While most examples of autoscopy are fairly brief, long-lasting autoscopy has also been recorded. Zamboni et al. provide a detailed description of this in a 2005 paper. Their patient, B.F., was a young woman who developed eclampsia in pregnancy and was comatose for two days. As she started to recover, it was evident that she was cortically blind and had a partial paralysis on both sides as well as an unawareness of her left side and of the left side of s.p.a.ce, a hemi-neglect. With further recovery, her visual fields became full and she could discriminate color, but she was profoundly agnosic, unable to recognize objects or even shapes. At this stage, Zamboni et al. wrote, their patient first started seeing her own image as if reflected in a mirror, about a meter in front of her. The image was transparent, as though it were set "in a sheet of gla.s.s," but a bit blurry. It was life-sized and consisted of a head and shoulders, though if she looked down, she could see its legs, too. It was always dressed exactly like her. It disappeared when she closed her eyes and reappeared the moment she opened them (although, as the novelty wore off, she was able to "forget" the image for hours at a time). She had no special feelings for the image and never attributed any thoughts or feelings or intentionality to it.

As B.F.'s agnosia disappeared, the mirror image gradually faded, and it had vanished entirely by six months after the original brain injury. Zamboni et al. suggested that the unusual persistence of this mirror image may have been a.s.sociated with her severe visual loss, along with disturbances of multisensory integration (visual, tactile, proprioceptive, etc.) at higher levels, perhaps in the parietotemporal junction.

An even stranger and more complex form of hallucinating oneself occurs in "heautoscopy," an extremely rare form of autoscopy where there is interaction between the person and his double; the interaction is occasionally amiable but more often hostile. Moreover, there may be deep bewilderment as to who is the "original" and who the "double," for consciousness and sense of self tend to shift from one to the other. One may see the world first with one's own eyes, then through the double's eyes, and this can provoke the thought that he-the other-is the real person. The double is not construed as pa.s.sively mirroring one's posture and actions, as with autoscopy; the heautoscopic double can do, within limits, whatever it wants to (or it may lie still, doing nothing at all).

"Ordinary" autoscopy-such as Linnaeus and Lullin experienced-seems relatively benign; the hallucination is purely visual, a mirroring which appears only occasionally, has no pretensions to autonomy, no intentionality, and attempts no interactions. But the heautoscopic double, mocking or stealing one's ident.i.ty, may arouse feelings of fear and horror and provoke impulsive and desperate acts. In a 1994 paper, Brugger and his colleagues described such an episode in a young man with temporal lobe epilepsy:

The heautoscopic episode occurred shortly before admission. The patient stopped his phenytoin medication, drank several gla.s.ses of beer, stayed in bed the whole of the next day, and in the evening he was found mumbling and confused below an almost completely destroyed large bush just under the window of his room on the third floor.... The patient gave the following account of the episode: on the respective morning he got up with a dizzy feeling. Turning around, he saw himself still lying in bed. He became angry about "this guy who I knew was myself and who would not get up and thus risked being late at work." He tried to wake the body in the bed first by shouting at it; then by trying to shake it and then repeatedly jumping on his alter ego in the bed. The lying body showed no reaction. Only then did the patient begin to be puzzled about his double existence and become more and more scared by the fact that he could no longer tell which of the two he really was. Several times his bodily awareness switched from the one standing upright to the one still lying in bed; when in lying in bed mode he felt quite awake but completely paralysed and scared by the figure of himself bending over and beating him. His only intention was to become one person again and, looking out of the window (from where he could still see his body lying in bed), he suddenly decided to jump out "in order to stop the intolerable feeling of being divided in two." At the same time, he hoped that "this really desperate action would frighten the one in bed and thus urge him to merge with me again." The next thing he remembers is waking up in pain in the hospital.

The term "heautoscopy" (sometimes spelled heautoscopy), introduced in 1935, is not always regarded as a useful one. T. R. Dening and German Berrios, for example, write, "We see no advantage in this term; it is pedantic, almost unp.r.o.nounceable, and not widely used in ordinary practice." They see not a dichotomy but a continuum or spectrum of autoscopic phenomena, in which the sense of relationship to one's autoscopic image may vary from minimal to intense, from indifferent to impa.s.sioned, and the sense of its "reality" may be equally variable and inconsistent. In a 1955 paper, Kenneth Dewhurst and John Pearson described a schoolteacher who, at the start of a subarachnoid hemorrhage, saw an autoscopic "double" for four days:

It appeared quite solid as if seen in a mirror, dressed exactly as he was. It accompanied him everywhere; at meal-times it stood behind his chair and did not reappear till he had finished eating. At night it would undress and lie down on the table or couch in the next room of his flat. The double never said anything to him or made any sign, but only repeated his actions: it had a constant sad expression. It was obvious to the patient that this was all a hallucination, but nevertheless it had become sufficiently a part of himself for the patient to draw a chair up for his double when he first visited his private doctor.

