What Dreams Are (Really) Made Of

The psychiatrist and neuroscientist Allan Hobson suggests replacing the traditional Freudian view—that dreams stem from unacceptable, hidden wishes and fears—with a more commonsense theory. Dreams, he says, are caused by spontaneous electrochemical signals in the brain, and their meaning is transparent, not obscure


IN THE SPRING OF 1900, SHORTLY AFTER THE PUBLICAtion of The Interpretation of Dreams, Sigmund Freud wrote a letter to a friend. “Do you suppose,” he asked, “that someday one will read on a marble tablet on this house: ‘Here, on July 24, 1895, the secret of the dream revealed itself to Dr. Sigm. Freud’?”

Freud’s faith in his theory never wavered. Nine years later he told an American lecture audience that “the interpretation of dreams is in fact the royal road to a knowledge of the unconscious. It is the securest foundation of psychoanalysis and the field in which every worker must acquire his convictions and seek his training.”

Two decades after that, looking back on The Interpretation of Dreams in his old age, Freud still felt pride of authorship.

“It contains,” he wrote, “even according to my present-day judgement, the most valuable of all the discoveries it has been my good fortune to make. Insight such as this falls to one’s lot but once in a lifetime.”

The world has echoed that verdict.

Virtually all scholars of psychoanalysis agree with Freud that his dream book was his most important work. Perhaps more significant, Freud’s dream theory has become an inescapable part of modern culture. Even people who reject much of Freudian theory as dubious or bizarre, who would never give credence to talk of Electra complexes or penis envy, make an exception for dreams.

We all accept, as a commonplace, that dreams bubble up from a troubled subconscious, that they represent hidden and mysterious wishes, and that they require deciphering. In both popular culture and high culture these notions are generally accepted without argument. As Walt Disney’s Cinderella put it, “A dream is a wish your heart makes when you’re fast asleep. Move from Hollywood to Harvard and the view doesn’t change much. In a discussion oi a book called Dream Time, Sven Birkerts, a Cambridge-based literary critic, notes in passing, “Ah, but we are all now children of Freud. We know that nothing in dreams is really accidental.”

In 1977 Freud’s dream theory was finally commemorated with the plaque he had hoped for. In that same year, by coincidence, a campaign began that would enlist all the tools of modern neuroscience in an effort to dethrone Freud and to vanquish the cult of the dream.

Chief among the would-be debunkers is a Harvard psychiatrist and neuroscientist named Allan Hobson. In a steady series of books and lectures and research papers and debates he has argued that the psychoanalytic theory of dreams is a museum piece, as outdated as theories of possession by demons. Not surprisingly, he has an alternative theory.

Hobson is fifty-seven, trim, and a bit above medium height, with wispy white hair and a long, thin nose, which was rearranged by muggers a couple of decades ago. He is an animated man, a good talker, whose voice rises and whose delivery speeds up as he works his way to a punch line. He made his reputation as a scientist, but his fascination whth dreams has led him into diverse realms.

On one wall of his lab hangs a fan letter from Federico Fellini. Hobson is a film buff who has spent time on the set with Fellini and written an essay on Bergman’s dream imagery. Now, on a winter morning in his office at the Massachusetts Mental Health Center, a facility affiliated with Harvard Medical School, he is recounting a dream from the previous night.

Hobson was talking with a woman at what seemed to be a reunion of his medical school class. What was strange, what made the dream “dreamy,” was that he couldn’t quite place her. “Only eight women were in my class, and I know them all,” he says. “Some of the data suggested classmate A, but the woman’s actual appearance was closer to that of classmate B, though not precisely.”

What is the psychological significance of that confusion? Hobson’s voice grows loud and indignant. “That was the best I could do. It’s the best that my mind could do under the circumstances.” He pauses for breath and adds a soupgon of incredulity to the mix, his voice now almost a squeak. “It’s not my mother, or somebody else that’s stuck in there dressed like my medical school classmates. My mind, or brain-mind, was making the best of a bad job. It was trying to fit the thing together into some whole meaning, and it didn’t work.”

That picture of the dreamer as a kind of sorcerer’s apprentice, racing madly to keep up with a flood of imagery, is central to Hobson’s theory. Every night, he savs, the dreaming brain automatically generates a barrage of signals that we do our best to assemble into a coherent story. The imagery itself has no “message,” but the mind, waking or dreaming, cannot help investing its world with meaning. “I walk out this door,” Hobson says, swinging his chair around his tiny office, “and I see the coatrack standing there. It’s got my coat on it, my hat, but when I look at it, I see a person.”

He snorts in derision at his own gullibility. “It’s happened fifty times. I fill in, I project. I know that it happens, but I look around startled. That’s clear evidence that in the waking state I’m taking bits of form and filling in the holes. And that’s what happens in dreams.”

This view stands conventional thinking about dreaming on its head. Dreams are caused by electrochemical signals darting helter-skelter around the brain, like untied balloons released in a room. The familiar expression “I had a dream” should probably be reversed: “A dream had me.”

Why Dreams Are Bizarre

LIKE THE ANONYMOUS NOVELIST WHOM NOEL Coward described as “every other inch a gentleman,” Allan Hobson is intermittently a modest fellow. His theory is bold—it aims to supplant freud, after all—but many of his claims for it are surprisingly limited.

To begin with, Hobson has restricted his attention to the formal properties of dreams, the features that all dreams share. He wants to know why dreams are bizarre, why they are vivid, and why they are hard to remember. The specifics of a given dream—why I dreamed of my grandfather last night—lie outside his reach.

The theory is not a ploy to dismiss dreams. Hobson is inordinately fond of them and has kept a dream journal off and on since 1973. One night recently I heard a family friend who had come to Hobson’s house for dinner proffer a dream the way a guest of another household might present a bottle of wine or a dessert.

Hobson does not deny that dreams have meaning. They are revealing, he says, much as interpretations of Rorschach inkblots can be. The particular narrative that a dreamer fashions from randomly generated signals does reflect his preoccupations and hopes and fears.

The dispute isn’t over whether dreams have a meaning but over where their meaning lies. Hobson’s dreamer reveals himself by what he adds to a jumble of apparently unrelated elements. Freud’s view was just the opposite. The unconscious, he said, teems with secret, forbidden wishes that we cannot bear to acknowledge. To guard our sleep, a censor disguises and subtracts information from our dreams so that we can endure them. Dreams seem strange and full of gaps and scene shifts because the censor has gotten to the newsstand ahead of us, tearing out incriminating pages, blacking out key sentences, disguising photographs.

Hobson concedes that we all walk around with painful memories that we do our best to banish. But he emphatically rejects Freud’s view that those repressed memories are the cause of dreams. Instead, he says, dreams are caused by the brain’s spontaneous self-activation while we sleep.

On the most general level, Hobson and Freud are in accord. Like Freud, Hobson believes that dreams are psychologically significant. Like Freud.

Hobson rejects the dismissive view of one of Freud’s scientifically minded predecessors that the dreaming brain is analogous to “the ten fingers of a man who knows nothing of music wandering over the keys of the piano.”

But on the specific nature of dreams Hobson has little use for Freud. Dreams are not obscure but transparent; they are not censored but unedited; dreaming is not triggered by daily events that resurrect buried memories but is a process as automatic as breathing. Most important, the characteristic strangeness of dreams is not a result of the dreamer’s inability to face up to unpleasant memories. The explanation, according to Hobson, is simply that the dreaming brain is working under adverse conditions, deprived of any access to information from the outside world while laboring to fashion a tale from a cascade of internally generated signals.

