Thomas Szasz takes on his critics: is mental illness an insane idea?

Szasz Under Fire: The Psychiatric Abolitionist Faces His Critics, edited by Jeffrey A. Schaler, Chicago: Open Court, 450 pages, $36.95 paper

IN 1980 THOMAS Szasz testified for the prosecution in the trial of Darlin June Cromer, a 34-year-old white woman charged with kidnapping and murdering Reginald Williams, a 5-year-old black boy. There was no question that Cromer, who attracted suspicion because she had a history of talking about "killing niggers" and trying to lure black children into her car, had abducted Reginald from an Oakland, California, supermarket, strangled him, and buried his body near her home. She had told police as much when they questioned her. Neither was her motive in doubt. She explained that "it is the duty of every white woman to kill a nigger child," telling a jail psychologist she hoped to ignite a race war.

But as the San Francisco Chronicle reported, Cromer's attorney argued that "his client killed because she is consumed by schizophrenic paranoia--not hate for blacks." Or as the lawyer put it, "This case does not involve racism; it involves insanity." To help undermine this claim, the prosecutor enlisted the assistance of Szasz, the iconoclastic psychiatrist famous for rejecting the insanity defense, involuntary commitment, and the very concept of mental illness. At the trial, Szasz explained the difference between a medical diagnosis and a psychiatric diagnosis: "Medical diagnoses deal with objective and demonstrable lesions of the body, broken bones, diseased livers, kidneys, and so on. Psychiatric diagnoses deal with behaviors that human beings display, and they have to be interpreted in moral, cultural, and legal terms and, therefore, different interpreters will arrive at different judgments." He pointed out that "homosexuality was recognized as a mental disease until a few years ago" and that smoking, previously considered a habit, had recently been classified as a mental disorder.

Asked "what [Cromer] was suffering from, if anything," on the day of the murder, Szasz offered the following opinion based on her records: "She was suffering from the consequences of having lived a life very badly, very stupidly, very evilly.... From the time of her teens, for reasons which I don't know ... whatever she [has] done, she has done very badly. She was a bad student.... She was a bad wife. She was a bad mother. She was a bad employee insofar as she was employable. Then she started to engage [in taking] illegal drugs, then she escalated to illegal assault, and finally she committed this murder.... Life is a task. You either cope with it or it gets you.... If you do not know how to build, you can always destroy."

Szasz did not really try to explain why Cromer killed Reginald. Indeed, the main strength of his testimony was his acknowledgment of how difficult it is to get inside the head of a criminal--or anyone else. Cromer was the way she was and did the things she did "for reasons I don't know." By contrast, the defense experts confidently asserted that her crime was caused by a mental illness.

In Szasz UnderFire, a new collection of exchanges with his critics, Szasz, a reason contributing editor and an emeritus professor of psychiatry at the SUNY Upstate Medical University in Syracuse, explains why, unlike the defense experts, he did not conduct a "psychiatric examination" of Cromer. "I regard the practice as the epitome of junk science and refuse to participate in it," he writes. Not only is there "no objective test for mental illness," but psychiatrists are supposed to determine a defendant's state of mind at the time of the crime by talking to him many months later, a pretense Szasz considers "prima facie absurd." This was one of the main points he made in his testimony, which was condemned by psychiatrists outraged that he had dared to question the premises of their profession. The jurors, who convicted Cromer, apparently agreed with Szasz about the reliability of psychiatric testimony.

As illustrated by the case of Andrea Yates, whose 2002 conviction for drowning her children was recently overturned because of false testimony that may have undermined her insanity defense, the questions raised at Cromer's trial are as relevant today as they were a quarter century ago. Do some people have mental impairments, either temporary or permanent, that prevent them from controlling their behavior ? If so, how can such people be identified, and how should they be treated ?

Although insanity pleas are offered in only about I percent of criminal cases and are usually unsuccessful, they have played a role in a number of sensational trials, including those of would-be presidential assassin John Hinckley and D.C. sniper Lee Boyd Malvo. Furthermore, issues of mental impairment are important not just in cases where defendants offer insanity pleas but in every context where there is potential for psychiatric coercion, including legally mandated addiction treatment and civil commitment of people deemed a threat to themselves or others. Each year in the United States more than 1 million people are committed to mental hospitals; some two-thirds of these commitments are officially voluntary, but that status can change once a "patient" tries to get out. Since mental illness is a widely accepted rationale for both relieving people of responsibility and depriving them of liberty--the twin dangers to which Szasz has been alerting us for more than four decades--the psychiatric is unavoidably political.

