions" and "dysfunctional beliefs" is understandable, since his more humanistic approach (the effectiveness of which, he concedes, is uncertain and hard to measure) emphasizes delving into clients' life stories in the hope that redemption can come through "narrative coherence." I confess that I find cognitive therapy more intuitively appealing. But whatever their relative merits, neither school seems especially compatible with the belief that depression is a medical condition just like cancer or diabetes.
That message, although promoted by drug commercials, public health literature, and legislation governing medical coverage, remains controversial, largely because almost all of us have some experience with depression, and we resist the idea that everyone is sick, which seems counterintuitive if not nonsensical. By contrast, the delusions and hallucinations that spring to mind when people think of schizophrenia seem so far outside everyday experience that the condition is easier to imagine as a brain disorder different not only in degree but in kind from garden-variety disturbances. It is commonly cited as a scientific explanation for anti-social behavior ranging from shouting on a street corner to mass murder at a shopping center.
But as Jonathan Metzl shows in The Protest Psychosis, the image of schizophrenics as belligerent and potentially violent is a relatively recent development. Based on a detailed examination of records from a state mental hospital in Michigan, combined with a review of diagnostic guidelines, medical journals, popular periodicals, movies, and music, Metzl concludes that the definition of schizophrenia underwent a marked shift in the 1960s and '70s. "Prior to the civil rights movement," he writes, "mainstream American medical and popular opinion often assumed that patients with schizophrenia were largely white, and generally harmless to society. From the 1920s to the 1950s, psychiatric textbooks depicted schizophrenia as an exceedingly broad, general condition, manifest by 'emotional disharmony' that negatively impacted white people's abilities to 'think and feel.'?" In the popular imagination and inside the walls of mental hospitals during this period, a typical schizophrenic might be a troubled middle-class housewife who today would be diagnosed with depression. After the 1950s, Metzl says, schizophrenia increasingly came to be identified with angry, paranoid black men, largely because of racial anxieties stoked by the turbulent politics of the time.
This shift is vividly illustrated by a 1974 ad in the Archives of General Psychiatry, reproduced in Metzl's book, that shows a black urban rioter with a clenched fist under the headline "Assaultive and Belligerent?" The ad informs psychiatrists who encounter such patients that "cooperation often begins with Haldol," a pacifying antipsychotic medication approved by the Food and Drug Administration in 1967. Metzl does not claim that every black man diagnosed with schizophrenia during this period was a political dissident repressed under the guise of medical treatment. Many of them, he suggests, were indeed seriously disturbed and in need of professional help. But he makes a convincing case that diagnoses were driven by institutionalized racism, with expressions of hostility against white people, including the ideologies espoused by the Black Power movement, treated as psychiatric symptoms.
Although schizophrenia supposedly afflicts about 1 percent of the population, regardless of race or gender, psychiatrists are more apt to perceive it in black men than in other groups. "In 1973," Metzl writes, "a series of studies in the Archives of General Psychiatry discovered that African American patients were 'significantly more likely' than white people to receive schizophrenia diagnoses, and 'significantly less likely' than white patients to receive diagnoses for other mental illnesses such as depression or bipolar disorder. Throughout the 1980s and 1990s, a host of articles from leading psychiatric and medical journals showed that doctors diagnosed the paranoid subtype of schizophrenia in African American men five to seven times more often than in white men, and also more frequently than in other ethnic minority groups."
Metzl does not want to leave the impression that he is questioning the existence of schizophrenia. "In no way is my telling of this history meant to suggest that schizophrenia is a socially fabricated disease or, worse, that people's suffering is somehow inauthentic," he writes. "As a psychiatrist, I have seen the tragic ways in which hallucinations, delusions, social withdrawal, cognitive decline, and profound isolation rupture lives, careers, families, and dreams in profoundly material ways."
If schizophrenia, as it is currently defined, is a bona fide disease, it follows that the old definition, the one that applied the label to grandiloquent novelists and apathetic housewives, was mistaken. It also follows that many of the angry black men identified as paranoid schizophrenics in the 1960s and '70s were misdiagnosed. And unless black men are especially prone to schizophrenia for some reason, the fact that they continue to be given this label at a higher rate than other groups means psychiatrists are continuing to make diagnostic mistakes, either seeing schizophrenia in blacks when it does not exist or failing to see it in whites when it does.
Yet Metzl refuses to make such judgments. He tells the story of a 31-year-old white middle-class woman who was brought to Michigan's Ionia State Hospital for the Criminally Insane in 1941 because she "got confused and embarrassed her husband." In addition to confusion, her main symptoms were "a delusional sadness and irascible feelings of guilt." Although she denied hallucinating or hearing voices, she was diagnosed with paranoid schizophrenia. In Metzl's view, to say this woman was misdiagnosed—that she suffered from depression or obsessive-compulsive disorder, say, rather than paranoid schizophrenia—would be to commit the intellectual offense of "presentism," anachronistically applying today's standards to a different time. According to Metzl, someone who declares that Mary Todd Lincoln or Vincent Van Gogh suffered from bipolar disorder, a concept that was not developed until after their deaths, is also guilty of presentism.
By contrast, it makes perfect sense to speculate about whether Friedrich Nietzsche's bizarre behavior toward the end of his life was related to syphilitic dementia. Although Nietzsche died years before the specific bacterial cause of syphilis was identified and its neurological effects were confirmed, either he had syphilis or he didn't, and either it damaged his brain or it didn't. If we could test his blood and examine his brain tissue, we would know for sure. Presentism does not enter into it. Szasz cites syphilitic dementia as the classic example of a mental illness (general paresis) that turned out to be a brain disease. Although we are supposed to believe the same thing is true of the conditions described in the DSM, if the evidence were conclusive they would not be listed there. As Szasz says, they would be treated by neurologists instead of psychiatrists.
Metzl is not interested in such distinctions. "Schizophrenia is shaped by social, political, and, ultimately institutional factors in addition to chemical or biological ones," he writes. "Too often, we assume that medical and cultural explanations of illness are distinct entities, or engage in frustratingly pointless debates about whether certain mental illnesses are either socially constructed or real." He says "this polarizing dichotomy serves no one, and makes it harder to see how mental illness is always already both."
It is hard to imagine someone making a similar speech about cancer or diabetes. "Unlike the conditions treated in most other branches of medicine," observes Marcia Angell, former editor of The New England Journal of Medicine, in a June New York Review of Books essay, "there are no objective signs or tests for mental illness—no lab data or MRI findings—and the boundaries between normal and abnormal are often unclear. That makes it possible to expand diagnostic boundaries or even create new diagnoses, in ways that would be impossible, say, in a field like cardiology." In other words, mental illnesses are whatever psychiatrists say they are. If someone is diagnosed with depression or schizophrenia based on the currently accepted behavioral markers, assuming the criteria are correctly applied, it does not make sense to say he does not really have depression or schizophrenia, since there is no test to disconfirm the diagnosis. And if the criteria change so that they no longer apply to him, his disease disappears or becomes something else; it has no independent existence.
No wonder the psychiatrist who was in charge of producing the current DSM despairs that defining mental disorders is "bullshit." Given the potential for ineffective, harmful, and involuntary treatment, this state of affairs is not just frustrating or embarrassing; it is downright dangerous.
Senior Editor Jacob Sullum is a nationally syndicated columnist.