To people who take their cues from the intellectual fashions of academe, any speculation about the decline and fall of psychoanalsyis must seem premature or downright perverse. Freud’s name, along with those of Nietzsche and Heidegger, has never evoked more automatic reverence than it does today, and the Continental thinkers, from Jürgen Habermas and Paul Ricoeur to Jacques Derrida and Jacques Lacan and the late Roland Barthes, who for the moment strike literary commentators as most advanced, are all Freudians in their various ways. Partisans of psychoanalysis can take comfort, furthermore, from an unabated outpouring of “applied analysis” in the form of psycholiterary, psychobiographical, and psychohistorical studies which, if not always a credit to the tradition, attest to the continuing seductiveness of Freud’s ideas.
But given the diminished standing of psychoanalysis as a psychiatric modality and a theory of mind, it is questionable how much longer the Freudian vogue can last. Bodies of purportedly scientific thought, however dazzling they may be to people who take ideological inspiration from them, ultimately depend for survival on empirical support for their claims. So, indeed, do all disciplines that intend to describe and account for experience in nontranscendent terms, whether or not they aspire to be recognized as sciences. After nearly a century, psychoanalysis has received only trifling and debatable corroboration—and much devastating criticism. If that criticism has yet to make an impact on literary intellectuals, we can anticipate that even they will eventually get the point.
In America at least, there are many signs that psychoanalysis has been falling out of favor. The number of applicants for therapy and training has declined, and perspicacious Freudians have been complaining that the brightest, most scientifically creative young doctors are no longer drawn toward their specialty. One of the central concepts of Freudianism, neurosis, has been pronounced devoid of meaning by the canonical Diagnostic and Statistical Manual of the American Psychiatric Association.1 The authority of Freud, uniquely important to the morale of his followers, has been eroded by a flood of books detailing his logical and empirical mistakes, his frailties of temperament, and his deafness to reasonable objections. The proliferation of psychoanalytic schools and splinter groups has undermined confidence that the movement as a whole rests on commonly secured scientific grounds. Feminists have convincingly challenged the Freudian picture of their sex as condemned by anatomy to masochism and penis envy. And decisive advances in our knowledge of the waking and sleeping brain, providing revolutionary insights into dream activity, psychotic states, the localization of functions, and the chemistry of pain and pleasure, have not only supplied more reliable means of characterizing the mind, but have established an embarrassing contrast with the vacuous Freudian research tradition, which has failed to yield a single authenticated discovery.
Although these developments, taken collectively, would appear to be ominous, most psychoanalysts continue to act as if no sum of setbacks and concessions could fundamentally damage their institution. In displaying such confidence they follow Freud himself, who on diverse occasions proposed that the whole of his metapsychology could be safely discarded, that neurophysiology would one day obviate his psychological postulates, that factors beyond the purview of psychoanalysis attach people to their neuroses, and that an analyst can have little expectation of undoing a complex. Those statements contradicted promises Freud made between and after his seizures of modesty. To the faithful, however, they appear not as confessions of frustration or as reflections on the vigorous claims that they negate, but as signs of an enviable open-mindedness. By the same token, some analysts feel that with each new debunking of a Freudian hypothesis, their discipline makes a smaller target for future snipers.
Psychoanalysis can indeed make do without any given item of doctrine. The reason, however, lies not in the strength of its other, demonstrated, claims, but in the ambiguous state of its theory and in the public’s willingness to grant that an entrenched psychiatric institution must possess a core of valid findings. That willingness is in the process of crumbling. As it does so, we can only become more acutely aware that Freud’s putative science of mind lacks a firm empirical basis, is riddled with gratuitous assumptions, and relies for its prestige on therapeutic expectations that seem to be met at least as well by rival and less tortuous forms of treatment.
I myself, among other one-time Freudians, supposed for years that the well-foundedness of psychoanalysis as a theory could be judged apart from any disallowed claims for psychoanalysis as a therapy. I professed agnosticism’ toward the therapy while maintaining that the theory had proved its independent merits as a psychology. Freudian theory, however, has always been tied epistemologically to the “clinical findings” of individual psychotherapy, and its many counterintuitive postulates took hold as a means of accounting for the triumphant therapeutic results claimed by psychoanalysts. If those results were to prove exaggerated, and if it were shown that the Freudian clinical situation is epistemologically compromised by the therapist’s presuppositions, then the whole necessity for positing the deep structures and mechanisms of the Freudian unconscious would dissolve.
