Musicophilia

by Oliver Sacks

  * * *

  MOST OF MY PATIENTS and correspondents with musical hallucinations have had hearing loss, in many cases severe. Many but not all of them have also had some sort of “noise in the ear”— rumbling, hissing, or other forms of tinnitus, or, paradoxically, recruitment— an abnormal and often unpleasant loudness of certain voices or noises. Sometimes additional factors seem to push the person over a critical limit— an illness, or surgery, or a further decrement in hearing.

  That said, about a fifth of my correspondents have no significant hearing loss— and only about 2 percent of those who do have hearing loss develop musical hallucinations (but given the number of aging people with advancing deafness, this means, potentially, hundreds of thousands of people are candidates for musical hallucinations). A majority of my correspondents are elderly, and there is a considerable overlap between the elderly and the hard of hearing. So while neither age nor hearing loss alone is sufficient to cause auditory hallucinations, the conjunction of an aging brain with hearing impairment or other factors may push a frail balance of inhibition and excitation towards a pathological activation of the auditory and musical systems of the brain.

  Some of my correspondents and patients, however, are neither elderly nor hard of hearing; one was a boy of nine.

  There are very few documented cases of musical hallucinations in younger children— though it is not clear whether this represents the actual rarity of such hallucinations in children or their unwillingness or inability to speak of them. But Michael B. had very clear musical hallucinations. His parents said that these were constant, “unrelenting, from morning to night…. He hears one song after another. When he gets tired or stressed out, the music gets louder and distorted.” Michael first complained of this when he was seven, saying, “I’m hearing music in my head…. I have to check the radio to see if it’s really on.” But it seems likely that he first experienced it even earlier, for when he was five, while traveling in the car he would sometimes scream, cover his ears, and ask for the radio to be turned off— when it was not on.

  Michael could not turn his musical hallucinations down or off, though he could suppress or replace them to some extent by hearing or playing familiar music, or by the use of a white-noise generator, especially at night. But as soon as he woke in the morning, he said, the music turned itself on. It can become almost unbearably loud if he is under pressure— he may scream at such times and seems to be in what his mother calls “acoustic agony.” He cries, “Take it out of my head. Take it away!” (This reminded me of a story that Robert Jourdain relates about Tchaikovsky as a child, who was reportedly once found weeping in bed, saying, “This music! It is here in my head. Save me from it!”)

  For Michael, there is never any holiday from the music, his mother stresses. “He has never been able to enjoy the beauty of a quiet sunset, take a silent walk in the woods, quietly reflect, or read a book without hearing a band playing in the background.”

  But he has recently started taking drugs to reduce cortical, and specifically musical, excitability, and he is beginning to show some response to this medication, though his music remains overwhelming. His mother recently wrote to me, “Last evening, Michael was very happy because his internal music stopped for about fifteen seconds. This has never happened before.”11

  * * *

  BESIDES PEOPLE who are tortured by loud and intrusive musical hallucinations, there are others whose musical hallucinations are so soft, so easily disregarded, that they may feel these are not worth seeking treatment for. Such was the case with Joseph D., an eighty-two-year-old retired orthopedist. He was moderately deaf and had given up playing his Steinway a few years earlier, because it sounded “tinny” with his hearing aids and “washed out” without them. He had also, in his increasing deafness, been “banging” on the piano— “My wife kept yelling at me, ‘You’ll break the piano!’ ” The onset of tinnitus (“like steam coming out of a radiator”) started two years before he came to see me, and this was followed by a low humming sound (“I thought it was the refrigerator or something in the kitchen”).

  About a year later, he started to hear “collections of notes, scales up and down, little twists and turns of two or three notes.” These would come on suddenly, repeat themselves for hours, and then disappear just as suddenly. Then, a few weeks later, he heard musical passages (which he recognized as themes from a Beethoven violin concerto) repeated again and again. He never heard the entire concerto, just this medley of themes. He could not specify hearing the sound of a piano or an orchestra— “It is only melody,” he said. He could not banish this by force of will, but it was usually quite soft, easily disregarded or overcome by external sound; it would disappear if he was physically or mentally engaged.

  Dr. D. was struck by the fact that, though his perception of real music was now distorted or washed out because of his hearing loss, his hallucinations were clear, vivid, and undistorted (he had tested this, at one point, by humming into a tape recorder along with his hallucination and then comparing the tape to an original recording— the two coincided exactly in pitch and tempo). Humming itself can produce a sort of echo, a repetition, in his mind.

  I asked him whether he ever got pleasure from his musical hallucinations, and he said, emphatically, “No!”

  Dr. D. was getting used to his hallucinations, which were mercifully mild. “At first I thought I was falling apart,” he said, “but now I just regard it as baggage. As you get older, you accumulate baggage.” Nevertheless, he was glad the baggage was only these relatively soft hallucinations.

