Mind and Body
DOES multiple personality matter to metaphysics? I do not think so. Metaphysics asks: What is a person, a soul, a self? It does not ask who I am, but what I am. What constitutes me as a person? One answer is well known to English empirical philosophy, for it is at least as old as John Locke. It is almost part of the general culture today: a person is constituted by consciousness and memory. Here is how it crops up in a popular science magazine: “The ability to retrieve a memory decays exponentially, and after only a month more than 85% of our experiences will have slipped beyond reach, unless boosted by artificial aids such as diaries and photographs. Given that our memories are our identities, this is a terrifying rate of loss.”1 It is easy enough to make fun of this. What? I’m losing my identity by the minute? Now that is terrifying! Or perhaps we should reach the opposite conclusion. Our memories are not (all there is to) our identities.
When did it first strike someone that multiple personality might be relevant to philosophical issues? The earliest example I have found is in a pithy editorial by Thomas Wakley, longtime editor of the British medical journal The Lancet. He began the issue of Saturday, 25 March 1843 by dismissing the philosophers addicted to pure reason and ignorant of matters of fact:
From the fact that the philosophy of the human mind has been almost wholly uncultivated by those who are best fitted for its pursuit, the study has received a wrong direction, and become a subtle exercise for lawyers and casuists, and abstract reasoners, rather than a useful field of scientific observation. Accordingly, we find the views, even of the most able and clear-headed metaphysicians, coming into frequent collision with the known facts of physiology and pathology. For example, that “consciousness is single” is an axiom among the mental philosophers, and the proof of personal identity is made by those gentlemen to rest chiefly on the supposed universality or certainty of that allegation. But what would they say to the case of a somnambulist who evinced what is regarded as double consciousness—the operation of the mind being perfectly distinct in the state of somnambulism from its developments in the wakeful condition? With reference to such an individual, the proof of his personal identity must rest with others, not with himself, for his memory in one state takes not the smallest cognisance of what he thought, felt, perceived, said, or did, in the other.2
Wakley refers to the Lockean tradition, but curiously, Locke himself would be unmoved. For on his clearly stated criteria, we have one and the same “man” (that is, woman), and two distinct persons. Now maybe that is a preposterous conclusion, but it is what Locke would consistently have maintained. Locke himself was a physician who might well have illustrated his theory of personal identity by the phenomenon of double consciousness—but no doubles were reported in 1693. Somnambulism was familiar to Locke, but the person in the somnambulistic state had not yet acquired the ability to conduct two distinct existences—or the physicians of the day had not yet discovered the ability to cultivate such phenomena.
Was Wakley correct? Does double consciousness or multiple personality show anything about what it is to be a person, or about the human mind, or about the nature of the self, or about the subject? I do not think so, or at any rate, what it shows is only oblique. At most the progress of multiple personality in Western history teaches something about what ordinary people or experts are prepared to say, and how they are willing to interact with people of unsound mind. We do not find nature’s illustrations of different possible conditions of the human mind, illustrations to which every philosopher of mind should attend. What we find are facts about communities whose central figures are the experts and their patients, but whose circle quickly expands to families, law and order, employers. Thanks to media exposure, the circle has expanded to “everyone” in North America, for everyone knows about multiples now. Television will not put the truly mad on display. Gone are the days when the cruel show at Bedlam was fun. No, we want to feast our eyes on the oddly dysfunctional, not on the crazed or catatonic. Only people with bizarre but manageable mental disorders are broadcast to the world. If multiple personality is a natural experiment, I contend, it is an experiment on the community.
I must make a distinction. Multiple personality shows nothing direct about the mind. That is, it does not furnish any evidence for any substantive philosophical thesis about mind (or self, etc.). The phenomena may certainly illustrate some claim about the mind that is held for reasons quite independent of the phenomena. If so, would not the phenomena be supporting evidence for the philosophical claim? No. I maintain that they furnish no evidence at all. They add nothing but color. The sheer fact that there is a real-life illustration often seems like evidence, but the doctrines that are illustrated are rooted in principles unrelated to multiple personality, and unsupported by its existence. I shall argue my case by citing three very different contemporary philosophers. They have paid very close attention to medical literature or phenomena of multiple personality. They can hardly be subject to Wakley’s strictures. One of them, Stephen Braude, has been intimately involved in the circles of patients and experts around the ISSMP&D. Another, Daniel Dennett and his collaborator Nicholas Humphrey have conducted almost an ethnography of the multiple movement. A third, Kathleen Wilkes, has immersed herself in older literature; whereas Braude and Dennett are American, with plenty of multiples and their clinicians available to talk with, she is English and her knowledge of multiplicity, at least at home, must be gleaned from books.
