Why do people talk to themselves, and when does it become a problem?
Talking to your yogurt again,” my wife, Pam, said. “And what does the yogurt say?”
She had caught me silently talking to myself as we ate breakfast. A conversation was playing in my mind, with a research colleague who questioned whether we had sufficient data to go ahead and publish. Did the experiments in the second graph need to be repeated? The results were already solid, I answered. But then, on reflection, I agreed that repetition could make the statistics more compelling.
I often have discussions with myself—tilting my head, raising my eyebrows, pursing my lips—and not only about my work. I converse with friends and family members, tell myself jokes, replay dialogue from the past. I’ve never considered why I talk to myself, and I’ve never mentioned it to anyone, except Pam. She very rarely has inner conversations; the one instance is when she reminds herself to do something, like change her e-mail password. She deliberately translates the thought into an external command, saying out loud, “Remember, change your password today.”
Verbal rehearsal of material—the shopping list you recite as you walk the aisles of a supermarket—is part of our working memory system. But for some of us talking to ourselves goes much further: it’s an essential part of the way we think. Others experience auditory hallucinations, verbal promptings from voices that are not theirs but those of loved ones, long-departed mentors, unidentified influencers, their conscience, or even God.
Charles Fernyhough, a British professor of psychology at Durham University, in England, studies such “inner speech.” At the start of “The Voices Within” (Basic), he also identifies himself as a voluble self-speaker, relating an incident where, in a crowded train on the London Underground, he suddenly became self-conscious at having just laughed out loud at a nonsensical sentence that was playing in his mind. He goes through life hearing a wide variety of voices: “My ‘voices’ often have accent and pitch; they are private and only audible to me, and yet they frequently sound like real people.”
Fernyhough has based his research on the hunch that talking to ourselves and hearing voices—phenomena that he sees as related—are not mere quirks, and that they have a deeper function. His book offers a chatty, somewhat inconclusive tour of the subject, making a case for the role of inner speech in memory, sports performance, religious revelation, psychotherapy, and literary fiction. He even coins a term, “dialogic thinking,” to describe his belief that thought itself may be considered “a voice, or voices, in the head.”
Discussing experimental work on voice-hearing, Fernyhough describes a protocol devised by Russell Hurlburt, a psychologist at the University of Nevada, Las Vegas. A subject wears an earpiece and a beeper sounds at random intervals. As soon as the person hears the beep, she jots notes about what was in her mind at that moment. People in a variety of studies have reported a range of perceptions: many have experienced “inner speech,” though Fernyhough doesn’t specify what proportion. For some, it was a full back-and-forth conversation, for others a more condensed script of short phrases or keywords. The results of another study suggest that, on average, about twenty to twenty-five per cent of the waking day is spent in self-talk. But some people never experienced inner speech at all.
In his work at Durham, Fernyhough participated in an experiment in which he had an inner conversation with an old teacher of his while his brain was imaged by fMRI scanning. Naturally, the scan showed activity in parts of the left hemisphere associated with language. Among the other brain regions that were activated, however, were some associated with our interactions with other people. Fernyhough concludes that “dialogic inner speech must therefore involve some capacity to represent the thoughts, feelings, and attitudes of the people with whom we share our world.” This raises the fascinating possibility that when we talk to ourselves a kind of split takes place, and we become in some sense multiple: it’s not a monologue but a real dialogue.
Early in Fernyhough’s career, his mentors told him that studying inner speech would be fruitless. Experimental psychology focusses on things that can be studied in laboratory situations and can yield clear, reproducible results. Our perceptions of what goes on in our heads are too subjective to quantify, and experimental psychologists tend to steer clear of the area.
Fernyhough’s protocols go some way toward working around this difficulty, though the results can’t be considered dispositive. Being prompted to enter into an inner dialogue in an fMRI machine is not the same as spontaneously debating with oneself at the kitchen table. And, given that subjects in the beeper protocol could express their experience only in words, it’s not surprising that many of them ascribed a linguistic quality to their thinking. Fernyhough acknowledges this; in a paper published last year in Psychological Bulletin, he wrote that the interview process may both “shape and change the experiences participants report.”