  * * *

  WHEN I SPOKE, a few years ago, to a class of perhaps twenty college undergraduates and inquired whether any of them had ever experienced musical hallucinations, I was astonished when three of them said that they had. Two told me rather similar stories of how, playing different sports, they were each briefly knocked unconscious and, upon recovering consciousness, “heard music” for a minute or two— music which they felt came from an external source, perhaps the public-address system, or maybe another student with a radio. A third student told me how he lost consciousness and had a seizure during a karate match when his opponent immobilized him in a too-tight neck hold. Coming around after this, he heard “sweet music,” which seemed to come from outside him, for a couple of minutes.

  Several correspondents have told me about musical hallucinations that occur only when they are in a particular position, usually recumbent. One was a ninety-year-old man described by his physician as healthy and with a “brilliant” memory. When guests at his ninetieth-birthday party toasted him with “Happy Birthday to You” (in English, though both he and they were German), he continued to hear this song, but only when recumbent. It would last for three or four minutes, stop for a while, then restart. He could neither stop it nor provoke it himself, and it never occurred while he was sitting or standing. His physician was struck by certain EEG changes in the right temporal region that were seen only when the patient was lying down.

  A thirty-three-year-old man also experienced musical hallucinations only when recumbent: “Just the movement of lying down on my bed would trigger them, and in a fraction of a second the music appeared…. But if I tried to stand up or even sit up, or even to raise my head slightly, the music would disappear.” His hallucinations were always songs, sometimes sung by individual voices, sometimes by a choir— he called them “my little radio.” This correspondent ended his letter by saying that he had heard about the case of Shostakovich, but that unlike him, he did not have any fragments of metal in his head.12

  Strokes, transient ischemic attacks, and cerebral aneurysms or malformations can all lead to musical hallucinations, but these tend to die down with the subsiding or treatment of the pathology, whereas the majority of musical hallucinations are very persistent, though they may fade a little over the years.13

  A wide range of medications (some that affect the ear itself, like aspirin and quinine, and others that affect the central nervous s
ystem, like propranolol and imipramine) may cause transient musical hallucinations, as can certain metabolic abnormalities, epileptic conditions, or migraine auras.14

  In most cases of musical hallucination, there is a sudden onset of symptoms; then the hallucinatory repertoire expands, becoming louder, more insistent, more intrusive— and the hallucinations may continue even if one can identify and remove the predisposing cause. The hallucinations have become autonomous, self-stimulating, self-perpetuating. At this point, they are almost impossible to stop or inhibit, though some people may be able to shift them to another tune in the “jukebox,” provided it has some similarity of rhythm, melody, or theme. Along with this stickiness or stubbornness, there may develop an extreme susceptibility to new musical inputs, so that whatever is heard is instantly replayed. This sort of instant reproduction has some resemblance to our reaction to catchy tunes, but the experience for someone with musical hallucinations is not mere imagery, but often physically loud, as-if-heard “actual” music.

  These qualities of ignition, kindling, and self-perpetuation are epilepsy-like characteristics (though similar physiological qualities are also characteristic of migraine and of Tourette’s syndrome).15 They suggest some form of persistent, uninhibitable spreading electrical excitement in the musical networks of the brain. Perhaps it is not coincidental that drugs like gabapentin (originally designed as an antiepileptic) are sometimes also useful for musical hallucinations.

  * * *

  HALLUCINATIONS OF MANY SORTS, including musical ones, may also occur if the senses and the perceptual systems of the brain have too little stimulation. The circumstances have to be extreme— such sensory deprivation is not likely to occur in ordinary life, but it may occur if one is immersed for days on end in deep stillness and silence. David Oppenheim was a professional clarinet player and a university dean when he wrote to me in 1988. He was sixty-six, with some mild high-frequency hearing loss. A few years earlier, he wrote, he had spent a week at a monastery retreat deep in the woods, where he took part in a sesshin, an intense meditation practice of nine or more hours a day. After two or three days of this, he started to hear faint music, which he took to be people singing around a distant campfire. The following year he returned, and once again he heard the distant singing, but soon the music got louder and more specific. “At its height the music is quite loud,” he wrote. “It is repetitive and orchestral in nature. It is all slow passages from DvoYák and Wagner…. The presence of this musical trackmakes the meditation impossible.”

  I can summon up DvoYák, Wagner, or anyone else, when not in meditation, but I don’t “hear” them…. In the sesshin version, I hear them.

  There is obsessive repetition of the same musical material, over and over for days at a time…. The “inner” musician cannot be stopped or shut up, but he can be controlled and manipulated…. I was successful in banishing the “Pilgrims’ Chorus” from Tannhäuser by changing it into the slow movement from Mozart’s lovely Symphony 25 in A major, because they start out with the same intervals.

  Not all of his hallucinations were of familiar music— some of it he “composed” but, he added, “I don’t compose at all in my everyday life. I used the word to indicate that at least one of the pieces that played in my head was not DvoYák or Wagner, but new music that I was somehow making up.”

  I have heard similar accounts from some of my friends. Jerome Bruner tells me that when he sailed the Atlantic solo and there were calm days with little to do, he sometimes “heard” classical music “stealing across the water.”