But first let us turn to two classics, two of the most powerful philosophical minds at work a century ago: William James and Alfred North Whitehead. James’s Principles of Psychology includes an incisive review of the literature of what he called alternating personality.3 He knew the French literature intimately. He also personally interviewed the famous American case of fugue, namely, Ansel Bourne.4 Finally, he was always close to the Boston investigators of psychic phenomena, who had a lot to do with the upsurge and persistence of interest in multiple personality in New England. James’s discussion is at the end of the chapter titled “The Consciousness of Self”—this follows the more famous chapter 11, “The Stream of Thought.” “This long chapter” on consciousness of self concludes with three types of what he calls “mutations of self”: losses of memory or false memories, alternating personality, and mediums. Following Ribot, he took alternating personality to be above all a disturbance of memory, since one personality knows nothing of another personality expressed by the same body at earlier times. James was willing to write that “the anaesthetic and ‘amnesic’ hysteric is one person.” She becomes a different person when “you restore her inhibited sensibilities and memories by plunging her into the hypnotic trance—in other words when you rescue them from their ‘dissociated’ and split-off condition and make them rejoin the other sensibilities and memories.”5 But he lays no great or philosophical weight on the word “person” here, no more than when we commonly say that someone is a different person after a couple of drinks. Indeed William James is a model for all philosophers who would address the mind. He records alternating personality as a phenomenon that leads on to “questions which cannot now be answered.”6 James drew no philosophical inferences whatsoever from alternating personality.
The late philosophy of Alfred North Whitehead reads very differently from that of William James. It demands long periods of absorbed study. I ask absolution from his admirers for making a few superficial remarks about the way that Whitehead used multiplicity in his book Process and Reality. In his view each thing that we commonly think of as an entity is a society. An electron is a society of electron occasions. “Our epoch is to be considered a society of electron occasions.”7 It follows that any organism is a society. But people are special:
In the case of the higher animals there is central direction, which suggests that in their case each animal body harbors a living person, or living persons. Our own self consciousness is direct awareness of such persons. There are limits to such unified control, which is indicated by dissociation of personality, multiple personalities in successive alternations, and even multiple personalities in joint possession.8
From Whitehead’s perspective, multiple personalities are come by all too easily. For, as he continued, “what needs to be explained is not dissociation of personality but unifying control, by reason of which we not only have unified behavior, which can be observed for others, but also consciousness of a unified experience.”9 Whitehead’s use of multiplicity is impeccable. He used it to illustrate a thesis, not to argue it. No phenomena known to Whitehead—certainly not those he had learned in Boston from Morton Prince or psychical research—constitute evidence for Whitehead’s cosmology. This, in my opinion, is a desirable relationship between the philosophy of mind and multiple personality. Whitehead’s philosophy has a ready-made slot for the multiple personality but can gain no support from it. His cosmology neither predicts nor explains any detail of the phenomena. Conversely, the clinical structure of multiple personality disorder is totally independent of Whitehead’s cosmology.
More recent philosophers have tried to use multiple personality as evidence. One of these is Daniel Dennett, author of Consciousness Explained, one of the most widely read recent books of philosophy of mind. Nicholas Humphrey is, among other things, a practicing psychiatrist. The two men explored the multiple community of clinicians and clients, and their joint work led to a deeply argued essay, “Speaking for Ourselves.” They observe how a termite colony can appear to act as if with a single purpose, even though each termite is doing its own thing. Their point is that what seems like collective agency does not need a master supervisor. “Most systems on earth that appear to have central controllers (and are usefully described as having them) do not.”10 Humphrey and Dennett use this fact as a partial model of what it is to be a person—a being with many subsystems. But how to characterize the sheer personhood? They offer an analogy, none other than the United States. We can speak of America’s characteristics, its brashness, its Vietnam memories, its fantasy of being forever young. But there is no controlling entity that embodies these qualities. “There is no such thing as Mr. American Self, but as a matter of fact there is in every country on earth a Head of State.” The American president is expected to inculcate and represent national values, and to be “the spokesman when it comes to dealing with other nation states.” A nation, our authors conjecture, needs a head to get on reasonably well as a nation.
