Angelo, a London-born scientist in his early 30s with sandy brown hair, round wire-frame glasses and a slight, unobtrusive stammer, vividly recalls the day he began to hear voices. It was Jan. 7, 2001, and he had recently passed his Ph.D. oral exams in chemistry at an American university, where, for the previous four and a half years, he conducted research into infrared electromagnetism. Angelo was walking home from the laboratory when, all of a sudden, he heard two voices in his head. "It was like hearing thoughts in my mind that were not mine," he explained recently. "They identified themselves as Andrew and Oliver, two angels. In my mind's eye, I could see an image of a bald, middle-aged man dressed in white against a white background. This, I was told, was Oliver." What the angels said, to Angelo's horror, was that in the coming days, he would die of a brain hemorrhage. Terrified, Angelo hurried home and locked himself into his apartment. For three long days he waited out his fate, at which time his supervisor drove him to a local hospital, where Angelo was admitted to the psychiatric ward. It was his first time under psychiatric care. He had never heard voices before. His diagnosis was schizophrenia with depressive overtones.

Angelo remembers his time at the hospital as the deepening of a nightmare. On top of his natural confusion and fear over the shattering of his psychological stability, Angelo did not react well to the antipsychotic he'd been prescribed, risperidone, which is meant to alleviate the symptoms of schizophrenia by reducing the level of dopamine in the brain. In Angelo's case, the pills had a predominantly negative effect. His voices remained strong and disturbing — an unshakable presence, quiet only in sleep — while he grew sluggish and enervated. "If you think of the mind as a flowing river of thoughts," he told me in an e-mail message, "the drug made my mind feel like a slow-moving river of treacle." Several days into his stay, Angelo's parents flew to the United States from London and took him back home.

More than six years later, Angelo still lives at his parents' house. He currently takes a cocktail of antidepressants and antipsychotics, with tolerable side effects, and sees a psychologist every two months to monitor his medication. The pills help Angelo to manage his voices, but they have not been able to eradicate them. Shortly after his return to London, he made an attempt to resume his career, accepting a research position at the university where he had received his undergraduate degree. He lasted eight months (his neighbors heard him screaming at his voices and called the police), checked himself into the hospital for six weeks and returned home. Despite these setbacks, Angelo has maintained his optimism. He is eager to discover new ways to combat his voices. Not long ago, he found one. In November, his psychologist informed him of a local support group for people who hear voices, from which he thought Angelo might benefit. Angelo began to attend the group late last year.

I first met Angelo at a meeting of the group in mid-January. (I was given permission to sit in on the condition that I not divulge the participants' last names.) The meeting took place in the bright, cheerfully decorated back room of a community mental-health center in North Finchley, an affluent, grassy suburb in the northern reaches of London. The gathering was small but eclectic. In addition to the group's facilitators — Jo Kutchinsky, an occupational therapist, and Liana Kaiser, a social-work student — five men and women assembled in a circle of bulky wool-knit chairs around a worn coffee table. Besides Angelo, there was Stewart, a young, working-class Londoner with a shaved head and a hoop earring; Jenny, an affable woman in her 50s who spoke of her fondness for arts and crafts; Michelle, a heavyset woman who dominated the session with her forceful opinions; and David, a 60-something man with a thick gray beard and a pageboy haircut who slumped in his seat and dozed throughout much of the meeting.

Angelo was the newest member of this group — it was his third visit — and he did not seem inclined to participate fully. When Kutchinsky opened the meeting by asking each member to discuss the previous week's experience hearing voices, he softly mentioned that his voices made it difficult to read, then quickly ceded the floor. What followed was sometimes painful. Stewart in particular was visibly agitated. His hallucinatory life, as he described it, was chaotic and irrepressible. He heard voices pleading to him for help; he heard the voices of strangers; he heard the voice of his father. Sometimes he heard the voices of military commandos, who offered to defend him against this confusion. "I haven't been well for a long time," he said glumly. Yet most of the members spoke of their voices in the way that comedians speak of mothers-in-law: burdensome and irritating, but an inescapable part of life that you might as well learn to deal with. When David's name was called, he lifted his head and discussed his struggle to accept his voices as part of his consciousness. "I've learned over time that my voices can't be rejected," he said. "No matter what I do, they won't go away. I have to find a way to live with them." Jenny discussed how keeping busy quieted her voices; she seemed to have taken a remarkable number of adult-education courses. Michelle expressed her belief that her voices were nothing more exotic than powerfully negative thoughts. "Negative thoughts are universal," she said. "Everyone has them. Everyone. What matters is how you cope with them: that's what counts."