Szasz Under Fire, edited by the Szaszian psychologist Jeffrey Schaler, appropriately focuses on this theme. Szasz's debating partners include psychiatrists, psychologists, bioethicists, and legal scholars, most of whom seem to have reservations about psychiatry's tendency to treat every facet of human behavior--happiness and sadness, energy and lethargy, neatness and sloppiness, shyness and boldness, inattentiveness and obsessiveness, thievery and honesty, promiscuity and celibacy, thinness and fatness--as a symptom of mental illness. At the same time, they ostensibly part company with Szasz when it comes to "severe" mental illnesses, which are usually said to include schizophrenia, bipolar (manic-depressive) disorder, and major depression. I say "ostensibly" because the contributors to Szasz Under Fire generally seem to believe that schizophrenia and a few other disorders that psychiatrists diagnose are in fact brain diseases, which would mean they are not "mental illnesses," any more than a brain tumor or Huntington's disease is.

"I struggle with ambivalence about [Szasz]," confesses E. James Lieberman, a professor of psychiatry at the George Washington University School of Medicine. "He's on the right track, but he goes too far and too straight." This book mainly deals with the ways in which Szasz's critics think he "goes too far," and in doing so it illuminates vital questions about the nexus between psychiatry and the law.

Discussions of Szasz's ideas tend to begin with his insistence that mental illness is, strictly speaking, a contradiction in terms, a literalized metaphor that confuses more than it clarifies. Not surprisingly, much of Szasz Under Fire continues the conceptual and semantic battles provoked by Szasz's 1961 classic The Myth of Mental illness. His critics offer various alternatives to the Szaszian perspective, which insists upon an objectively measurable bodily defect as the sine qua non of a true disease. Among other things, they argue that some so-called mental illnesses are genuine brain diseases, although their precise etiologies have not been figured out yet; that if mental illness is a myth, so is physical illness, because both categories have fuzzy boundaries and are to a large extent culturally determined; that viewing mental illness as a myth is a fiction that is necessary to maintain the integrity of psychotherapy as a moral enterprise; and that the distinction between mental and physical disease is misleading, since (as the American Psychiatric Association puts it) "there is much that is 'physical' in mental disorders and much 'mental' in 'physical' disorders."

This book is the first in an Open Court series featuring debates between important thinkers and their critics, and Szasz should be commended for responding directly and at length to people who disagree with him, given how easy it is for a writer of his stature to retreat to an echo chamber populated by loyal followers. He sticks to his guns and scores many points, but his responses are not always completely satisfying. That's a shame, because many readers will look to this book for an introduction to his ideas and may be put off by the questions he neglects or only partially answers.

Consider Szasz's response to the late Robert E. Kendell, former president of the U.K.'s Royal College of Psychiatrists. Kendell writes that "it is impossible to identify any characteristic feature of either the symptomatology or the etiology of so-called mental illnesses which consistently distinguishes them from physical illnesses." This assertion seems to go to the heart of Szasz's insistence that mental illnesses are not real diseases. But rather than refute it, he replies, "This is true, but not enough." Enough for what isn't exactly clear. Szasz then cites three distinctions between physical and mental illness that are generally valid but do not hold in every case: I) "Typically, physical illnesses are identified by observing the patient's body," while "typically, mental illnesses are identified by observing the patient's verbal pronouncements." 2) There are "objective, physical-chemical markers" to ascertain whether someone has a particular brain disease but "no such markers" to ascertain whether he has a particular mental illness. 3) "The typical medical patient" is treated only with his informed consent, while "the typical mental patient" is treated without his consent.

Although that last claim does not apply to the millions of Americans who voluntarily seek antidepressants or psychotherapy as a way of improving their lives, it arguably describes hospitalized mental patients, keeping in mind the blurry line between voluntary and involuntary commitment. The combination of subjective diagnosis and involuntary treatment poses obvious dangers. As Szasz says, "There is no way that someone can disprove the 'diagnosis' that he 'suffers' from schizophrenia." Still, the hallmarks of true disease that Szasz mentions do not always apply. In his contribution to Szasz UnderFire, Ronald Pies, a professor of psychiatry at the Tufts University School of Medicine, cites migraine headaches as an example of a physical condition that is diagnosed based entirely on "the patient's verbal pronouncements" (complaints of pain, nausea, light flashes, etc.). Then again, migraine sufferers are not treated against their will.