In 1844, a century before the term was coined, A. L. Wigan, a physician, described an extreme case of heautoscopy with tragic consequences:

I knew a very intelligent and amiable man, who had the power of thus placing before his eyes himself, and often laughed heartily at his double, who always seemed to laugh in turn. This was long a subject of amus.e.m.e.nt and joke, but the ultimate result was lamentable. He became gradually convinced that he was haunted by [his other] self. This other self would argue with him pertinaciously, and to his great mortification sometimes refute him, which, as he was very proud of his logical powers, humiliated him exceedingly. He was eccentric, but was never placed in confinement or subjected to the slightest restraint. At length, worn out by the annoyance, he deliberately resolved not to enter on another year of existence-paid all his debts-wrapped up in separate papers the amount of the weekly demands-waited, pistol in hand, the night of the 31st of December, and as the clock struck twelve fired it into his mouth.

The theme of the double, the doppelganger, a being who is partly one, partly Other, is irresistible to the literary mind, and is usually portrayed as a sinister portent of death or calamity. Sometimes, as in Edgar Allan Poe's "William Wilson," the double is the visible and tangible projection of a guilty conscience that grows more and more intolerable until, finally, the victim turns murderously on his double and finds that he has stabbed himself. Sometimes the double is invisible and intangible, as in Guy de Maupa.s.sant's story "Le Horla," but this double nonetheless leaves evidence of his existence (for instance, he drinks the water that the narrator sets out in his night bottle).

At the time he wrote this, de Maupa.s.sant often saw a double himself, an autoscopic image. As he remarked to a friend, "Almost every time when I return home I see my double. I open the door and see myself sitting in the armchair. I know it's a hallucination the moment I see it. But isn't it remarkable? If you had not a cool head, wouldn't you be afraid?"

De Maupa.s.sant had neurosyphilis at this point, and when the disease grew more advanced, he became unable to recognize himself in a mirror and, it is reported, would greet his image in a mirror, bow, and try to shake hands with it.

The persecuting yet invisible Horla, while perhaps inspired by such autoscopic experiences, is a different thing altogether; it belongs, like William Wilson and Golyadkin's double in Dostoevsky's novella, to the essentially literary, Gothic genre of the doppelganger, a genre which had its heyday from the late eighteenth century to the turn of the twentieth.

In real life-despite the extreme cases reported by Brugger and others-heautoscopic doubles may be less malign; they may even be good-natured or constructive moral figures. One of Orrin Devinsky's patients, who had heautoscopy in a.s.sociation with his temporal lobe seizures, described this episode: "It was like a dream, but I was awake. Suddenly, I saw myself about five feet in front of me. My double was mowing the lawn, which is what I should have been doing." This man subsequently had more than a dozen such episodes just before seizures, and many others that were apparently unrelated to seizure activity. In a 1989 paper, Devinsky et al. wrote:

His double is always a transparent, full figure that is slightly smaller than life size. It often wears different clothing than the patient and does not share the patient's thoughts or emotions. The double is usually engaged in an activity that the patient feels he should be doing, and he says, "that guy is my guilty conscience."

Embodiment seems to be the surest thing in the world, the one irrefutable fact. We think of ourselves as being in our bodies, and of our bodies as belonging to us, and us alone: thus we look out on the world with our own eyes, walk with our own legs, shake hands with our own hands. We have a sense, too, that consciousness is in our own head. It has long been a.s.sumed that the body image or body schema is a fixed and stable part of one's awareness, perhaps in part hardwired, and largely sustained and affirmed by the continuing proprioceptive feedback from joint and muscle receptors regarding the position and movement of one's limbs.

There was general astonishment, therefore, when Matthew Botvinick and Jonathan Cohen showed in 1998 that a rubber hand, under the right circ.u.mstances, could be mistaken for one's own. If a subject's real hand is hidden under a table while the rubber hand is visible before him, and both are stroked in synchrony, then the subject has the convincing illusion, even though he knows better, that the rubber hand is his-and that the sensation of being stroked is located in this inanimate though lifelike object. As I found when I looked through the "eyes" of a robot, knowledge in such a situation does nothing to dispel the illusion. The brain does its best to correlate all the senses, but the visual input here trumps the tactile.