Hobson’s and Freud’s shared belief that dreams are meaningful has ancient roots. The Bible tells of Pharaoh’s dream of seven lean cattle following and then devouring seven fat cattle and Joseph’s interpretation that the dream forecast seven years of famine following seven years of plenty.

That view of dreams as hidden prophesies endures in popular culture. The accompanying view, that dreams are messages from gods or angels, has lost its hold. Dreams are indeed messages, we still believe, but thanks to Freud we now look inward to find their source. Freud was not the first intellectual to champion dreams, but, especially among scientists, he was in a minority. In the opinion of most scientists of his day, dreams were mental froth. And even among the psychologists and writers whose views of dreams anticipated Freud, no one had produced a theory with the scope and detail of The Interpretation of Dreams.

Freud’s theory was that the dream we recall, the “manifest” dream, is only a distorted version of the true dream, the “latent dream. This latent dream contains the unacceptable wish that instigated the dream in the first place. The point of the nearly 500 pages of The Interpretation of Dreams is to explain how the two forms of the

dream are related, and how the latent dream can be uncovered.

Hobson’s theory is far less ornate. Where a physiological explanation is at hand, he says, a psychological explanation is unnecessary. “The nonsensical features of dreams are not a psychological defense,”Hobson insists, “any more than the disoriented ramblings of a patient with Alzheimer’s disease are.”

Such barbs are aimed at Freud, but Hobson isn’t a doctrinaire follower of any therapeutic school. “The scientific evidence is very strong in favor of the idea that it is therapists, and not therapies, that help people,” he says. “There’s very little evidence that one school or one technique is better than another. The only exception is behavior therapy, which is probably the treatment of choice for phobias.”

In some ways, though, Hobson’s view of dreaming is similar to Jung’s. Hobson doesn’t have much use for Jung’s archetypal symbols, but he does follow Jung in seeing dreaming as creative rather than neurotic. And he agrees with Jung that dreams are undisguised.

The meaning, he says, is right out on the surface, shouting to the rooftops. “When I’m up for promotion or tenure and there’s a really unbelievable administrative botch of the whole thing,” he says, “I have five years of dreams where I’m missing trains, missing boats, I don’t have my papers, my dossiers not in order.”

He has rattled off that list of calamities at tobacco-auctioneer speed and now can hardly sit still. “This is a transparent reflection of my concern about my credentials, ” he roars. “No problem! It’s not a disguise of my fear of failure or my anxiety that I’m going to succeed. My anxiety is, I’m afraid either those bureaucrats won’t get my records straight or they’ll say no promotion.”

To venture more-elaborate explanations, Hobson says, is a kind of showing off, an entertaining but empty display ot intellectual ingenuity. This is a point Hobson circles back to repeatedly, sometimes sounding as enamored of “plain talk and common sense as a Fourth of July orator praising the homespun wisdom of the American people.

That is an odd stance for a scientist to take (common sense surely tells us that, say, the earth is flat), and in his more considered moments Hobson instead invokes one of the patron saints of science, William ot Occam. That fourteenth-century philosopher spelled out the doctrine now known as Occam’s razor, which says that a simple explanation that fits the facts is preferable to a complex one. Never introduce more than is required for an explanation, Occam declared, or, in Hobson’s free translation, “Given two alternative theories, one of which is straightforward and the other convoluted, you pays your money and you takes your choice.”

Hobson wields Occam’s razor with the flair and selfrighteousness of a knight of old brandishing his sword. Does Freud say that we dream because at night, when our defenses are down, lurid thoughts escape from the dungeon of the unconscious? Whhssst! cuts the razor. We dream because the sleeping brain automatically sparks itself into life every ninety minutes or so. Does Freud say that dreams are bizarre because we censor and disguise their true message? Whhssst! Dreams are bizarre because they’re constructed from random bits and pieces. Does Freud say that we forget dreams largely because even in censored form they’re too painful to acknowledge? WMsssst! We forget them because the dreaming brain happens to be deprived of certain chemicals that are essential for storing memories.

The Origins of Hobson’s Theory

THE SCIENCE UNDERLYING HOBSON’S THEORY of dreaming stems from a discovery by a most unlikely Archimedes, a ne’er-do-well graduate student named Eugene Aserinsky. In 1952 Aserinsky was studying physiology at the University ot Chicago. In the dozen years before that he had tried college but left without a degree, begun dental school but dropped out, served in the Army, and been a social worker. He had never earned even a bachelor’s degree. Now he was, in his words, a “stray cat" whom a kindly professor had taken in, and he was working on a “nonsensical idea" that no one else was interested in.

For no very clear reason Aserinsky wanted to knowhow a person’s eves move while he is asleep, The best way to tackle the problem, he decided, was to observe a sleeper for a full night. The most convenient research Subject available was his eight-year-old son. Armond.

Aserinsky found an ancient, broken-down electroencephalograph machine, abandoned in a university basement. His plan was to tape electrodes near Armond’s eyes and use the electroencephalograph, a machine akin to a lie detector, to record any eye movements.

For week after frustrating week the machine malfunctioned. “It would break down with one ailment and I would fix that, and it would break down with something else, Aserinsky recalls. Throughout this period the pens attached to the EEG would occasionally interrupt their slow, wavy tracing of Armond’s eye movements and begin marking spiky peaks and valleys.

The interruptions seemed to show that the brain was occasionally as active in sleeping as in waking. That didn’t make sense, and Aserinsky figured he still hadn’t fixed his machine. Scientists thought of the sleeping brain as like a house late at night, the day’s hubbub of activity replaced by the quiet hum of rest. We wake refreshed, conventional wisdom had it, because the brain has had a break from work. Aserinsky’s research adviser was one of the leading proponents of this view. Either Aserinsky had made a startling discovery or his machine was still broken, and he didn’t know which.

He phoned the manufacturers. They couldn’t help. He managed to reach the scientist who was the reigning authority on the EEG, and this man advised Aserinsky to abandon the project. “If I had a suicidal nature, this would have been the time, Aserinsky says. Even today, safe in retirement, his tone as he tells the story recalls the panicky young man he was. “I was married, I had a child. I’d been in universities for twelve years with no degree to show for it. I’d already spent a couple of years horsing around on this, I was absolutely finished.”

Finally he saw the solution. He could record the movements of each eye independently. Eyes move in tandem, and if the pens did too, that would suggest that the spiky patterns probably weren’t caused by mechanical problems. This strategy of double-checking the machine turned out to be an old idea, but, Aserinsky says, “it saved my life.”

Episodes of rapid eye movement, Aserinsky was soon convinced, came periodically throughout the night. “Well, it was a pretty quick jump to think of dreaming,”Aserinsky says. “But that wasn’t an idea I readily accepted. As a physiologist, I was more interested in blood and guts than in behavior.”

Aserinsky now recruited a number of volunteers; he woke them up when their eyes began twitching and they reported that they had indeed been dreaming. His adviser asked for a demonstration but, wary of cheating, turned down Aserinsky’s offer to recruit a volunteer and insisted that his own daughter be the test subject. She fell asleep. Soon after, her father’s theory that the sleeping brain is resting was “totally demolished,” Aserinsky says. “It doesn’t exist anymore, except in the Annals of Peculiar Notions.”