The therapeutic claims of psychoanalysis, it must be understood, are differential. That is, Freudians maintain that their regimen, by far the most expensive and time-consuming of some two hundred competing therapies, demonstrates its value by effecting permanent improvements, as opposed to the mere symptomatic relief, or even outright “symptom substitution,” that results from quicker treatments. The idea is that psychoanalysis alone roots out the source of the patient’s suffering by giving him conscious access to his long-repressed traumatic experiences in the first years of life.2 Although not every psychoanalyst feels confident enough to make that boast, every orthodox practitioner depends on it for his livelihood. For if one therapy worked about as well as another, only people with severely impaired reasoning, or with motives other than a wish to be speedily cured, would choose the one that is most disruptive of their budgets and of their work schedules.
Because of intrinsic difficulties in comparing long-term outcomes of diverse therapies, one cannot say for certain whether any modality is superior to another. It is significant, however, that existing studies, even when they have made no allowance for the vastly longer duration of Freudian treatment, have failed to note any meaningful advantage of that treatment over its myriad competitors.3 Researchers have established a likelihood that psychoanalysis and all other psychotherapies are statistically better than no intervention at all, but this scarcely constitutes an endorsement of Freudianism. On the contrary, if all psychotherapies were to be judged about equally effective, psychoanalysis would rank as the least efficient of therapies, bar none. As for “symptom substitution,” it appears to be at least as common in former patients of psychoanalysis as in others. Indeed, it pervades Freud’s own case histories, which are to a remarkable extent a record of confessed or implied failure.4
More importantly, the tentatively supported equivalence of therapeutic outcomes places in jeopardy the claim of psychoanalysis to succeed because its psychological theory is singularly correct. An overwhelmingly plausible alternative view is that all therapies succeed, insofar as they do, for reasons other than the unique causal factors specified in their accompanying theories. It is easy to imagine such reasons—for instance, that patients who seek therapy have already decided to take themselves in hand, or that any explanation offered in tandem with a promise of symptom relief can be happily embraced, or simply that a hired friend is better than none.5 In short, theories that enter into therapies probably do so in the function of welcome placebos. If so, the theories not only remain unsubstantiated by therapeutic success; they are positively erroneous in isolating curative factors that are wide of the mark.
The fact that psychoanalysis frequently “works” is therefore less corroborative than some contented former patients think it is. Faith healing “works,” too, as Freud ruefully acknowledged in deferring to the superior results achieved at Lourdes. Just as a successful laying on of hands demonstrates nothing about Christ’s mercy, so the propositional content of psychoanalysis remains undemonstrated by a successful case history—or by any number of them. If psychoanalysis is to justify its distinctly exotic theory, it must show that the unique features of that theory are well authenticated by facts that do not lend themselves to any simpler explanation.
Unfortunately, however, the situation that produces nearly all psychoanalytic evidence—the clinical interview—is epistemologically contaminated to an extreme degree. It would be hard to find a data-gathering arrangement less conducive to the empirical ideal of neutralizing the investigator’s bias. Psychoanalysts, like other therapeutic practitioners, perceive their patients through the categories of the theory that outsiders regard as questionable—and the theory in this instance is notorious for its encouragement of premature conclusions. A Freudian can as readily find “evidence” of libidinal cathexes and repressed imagos as a Jungian can locate the anima, the shadow, and the persona. Neither group, we can be sure, will ever stumble across “evidence” for the existence of the other’s postulated entities.
Freudian analysts would have us believe that their patients get better because they have raised repressed memories to consciousness and achieved a maturing insight into their formerly buried conflicts. But as Freud himself sometimes feared, there are indications that some of the materials eventually “recovered” from a patient’s “repressed unconscious” may be artifacts of insistent suggestion on the analyst’s part. Doubts have been raised as to whether the memory impressions of early infancy survive at all6—to say nothing of whether they possess pathogenicity for outbreaks of neurosis several decades later. Moreover, in the words of two pro-Freudian but understandably troubled reviewers of research, “Investigators have found that individuals will enthusiastically accept bogus interpretations as accurate descriptions of their own personalities.”7 Other researchers have found experimental evidence suggesting that introspective confirmation of the causes of one’s own thoughts and feelings is beyond the capacity of exceptional as well as ordinary people.8 Without such confirmation from the patient, an analyst’s inferences that certain early traumatic experiences were pathogenic must remain conjectural. Putting these considerations together with the failure of psychoanalysis to indicate any substantial advantage over other treatments, we can say that no grounds whatever exist for believing that recovery of the repressed is specifically therapeutic—or indeed that “the repressed” is an applicable term.
In linking therapeutic success to indifferent or spurious factors, psychoanalysis is no more disingenuous than any number of therapies that invite their clients to merge with the collective unconscious, regress to infancy, relive their births, or identify their previous incarnations. It does stand out, however, as the therapy that places fundamental stress on analysis of the resistance. The resistance in question issues in a reluctance to accept the therapist’s most assured guesses about the patient’s psychodynamics and their relation to “remembered” or reconstructed infantile experiences. Thus, the patient’s balking at interpretive hints that may be misguided or even silly is taken as a sign of resurgent conflict with parents and siblings. The dissolving of this allegedly atavistic uncooperativeness—a recalcitrance which may in fact attest to the patient’s unsurrendered common sense—is considered to be the analyst’s infinitely painstaking task. Uniquely among recipients of psychotherapy, then, the Freudian client has his treatment prolonged and pays handsomely for the privilege of having some of his objections set aside, even though they may have been entirely warranted.