Henrik Ehrsson, in Sweden, has developed a great range of such illusions, using the simplest equipment-video goggles, mannequins, and rubber arms. By disrupting the usual unity of touch, vision, and proprioception, he has induced uncanny experiences in some people, convincing them that their bodies have shrunk or grown enormous, even that they have swapped bodies with someone else. I experienced this for myself when I visited his laboratory in Stockholm for a number of experiments. In one, I was convinced that I possessed a third arm; in another, I felt embodied in a two-foot-high doll, and as I looked through "its" eyes via video goggles, normal objects in the room appeared enormous.

It is evident, from all of this work, that the brain's representation of the body can often be fooled simply by scrambling the inputs from different senses. If sight and touch say one thing, however absurd, even a lifetime of proprioception and a stable body image cannot always resist this. (Individuals may be more or less susceptible to such illusions, and one might imagine that dancers or athletes, who have an exceptionally vivid sense of where their bodies are in s.p.a.ce, may be harder to fool in this way.)

The body illusions Ehrsson is exploring are very much more than party tricks; they point to the ways in which our body ego, our sense of self, is formed from the coordination of senses-not just touch and vision but proprioception and perhaps vestibular sensation, too. Ehrsson and others favor the idea that there are "multisensory" neurons, perhaps at a number of places in the brain, which serve to coordinate the complex (and usually consistent) sensory information coming into the brain. But if this is interfered with-by nature or experiment-our seemingly una.s.sailable certainties about the body and the self can vanish in an instant.

1. The term "out-of-body experience" was introduced in the 1960s by Celia Green, an Oxford psychologist. While there had been stories of out-of-body experiences for centuries, Green was the first to systematically examine a large number of firsthand accounts, from more than four hundred people whom she located by launching a public appeal through the newspapers and the BBC. In her 1968 book, Out-of-the-Body Experiences, she a.n.a.lyzed these in detail.

2. Several of Celia Green's subjects described similar feelings. "My mind was clearer and more active than before," one wrote; another spoke of being "all-knowing and understanding." Green wrote that such subjects felt they "could obtain an answer to any question they chose to formulate."

3. August Strindberg noted, in his autobiographical novel Inferno, an odd body double, an "other" who mirrored his every movement.

This unknown man never uttered a word; he seemed to be occupied in writing something behind the wooden part.i.tion that separated us. All the same, it was odd that he should push back his chair every time I moved mine. He repeated my every movement in a way that suggested that he wanted to annoy me by imitating me.... When I went to bed the man in the room next to my desk went to bed too.... I could hear him lying there, stretched out parallel to me. I could hear him turning the pages of a book, putting out the lamp, breathing deeply, turning over and falling asleep.

Strindberg's "unknown man" is identical with Strindberg in one sense: a projection of him, at least of his movements, his actions, his body image. Yet, at the same, he is someone else, an Other who occasionally "annoys" Strindberg, but perhaps, at other times, seeks to be companionable. He is, in the literal sense of the term, Strindberg's "Other," his "alter ego."

15

Phantoms, Shadows, and Sensory Ghosts

While hallucinations of sight and sound-"visions" and "voices"-are described in the Bible, in The Iliad and The Odyssey, in all the great epics of the world, none of these so much as mentions the existence of phantom limbs, the hallucinatory feeling that one still has a limb even though it has been amputated. Indeed, there was no term for these before Silas Weir Mitch.e.l.l gave them their name in the 1870s. And yet they are common-more than a hundred thousand people in the United States have amputations every year, and the vast majority of them experience phantoms after their amputations. The experience of phantom limbs must be as old as amputation itself, and amputations are not new-they were performed thousands of years ago: the Rig Veda tells the story of the warrior queen Vishpla, who went to battle with an iron prosthesis after she lost a leg.

In the sixteenth century, Ambroise Pare, a French military surgeon who was called upon to amputate dozens of injured limbs, wrote, "Long after the amputation is made, patients say that they still feel pain in the amputated part ... which seems almost incredible to people who have not experienced this."

Descartes, in his Meditations on First Philosophy, observed that, just as the sense of vision was not always reliable, so "errors in judgment" could occur in the "internal senses" as well. "I have sometimes been informed," he wrote, "by parties whose arm or leg had been amputated, that they still occasionally seemed to feel pain in that part of the body which they had lost-a circ.u.mstance that led me to think that I could not be quite certain even that any one of my members was affected when I felt pain in it."