More fundamentally, neither experiment can do more than provide a rough phenomenology of inner speech—a sense of where we experience inner speech neurologically and how it may operate. The experiments don’t tell us what it is. This hard truth harks back to William James, who concluded that such “introspective analysis” was like “trying to turn up the gas quickly enough to see how the darkness looks.”
Nonetheless, Fernyhough has built up an interesting picture of inner speech and its functions. It certainly seems to be important in memory, and not merely the mnemonic recitation of lists, to which my wife and many others resort. I sometimes replay childhood conversations with my father, long deceased. I conjure his voice and respond to it, preserving his presence in my life. Inner speech may participate in reasoning about right and wrong by constructing point-counterpoint situations in our minds. Fernyhough writes that his most elaborate inner conversations occur when he is dealing with an ethical dilemma.
Inner speech could also serve as a safety mechanism. Negative emotions may be easier to cope with when channelled into words spoken to ourselves. In the case of people who hear alien voices, Fernyhough links the phenomenon to past trauma; people who live through horrific events often describe themselves “dissociating” during the episodes. “Splitting itself into separate parts is one of the most powerful of the mind’s defense mechanisms,” he writes. Given that his fMRI study suggested that some kind of split occurred during self-speech, the idea of a connection between these two mental processes doesn’t seem implausible. Indeed, a mainstream strategy in cognitive behavioral therapy involves purposefully articulating thoughts to oneself in order to diminish pernicious habits of mind. There is robust scientific evidence demonstrating the value of the method in coping with O.C.D., phobias, and other anxiety disorders.
Cognitive behavioral therapy also harnesses the effectiveness of verbalizing positive thoughts. Many athletes talk to themselves as a way of enhancing performance; Andy Murray yells at himself during tennis matches. The potential benefits of this have some experimental support. In 2008, Greek researchers randomly assigned tennis players to one of two groups. The first was trained in motivational and instructional self-talk (for instance, “Go,” “I can,” “Shoulder, low”). The second group got a tactical lecture on the use of particular shots. The group trained to use self-talk showed improved play and reported increased self-confidence and decreased anxiety, whereas no significant improvements were seen in the other group.
Sometimes the voices people hear are not their own, and instead are attributed to a celestial source. God’s voice figures prominently early in the Hebrew Bible. He speaks individually to Adam, Eve, Cain, Noah, and Abraham. At Mt. Sinai, God’s voice, in midrash, was heard communally, but was so overwhelming that only the first letter, aleph, was sounded. But in later prophetic books the divine voice grows quieter. Elijah, on Mt. Horeb, is addressed by God (after a whirlwind, a fire, and an earthquake) in what the King James Bible called a “still small voice,” and which, in the original Hebrew (kol demamah dakah), is even more suggestive—literally, “the sound of a slender silence.” By the time we reach the Book of Esther, God’s voice is absent.
In Christianity, however, divine speech continues through the Gospels—the apostle Paul converts after hearing Jesus admonish him. Especially in evangelical traditions, it has persisted. Martin Luther King, Jr., recounted an experience of it in the early days of the bus boycott in Montgomery, in 1956. After receiving a threatening anonymous phone call, he went in despair into his kitchen and prayed. He became aware of “the quiet assurance of an inner voice” and “heard the voice of Jesus saying still to fight on.”