In the following years discoveries about rapid-eyemovement sleep tumbled out of laboratories around the world. Wake someone up during REM sleep and about 80 percent of the time he or she will report vivid, elaborate, hallucinatory dreams; wake the person during one of the bursts of particularly intense eye movement that punctuate REM sleep and the odds rise to 95 percent. But if the sleeper isn’t wakened, the dream will almost certainly be lost. Dreams melt quickly: 95 percent of what we dream, perhaps 99 percent, is never remembered.

REM sleep begins some ninety minutes after we fall asleep. The brain begins running at full speed, blood pressure rises, breathing quickens, and the heart beats faster. Muscles become totally relaxed and unresponsive, though eyes and extremities may twitch. The dreamer is floating free in a self-created universe, his churning brain trying to keep its bearings without any cues from the outside world.

Episodes of REM-sleep are separated by calmer, deeper periods of sleep. We may dream during these hiatuses, but such dreams are rarer than REM-sleep ones and tend to be briefer and less bizarre. And every ninety minutes we automatically shift back into REM sleep.

We pass through four or five such dream episodes a night. They grow longer as the night goes on, and total about two hours. Because bed partners tend to fall asleep at roughly the same time and to wake each other by jostling or snoring, their pathways to REM sleep are roughly synchronized. “It may be biologically trivial hut it is nonetheless charming,” Hobson says, that “by sleeping together, couples increase the chances of dreaming together.”

REM sleep has been found in all mammals studied to date except the spiny anteater, and, to a limited extent, in birds and some reptiles. (Any cat or dog owner watching his pet’s twitching eyes and paws could have anticipated Aserinsky’s discovery.) A newborn baby spends about eight hours a day in REM sleep. And before birth, at about thirty weeks after conception, the developing infant appears to spend almost all its time in REM sleep.

Just what infants (let alone animals) could be dreaming about is unclear. David Foulkes, a psychologist at Emory University, in Atlanta, has done the best work on the dreams of children. By monitoring children in a sleep lab and waking them at intervals, he found that children aged three to seven rarely reported that they’d been dreaming. After the age of seven children seem to dream about as often as adults.

The nature of dreams, as well as their frequency, changes with age. The earliest dreams are brief and almost devoid of action—a child might dream of herself asleep in a bathtub. At age five, six, or seven dreams become much longer but the dreamer still figures only rarely as an active participant in the dreams. By age eight or nine children’s dreams begin to become as complex and lengthy as adult ones.

Any of the REM-sleep discoveries could have called Freud’s dream theory into question. If dreams are caused by wishes, as Freud proposed, why should those wishes come every ninety minutes? If dreams are caused by repressed sexual desires, what unmentionable fantasies is a newborn baby entertaining? What of Fido asleep in front of the fireplace?

The challenge to Freud is not so much that the sleeping brain turns out to be active (his hardworking censor fits nicely with that finding) as that it is active at recurrent, predictable intervals. We dream with clockwork regularity. That poses no problem for physiology, which has long focused on explaining the body’s rhythms. For psychology, however, and especially for a theory that dreams reflect individual and idiosyncratic hopes and fears, that regularity is a major mystery.

But the scientific assault on Freud waited another generation. The fortress, apparently, was strong, and didn’t have to be abandoned just because of some sniper fire from the physiologists’ camp. What was needed, in addition to criticism of Freud, was a scientifically based theory that could serve as an alternative to Freudian ideas.

Developing one was the mission that Allan Hobson saw for himself. The confrontational style that was required came naturally. He says, “One of the most important things that has happened to me is that I went to school in England when I was nineteen and was exposed to formal debate.”

He absorbed the lessons well. To this day he can address a single listener in tones more appropriate to a prosecutor making a closing argument to a jury. After summarizing a critic’s charges, he will say, “I submit to you that that is absurd.” “What is most objectionable,” he will cry, as he conjures up a flock of dream-interpreting psychoanalysts, “is that they do it under the mantle of science when it’s not science at all. That is a lie.”

Hobson seems genuinely fond of confrontation. “Some people count their blessings with the number of enemies they have,” one of his colleagues observes, “and I think Allan is like that.”But, surprisingly, he is on good terms with most of these “enemies.” The sparring is serious, but Hobson seems not to take it personally. “He’s the best psychologist working on dreams,” he says of one researcher, and adds in the next breath, “He thinks my theory is bunk, just totally useless.”

To the debater’s combativeness Hobson adds a showman’s flair. In 1977 he helped design an art exhibit cum science experiment that drew 10,500 visitors. The main attraction of the Dreamstage show was a volunteer sleeping behind a one-way mirror while hooked up to gadgets that continuously monitored his brain waves, eye position, and muscle tone. An audience sat in a darkened adjacent room watching colored lights paint those waves of information along the walls. At the same time, a synthesizer converted the waves into music, in effect a kind of improvised jazz composed by the sleeper. When he rolled over or began dreaming, the music grew louder and faster, and crowds of visitors scurried to the one-way mirror to see what was happening.

The Assault Begins

THIS Hobson ATTACK and ON his longtime FRUID BEGAN collaborator, IN 1977, Robert WHEN McCarley, a psychiatrist who teaches at Harvard, published two papers on dreaming in the American Journal of Psychiatry. The articles, written in dry and rigorous prose, were explicitly intended as assaults on psychoanalysis. “I would admit to having created some heat where light might have been more useful, Hobson savs, “but I can tell you, they weren r paying any attention until I turned the heat up a bit. Hobson and McCarley caught the eye of the psychiatric community. The articles generated more letters to the editor than any papers the journal had ever before published. Most were from outraged analysts, who perceived correcth that Hobson and McCarley were deeply skeptical of the theories that guided their profession.

The two articles amounted to a one-two punch. First came a critique of Freud’s “antique neurobiology.” Freud’s dream theory, Hobson and McCarley argued, was based on the brain science of the 1890s, which is now universally agreed to be obsolete. Since those biological ideas had proved false, a psychology built on them must also be mistaken.

The first decade of Freud’s career was devoted to neurobiology and neurology. Freud produced a spate of technical papers on such topics as the nerve cells of crayfish. In 1895, at the age of thirtyeight, he began an ambitious essay now known as the “Project for a Scientific Psychology.” Memorycognition, dreaming, and more were all to be explained biologically, in terms of the activity of brain cells, “The intention,” Freud announced, “is to furnish a psychology that shall be a natural science.”

That goal was never achieved. After a few months of frenzied work in 1895, Freud left the “Project” unfinished. Abandoning neuroscience, he turned his efforts to psvchology. His theory of dreams, Freud later said, was based on a lengthy, painful self-analysis in the mid-1890s rather than on any theory of how the brain works. By probing his own emotions and earliest memories with ruthless honesty, psychoanalytic history has it, Freud unearthed such prizes as the Oedipus complex, the stages of sexual development, and the source of “accidental slips of the tongue. His most important tool was free association, mainly with respect to the material from dreams. Hobson and McCarley didn’t buy it. The seltanalysis story, they insisted, was a myth. Freud’s dream theory was simply a translation of the “Project" into a form that concealed its origins in neurobiology.