I had trouble gauging Angelo's reaction throughout these testimonies, so afterward I pulled him aside and asked him what he thought. "It's interesting to hear people's stories," he said. "Before I started coming, I hadn't realized just how long some people have suffered. I've heard voices for six years. Some people have heard them for 15 or 20. It's amazing." I asked him if this knowledge reassured or frightened him. "It's a bit scary, in a way. I think, I could be this way for a long time." Still, he appeared to appreciate the camaraderie. For years, he had been socially isolated. He spends most of his time with his parents and a sympathetic older sister. His neighbors know only that he is "off work." It was comforting, he said, to speak at last with people who understood.

The meeting that I attended in London is one of dozens like it affiliated with a small but influential grass-roots organization known as Hearing Voices Network. Based in Manchester, Hearing Voices Network (H.V.N.) has since its inception, in 1991, developed a range of services related to the phenomenon known as auditory hallucination: a hot line for people who suffer from the experience, a series of educational workshops for mental-health professionals and 170 support groups across Britain, with more in development. H.V.N., which openly challenges the standard psychiatric relationship of expert physician and psychotic patient, might be said to take the consumer movement in mental health care to its logical endpoint. Although H.V.N. groups meet in a variety of settings — from psychiatric wards to churches to the organization's headquarters — all must be run by, or there must be active plans for them to be run by, voice-hearers themselves. What's more, H.V.N. groups must accept all interpretations of auditory hallucinations as equally valid. If an individual comes to a group claiming that he is hearing the voice of the queen of England, and he finds this belief useful, no attempt is made to divest him of it, but rather to figure out what it means to him.

H.V.N.'s ecumenical approach makes it a difficult organization to pin down. I have met members who believed that their voices were a result of a biochemical glitch, requiring all the tools modern pharmacology has to offer; I have met those who believed their voices were signs from the spirit realm — a cherished gift. Yet the organization's clearest rhetorical note is oppositional and antipsychiatric.

For more than a half-century, auditory hallucinations have primarily been studied and discussed in terms of severe mental illness, most notably schizophrenia, and linked to bizarre delusions, disordered thought and emotional dissociation. Approximately 75 percent of patients diagnosed with schizophrenia hear voices, and for the majority the experience is overwhelmingly negative. Those voices may issue commands, comment sarcastically on everyday actions or berate, curse and insult the hearer. As many as one-third of people with schizophrenia attempt suicide; as many as one-fifth hear voices that command them to do so. H.V.N. does not dispute that auditory hallucinations are frequently painful: many of the organization's leading members have endured harrowing voices themselves and, at one time or another, sought psychiatric help.

What H.V.N. does dispute is that the psychological anguish caused by hearing voices is indicative of an overarching mental illness. This argument, disseminated through a quarterly newsletter, numerous pamphlets and speeches and alternative mental-health journals, are as voluminous and diverse as its membership. But H.V.N.'s brief against psychiatry can be boiled down to two core positions. The first is that many more people hear voices, and hear many more kinds of voices, than is usually assumed. The second is that auditory hallucination — or "voice-hearing," H.V.N.'s more neutral preference — should be thought of not as a pathological phenomenon in need of eradication but as a meaningful, interpretable experience, intimately linked to a hearer's life story and, more commonly than not, to unresolved personal traumas. In 2005, Louise Pembroke, a prominent member of H.V.N., proposed a World Hearing Voices Day (held the next year) that would "challenge negative attitudes toward people who hear voices on the incorrect assumption that this is in itself a sign of illness, an assumption made about them that is not based on their own experiences, is stigmatizing, isolating and makes people ill."

H.V.N.'s insistence that it is not just the psychotic who hear voices does not, in fact, contradict psychiatric orthodoxy. According to the Diagnostic and Statistical Manual of Mental Disorders, the so-called bible of psychiatry, auditory hallucinations are only a potential symptom of mental illness — they must appear with other symptoms, persist for a specified length of time and impede day-to-day functioning in order to become part of a diagnosable syndrome. In a 2001 debate on whether voices are by definition pathological, Tony David, a neuropsychiatrist at the Institute of Psychiatry in London, noted that a "voice-hearer who is not in any distress, who lives a fruitful and productive life according to commonsense criteria, would never enter the arena in which the possibility of mental illness was up for discussion." Nor does psychiatry insist that the syndrome in question when a voice-hearer is in distress is invariably schizophrenia. Approximately 20 percent of patients suffering from mania and 10 percent of patients suffering from depression hear voices. Auditory hallucinations can also be caused by "organic" conditions, like Parkinson's, Alzheimer's, temporal-lobe epilepsy, hyperthyroidism and migraine headaches, and have long been known to occur in the twilight consciousness between wakefulness and sleep.