Fernyhough relates some arresting instances of conversations with God and other celestial powers that occurred during the Middle Ages. In fifteenth-century France, Joan of Arc testified to hearing angels and saints tell her to lead the French Army in rescuing her country from English domination. A more intimate example is that of the famous mystic Margery Kempe, a well-to-do Englishwoman with a husband and family, who, in the early fifteenth century, reported that Christ spoke to her from a short distance, in a “sweet and gentle” voice. In “The Book of Margery Kempe,” a narrative she dictated, which is often considered the first autobiography in English, she relates how a series of domestic crises, including an episode of what she describes as madness, led her to embark on a life of pilgrimage, celibacy, and extreme fasting. The voice of Jesus gave her advice for negotiating a deal with her frustrated and worried husband. (She agreed to eat; he accepted her chastity.) Fernyhough writes imaginatively about the various registers of voice she hears. “One kind of sound she hears is like a pair of bellows blowing in her ear: it is the susurrus of the Holy Spirit. When He chooses, our Lord changes that sound into the voice of a dove, and then into a robin redbreast, tweeting merrily in her ear.”
Forty years ago, Julian Jaynes, a psychologist at Princeton, published a landmark book, “The Origin of Consciousness in the Breakdown of the Bicameral Mind,” in which he proposed a biological basis for the hearing of divine voices. He argued that several thousand years ago, at the time the Iliad was written, our brains were “bicameral,” composed of two distinct chambers. The left hemisphere contained language areas, just as it does now, but the right hemisphere contributed a unique function, recruiting language-making structures that “spoke” in times of stress. People perceived the utterances of the right hemisphere as being external to them and attributed them to gods. In the tumult of attacking Troy, Jaynes believed, Achilles would have heard speech from his right hemisphere and attributed it to voices from Mt. Olympus:
The characters of the Iliad do not sit down and think out what to do. They have no conscious minds such as we say we have, and certainly no introspections. When Agamemnon, king of men, robs Achilles of his mistress, it is a god that grabs Achilles by his yellow hair and warns him not to strike Agamemnon. It is a god who then rises out of the gray sea and consoles him in his tears of wrath on the beach by his black ships. . . . It is one god who makes Achilles promise not to go into battle, another who urges him to go, and another who then clothes him in a golden fire reaching up to heaven and screams through his throat across the bloodied trench at the Trojans, rousing in them ungovernable panic. In fact, the gods take the place of consciousness.
Jaynes believed that the development of nerve fibres connecting the two hemispheres gradually integrated brain function. Following a theory of Homeric authorship that assumed the Odyssey to have been composed at least a century after the Iliad, he pointed out that Odysseus, who is constantly reflecting and planning, manifests a self-consciousness of mind. The poem’s emphasis on Odysseus’ cunning starts to seem like the celebration of the emergence of a new kind of consciousness. For Jaynes, hearing the voice of God was a vestige of our past neuroanatomy.
Jaynes’s book was hugely influential in its day, one of those rare specialist works whose ideas enter the culture at large. (Bicamerality is an important plot point in HBO’s “Westworld”: Dolores, an android played by Evan Rachel Wood, is led to understand that a voice she hears, which has urged her to kill other android “hosts” at the park, comes from her own head.) But Jaynes’s thesis does not stand up to what we now know about the development of our species. In evolutionary time, the few thousand years that separate us from Achilles are a blink of an eye, far too short to allow for such radical structural changes in the brain. Contemporary neurologists offer alternative explanations for hearing celestial speech. Some speculate that it represents temporal-lobe epilepsy, others schizophrenia; auditory hallucinations are common in both conditions. They are also a feature of degenerative neurological diseases. An elderly relative with Alzheimer’s recently told me that God talks to her. “Do you actually hear His voice?” I asked. She said that she does, and knows it is God because He said so.
Remarkably, Fernyhough is reluctant to call such voices hallucinations. He views the term as pejorative, and he is notably skeptical about the value of psychiatric diagnosis in voice-hearing cases:
It is no more meaningful to attempt to diagnose . . . English mystics (nor others, like Joan, from the tradition to which they belong) than it is to call Socrates a schizophrenic. . . . If Joan wasn’t schizophrenic, she had “idiopathic partial epilepsy with auditory features.” Margery’s compulsive weeping and roaring, combined with her voice-hearing, might also have been signs of temporal lobe epilepsy. The white spots that flew around her vision (and were interpreted by her as sightings of angels) could have been symptoms of migraine. . . . The medieval literary scholar Corinne Saunders points out that Margery’s experiences were strange then, in the early fifteenth century, and they seem even stranger now, when we are so distant from the interpretive framework in which Margery received them. That doesn’t make them signs of madness or neurological disease any more than similar experiences in the modern era should be automatically pathologized.