Amid all this terminological disputation, it is important, though not always easy, to keep in mind the real-world consequences of these ideas. Defenders of psychiatry can be remarkably blithe about those consequences. The American University sociologist Rita Simon, co-author of a book on the insanity defense, uses barely two pages of Szasz Under Fire to defend the practice, which she does simply by asserting that "a few individuals," because of "a mental disability or disease," lack "the minimal capacity for rational and voluntary choices on which the law's expectation of responsibility is predicated." She leaves unexamined the question of how the legal system should determine whether it is confronting such an individual. At the end of her very brief essay, she offers more cause for doubt by approvingly quoting legal scholar Alan Stone's statement that the insanity defense "purports to draw a line between those who are morally responsible and those who are not, those who are blameworthy and those who are not, those who have free will and those who do not." As Szasz notes in his reply, purport usually suggests a pretense of some sort.

In any event, how should a court decide whether someone like Darlin June Cromer is a paranoid schizophrenic or merely a racist murderer? And if it decides she is a schizophrenic, is that diagnosis enough to show that she should not be held responsible for strangling a 5-year-old-boy? In theory, the jury in her case could have decided that she was a schizophrenic but still responsible for her actions, since the legal definition of insanity--which generally requires that the defendant either did not know what he was doing, did not know it was wrong, or could not stop himself--is not the same as the criteria for a psychiatric diagnosis. Cromer's repeated attempts to kidnap black children, her explanation of her motive, and the fact that she tried to hide the body all suggest she knew what she was doing, which presumably helps explain why the jury found her guilty. But the defense argued that her appalling actions and statements all were symptoms of her disease--indeed, that they were so appalling they had to be. "If she isn't crazy," one of the expert witnesses asked, "who is?"

Similarly, Lieberman complains in his essay that, when it comes to individual responsibility, Szasz "makes no exception for a woman who drowns her five children" This reference to Andrea Yates implies that the nature of her act proves she was not responsible for it--a standard that would give a pass precisely to those guilty of the most horrendous crimes.

Still, surely there are people who commit what would ordinarily be considered crimes when they are mentally incapacitated: a sleepwalker who assaults a stranger while acting out a dream, say, or a Huntington's patient who throws a dish at his caretaker. Szasz's reply to Simon would have been stronger if he had explained how such cases should be handled. And if some people diagnosed as schizophrenics do in fact suffer from an incapacitating brain disease--a possibility Szasz concedes--presumably they too should be held less culpable than people in full possession of their faculties.

But as Szasz notes, if psychiatrists were interested merely in identifying and treating the brain diseases underlying certain forms of insanity, their field ultimately would be swallowed by neurology. Their agenda is far more ambitious than that, as illustrated by their attachment to the calculatedly ambiguous term mental disorder, which the American Psychiatric Association continues to use even while complaining that it "unfortunately implies a distinction between 'mental' disorders and 'physical' disorders that is a reductionistic anachronism of mind/body dualism." Given the sweep of the APA's Diagnostic and Statistical Manual of Mental Disorders (DSM), which takes in misbehavior ranging from rudeness to murder, it's fair to read this caveat to mean that anything bad people think, feel, say, or do can be interpreted as a symptom of a disease. In practice, psychiatrists often distinguish between "severe" disorders thought to have a physiological basis and the myriad sins, foibles, bad habits, and eccentricities cataloged by the DSM. But their training, terminology, diagnostic framework, and billing practices imply that all these are medical problems appropriately handled by physicians. As Kendell notes, "the inexorable expansion of the concept of mental illness" despite a "fragile empirical basis" leaves psychiatrists "vulnerable to accusations of unjustified medicalization of deviant behavior and the vicissitudes of daily life."

The DSM's broad scope is consistent with a perspective that sees brain defects at the root of all misbehavior and psychological problems. According to this view, the contents of the mind are determined by the structure of the brain; that structure, in turn, is shaped by genetics and experience. When something goes seriously wrong with either or both, the result is a disorder that is just as rooted in biology as a so-called physical disease. In this light, it makes perfect sense for Harvard psychiatrist Alvin Pouissant to argue that "extreme racism" should be considered a mental illness. So far the APA has rejected this suggestion, which would render racist murderers like Darlin June Cromer insane by definition and open the door to psychiatric treatment of unconventional opinions, a la Soviet Russia and Communist China. On the brighter side, the equation of broadly defined mental disorders with physical illness has the potential to let everyone--not just the Darlin June Cromers of the world--off the hook (although the legal system might continue to punish for the sake of deterrence, as opposed to justice). If we consistently apply the assumption that bad behavior is caused by defective brains, the question is not whether a particular murderer has a brain disease but whether there can be such a thing as a murderer who doesn't.