That said, H.V.N.'s insistence that voice-hearers should attend carefully to what their hallucinations say is far from traditional. Prolonged exposure to untreated psychosis is held by many experts to be damaging to an individual's ability to hold down a job or to maintain a meaningful relationship and by others to be damaging to brain function — what clinicians refer to as "psychosocial toxicity" and "neurotoxicity," respectively. And though psychiatrists acknowledge that almost anyone is capable of hallucinating a voice under certain circumstances, they maintain that the hallucinations that occur with psychoses are qualitatively different. "One shouldn't place too much emphasis on the content of hallucinations," says Jeffrey Lieberman, chairman of the psychiatry department at Columbia University . "When establishing a correct diagnosis, it's important to focus on the signs or symptoms" of a particular disorder. That is, it's crucial to determine how the voices manifest themselves. Voices that speak in the third person, echo a patient's thoughts or provide a running commentary on his actions are considered classically indicative of schizophrenia.

Interpreting voices in relation to a patient's past has a checkered history in the treatment of psychosis. Though Freud discouraged the application of psychoanalysis to psychotic patients, it nonetheless became, for 25 years after World War II, a widespread treatment for schizophrenia in the English-speaking world. This episode in psychiatry is now widely acknowledged to have been a medical and moral disaster; crippling psychoses were routinely blamed on insufficiently nurturing and "schizophrenogenic" mothers. "The psychoanalytic approach to psychosis was toxic," says Peter Weiden, a professor of psychiatry at SUNY Downstate Medical Center in Brooklyn. "Clinicians of that time were often highly antagonistic toward family members. They blamed the parents, left them out of the treatment process and isolated the patient from his family." Over the past 30 years, the biomedical model displaced the psychoanalytic one, bolstered by advances in pharmacology, modern genetic and neurological research and the completion of large-scale empirical studies that concluded that psychoanalysis was useless at best and actively destructive at worst. Today, medication is typically prescribed to extinguish, or at least mitigate, voices (about 80 percent of patients experience a reduction in voices, Weiden says, from medication alone, though this does not always translate into an equal improvement in day-to-day functioning); psychotherapy is usually admitted as an adjunct, to deal with issues of social functioning and stigma.

There are signs, however, that psychotherapy is again encroaching on the biomedical paradigm in the treatment of psychoses. Since the 1990s, a growing number of researchers and clinicians, predominantly based in England, have been comparing voice-hearing in psychotic patients with voice-hearing in nonpatients, measuring the incidence of hallucinations in the general population, and using cognitive behavioral therapy (C.B.T.), a popular, short-term treatment for depression and anxiety, to help them manage their responses to the voices they continue to hear. C.B.T. typically asks patients to scrutinize how they interpret their symptoms rather than focusing on an illness as an underlying cause. "The matter of whether it's effective, and to what extent," Lieberman says, is still being investigated. So far, the use of C.B.T. in the treatment of psychoses is much more prevalent in the U.K. than in the U.S. In large part, Lieberman says, this is because "the motivation to research the treatment has mostly come from investigators in England." But, he added, "you could also read into the situation the influence of a strong antipsychiatry or antimedication movement in England — there's more of an interest in getting nonmedication treatments into clinical use."

In England, this new cognitive approach to psychosis and the efforts of Hearing Voices Network are independent of each other, and are sometimes at odds. H.V.N.'s leading members, for instance, frequently criticize even sympathetic academic researchers for being insufficiently political. Yet both approaches share a similar purpose in seeking to place voice-hearing within the continuum of normal human experience — one, in order to better treat patients, the other, out of a firm conviction that hearing voices need not interfere with leading an otherwise "normal" life. Over the years, they have forged something of an alliance; psychologists, though they may not embrace H.V.N.'s more polemical views, frequently refer their patients to H.V.N. groups, while H.V.N. frequently cites the research of psychologists. And both H.V.N. and the cognitive approach to psychosis can be traced, to varying degrees, to the same radical figure.