In his unwillingness to draw a clear line between normal perceptions and delusions, Fernyhough follows ideas popularized by a range of groups that have emerged in the past three decades known as the Hearing Voices Movement. In 1987, a Dutch psychiatrist, Marius Romme, was treating a patient named Patsy Hage, who heard malign voices. Romme’s initial diagnosis was that the voices were symptoms of a biomedical illness. But Hage insisted that her voice-hearing was a valid mode of thought. Not coincidentally, she was familiar with the work of Julian Jaynes. “I’m not a schizophrenic,” she told Romme. “I’m an ancient Greek!”
Romme came to sympathize with her point of view, and decided that it was vital to engage seriously with the actual content of what patients’ voices said. The pair started to publicize the condition, asking other voice-hearers to be in touch. The movement grew from there. It currently has networks in twenty-four countries, with more than a hundred and eighty groups in the United Kingdom alone, and its membership is growing in the United States. It holds meetings and conferences in which voice-hearers discuss their experiences, and it campaigns to increase public awareness of the phenomenon.
The movement’s followers reject the idea that hearing voices is a sign of mental illness. They want it to be seen as a normal variation in human nature. Their arguments are in part about who controls the interpretation of such experiences. Fernyhough quotes an advocate who says, “It is about power, and it’s about who’s got the expertise, and the authority.” The advocate characterizes cognitive behavioral therapy as “an expert doing something to” a patient, whereas the movement’s approach disrupts that hierarchy. “People with lived experience have a lot to say about it, know a lot about what it’s like to experience it, to live with it, to cope with it,” she says. “If we want to learn anything about extreme human experience, we have to listen to the people who experience it.”
Like other movements that seek to challenge the authority of psychiatry’s diagnostic categories, the Hearing Voices Movement is controversial. Critics point out that, while depathologizing voice-hearing may feel liberating for some, it entails a risk that people with serious mental illnesses will not receive appropriate care. Fernyhough does not spend much time on these criticisms, though in a footnote he does concede the scant evidentiary basis of the movement’s claims. He mentions a psychotherapist sympathetic to the Hearing Voices Movement who says that, in contrast to the ample experimental evidence for the efficacy of cognitive behavioral therapy, “the organic nature of hearing voices groups” makes it hard to conduct randomized controlled trials.
Fernyhough is not only a psychologist; he also writes fiction, and in describing this work he emphasizes the role of hearing voices. “I never mistake these fictional characters for real people, but I do hear them speaking,” he writes in “The Voices Within.” “I have to get their voices right—transcribe them accurately—or they will not seem real to the people who are reading their stories.” He notes that this kind of conjuring is widespread among novelists, and cites examples including Charles Dickens, Joseph Conrad, Virginia Woolf, and Hilary Mantel.
Fernyhough and his colleagues have tried to quantify this phenomenon. Ninety-one writers attending the 2014 Edinburgh International Book Festival responded to a questionnaire; seventy per cent said that they heard characters speak. Several writers linked the speech of their characters to inner dialogues even when they are not actively writing. As for plot, some writers asserted that their characters “don’t agree with me, sometimes demand that I change things in the story arc of whatever I’m writing.”