The breadth of the territory claimed by psychiatry would not be nearly as troubling if it were not so often settled by force. In addition to peering backward in time to determine a defendant's state of mind when he committed his crime, psychiatrists are expected to predict the future, assessing whether a given individual is likely to harm himself or others. Based on that judgment, innocent people can be forcibly "treated" and deprived of their liberty indefinitely. As with the insanity defense, defenders of psychiatry tend to minimize both the frequency of civil commitment and the importance of the psychiatrist's role in it.

Lieberman, for instance, casually remarks that "one rarely hears of someone being committed involuntarily to a mental hospital." Szasz rightly calls this an "astounding assertion," citing an estimate from the 1996 book Mental Health and Law that "each year in the United States well over one million persons are civilly committed to hospitals for psychiatric treatment." The book adds that "it is difficult to completely separate discussions of voluntary and involuntary commitment because voluntary status can be converted efficiently to involuntary status, once the patient has requested release."

Given their defense of involuntary treatment as not only justified but morally mandatory, psychiatrists seem weirdly reluctant to acknowledge their role in it. Pies approvingly cites a passage from Robert Simon's Psychiatry and Law for Clinicians that says "mental health professionals must understand that it is not they who make commitment decisions about patients. Commitment is a judicial decision that is made by the court or by a mental health commission. The clinician files a petition or medical certification that initiates the process of involuntary hospitalization." A prosecutor might with equal plausibility deny that he is the one who puts a defendant away for life; after all, it's the jury that convicts and the judge who imposes the sentence. True enough as far as it goes, but the prosecutor plays a crucial role. The same is true of the psychiatrist who "initiates the process of involuntary hospitalization"--even more so, since people who are committed do not receive the same protections as criminal defendants. "How do judge and mental patient meet?" Szasz asks. "The psychiatrist introduces them to one another. How does the judge know whom to commit? The psychiatrist tells him."

To get a sense of why psychiatrists might want to disclaim responsibility for civil commitment, consider the case of Rodney Yoder, an Illinois man whose cause Szasz has championed. Yoder, who completed a prison sentence for assaulting his ex-wife in 1991, has been locked in a mental hospital ever since, based on a series of dubious and contradictory diagnoses. To judge by a 2002 report in Time and other press accounts, Yoder's "illness" boils down to an abrasive personality that rubbed the wrong people the wrong way. Ostensibly, he is kept behind bars because he represents an intolerable threat to the public, even though a psychiatrist who examined him in prison said he wasn't a danger and did not meet the standard for involuntary hospitalization. In a 2001 letter about Yoder to then-Illinois Gov. George Ryan, the psychiatrist Loren Mosher said "the state is practicing preventive detention in the guise of mental-health 'treatment.'"

As Szasz notes, in 1997 the U.S. Supreme Court endorsed such detention for sex offenders who have completed their sentences but who, because of a "mental abnormality" or "personality disorder," are deemed likely to commit new crimes. These offenders, who are confined indefinitely in mental hospitals after serving their time, are considered sane enough to be convicted and punished but not sane enough to be released. Given the plethora of mental abnormalities and personality disorders identified by psychiatrists, this practice could be extended to many other criminals--for example, those suffering from antisocial personality disorder, "a pervasive pattern of disregard for and violation of the rights of others." The Bureau of Justice statistics estimates that 16 percent of prison and jail inmates are mentally ill, and that includes only those who "reported either a mental condition or an overnight stay in a mental hospital."

Civil commitment does not require dangerousness to others; dangerousness to oneself will do. A diagnosis of schizophrenia, which is said to afflict about 1 percent of the population, is much more likely to result in hospitalization (or, as Szasz would say, imprisonment) than is a diagnosis of, say, major depression: Data from the Substance Abuse and Mental Health Services Administration indicate that in 1997 schizophrenia accounted for more than a quarter of in-patient psychiatric admissions of people with "serious functional impairment," compared to 38 percent for affective disorders and 12 percent for substance-related disorders--both of which, according to the DSM, are far more common than schizophrenia in the general population. Still, the rationale for most of these admissions presumably is the patient's own welfare, as opposed to public safety. And since psychiatric diagnoses, unlike the typical medical diagnosis, generally imply that the "patient" either does not properly understand his own interests or is not capable of acting on them, the threat of involuntary treatment always hangs in the background.