By curious coincidence, Whitehead had used almost the same analogy. Noting that we need unifying control in order to be persons, he wrote, “It is obvious that we must not demand another mentality presiding over these other actualities (a kind of Uncle Sam, over and above all the U.S. citizens).”11 In the same spirit, Humphrey and Dennett’s president is importantly not Uncle Sam but just another citizen, temporary head citizen.
What we think of as a person is, according to Humphrey and Dennett, many subsystems. It is nevertheless possible to have one subsystem that is crucially important in various ways, including the ways in which it has relationships with other people. According to the presidential analogy it is a chief representative for the public view of the collection of subsystems. The analogy suggests a neat way to think about multiple personality. There are several functioning, or malfunctioning, subsystems that take turns as the representative, as the president, particularly in dealing with distinct aspects of the system of subsystems. For the background philosophy we must turn to Dennett’s best-known book, Consciousness Explained, where personality disorder is described as one “of the terrible experiments that nature conducts.”12
What do such experiments teach? Dennett’s skepticism about the very idea of the self is now quite well known. His theory of consciousness discredits the attitude to the self that he satirizes as “All or Nothing and One to a Customer.”13 He offers multiple personality as a good illustration of the way that his own theory challenges that attitude. On the same page he mentions a tale of forty-year-old twins who are never apart, who continue each other’s sentences and perform acts jointly. One person with two bodies—fractional personality disorder! The power of “FPD” as an illustration in no way depends on whether the report of such twins is true or false. Dennett’s view of the person allows such a description to make sense. Multiple personality is no more a surprise to Dennett than it was to Whitehead. He is astonished not at multiplicity but at the horrendous conditions in which some children grow up, and which, according to some clinicians, lead children to dissociate.
These children have often been kept in such extraordinarily terrifying and confusing circumstances that I am more amazed that they survive psychologically at all than I am that they manage to preserve themselves by desperate redrawing of their boundaries. What they do, when confronted with overwhelming conflict and pain, is this: They “leave.” They create a boundary so that the horror doesn’t happen to them: it either happens to no one, or to some other self, better able to sustain its organization under such an onslaught—at least that’s what they say they did, as best they recall.14
Is this the result of what Dennett calls “a terrible experiment of nature”? He was not the first to think of multiple personality as an experiment of nature; multiple personality has often seemed to furnish a great experiment for the study of the human mind. In 1944 the authors of a classic early survey of multiple personality ended by citing Francis Bacon, and saying that “cases of multiple personality are natural experimentum lucifera.”15 Ernest Hilgard, the great student of hypnotism, wrote in the same vein, “Overt multiple personalities of the kinds [that the dissociation theorists] studied appear to be rather rare experiments of nature.”16
In his Remarks on the Foundations of Mathematics, Wittgenstein observed that if a picture of an experiment is compelling, then the picture is not functioning as an experiment at all.17 He was making a point about the use of pictures in mathematical proofs, but what he said is true in general. It is not as experiments that cases of multiple personality do anything for Dennett’s philosophy of mind. They serve only as illustrations. But what do they illustrate? Patients have been diagnosed with double consciousness or multiple personality for two centuries now. But they began to talk the way they do now—using the symptom language noticed by Dennett—only very recently. Today they all say such things, or at least suspect that they ought to say them. That is how they learn to describe themselves in therapy. They do not recall themselves dissociating so much as recall various bits of horror in the personae of a number of alters who experience it. What patients say about themselves has changed radically in the past two decades.