The second paper described Hobson and McCarley’s own theory. In the years since it was written. Hobson has continued to refine his model. (McCarley. still a friend and ally, has gone his own way.) The fullest account appears in Hobson’s 1988 book The Dreaming Brain.

Dreaming is so familiar that we. tend to overlook its strangeness. In Hobson’s summary,

We see things, but the lights are out; we imagine running, flying, or dancing the tango, but are paralyzed; we explain the bizarre proceedings to our full satisfaction, but the logic by which we do so is as bizarre as the proceedings; we have intense emotional involvement in the action, but we forget the whole business as soon as it is over. What is going on?

That question, like What is time?, is easy to ask but maddeningly hard to answer. The brain, estimated to contain between 20 billion and 100 billion nerve cells, is one of the most complicated regions in the universe. The brain’s nerve cells communicate in chemical messages called neurotransmitters, and each nerve cell, or neuron, is in simultaneous communication with upwards of 10,000 others. Each cell sends between two and a hundred messages every second, ceaselessly, day and night.

By a process that no one claims to understand, this electrified mound of gray-white Jell-O—like matter somehow becomes conscious. Brain becomes mind. But a daunting chasm separates the two. Physiologists assess one side of the territory, psychologists and therapists the other.

Hobson has a more ambitious (some would say ludicrously ambitious) goal, He wants to travel back and forth across the body-mind chasm, using dreams as the bridge. Formidable as that task would seem to be, the strategy is straightward, because the discovery of REM sleep provides a physiological handle on a psychological state. But the human brain is difficult to study experimentally. So research into the workings of the brain must detour by way of animal subjects.

Most of what sleep physiologists know about the living brain they’ve learned from cats, That may sound like a peculiar choice for research intended to explain the mystery of thought, but it is a practical one. Cat and human brains are roughly similar in design, and for a student of sleep a better subject would be hard to find.

A generation or so ago new tools were developed that provide more-detailed pictures of the brain at work than EEGs could offer. These are microelectrodes that record not electrical activity in general but the activity of single cells in particular. The tiny probes revealed that many neurons in the visual areas of the brain fire at least as often in REM sleep as they do in waking.

That was a surprise. When a wide-awake cat eyes the world, its visual cortex lights up with activity. Let that cat fall asleep, eyes shut tight in a black room, and the same cells will light up just as intensely. The brain interprets its own internally generated signals as if they had come from outside.

Similarly, brain cells that have to do with physical activity fire as intensely in REM sleep as in waking. “As far as the neurons are concerned,” Hobson says, “the brain is both seeing and moving in REM sleep.”

None of this was known in Freud’s day. For Freud, for example, the question of why dreams are so intensely visual was a tricky one. Dreams represent a regression to infancy, he argued, and therefore a return to a mental life dominated by imagery rather than thought. For Hobson, matters are simpler. Dreams are visual because the dreaming brain is bombarded by internally generated signals that make it think it is seeing. (And sensations of taste and smell and pain are rare in dreams because the appropriate regions of the brain aren’t as effectively activated.)

Why do dreamers find themselves trying to run but unable to move? Freud suggested that we are stalemated because our conscious wishes and our unconscious desires are in conflict. Hobson refers instead to physiological studies showing that in REM sleep we are effectively paralyzed. Though our dreams are full of effortless movement, things often go wrong when we try to exert our will and move voluntarily. Then a stalemate does occur, in the best Freudian tradition, but it is between the mind giving the command “Run!” and muscles that are blocked from acting. The dreamer can’t move, so he can’t flee the dragon chasing him.

Much of Hobson’s research has sought answers to a different layer of questions. He, and many others, wanted to find the brain cells that trigger REM sleep. The search focused on the brainstem, a structure atop the spinal column that regulates such “primitive” functions as body temperature and appetite. A French researcher named Michel Jouvet had already homed in on a region of the brainstem called the pons, in 1962. But which cells in the pons were the crucial ones?

Like most questions in neurophysiology, this one was difficult to answer. When you probe with microelectrodes, a tiny miss can land you in a region of cells with a role entirely different from that of the target cells. Moreover, the components of the brain are interconnected in a bewilderingly complex fashion. To test a theory that certain brain cells are essential to REM sleep, for instance, you might try destroying them. But, in the glum words of one neuroscientist, “when you make a hole, you’re going to cut down all the telephone wire that goes through that area as well as the telephone pole.”

In 1973 Hobson made an accidental discovery, a “thrilling find that he calls “the highlight of my scientific career. While looking for cells that fired only in REM sleep, he found a cell that stopped firing in REM sleep.

Though this was the opposite of what he had sought, Hobson kept watching. When REM sleep ended, the cell began firing. When REM sleep resumed, the cell stopped again.

Hobson’s microelectrode had missed its target by a millimeter. His mistake had taken him to a region of the brainstem called the locus ceruleus. Other laboratories made similar findings, locating additional populations of so-called REM-off cells elsewhere in the brainstem, and a new picture of the dreaming brain emerged.

The idea is that the brainstem contains clusters of cells that trigger REM sleep and other clusters of cells that turn REM sleep off. Whether we are dreaming or not depends on which group of cells has the upper hand. Hobson has described this cellular interplay as “a sort of continuous war whose effects spread from the brain stem throughout the brain, taking the mind hostage. This battle for the mind occurs regularly—and silently—every night in our sleep. And the only outward sign may be the fleeting recollection of a dream as we read the morning newspaper!”

The war is regulated by a curious clock that operates in ninety-minute cycles. Hobson’s colleague McCarley developed a theory about its workings after studying the problems of Canadian fur trappers in the 1800s.

Trappers sold fur pelts from lynx and snowshoe hares, but because the lynx preyed on the hares, the populations of the two species fluctuated in balance. When the hare population was large, the lynx population grew large. Eventually, more and more predators meant fewer and fewer hares. Less food for the predators meant, eventually, fewer predators, which meant more hares and thus, eventually, more predators. And so on and on, the two populations rising and falling in cycles, the timing of which depended chiefly on the reproduction rates of the lynx and hares.

All that had been worked out in detail by nineteenthcentury mathematicians. The pleasant surprise for McCarley was that he could use essentially the same model to explain the periodic onset of dreaming. Instead of lynx and hares, he had cells that turn on in REM sleep and cells that turn off in REM sleep. Moreover, the two populations of brain cells functionally silence each other. Just as prey and predators compete for territory, the two populations of cells compete for control of the mind. In slow oscillations repeated through the night, first one group holds sway and we sleep deeply, and then the other takes over and we move into REM sleep, and dream.

The model is appealingly tidy, not least because it has several testable consequences. The cells that turn on REM sleep seem to do so by releasing a neurotransmitter called acetylcholine, which is broken down by an enzyme in the brain. If you inject a substance that mimics acetylcholine into the brainstem, you should increase REM sleep. If you block the enzyme that breaks down acetylcholine, you should also increase REM sleep.

And so you do. The model has a flip side, too, that can equally well be tested. The brain cells that turn off REM sleep release their own neurotransmitters. Like playground monitors whose job is to keep rowdy children quiet, these substances tamp down the activity of the cells they come in contact with. Enhance the effect of these neurotransmitters and you should see less REM sleep. Break them down—blindfold the playground monitor— and you should see more REM sleep.