When H.V.N. is accused of being hostile to psychiatry, its members sometimes point out that the organization was, in effect, founded by a psychiatrist — albeit a singularly unorthodox one. In 1986, Marius Romme, a professor of psychiatry at Maastricht University in the Netherlands, was referred a patient, Patsy Hage, who suffered from chronic auditory hallucinations and fell into a deep, suicidal depression. Hage took comfort only from reading "The Origin of Consciousness in the Breakdown of the Bicameral Mind," an eccentric book, published in 1976, by Julian Jaynes, a Princeton psychologist, in which he argues that before around 2,000 B.C., all humans were guided by hallucinated verbal commands caused by a physical split between the right and left hemispheres of the brain.

Romme wondered whether Hage might benefit from communicating this theory with other voice-hearers and arranged to appear with his patient on a popular Dutch television program. They invited people who heard voices to contact them: 450 people called in, one-third of whom claimed they were able to live alongside their voices without much difficulty. It struck Romme that this smaller group, the existence of which surprised him, might serve as a therapeutic resource. He asked 20 men and women who had learned to manage their voices to serve as speakers at a conference for voice-hearers. The governing principle of the meeting was that all interpretations of voice-hearing, no matter how unusual, would be accepted.

Today, H.V.N.'s members speak of that first conference as the birthplace of their organization — and, indeed, of a worldwide Hearing Voices movement. There are currently self-help organizations for people who hear voices in more than 15 countries, including Germany, Japan and Australia. (The group has only recently begun to make inroads in the U.S.) Meanwhile, Romme has emerged as a spirited leader-activist, increasingly speaking of psychiatry in terms of cultural and personal oppression. People who hear voices, he has declared, "are like homosexuals in the 1950s — in need of liberation, not cure." H.V.N. often echoes this protesting stance. Its annual conferences, held in Manchester, project an antiestablishment tenor. Ron Coleman, a prominent member, sports a tattoo that reads "Psychotic and Proud" and looks forward to a day when he can "walk the streets talking to his voices and not be denied his freedom."

In its publicly disseminated material, however, H.V.N. tends to focus less on Romme's rhetoric than on his research. In an article published with a group of colleagues in The Journal of Nervous and Mental Disease in 1998, Romme reported on the hallucinations of three groups: patients with schizophrenia, patients with dissociative disorder and nonpatients. All three groups heard a mixture of positive and negative voices (though the patients heard more negative voices than nonpatients), reported traumatic experiences in their past and heard both external and internal voices. What for Romme distinguished the voices of the patients most from those of the nonpatients was that the latter felt that they had control over their voices; they rarely sought to eradicate or ignore them and devised ways of coping with and understanding them. "Helping the patient to accept the voices and actively developing effective coping strategies with the patient," Romme had noted earlier, "may well prove an effective adjunct to psychiatric rehabilitation."

The concept of "coping" is central to H.V.N., based on its belief that people feel better not when their voices are extinguished but when the person hearing voices learns to listen to his hallucinations without anguish. Jacqui Dillon, the national chairwoman of H.V.N., embraces this credo based on personal experience. Dillon, a mother of two, has heard voices for more than 30 years and has never taken medication for them. Mostly, she says, her voices are supportive and even witty, though occasionally they are cruel — they swear and tell her to harm herself. But she no longer heeds their commands or allows them to bother her. Instead, she takes them as symbols of her unconscious thoughts. "Sometimes voices carry messages that you don't want to hear," she told me. "Nevertheless, you don't shoot the messenger. You listen to him."

When Romme's gospel of hallucinatory "acceptance" emerged in the early 1990s, it sparked a notably sharp rebuke in the mainstream British Medical Journal. Reviewing "Accepting Voices" (1993) — a hodgepodge of research findings, coping strategies and firsthand testimonies edited by Romme and his wife and colleague, Sandra Escher — Raymond Cochrane, a professor of psychology at the University of Birmingham, wrote, "Anything that may encourage people to accept the reality of delusional beliefs, and even attribute to these beliefs some mystical supernatural power, can only prolong the existence of these beliefs and make recovery from schizophrenia more protracted and more uncertain."