Dennett speaks of the terrible experiments that nature conducts. What exactly are these experiments? It is not as if nature produces for us adults on desert islands who say the things that Dennett says they say. The events involve a patient who is in therapy, often for several years, and who comes to say the things she says. The experiment is so strongly controlled that if she does not say those things, she may even be released from therapy for being too resistant, for denial. The question is not whether children are abominably treated. The question is not whether they will grow up with grave psychological difficulties if their childhood is vile. The question is: Is the subsequent prototypical multiple behavior one of nature’s experiments? Or is it rather the way in which a certain class of adults in North America will behave when treated by therapists using certain practices, and having certain convictions? Nothing I have said calls into question the lucid and probing investigation of the multiple movement reported by Humphrey and Dennett. Nor does it take issue with any of the fundamental precepts of Dennett’s philosophy of the human mind. They stand on their own, and that is my only point here. Multiple personality may furnish a graphic illustration of Dennett’s philosophy, but nothing in the detailed phenomena of present-day multiple personality teaches us anything about his theory of subsystems. His philosophy is no more supported by the phenomena than is Whitehead’s.
Humphrey and Dennett were careful observers, for a shortish time, of the multiple scene. Stephen Braude is more like a participant-observer. His book, First Person Plural: Multiple Personality and the Philosophy of Mind, was published in 1991, the same year as Dennett’s Consciousness Explained, but its philosophy is exactly the opposite of Dennett’s. Where Dennett heartily dismisses, or explains away, the idea of a single underlying self, Braude firmly believes in the necessity of such an entity. This already performs for us an important service by exactly reversing the inferences to be drawn from multiple personality. I have been urging that the illness has nothing to teach the philosophy of mind. But at least the very existence of the phenomenon must (it seems) be inconsistent with ideas such as a metaphysical soul, a necessarily unified self, or a transcendental ego. Hence (it seems) the existence of multiple personality does bear on traditional philosophical issues of great importance. So I’m wrong: multiplicity does bear directly on philosophical questions about the self? That was exactly what Ribot argued, and to some extent Dennett too. But Braude argues the opposite way. He contends that the very phenomena of multiple personality demand a unity underneath the multiplicity. Starting with almost exactly the same suppositions as Ribot, he concludes that there must be a transcendental ego. Who is right, Ribot or Braude? One possibility is that one of the two men is right. The other is that both are wrong: no conclusions about the self can be derived from the phenomena of multiple personality. I take the latter view. Ribot and Braude cancel each other out, each reminding us how slippery is the argument favored by the other.
Braude thinks there is an underlying self, but he disowns the most obvious model for this idea. You might think that there is a true person waiting to be discovered, a true person who has been there all along, and who must be revealed in therapy. I have mentioned the debate on which was the true state of Azam’s Félida: the first state or the second, the one she finally settled into. Early American writers, Prince and clinicians influenced by him, seem to have had a picture of the true personality. Which alter is the true Miss Beauchamp? Foster her, once she has been discovered, and tell the others to exit (Prince did just that with one alter, who obeyed). Braude argues that there need be no original person who split and is to be reclaimed.
One of Braude’s arguments for an underlying ego—not the true self, but a central core of all selves—starts from the observation that the alters of one individual have a lot of overlapping basic skills. They can walk and cross the street and tie their shoelaces. Even those rare multiples who have had to do a great deal of relearning in each state retained nearly all ordinary skills. They relearned only what could be rather ostentatiously learned, such as penmanship, or piano, or Greek, or male athletics, skills that are themselves manifestations of a desired social status. The pert and lively person could do these things better than her normal inhibited self. Meanwhile, she could still make small talk, find change at the grocer’s, or drive a car. Of course a child alter may be unable to do some of these things, but they are precisely the aspects of the grown-up world that the patient is trying to avoid. Unless she is making a scene, the child alter preserves the skills needed to cross a busy street. There must be some substratum that explains the overlapping skills of alters and enables the alters to interact when they become co-conscious. There is some underlying unified self in which this mental theater is engaged.
Braude is glad to say that there is more than one self per multiple person. Thus far, he and Dennett agree. After that, there’s trouble. He calls Dennett’s type of view “colonialist,” in virtue of the termite colony metaphor. Colonialism, writes Braude, is a view according to which “there is no ultimate psychological unity, only a deep and initial multiplicity of subjects, ‘selves’ (or, for those smitten with recent work in cognitive science and artificial intelligence) ‘modules’ or subsystems within a person.”18 Dennett has himself protested against this description in terms of modules.19 One of Braude’s key points against colonialism is the network of overlapping basic skills partially shared from alter to alter. This observation is correct but does not seem as forceful a counter to Dennett’s position as he thinks. For Dennett does not portray a multiple as several subsystems, each of which mysteriously has the same street-crossing capacity as every other. On the contrary, there may be a subsystem that handles street crossing for the one body, a body variously represented, at different times, by different subsystem “heads of state” that collaborate with the vast majority of street-crossing and business-conducting subsystems. After a palace revolution, the new head of state retains most but not all of the old bureaucracy of government.