That is what Hobson predicts, but the results of tests of this half of his model aren’t in yet. In any event, REM sleep in cats, which is what is being studied in these experiments, is not dreaming in human beings. And experiments with people have to be oblique, for reasons relating to ethics. In 1978, however, researchers at the National Institute of Mental Health devised a way to test Hobson’s model on human beings. A substance that imitates acetylcholine was given to volunteers intravenously, while they slept. As expected, they quickly entered a long, intense phase of REM sleep. More tellingly, when they were awakened, they reported that they had been dreaming. For the first time, dreams had been triggered artificially.

Hobson’s Critics Have Their Say

HOBSON’S THEORY OF DREAMING HAS SOMETHING to offend everyone. His many scientific critics say that it is premature. The great majority of them concede that the research itself is careful and solid, but they insist that not enough is known about the brain to justify theories of how and why we dream.

Hobson is happy to concede that his theory of dreaming is far from proved. “I agree it’s incomplete,” he says. ”It’s very important to admit that. But I’d also say that you have to make a distinction between the specific burdens of a dream theory and a general theory of consciousness. If you’re interested in why dreams are bizarre, my theory has to interest you. If you want it to solve the whole mind-body question at a single stroke and create a completely detailed theory of human consciousness, you’re going to be disappointed.”

“In one generation we’ve got a rough blueprint of a theory,” Hobson continues. “What do you want? Let’s go on and build a house. That’s going to take a while, probably on the order of hundreds of years. But the door is open to establishing a physical theory of consciousness.”

That’s daring and vision if you like Hobson, grandstanding if you don’t. He concedes that his manner has provoked some of his peers. In one notorious instance, involving a dispute over the properties of a particular group of brain cells, Hobson took years to acknowledge that his critics had been right and he had been wrong.

The skepticism has lingered. “He’s extraordinarily clever, one of the most creative people in the field,” one critic who wishes to remain anonymous acknowledges, “but if you really look into the details of what he says, you say, “This isn’t true, and that’s not true,’ and the whole thing just falls apart.”

Hobson is unruffled by such charges. “If you’re bold and ambitious,” he says, “you can expect a lot of people not to like that.”He points out that he has proposed a conceptual model for thinking about dreaming, and he notes (correctly) that even most of his detractors, while criticizing him on details, accept the framework he has suggested. “Allan’s an unusual scientist, in that he does propose fairly general theories,” says Robert Moore, the chairman of the neurology department at the State University of New York at Stony Brook. “I think people who say he’s a flimflam artist are the ones who aren’t smart enough to do that kind of thing themselves.”

But Hobson’s scientific critics can also muster substantive objections. Over the years, Hobson has modified his model of the dreaming brain significantly. He argued originally, for example, that one particular localized group of cells triggered REM sleep, but he now maintains that the trigger is distributed in multiple locations. Even McCarley, Hobson’s collaborator of sixteen years, has reservations about this approach.

And some predictions implicit in the model haven’t panned out. Several distinct groups of cells that turn off in REM sleep have been identified, for example. According to Hobson, those cells secrete neurotransmitters that block REM sleep. Destroy the cells, therefore, and you would expect to disrupt the REM-sleep cycle. But when such cell groups were destroyed in laboratory animals, the REM-sleep cycle continued.


To illustrate how his theory of dreaming departs from the classical psychoanalytic view, as he sees it, Allan Hobson offers dual interpretations of a single dream. The dream transcript, and Hobson’s analyses, are excerpted from his book The Dreaming Brain.

The Dream: My wife, Joan, and I are at the Museum of Fine Arts in Boston to attend a concert in the larger Remis auditorium. It is someone (perhaps John Gibbons) playing a Mozart piano (concerto?) on a large Steinway (no orchestra, but image vague anyway). [The piano is reminiscent of the large Steinway grand in the great hall of the Phillips Collection in Washington, which I visited the previous Saturday. ] As is usual on such “museum” occasions, I am restive, feeling like the third wheel on Joan’s business bicycle, and hence inattentive.

I decide to explore and go down to the smaller, older theater (near the Egyptian sarcophagi). This theater is now limited to small lectures but twenty years ago was the place where, as young members, Joan and I attended museum programs of the type that are now in Remis, and now under Joan’s direction.

I hear music and the faint bustle of excitement. Opening the door a crack, I am amazed to realize that Mozart himself is on stage, playing the same concerto (again without orchestra) on an antique harpsichord from the museum collection (not the Mozart pianoforte). Although the door is open only for an instant, I notice Mozart’s rich red brocaded frock coat (the curlicues are gold-embossed) and his white powdered wig. He has a beatific smile, and the arpeggios stream through the door into my ear. I also notice that Mozart has gotten a bit overweight, and wonder why.

I close the door with a shhh!, and try to figure out how to tell Joan of my discovery.

Then I wake up.


Where does the energy for the dream process come from?

What is the nature of the energy during this dream?

What is the cause of the sensory aspects of this dream?

In what direction does information processing proceed?

What accounts for the bizarreness of this dream; the corpulence of Mozart, for example?

What does the Mozart dream mean?


From an external stimulus that could not be discharged as a response because of conflict.

An unconscious idea: for example, a wish to kill my father so I can have my mother to myself.

Hearing music and seeing concert halls is a regression, to the sensory side, from the unacceptable ideational dream stimulus (such as a wish to kill my father).

The report is only the manifest content, which has been designed to conceal the dream’s true meaning.

Mozart is an obvious symbol of a powerful, venerated but unapproachable male—that is, my father. The fact that he is overweight clinches the argument (but my father is not overweight!).

Opaque: You hate your father and you want to kill him! But you can’t face that base wish, so you turn him into a great man and laud him.


It is intrinsic to the brain.

From the neurons of the brain, each of which is an electrical generator and capacitaror.

The music I “heard” in the dream is quite familiar to me. I often listen ro Mozart piano concertos while driving. The scenes (in the museum) are also familiar.

I go there often with my wife, who is a program director. I have just been to the Phillips Gallery.

From low-level neural signals impinging on my mind and auditory cortex, I elaborate the sensations that cohere in a plot around the related themes: concert, Mozart, museum.

Mozart is Mozart. I have seen the film Amadeus at a special museum showing, so he fits, although in the dream it was not a film.

The body-type file has been opened and incongruous plot features are the result: my belly has begun to bulge!

Transparent: I would love to see Mozart, to have my wife “score” by attracting him to the museum, and to discover him there so that I could report her coup to others.

At a deeper level I am prepared to admit that this dream might have a psychoanalytic “meaning"; I am ambitious. I do admire Mozart. I would, consciously, like to be as brilliant as. Mozart. Some of my most devoted f riends have even called me “Mozart. Bur to propose that Mozart is a stand-in for my father seems less plausible than to assume that, under the suspended cognitive rules of REM sleep, Mozart is Mozart. I saw him and heard him, discovered him even, in an obscure corner of the museum. I found the dream pleasant, surprising, and gratifying. I also enjoyed telling my wife this story. From a social point of view, my dream was a belated wedding present!

Copyright © 1988 by J. Allan Hobson. Reprinted by permission of Basic Books, Inc.

Is that because the system is redundant? This is Hobson’s explanation, and he points out that no one has destroyed all the cells at once. Or is Hobson simply wrong?