But by and large, as the neurologist George Riddoch brought out (in 1941), a curious atmosphere of silence and secrecy seems to surround the subject. "Spontaneous description of phantoms is rarely offered," he wrote. "Dread of the unusual, of disbelief, or even of the accusation of insanity may be behind this reticence."

Weir Mitch.e.l.l himself hesitated for years before writing professionally on the subject; he introduced it first in the form of fiction (he was a writer as well as a physician), in "The Case of George Dedlow," published anonymously in the Atlantic Monthly in 1866. As a neurologist working at a military hospital in Philadelphia during the Civil War (the place was informally known as the "Stump Hospital"), Mitch.e.l.l saw dozens of amputees and, driven by his own curiosity and compa.s.sion, he encouraged them to describe their experiences. It would take him several years to fully digest what he had seen and heard from his patients, but in 1872, in his cla.s.sic Injuries of Nerves, he was able to provide a detailed description and discussion of phantom limbs-the first such in the medical literature.1

Mitch.e.l.l devoted the final chapter of his book to phantom limbs, introducing the subject as follows:

No history of the physiology of stumps would be complete without some account of the sensorial delusions to which persons are subject in connection with their lost limbs. These hallucinations are so vivid, so strange, and so little dwelt upon by authors, as to be well worthy of study, while some of them seem to me especially valuable, owing to the light which they cast upon the subject of the long-disputed muscular sense.

Nearly every man who loses a limb carries about with him a constant or inconstant phantom of the missing member, a sensory ghost of that much of himself.

After Mitch.e.l.l had brought attention to the subject, other neurologists and psychologists were drawn to study phantom limbs. Among them was William James, who sent a questionnaire to eight hundred amputees (he was able to contact them with the help of prosthetic manufacturers), and of these, nearly two hundred answered the questionnaire; a few he was able to interview personally.2

Where Mitch.e.l.l's observations, working with Civil War amputees, were of fresh, just developed phantom limbs, James was able to study a much more varied population (one man, in his seventies, had had a thigh amputation sixty years earlier), and so he was in a better position to describe the changes in phantom limbs over years or decades, changes which he described in detail in his 1887 paper on "The Consciousness of Lost Limbs."

James was especially interested in the way that initially vivid and mobile phantoms often tended to shorten or disappear with time. This surprised him more than the presence of phantoms, which he felt was only to be expected with continuing activity in the areas of the brain that represented sensation and movement in the lost limb. "The popular mind wonders how the lost feet can still be felt," James wrote. "For me, the cause for wonder are those in which the lost feet are not felt." Hand phantoms, he observed, unlike leg or arm phantoms, rarely disappeared. (We now know that this is because the fingers and hands have a particularly ma.s.sive representation in the brain.) He did, however, note that the intervening arm might disappear, so that the preserved phantom hand now seemed to sprout from the shoulder.3

He was also struck by the way in which an initially mobile phantom could become immobile or even paralyzed, so that "no effort of will can make it change [its position]." (In rare cases, he said, "the very attempt to will the change has grown impossible.") James saw that fundamental questions were raised here about the neurophysiology of "will" and "effort," though he could not answer them. And they were not to be answered for more than a century, until V. S. Ramachandran clarified the nature of "learned" paralysis in phantom limbs in the 1990s.

Phantom limbs are hallucinations insofar as they are perceptions of something that has no existence in the outside world, but they are not quite comparable to hallucinations of sight and sound. While losing one's eyesight or hearing may lead to corresponding hallucinations in 10 or 20 percent of those affected, phantom limbs occur in virtually all who have had a limb amputated. And while it may be months or years before hallucinations follow blindness or deafness, phantom limbs appear immediately or within days after an amputation-and they are felt as an integral part of one's own body, unlike any other sort of hallucination. Finally, while visual hallucinations such as those of Charles Bonnet syndrome are varied and full of invention, a phantom closely resembles the physical limb that was amputated in size and shape. A phantom foot may have a bunion, if the real one did; a phantom arm may wear a wrist.w.a.tch, if the real arm did. In this sense, a phantom is more like a memory than an invention.

The near universality of phantom limbs after amputation, the immediacy of their appearance, and their ident.i.ty with the corporeal limbs in whose stead they appear suggest that, in some sense, they are already in place-revealed, so to speak, by the act of amputation. Complex visual hallucinations get their material from the visual experiences of a lifetime-one has to have seen people, faces, animals, landscapes to hallucinate them; one has to have heard pieces of music to hallucinate them. But the feeling of a limb as a sensory and motor part of oneself seems to be innate, built-in, hardwired-and this supposition is supported by the fact that people born without limbs may nonetheless have vivid phantoms in their place.4