But Romme's work influenced a number of researchers and clinicians, even those not inclined to agitate for the liberation of voice-hearers. "By the late 1980s, hearing voices had become such a stigmatized experience, people had forgotten that it is not just the insane who hear voices," says Douglas Turkington, a psychiatrist at the Royal Victoria Infirmary, in Newcastle. "The standard line was: 'Don't talk to patients about the experience. It'll only make it worse.' " Romme's report that there existed numerous people living in the community who heard voices and were not distressed, Turkington says, provided ballast for a psychotherapeutic approach to schizophrenia that he and a colleague, David Kingdon, a professor of psychiatry at the University of Southampton, had already embarked on. Adapting the techniques of cognitive behavior therapy, they started in the late 1980s to lead patients, through Socratic-style questioning, toward an understanding of their hallucinations as coming from their own minds. The therapy included mitigating patients' fears of madness by pointing out that even "normal" people can hear voices. Turkington and Kingdon's efforts are now widely credited with helping to reopen the door to psychotherapeutic approaches to psychotic symptoms.

Richard Bentall, a professor of psychology at the University of Manchester, embraced not only the clinical implications of Romme's research but also some of his political activism. Over the past 15 years, Bentall has garnered a reputation as an opponent of traditional psychiatric diagnostics and as an enthusiastic supporter of consumer-based therapies. In the early 1990s, as a professor at the University of Liverpool, Bentall supported the first H.V.N. group in that city, and last August he joined Romme and H.V.N. at a news conference held to announce the Campaign for the Abolition of the Schizophrenia Label, which they billed as "the last great civil rights movement."

For Bentall, schizophrenia is the diagnostic equivalent of a circus tent, sheltering a heterogeneous crowd of experiences and serving to stigmatize patients. Clinicians, he maintains, should be treating individual symptoms, not syndromes. He is not merely indulging in the kind of rhetoric favored by the antipsychiatrists of the 1960s and 1970s, with whom he is sometimes compared. Questioning the validity of schizophrenia as a label is not new — even the architects of the DSM acknowledge that it is useful only insofar as it guides research and treatment. And while Bentall's argument that specific psychotic symptoms should be studied in relation to the psychological mechanisms that give rise to them places him outside the mainstream, he publishes frequently in leading peer-reviewed journals and is often cited by psychologists (who are more inclined than their psychiatric counterparts to place auditory hallucinations on a continuum of everyday experience). He has also received support from the British government for his clinical work, most recently a $2.9 million grant from the U.K. Medical Research Council to investigate whether C.B.T. can prevent people who exhibit early signs of psychosis from developing a full-blown disorder.

In his 2003 book, "Madness Explained," Bentall draws on the theory that auditory hallucinations may have their roots in what psychologists call "inner speech." All of us, every day, produce a steady stream of silent, inward-directed speech: plans, thoughts, quotations, memories. People hear voices, Bentall argues, when they make faulty judgments about whether this inner speech is the product of their own consciousness or of something alien to their consciousness. Lapses in what researchers call "source monitoring" may occur for a number of reasons — because an individual is primed to expect a perception to occur, because the level of background noise makes it difficult to separate what is internal from what is external, because he or she is in a state of emotional arousal. But whatever the cause, Bentall writes, there is evidence to suggest that hallucinating "can be explained in terms of the same kinds of mental processes that affect normal perceptual judgments."

This theory raises the critical question of why making source-monitoring errors results in psychosis: why, when people mistake their private speech for someone else's, does it cause them to grow so distressed that they seek professional help? The answer Bentall gives echoes Romme's observation that a fundamental difference between voice-hearers in the community and voice-hearers under psychiatric care is that the latter think negatively about their experience. According to Bentall, how patients perceive auditory hallucinations can have a significant impact on how those hallucinations are experienced. Bentall cites a landmark 1994 study by Max Birchwood, at the University of Birmingham, and Paul Chadwick, of the Royal South Hants Hospital in Southampton, to support this view. Published in The British Journal of Psychiatry, it proposes that the anguish experienced by patients who hear voices is directly related to their beliefs in the malevolence and power of the voices. A more recent study, published in 2004 and led by Anthony Morrison, a colleague of Bentall's at the University of Manchester, also found that negative beliefs about voices are associated with an increase in the distress of the experience, and often these negative beliefs are reinforced by both mental-health services and the media.