“I think that to be useful a scientific theory has to be fairly specific, to go out on a limb and make predictions that could be refuted,”says Jerome Siegel, a physiologist at the University of California Los Angeles and a longtime rival of Hobson’s. “His initial version of the theory did make predictions, and they were refuted. Now his theory seems to have evolved to the point where it makes fewer predictions and the predictions are too vague to refute.”

Such charges may sound as if they could be resolved simply, by weighing facts impartially. But in the end, the quarrel over Hobson’s model of the dreaming brain comes down to the kind of judgment call we are all familiar with. Hobson looks at his theory and sees a trusty old car that might need an oil change or a new set of spark plugs. His scientific rivals look at the same car and see a patched-together jalopy that has outlived its usefulness. Much as he perturbs some of his fellow physiolo-

Much as he perturbs some of his fellow physiologists, the harshest attacks on I lobson come from outside science. For traditional psychoanalysts, Hobson’s approach was doomed from the start, because of his refusal to use their methods. “I can’t convince anybody that there are microbes in this world unless I apply a light microscope,”says Theodore Shapiro, the editor of the Journal of the American Psychoanalytic Association. “I can’t tell you anything about the ultrastructure of cells unless I apply an electron microscope. How can Hobson say anything about the insights to be derived from dreams if he doesn’t apply the psychoanalytic method?”

The New Dream Interpreters

L ATELY HOBSON HAS BEEN STIRRING UP A NEW group. These are therapists who hold no brief for Freud but are devoted students of dreams. They dislike Hobson’s theory because they feel that it denigrates dreaming. Their opposition is significant, for while these are lean days for psychoanalysis, the dream business is booming. Anyone interested in the meaning of dreams may choose among fifty or more psychology and self-help books. Dozens of colleges and universities around the country, including such unlikely schools as Notre Dame and the Stanford School of Business, offer courses on working with dreams.

The new dream interpreters don’t follow an explicit party line, but they tend to share several beliefs. Dreams can be triggered by any emotional concern, they maintain, and are not the result of repressed wishes alone. Dreams needn’t have to do with sex. Dreams don’t employ or reflect a universal symbolism. Dreams use symbols as metaphors to convey meaning rather than as disguises to obscure it.

Hobson has no quarrel with those ideas. Nonetheless, he rails against what he calls “the dream-cult people.”Doffing his scientist’s lab coat in favor of his psychiatrist’s tweed jacket, he mounts his attack. “I’ve never had an experience in therapy of feeling that a dream was a turning point of treatment,"he says, “or a revelation of a truth not otherwise suspected or known, or anything else of that nature. And I have had successful therapies where dreams were almost never discussed. I’d rather talk about dreams. I think they’re fun. I think that a full view of human experience includes them. And sometimes I’d have to grant that I learned something important from them. I , But is the dream uniquely valuable, uniquely informative? I would have to say a qualified no. I am just not sure it’s all that useful.”

That is heresy to the dream groups, who have pruned and trimmed the Freudian garden but who continue to huddle under the familiar old tree at its center. Freud’s approach to dream interpretation was off-target, they say, but his basic insight was valid: dreams are a royal road to the unconscious, messages to ourselves that convey truths we might otherwise miss. Perhaps the best-known dream psychologist, Gayle Delaney, has built her entire career around that belief. Delaney knows and likes Hobson, and she shares his anti-Freudianism, but she insists that dreams are invaluable in therapy. “Doing psychology without using dreams,”she says, “is like doing orthopedics without using x-rays.”

Skeptics look at Delaney and never get past her celebrity: she has written a best seller on dreams, and she was the host of a radio show on which people phoned in their dreams. And she lives and works in San Francisco. But she resents those who would dismiss her as “a California touchy-feely.”Not many minutes had passed in our conversation before she pointed out that she “graduated with highest honors from Princeton.”

Delaney has heard people tell her tens of thousands of dreams over the years. Dreamers have characteristic styles, she says. She believes that dreams use symbols and metaphors unique to each dreamer to convey important messages with an uncanny compactness. She repeats the dream of a woman named Barbara as an example: “I was in a pool swimming with my eight-year-old son on my back. I would swim under the water while my son’s head would stay above it. I did this in several short bursts, while my husband was supposed to take a picture of us in this position. But somehow he wasn ‘t getting the picture taken. I was beginning to feel as if I was going to drown if he didn’t get it soon. Each time I surfaced, I asked him, ‘Did you get the picture?’ Each time his answer was ‘Not yet.’”

To Delaney, the message was blatant. The dreamer felt she was drowning under her child-care responsibilities, and her husband didn’t get the picture. Did the dream reveal anything that Barbara didn’t already know? “Dreams take you to a point where you can feel things that you don’t allow yourself to feel or think while you’re awake,” Delaney says. “I would agree that at some level you know it—dreams don’t come out of the ether and tell you what God wants for you. Our dreams tell us what we should know but don’t let ourselves know.

Hobson bristles at such talk. He thinks this business of “approaching the dream in hushed tones” is cultishness. If the function of dreaming is to convey information, how do we explain that ninety-nine messages in every hundred are lost before they are delivered?

A therapist can learn about a client in any number of ways that have nothing to do with dreaming, Hobson says. Inkblots or free association or simply asking someone what he did the day before would serve equally well. “I do most of my therapy without any special recourse to dreams,” Hobson says. “I don’t feel I need to do that to find out what’s going on. And I rarely find out from discussing dreams what I didn’t already know.

Here therapists and psychoanalysts of all persuasions join forces to tackle Hobson. By focusing so much of his attention on the bizarreness of dreams, they say, Hobson dismisses them too glibly. Listen to foreigners telling stories, one analyst says, and it you don t bother to learn their language, their tales will sound like gibberish. “Dreaming is involuntary poetry, another analyst says—and sometimes understanding poetry is hard work.

Undoubtedly, these therapists continue, we have some conscious knowledge of our hopes and tears. But Hobson is missing the point if he thinks that dreams simple restate in symbolic or ornate language truths that we already grasp.

“The point of Freud’s work on dreams, says Mardi Horowitz, a professor of psychiatry at the University of California-San Francisco, “was that we tend to be consciously aware of our more proximal wishes, which are to solve our workaday problems, but we tend to be only dimly aware of why it’s so imperative that we solve those problems. Perhaps we have to solve our problems or we’ll be unloved, or some thought like that. Freud was presenting the idea that the deepest and most unrecognized wishes—unconscious wishes—might sometimes be found in dreams.”

Faced with such challenges from psychotherapists, Hobson retreats just long enough to slip back into his lab coat. There’s nothing wrong with therapists’ interpreting dreams, he says. Their interpretations may even be right. “But the burden of proof is on them to show either that their knowledge is richer or that their results are better,” he says.

Above all, they should make clear that their interpretations lack a foundation in science. “I’d have no quarrel if they’d tell patients, ‘This isn’t based on science; it’s more like interpreting literary texts.’ I say to them, ‘Stand up and be counted. Don’t say you re a doctor if you re an artist.’”

Dreams as Neurotic Symptoms

DESPITE THE INTENSITY OF HIS ANTI-FREUDIAN critique, Hobson’s approach to dreaming is in a sense an homage to Freud. Freud made dreams a subject worthy of serious study, after centuries in which they had been brushed aside as the stuff of fortune-tellers. And both Hobson and Freud, in their efforts to explain the bizarre discontinuities and images of dreams, have taken for granted that these are the central and obvious issue to explore.