The importance of voice-hearing to many writers might seem to validate the Hearing Voices Movement’s approach. If the result is great literature, it would be perverse to judge hearing voices an aberration requiring treatment rather than a precious gift. It’s not that simple, however. As Fernyhough writes, “Studies have shown a particularly high prevalence of psychiatric disorders (particularly mood disorders) in those of proven creativity.” Even leaving aside the fact that most people with mood disorders are not creative geniuses, many writers find their creative talent psychologically troublesome, and even prize an idea of themselves as, in some sense, abnormal. The novelist Jeanette Winterson has heard voices that she says put her “in the crazy category,” and the idea has a long history: Plato’s “mad poet,” Aristotle’s “melancholic genius,” and John Dryden’s dictum that “great wits are sure to madness near allied.” But, in cases where talent is accompanied by real psychological disturbance, do the creative benefits really outweigh the costs to the individual?
On a frigid night in January, 1977, while working as a young resident at Massachusetts General Hospital, I was paged to the emergency room. A patient had arrived by ambulance from McLean Hospital, a famous psychiatric institution in nearby Belmont. Sitting bolt upright, laboring to breathe, was the poet Robert Lowell. I introduced myself and performed a physical examination. Lowell was in congestive heart failure, his lungs filling with fluid. I administered diuretics and fitted an oxygen tube to his nostrils. Soon he was breathing comfortably. He seemed sullen and, to distract him from his predicament, I asked about a medallion that hung from a chain around his neck. “Achilles,” he replied, with a fleeting smile.
I’ve no idea if Lowell knew of Jaynes’s book, which had come out the year before, but Achilles was a figure of lifelong importance to him, one of many historical and mythical figures—Alexander the Great, Dante, T. S. Eliot, Christ—with whom he identified in moments of delusional grandiosity. In Achilles, Lowell seemed to find a heroic reflection of his own mental volatility. Achilles’ defining attribute—it’s the first word of the Iliad—is mēnin, usually translated as “wrath” or “rage.” But in a forthcoming book, “Robert Lowell, Setting the River on Fire: A Study of Genius, Mania, and Character,” the psychiatry professor Kay Redfield Jamison points out that Lowell’s translation of the passage renders mēnin as “mania.” As it happens, mania was Lowell’s most enduring diagnosis in his many years as a psychiatric patient.
In her account of Lowell’s hospitalization, Jamison cites my case notes and those of his cardiologist in the Phillips House, a wing of Mass General where wealthy Boston Brahmin patients were typically housed. Lowell wrote a poem about his stay, “Phillips House Revisited,” in which he overlays impressions of the medical crisis I had witnessed (“I cannot entirely get my breath, / as if I were muffled in snow”) with memories of his grandfather, who had died in the same hospital, forty years earlier.
There was a long history of mental illness in Lowell’s family. Jamison digs up the records of his great-great-grandmother, who was admitted to McLean in 1845, and who, doctors noted, was “afflicted with false hearing.” Lowell, too, suffered from auditory hallucinations. Sometimes, before sleep, he would talk to the heroes from Hawthorne’s “Greek Myths.” During a hospitalization in 1954, he often chatted to Ezra Pound, who was a friend—but not actually there. Among his contemporaries, recognition of Lowell’s mental instability was inextricably bound up with awe of his talent. The intertwining of madness and genius remains an essential part of his posthumous legend, and Lowell himself saw the two as related. Jamison quotes a report by one of his doctors:
Patient’s strong emotional ties with his manic phase were very evident. Besides the feeling of well-being which was present at that time, patient felt that, “my senses were more keen than they had ever been before, and that’s what a writer needs.”
But Jamison also shows that Lowell sometimes saw his episodes of manic inspiration in a more coldly medical light. After a period of intense religious revelation, he wrote, “The mystical experiences and explosions turned out to be pathological.” Splitting the difference, Jamison suggests that his mania and his imagination were welded into great art by the discipline he exerted between his manic episodes.
Lowell was discharged from Mass General on February 9th. Jamison quotes a note that one of my colleagues wrote to the doctors at McLean: “Thank you for referring Mr. Lowell to me. He proved to be just as interesting a person and a patient as you suggested he might be.” Later that month, Lowell had recovered sufficiently to travel to New York and do a reading with Allen Ginsberg. He read “Phillips House Revisited.” That September, he died. ♦