  Michael Sundue, a botanist, recently wrote to me about his own experience as a novice sailor:

  I was twenty-four years old, working as a member of a crew that was hired to deliver a sailboat. We were at sea for a total of twenty-two days. It was very boring. After the first three days, I had read every book that I brought with me. There was nothing to do for entertainment except to watch the clouds and take naps. For days and days, there was no wind, so we simply puttered along at a few knots by running the engine while the sails luffed. I would lie on my back on the deck or on a bench in a cabin and stare out through the Plexiglas window. It was during these long days of complete inactivity that I had several musical hallucinations.

  Two of the hallucinations began from the monotonous and ever-present sounds that were generated by the boat itself. These were the buzzing of the reefer (the small refrigerator), and the whistling sound of the rigging in the wind. Each of these sounds transformed into never-ending instrumental solos. The transformation into music was such that the original sound and its source were forgotten, and I lay there in my lethargic state for long stretches of time just listening to what sounded like stunning and beautiful compositions. Only after enjoying each of these in a sort of daydream-like state did I figure out what the sources of noise were. The instrumental sounds themselves were interesting in that they were not anything I typically listened to for enjoyment. The buzzing noise of the reefer to me sounded like a virtuoso heavy metal guitar solo, an onslaught of high-pitched strings played quickly through a distorting amplifier. The whistling sound of the rigging took on the form of highland Scottish bagpipes with a set of drones and a melody line. The sounds of both of these types of music are familiar to me, but neither are things that I would generally play on my home stereo.

  Around the same time, I would also hear my father’s voice call my name. As far as I know, there was no sound which triggered this. (At one point, I also had a visual hallucination of a shark’s fin emerging from the water. It took no time at all for my companions to discredit my claim that I had seen a shark. They laughed at me. I think from their reaction that seeing sharks was a very common reaction for inexperienced sailors to have.)

  ALTHOUGH COLMAN, in 1894, wrote specifically about “Hallucinations in the Sane, associated with local organic disease of the sensory organs, etc.,” the impression has long remained both in the popular mind and among physicians, too, that “hallucinations” mean psychosis— or gross organic disease of the brain.16 The reluctance to observe the common phenomenon of “hallucinations in the sane” before the 1970s was perhaps influenced by the fact that there was no theory of how such hallucinations could occur until 1967, when Jerzy Konorski, a Polish neurophysiologist, devoted several pages of his Integrative Activity of the Brain to the “physiological basis of hallucinations.” Konorski inverted the question “Why do hallucinations occur?” to “Why do hallucinations not occur all the time? What constrains them?” He conceived a dynamic system which, he wrote, “can generate perceptions, images, and hallucinations…the mechanism producing hallucinations is built into our brains, but it can be thrown into operation only in some exceptional conditions.” Konorski brought together evidence— weak in the 1960s, but overwhelming now— that there are not only afferent connections going from the sense organs to the brain, but “retro” connections going in the other direction. Such retro connections may be sparse compared to the afferent connections, and may not be activated under normal circumstances. But they provide, Konorski felt, the essential anatomical and physiological means by which hallucinations can be generated. What, then, normally prevents this from happening? The crucial factor, Konorski suggested, is the sensory input from eyes, ears, and other sense organs, which normally inhibits any backflow of activity from the highest parts of the cortex to the periphery. But if there is a critical deficiency of input from the sense organs, this will facilitate a backflow, producing hallucinations physiologically and subjectively indistinguishable from perceptions. (There is normally no such reduction of input in conditions of silence or darkness, because “off-units” fire up and produce continuous activity.)

  Konorski’s theory provided a simple and beautiful explanation for what soon came to be called “release” hallucinations associated with “de-afferentation.” Such an explanation now seems obvious, almost tautological— but it required originality and audacity to propose it in the 1960s.

  There is now go
od evidence from brain-imaging studies to support Konorski’s idea. In 2000, Timothy Griffiths published a detailed and pioneering report on the neural basis of musical hallucinations; he was able to show, using PET scans, that musical hallucinations were associated with a widespread activation of the same neural networks that are normally activated during the perception of actual music.

  * * *

  IN 1995 I received a vivid letter from June B., a charming and creative woman of seventy, telling me of her musical hallucinations:

  This first started last November when I was visiting my sister and brother-in-law one night. After turning off the TV and preparing to retire, I started hearing “Amazing Grace.” It was being sung by a choir, over and over again. I checked with my sister to see if they had some church service on TV, but they had Monday night football, or some such. So I went onto the deck overlooking the water. The music followed me. I looked down on the quiet coastline and the few houses with lights and realized that the music couldn’t possibly be coming from anywhere in that area. It had to be in my head.

  Mrs. B. enclosed her “play list,” which included “Amazing Grace,” “The Battle Hymn of the Republic,” Beethoven’s “Ode to Joy,” the drinking song from La Traviata, “A-Tisket, A-Tasket,” and “a really dreary version” of “We Three Kings of Orient Are.”

  “One night,” Mrs. B. wrote, “I heard a splendidly solemn rendition of ‘Old Macdonald Had a Farm,’ followed by thunderous applause. At that moment I decided that, as I was obviously completely bonkers, I’d better have the matter looked into.”