In addition, the scientific approach to psychology, which Hobson follows, is the one with which Freud himself began. Freud’s early attempt to establish psychology on a solid scientific foundation, his “Project for a Scientific Psychology,” was a major effort. “[It] does not precisely read like an early draft of psychoanalytic theory, notes the historian and Freud partisan Peter Gay, but Freud’s ideas on the drives, on repression and defense, on the mental economy with its contending forces of energies, and on the human animal as the wishing animal, are all adumbrated here.”

James Strachey, a translator of Freud and the editor of the standard English-language edition of his works, wrote, “The Project’s invisible ghost haunts the whole series of Freud’s theoretical writings to the very end. Throughout his life Freud clung to the hope that someday science would vindicate his early attempts to ground psychology in physiology.

Now Allan Hobson has stepped forward, proclaimed himself the voice of modern science, and announced that Freud had it all wrong. The role is an odd one for Hobson. As a college student, he was a Freud idolater who read and reread everything Freud wrote. His college honors thesis was on Freud and Dostoevsky. Even today Hobson happily acknowledges, “Freud is brilliant. And the dream theory is wonderful, it’s compelling and beautifully written and developed, and it’s rich.”

Freud’s ambition extended far beyond interpreting the dreams of individual patients. His real goal was a theory of the mind. The key to such a synthesis was a crucial analogy: dreams are caused by the festering of unacceptable wishes, just as neuroses are caused by the repressing of unacceptable emotions or memories.

“Freud regarded dreams as if they were neurotic symptoms,”writes Anthony Storr, a British psychiatrist sympathetic to Freud. “Since normal people dream, Freud’s theory of dreams supported the idea that neurotic and normal cannot be sharply distinguished, and paved the way for establishing psycho-analysis as a general theory of the mind which applied to everyone.”

Hobson’s charge that this revolutionary theory of the mind is in fact based on Freud’s abandoned “Project fora Scientific Psychology” is a serious one, since, as noted, everyone today agrees that the 1890s neurobiology of the “Project" is worthless. Freud and his contemporaries thought of the brain as a “passive reflex” machine that could act only in response to messages from outside. A dream was triggered by an event from daily life—say, a run-in with the boss—that somehow unearthed and activated a hidden wish from long ago. That is in sharp contrast with today’s picture of a self-activating brain that can both create and cancel its own energy.

Brain cells are now known to be of two types, excitatory and inhibitory. Excitatory cells transmit electrochemical impulses that increase the activity of the cells they contact; inhibitory cells decrease that activity. In Freuds day only excitatory cells were known. “This meant,” one scientist explains, “that once you got a notion in your head, it was doomed to run around in there forever until you finally decided to do something about it. Or, alternatively, until it found a way to trick you into unconsciously expressing it in some unintended action— like the famous ‘Freudian slip.’”

Similarly, it was thought that repressed wishes would boil and bubble endlessly in the cauldron of the unconscious, until they managed to emerge, suitably disguised, as dreams. In essence, Hobson argues that Freud’s dream theory came into being in a somewhat comparable way: the brain-based picture of the mind that he labored over in the “Project” never panned out, but eventually it emerged, suitably translated into psychological terms, as The Interpretation of Dreams.

But, oddly, this feature of Hobson’s argument seems not to faze the analytic community. “It never dawns on psychoanalysts,” says Frank Sulloway, a historian of science and a revisionist Freud scholar, “that if Freud was wrong about the general properties of dreams, he might also have been wrong about the interpretation of specific dreams. If you say that the whole dream theory is based on outmoded biology, they say, ‘We’ll give you that and keep everything else. It’s as if they lived in a building and someone said, ‘The first floor’s about to collapse,’ and they said, We don’t care; we live on the tenth floor.’”

Even on the tenth floor signs of trouble are visible. Significant numbers of strict freudian analysts are still treating patients, but psychoanalysis has been in decline since the 1960s. Hundreds of alternate forms of psychotherapy have sprung up, the psychoanalytic-training institutes are hard pressed for students, and the leading psychiatric journals have cut down on their psychoanalyticaily based articles. “It’s almost dead, says Robert Moore, the Stony Brook neurologist. ”I know of no institution looking for a chief of psychiatry that’s looking for a psychoanalyst.”

The decline of psychoanalysis is due largely to the rise of biology. Depression, manic depression, schizophrenia, and other mental illnesses seem to be yielding some of their secrets to neurobiology. With support from the National Institute of Mental Health, Congress has declared the 1990s the “decade of the brain.”

Freud’s dream-analysis technique, in particular, has fallen out of favor with some psychoanalysts. Despite Freud’s insistence that dream interpretation is “the securest foundation of psychoanalysis,” contemporary analysts seem to believe that following the weaving course of a patient’s free associations to each dream element takes too long and is too much trouble. Freud’s approach to dreams in therapy may eventually suffer the crudest of all fates—to be deemed not wrong but irrelevant.

Nevertheless, the decline of Freud within the therapeutic community seems not to have significantly affected his reputation in the rest of the academic world, where his standing is as high as it ever was. Many debunking books and articles have appeared, most of them the work of historians or philosophers, but they have not had lasting impact. They are published, they win prizes and respectful audiences, and then their message is forgotten.

Great numbers of literary critics, social scientists, and historians continue to march to a Freudian drummer. Every day sees the publication of a new psychobiography of someone or other, or a new psychoanalytically based work of literary scholarship. On campus, at least, Auden’s words remain apt: Freud is “no more a person now but a whole climate of opinion.”

Dreams as Creative Opportunities

AS INTELLECTUAL FASHIONS CHANGE, DRFAMING falls in and out of favor. It remains as mysterious as ever. For two hours in every twenty-four, for six years in an average lifetime, we are mad as hatters. Why? What is dreaming for?

Dreaming seems to be important. We can’t decide to dream or not to dream, and if some sleep researcher prevents US from dreaming one night, we make up for it the next.

Hobson, in many ways an optimistic man, has devised a much sunnier answer than Freud did to the question Why dream? “Instead of seeing the dream process as some sort of laundry for kinky thoughts,” he says, “I see it as a resourceful artist producing all kinds of wonderful new solutions.”

In this view, dreaming is a virtual parody of scientific thinking, in which every idea can be considered, even outlandish notions can be pursued, and anything is possible. And, indeed, numerous problems have been solved in dreams. Robert Louis Stevenson dreamed the plot of The Strange Case of Dr. Jekyll and Mr. Hyde. Elias Howe claimed that the crucial idea for his sewing machine came to him in a dream, after years of struggle. In all his failed models the needle’s eye was in the middle of the shaft. One night Howe dreamed he had been captured by a tribe of savages who carried spears with eye-shaped holes near the tip.

At least one dreamer earned a Nobel Prize. Otto Loewi, a German-born physiologist and pharmacologist, wanted to know how nerves send signals. Do they simply transmit electricity, like tiny wires, or do they also send chemical signals? Loewi was studying frogs, trving to learn why stimulating the vagus nerve causes the heart to slow. Unlike his peers, he believed that chemicals were somehow involved, but he couldn’t think how to prove he was right.

On Easter Sunday of 1920 the answer came to him in a dream. Loewi woke up, scrawled it down, and fell back asleep. In the morning he couldn’t read his writing and he couldn’t remember his dream.