Although Szasz emphasizes the contrast with medicine proper, which usually is predicated on the patient's consent, there are exceptions based on competence: Children, the severely retarded, and patients in the advanced stages of Alzheimer's do not make their own medical decisions. In North Dakota, the children of former federal judge Bruce Van Sickle, who has Alzheimer's, are engaged in a legal battle over whether he should remain in a nursing home. One of his sons says Van Sickle wants to go home, while his other three children say he is too far gone to know what he wants. Elsewhere Szasz has acknowledged the need for a legal process to determine competence in such cases. Some discussion of that issue would have been appropriate in response to Pies' hypothetical question regarding an elderly man who begins to behave strangely after falling and hitting his head. If involuntary treatment can be justified in such a case, can't it also be justified for a schizophrenic? Szasz's failure to address that question leaves him open to the charge of dodging an important issue. I think he would have to say that the two cases should be handled in a similar way--provided the schizophrenic's brain injury can be demonstrated as readily as the old man's.

But as Richard Bentall, a psychologist at the University of Manchester, shows in his contribution to this volume, the science regarding the etiology of schizophrenia is not nearly as clear as psychiatrists often imply. Bentall points to several weaknesses in the leading theory, which holds that schizophrenia is caused by an excess of the neurotransmitter dopamine. He notes that antipsychotic drugs that block dopamine receptors within hours do not affect behavior for weeks; that some drugs considered effective in treating schizophrenia do not zero in on dopamine receptors; and that antipsychotic drugs do not work for all schizophrenics but may work for people with different diagnoses. Perhaps most important, "an expensive and sustained search for dopamine abnormalities in schizophrenia patients has so far drawn a blank." Kendell likewise concedes that "in schizophrenia a structural abnormality can only be demonstrated in populations, not in all or even most individuals." By comparison, while the etiology of Alzheimer's disease remains murky and its initial diagnosis depends partly on behavior (along with brain scans and cognitive tests), the diagnosis can be confirmed in autopsies by the presence of brain plaques and tangles. The postmortem evidence, which confirms the initial diagnosis about 85 percent of the time, gives physicians confidence that they are looking at a discrete condition with a common physiological cause.

To a large extent, then, the issue of involuntary treatment comes down to a question of where the burden of proof should lie and how heavy it should be. Even those who are skeptical of psychiatric pretensions cannot easily dismiss Pies' invocation of "the young man, rocking back and forth in a pool of his own urine, responding to voices from 'a CIA computer' that are instructing him to kill himself." If such a person is indeed suffering from an incapacitating brain disease, it should be possible to allow his family to make treatment decisions on his behalf. At the same time, anyone who cares about liberty has to hesitate before imposing treatment on someone who insists he does not want it.

The psychiatrist E. Fuller Torrey, once a Szasz admirer, is now one of his most vocal critics, having concluded that "schizophrenia is a disease of the brain in the same sense that Parkinson's disease and multiple sclerosis are diseases of the brain. "Yet Torrey, a prominent advocate of involuntary psychiatric treatment, concedes "there is no single abnormality in brain structure or function that is pathognomonic for schizophrenia" and therefore "we do not yet have a specific diagnostic test."

That limitation should give Torrey pause in light of the concerns he expressed in his 1974 book The Death of Psychiatry, quoted in Szasz Under Fire. He argued that "it is better that we err on the side of labeling too few, rather than too many, as brain diseased. In other words, a person should be presumed not to have a brain disease until proven otherwise on the basis of probability. This is exactly the opposite of what we do now as we blithely label everyone who behaves a little oddly 'schizophrenic.' Human dignity rather demands that people be assumed to be in control of their behavior and not brain diseased unless there is strong evidence to the contrary."

While the identification of schizophrenics may be less casual today than it was three decades ago, psychiatric labels have multiplied since then, and a significant part of the population is still forcibly treated, whether in mental hospitals, through outpatient commitment, or in drug treatment programs fed by the criminal justice system. In other words, there is still a need to guard against invasions of liberty justified in the name of mental health. The approach Torrey suggested seems about right to me, although much hinges on what counts as "strong evidence" of brain disease. Szasz continues to make a powerful case that a psychiatric diagnosis is not enough.

Senior Editor Jacob Sullum (jsullum@reason. com), the author of Saying Yes: In Defense of Drug Use (Tarcher/Penguin), received the 2004 Thomas S. Szasz Award for Outstanding Contributions to the Cause of Civil Liberties.