So one must examine Braude’s vision of the life of the multiple directly, and not as the result of a successful refutation of Dennett. As I have said, he does not think that the alters are splits off a one true person. He is willing to say that a multiple has a number of different selves, although he sees that we must improve on this terminology. He would do this as follows. Multiples are genuinely different from most people. They have distinct “centers of apperception.” That is philosophical jargon with a long history, going back past Kant to Leibniz. The dictionary defines apperception as conscious perception with full awareness. To have different centers of apperception means, for Braude, having several “me’s.” Each me has a fairly ordinary collection of awarenesses, beliefs, memories, hopes, angers, and so forth; each me ascribes these beliefs itself to its own “I” in the first person: the beliefs are what Braude calls “indexical.” That is another philosophical word from very recent linguistic philosophy. Words like “here” and “now” and “me” and “they” refer only in a context of utterance. They are called indexical. When I say “I went to town,” I mean me; when you say it, you mean you.
Braude makes good use of this idea. He argues, for example, that there are not separate centers of apperception in hypnosis. In contrast, psychic mediums may have several distinct centers of apperception. They go into trances. They speak of beliefs, memories, and feelings in a thoroughly indexical way, associating different selves with the different voices that speak through them: the voice of your grandmother, Zoroaster, and so forth. Braude is therefore inclined to take the “disorder” out of multiple personality disorder: mediums are unusual; in some ways they resemble multiples, but they are not suffering from a disorder in need of treatment. And maybe some multiples could be just fine too. It should be said here that Braude published two previous books on psychical research about which he holds a careful but favorable opinion.20 He was particularly fascinated by psychokinesis—the use of the power of thought to make accordions float, as in the golden days of psychic events around 1900, or nowadays to predict sequences of random numbers produced by sophisticated electronics. He also takes mediums very seriously, not as people gifted at communicating with spirits, but as people with multiple selves, not necessarily disordered.
His theory has the merit that it does not make dissociation into an artificial continuum. There I have agreed with him, but we do not need his semantic terminology to do the work. His use of words like “indexicality” sounds like deep logic. He argues that the way multiples use pronouns such as “we” reflects an underlying epistemological stance. Nothing as profound as that, alas, is in question. What helps keep all those “me’s” going for a patient in a clinic or a medium in a séance may be nothing as logically fancy as the indexical use of “I” to refer to distinct centers of apperception. It is the old-fashioned practice of naming. We should replace Braude’s importation of technical semantics with some down-home reflections on the use and abuse of proper names.
Braude’s argument has grander ambitions than a mere refutation of colonialism. He holds that Kant was fundamentally correct, that there is a “transcendental unity of apperception” underlying all those distinct centers. Kant and Braude agree on the conclusion—a transcendental ego—but differ on the argument for it. Kant’s argument is notoriously difficult, but I think I know how it goes. I cannot figure out how Braude’s goes, so I must leave his version for the reader. Ribot and Braude start with essentially the same phenomena. From these data Braude wants to lead us to his conclusion, that there is a fundamental prior and perhaps transcendental ego. Ribot wanted us to reach the opposite conclusion, that there is no such thing. Neither one nor the other is correct. Multiple personality adds color to the arguments but furnishes no evidence for them.
Wilkes’s Real People takes a tack very different from those of Dennett and Braude. Those two are trying to say what the self, the person, consciousness, or whatever really is. She writes in the tradition called ordinary language philosophy, aiming at conceptual analysis. She wants to understand not objects but our concepts of objects. She wants to know how we think about things. A familiar concept is articulated by the usage of words. Concepts can be limned not only by what we actually do say, but by what we would say in various circumstances. Sufficiently strange events may leave us speechless. Concepts that snugly fit the sorts of things that usually happen may fall apart when asked to sort out really weird events. When we notice that happening, we have found out something about the bounds of application for our ideas. Wilkes is, however, affronted by the practice of so many writers in her tradition. They invent stories. Personal identity is a favorite domain of philosophical fiction: “What would we say if….” The lacuna is filled by various “bizarre, entertaining, confusing, and inconclusive thought experiments” that are supposed to push our concept of personal identity to the limit.