The next night, at three o’clock, the idea returned,” he wrote later. “I got up immediately, went to the laboratory, and performed the experiment.” Loewi’s inspiration was to stimulate the vagus nerve of one frog, thus slowing its heart, and then to transfer blood from that frog’s heart to a second frog. When the second heart slowed too, Loewi had proved that the vagus nerve acted by releasing a chemical. The chemical turned out to be acetylcholine, the very substance that is now known to be the neurotransmitter that triggered Loewi’s dream in the first place.

Hobson likes such stories, but he is wary of them too. “Nobody ever tells you about all the cockamamie ideas that didn’t work,” he says. “You don’t hear about the guy who went off for ten years and worked on this crazy idea that occurred to him in a dream. In fact, you don’t even know whether two days later it would have occurred to him at breakfast. He laughs, but then turns crabby. You never hear anybody raise those questions!” he shouts. “Because, again, that’s the dream mystique. You want to believe this thing’s wonderful.”

In fact, the notion of problem-solving in dreams is one that therapists of all sorts rush to embrace. The idea rests on the commonsensical premise that we work to make sense of our lives while we ‘re awake and the process continues while we’re asleep.

Perhaps the best-known advocate of this view is Rosalind Cartwright, the chairman of the psychology department at Rush University, in Chicago. The function of dreaming, she says, is to give the mind a chance to sort uninterrupted through emotional issues that we are too preoccupied to untangle during our waking lives. The process goes on automatically, whether or not we can recall our dreams later.

“In dreaming, you update the program of who you are every night,” Cartwright says. “If nothing much has changed in your life, you get a night off to play or be creative or tell jokes to yourself, or you just have a dull night of nothing much going on. But when you’re going through crises, you need to revise who you are, and you have to update that program in a dramatically new way.”

To test her theory, Cartwright has spent several years studying men and women who are going through divorces and are depressed. She has found that their dreams differ in key ways from those of people who are happily married. For people whose lives are going along smoothly, the first dream of the night is typically brief and dull: “I went shopping.” In Cartwright’s reading, the dream hasn‘t much work to do. For most of the depressed subjects in her studies, however, the first dream comes much sooner than it does for others, lasts much longer, and is far more complicated.

And the dreams are terrible, endless and masochistic rehashes of mistakes made and opportunities lost. One of the supposed benefits of anti-depressant medications, in fact, is that they suppress dreaming. But with a nudge from therapy sessions during the day, dreams change during the course of the night. Dreamers become more angry and less depressed. “When they do that,” Cartwright says, “they recover. It’s a predictor that dreaming has gone into high gear, has stirred up the feelings to be worked through. When I see those people at follow-up, they’re no longer depressed.”

Hobson is almost visibly ambivalent about such stories. On the one hand, he is perfectly happy to concede that dreaming has a constructive, creative side. It fits with his theory that the dreamer fashions a plot from whatever materials happen to be at hand, and it furnishes ammunition against “the peculiar modern, psychoanalytic tendency to view even the normal as somehow neurotic.”

On the other hand, as a good scientist, Hobson is fearful of venturing too far into speculative talk of “purpose.” And to talk about the purpose of dreaming is to pile intangible on intangible. He is more comfortable speculating about the purpose of REM sleep.

For example, we spend more time in REM sleep as infants in the womb than we ever will again. Why? Hobson suggests that infants prior to birth are literally “making up their minds, working on tasks that are somehow essential to cognitive development.

Some experiments with adults also seem to link REM sleep and intellectual work. Volunteers deprived of REM sleep by experimenters have more trouble solving complicated puzzles than do people deprived of non-REM sleep. And people trying to assimilate new information— students of a foreign language, for example—show an increase in REM sleep.

But because REM sleep is distributed so widely throughout the animal kingdom, no such explanation is adequate. Both opossums and moles spend substantial amounts of time in REM sleep, for instance, and neither species is noted for perspicacity.

The hardheaded scientific attitude is to dismiss dreams altogether. All mammals undergo REM sleep, the argument runs, human beings in particular have powerful minds, and the combina tion yields the strange experiences we call dreams. Dreaming isn’t for anything; it just happens. Occasionally Hobson talks in this vein, speculating that “dreaming is just the noise the brain makes while it automatically sorts and files,” but more often he takes a softer line.

He prefers to think of REM sleep as a time when the brain refreshes and readjusts itself, a notion that explains why we have so much trouble remembering our dreams.

The explanation is based on a striking observation: the brain cells that turn off in REM sleep release neurotransmitters that are crucial to attention, ing, and memory.

“You have these systems firing all day long like a metronome,” Hobson says, “with zillions and zillions of packers of neurotransmitter being bled out over the day, and your attention sort of runs down. At a certain point the sleep svstem kicks in, you go to bed, and before you know it these neuronal systems are shut down, four metabolic rate doesn t go to zero—you continue to manufacture enzymes and neurotransmitters, and all the packets just get filled up again, You wake up in the morning and you say, ‘God, I feel good. I feel sharp.'”

The brain is ready for another days work, but the night’s dreams are lost, because the dreamer was deprived of the very substances necessary to lay down memories. Hobson’s explanation of REM sleep has still another facet. The dreaming brain also carries out what he calls an “active maintenance" program to test its own circuitry. While the dreamer is barely connected with the outside world, dreaming provides a chance “to run through vour repertoire of instinctive behaviors with the clutch pressed in, to make sure they’re still working in the proper way.”

For Hobson, such armchair speculation is an intellectual version of sport. Ask him a question or give him an analogs to consider and he engages his whole body in the attempt to find an answer. He twists in his chair, throws his arms out, and casts his eyes around to address invisible listeners in every corner of the room. He starts an explanation headed one way, sees a grinning tackier lying in wait, and cuts back in the opposite direction. Sometimes he arrives far downfieid, visibly proud of himself. Occasionally he doubles back and forth so often that when he finally runs out of steam he has made hardly any progress.

The picture is far from that of the stereotype of the scientist as a bloodless logic machine. Indeed, the romantic accusation that science banishes beauty by dissecting it might better be directed at Freud than at Hobson.

than at Hobson. In a curious way, the two men have exchanged roles. Hobson, the neurophysiologist, who might be expected to play the level-headed spokesman for the hard sciences, has devised a theory of dreams that leaves a great deal of room for chance. Dreams are unpredictable and improvised, he argues. The dreamer’s own interpretation can be as valid as a therapist’s. Freud, the psychologist, who might have emphasized the variety of human experience, proposed a rigidly determinist, reductionist theory of dreams. Every feature of a dream, no matter how trivial, is deeply significant. Every dream, “invariably and indisputably,”has the same cause. Dreams can be interpreted only by following the psychoanalytic method.

The years between the two theories saw the most extensive scientific progress the world has ever know n. One might have expected steady progress in revealing the dream in its true guise. But the science of dreaming has brought a different, more humbling message: the dream was undisguised all along. Like Poes purloined letter, it lay hidden in plain sight.

Freud’s dream interpretations were brilliant, so brilliant that in the end he outsmarted himself. It now appears that his rival Carl Jung came nearer the mark. “I was never able to agree with Freud that the dream is a ‘façade’ behind which its meaning lies hidden—a meaning already known but maliciously, so to speak, withheld from consciousness,”Jung wrote. To me dreams are a part of nature, which harbors no intention to deceive, but expresses something as best it can, just as a plant grows or an animal seeks its food as best it can.“