According to Louise Johns, a psychologist at the Institute of Psychiatry in London, cognitive models of psychosis have had a significant impact on mental-health practice in Britain. Few psychologists trumpet C.B.T. as a panacea; it is considered an adjunct to, not a replacement for, standard medical intervention, most frequently in cases where patients do not respond well to medication or are chronically ill. Since the early 1990s, more than 20 randomized clinical trials have been conducted with C.B.T. for psychosis, showing that C.B.T. can be useful in helping patients cope with psychotic symptoms, increasing insight into psychosis and promoting compliance with medication. Based on the findings of these studies, the National Institute for Health and Clinical Excellence — the advisory body of the British National Health Service — in 2002 announced that all "individuals with schizophrenia who are experiencing persistent psychotic symptoms should be offered C.B.T."

Members of H.V.N. express a wary appreciation of these developments. Many are glad that the mental-health professions are taking a psychological approach to psychosis seriously (their literature even notes the increasing availability of cognitive behavioral therapy), but they worry that the organization's more radical, populist message is being obscured. As always, said Jacqui Dillon, the chairwoman of Hearing Voices Network, the mainstream is attending to what scientists have to say about how to treat voice-hearing rather than what nonscientists like her have to say about how to accept the phenomenon.

It was just before noon on a mild Friday in January when the North Finchley hearing-voices group reconvened after a 15-minute coffee break. A sixth participant had joined the group: Chris, light-haired, overweight and audibly short of breath, who has been a member for four years. He seemed completely at ease.

Earlier in the day, Kutchinsky and Kaiser printed out a list of coping strategies that another group's members had found useful, cutting each description into thin rectangles, which they now spread across the table, facedown. The participants were asked to choose one and discuss. Angelo picked first: "Hobbies." He cleared his throat, and in a gentle, measured voice, began: "Collecting, day or evening classes, visiting a library, computer skills, reading and sport. All these activities are not only fun and relaxing; they can fill voids in our lives and help to occupy us during the day or evening. They can improve concentration and reduce isolation. They can also boost our morale and confidence and give us a feel-good factor." Finished, Angelo lifted his head from the paper and looked around.

"Well, do you have any hobbies, Angelo?" Kutchinsky asked.

"I like to play chess," he said. "And, as I've said, I like to read. But it's difficult. I can really only handle something light or humorous. Like Dave Barry."

"I see. Does reading help to block the voices at all?"

Angelo's eyes seemed to darken. "No. I'm afraid nothing at all blocks the voices. Even if I play music really loud, it doesn't help at all."

The coping strategies that followed were within the same vein as the first — commonsensical lifestyle suggestions geared toward improving one's frame of mind, or sanding down the edges of the experience's effects. Liana chose "Exercise"; Jenny chose "Religious Activities"; David chose "Pamper Yourself" ("Put nice music on in the next room, put some scented candles around the room. You could even have a bath with your partner!"). The most novel strategy, and the only one that seemed to cause the group's members to perk up, came under the heading of "Mobile Phones." If you have the temptation to yell at your voices in public, one suggestion went, you should do so with a phone to your ear. That way you can feel free to let loose, and no one who sees you will think you're crazy. Chris in particular seemed to cozy to the suggestion. "I sometimes talk to my voices in public," he said matter-of-factly. "It's very upsetting. I have to bite my knuckles to suppress the urge."

Participants in H.V.N.'s self-help groups take comfort from strategies like these not least because they approach voices as you would approach any other painful but normal experience, like anxiety or stress. Many of the members of the North Finchley group, however, pursue pharmaceutical treatment in addition to coping strategies: they talk at the same time that they are on pills. Indeed, as Kutchinsky told me, they sometimes talk about their pills. They talk about the best dosage and about how to deal with the anxiety and sluggishness and impotence that pills sometimes cause. There is, of course, nothing wrong with this combination. It might even be ideal: who better to talk to about medication than people who are actually on it? And if you were struggling from an experience as complicated as hearing voices, why wouldn't you use every means at your disposal to deal with it?

As for Angelo, his concern is not to choose one option over another — but only to recover. "I have found the group interesting," Angelo wrote via e-mail three weeks after we met. "It has made me realize that many voice-hearers have had the problem for many years, and that many never stop hearing the voices, though some are successful in that regard. One lady has recently quit the group as she no longer hears voices. I also see that some hearers are quite high-functioning and are able to hold down a job despite the voices. I hope to do this myself. Perhaps the right combination of drugs will make this possible."

Daniel B. Smith is the author of "Muses, Madmen and Prophets: Rethinking the History, Science and Meaning of Auditory Hallucination," just published by the Penguin Press and from which this article is in part adapted.