To my mind, these alluring fictions have led discussion off on the wrong tracks; moreover, since they rely heavily on imagination and intuition they lead to no solid or agreed conclusions, since intuitions vary and imaginations fail. What is more, I do not think that we need them, since there are so many actual puzzle-cases which defy the imagination, but which we nonetheless have to accept as facts.21
Wilkes makes excellent use of a few famous reports of multiplicity, especially Morton Prince’s most celebrated exemplar, Miss Beauchamp. She writes of puzzle-cases that we “have to accept as facts.” We should be cautious about the facts, and about the belief that fact is not only stranger than, but also wholly distinct from, fiction. My worry occurs at almost every possible level. To begin with, there is lying. H. H. Goddard simply lied about his patient “Norma.” This is a salutary reminder that “the facts” may not be exactly as they stand in the case record.
Morton Prince’s prodigiously long book about Miss Beauchamp does not exhaust the facts. He never told us that his patient married a colleague of his who became a society psychiatrist in Palm Springs. We know a lot more about the shadowy Mr. Jones, probably Beauchamp’s first husband, than Prince ever told us. We know many more things about a famous scene of gothic horror that precipitated Beauchamp’s crisis. She was an assistant in a madhouse during a thunderstorm, when Jones appeared on a ladder at the window. That and more happened the day before a trial for “the crime of the century” opened in the next village, none other than that immortalized in verse: “Lizzie Borden took an ax / And gave her mother forty whacks. / When she saw what she had done / She gave her father forty-one.” So truth is stranger than fiction? Or is it fiction, not imagination superseded, as Wilkes implies, but imagination heightened?22 Some five hundred playwrights submitted scripts for the play of Prince’s book.23 The winning author’s play, The Case of Becky, directed by David Belasco, ran for six months on Broadway and was also made into a silent movie. Are we dealing with Docudrama, or Drama Doc?
Wilkes also refers to The Three Faces of Eve. Eve presented three different faces in three different autobiographies. Wilkes is well informed about these stories and as skeptical as I am. I go on at length because of Wilkes’s method. She proposes to try our various conceptions of the self, person, and so forth against what the title of her book calls “real people.” In the multiple personality chapter, she works chiefly with the four personalities of Miss Beauchamp. How many persons are there? she asks. She explores various things that we might say. She urges that our notion of the person seems to fall apart. Early in the book she had offered “six conditions of personhood.” She asks how well each of Beauchamp’s personalities fits each condition. On balance, at least three of the alters fare rather well.
The brunt of the argument suggests that we ought to conclude that during [a certain period of therapy] Prince had three people to deal with. Arguments in favour of affirming plurality are more numerous than those suggesting singularity. What we ought to say, and what we do say, however, may not always jibe.24
Note that this is the last conclusion that an up-to-date dissociative identity disorder clinician would encourage. Prince, they would say, was dealing with one person, three of whose alters were more integrated than the other three. But Wilkes’s analysis is scrupulous.25 I question only the presupposition that we are presented with one true story and asked to try out our language and our philosophical analysis upon that actual case, a real case, not a fictional one. In the few years since Wilkes published Real People the number of multobiographies and auto-multo-biographies may well have doubled again.26
I do not make the weak point that it would have been equally good if Prince’s case were fictional. I make the strong point that the whole language of many selves had been hammered out by generations of romantic poets and novelists, great and small, and also in innumerable broad-sheets and feuilletons too ephemeral for general knowledge today. Prince knew exactly how to describe his patient so that she would be a multiple. Is it any wonder that scanning his interminable report we conclude that there are several persons in one body? This is not a test of how we use language in order to describe real people. It is a consequence of how the literary imagination has formed the language in which we speak of people—be they real, imagined, or, the most common case, of mixed origin. When it comes to the language that will be used to describe ourselves, each of us is a half-breed of imagination and reality. Karl Miller summed up the matter well in his wonderfully rich book about the European novels of doubling:
Every life is made up, put on, imagined—including, hypocrite lecteur, yours. Sybil’s life was made up by Sybil, by her doctor, when she became a case, and again, when she became a book, by her author. Sixteen selves were imagined, but it is not even entirely certain that there were as many as two.27