If psychoanalysis, as its most fervent supporters have said, really is “the” cure for personality disorders, we must wonder why it has confined its benefits to the relatively healthy as well as the relatively affluent. The best indicators for success as a psychoanalytic patient, according to the Freudian investigators already cited, are “youth, education, intelligence, motivation, time, money, and a relative lack of profound personality disturbance.”9 Yet even among this elite, psychoanalysis has not been able to show differentially impressive results. And it has on its books an exceptional share of the walking wounded. Everyone around the Freudian community knows patients who have become addicted to their analyses, clinging to a fruitless, financially draining dependency for ten, fifteen, or even twenty years without being either “cured,” discharged as incurable, or referred to another form of treatment. Although Freud himself, in his late, pessimistic paper “Analysis Terminable and Interminable,” recommended the setting of an arbitrary concluding date for stubborn cases, a significant number of analysts have dealt with their own protracted failure by calling the patient’s efforts incomplete and by represcribing the very medicine that had failed to take effect.
When we turn from the state of the therapy to that of the theory, we find that the taint of subjectivism is pervasive. The problem begins with Freud, who took little care for self-consistency and whose writings, in the assessment of one distinguished psychoanalyst, were “formulated in a bewilderingly unsystematic way.”10 Because he believed, furthermore, that the clinical dialogue provided all the testing of hypotheses that scientific rigor could require, Freud continually overrated the empirical basis of his ideas—even to the point of maintaining that he had confounded genetic theory by discovering in his patients inherited memory traces from the dawn of civilization. Although few of his followers have accepted that judgment, nearly all of them have shared in the naive methodological assumption that lay behind it—namely, the belief that hypotheses about the deepest structure of the human mind can be confirmed by what patients reveal to their analysts through “free association” and reports of dreams.
In logic, there is no need for contemporary psychoanalysts to repeat Freud’s errors of fact or to emulate his overconfidence in drawing conclusions. In practice, however, analysts have not proved capable of subjecting Freud’s ideas to the unsparing criticism that would typify a genuinely empirical discipline. Though most analysts differ from Freud on various particulars, they show no eagerness to confront the general, long-familiar epistemological objections to his work. The reason, we may suppose, is that their own work rests not only on Freud’s scientific authority as a ground-breaker but also on the questionable rules of inference-drawing that he devised. The unwritten imperative of psychoanalytic reform is to salvage as much of Freud as possible, in appearance if not in substance. As we will see in a characteristic example, metapsychology has consequently choked itself with provisos and addenda that are meant to reverse Freud’s deterministic emphasis without seeming to do so. Much of psychoanalytic thought is by now a palimpsest of hazy, mutually jostling notions, not one of which has been shown by an adequately designed empirical study to be the most likely explanation of a given phenomenon.
To be sure, some researchers who began with Freudian sympathies and who neglected to test for plausible alternatives to, say, repression and castration anxiety believe they have proved the explanatory value of such concepts.11 Unreconstructed critics, however, have had little trouble exposing the fallaciousness of such proofs.12 And still others have reviewed the whole standing of psychoanalysis as knowledge and found it to be nugatory. In an important series of recent articles, the eminent philosopher of science Adolf Grünbaum has inquired whether psychoanalysis meets currently accepted canons of inductivism.13 His judgments will be gathered in a book to be entitled Is Psychoanalysis a Pseudo-Science? While Grünbaum does not rule out the logical possibility that some Freudian hypotheses may one day be supported, he finds that for now psychoanalysis rests on no solid evidence and that its purported clinical confirmations are open to the most fundamental doubts. Nor can we overlook the stunning pronouncement made in 1975 by the Nobel prizewinner in medicine P. B. Medawar: “[D]octrinaire psychoanalytic theory is the most stupendous intellectual confidence trick of the 20th century: and a terminal product as well—something akin to a dinosaur or a zeppelin in the history of ideas, a vast structure of radically unsound design and with no posterity.”14
Against this bleak prospect, defenders of psychoanalysis assert that forbearance is in order. Their movement, they say, is still young; it is as yet a “protoscience,” groping with forgivable clumsiness toward an adequate framing of its discoveries. We are asked to look forward to a time when conscientious Freudians will have made peace with the scientific mainstream by simplifying the theory and submitting the remaining core of propositions to the most demanding empirical tests. Then, presumably, we will have avoided discarding valuable, hard-won insights along with early misconceptions. Every champion of the movement can point to a favorite branch of revisionism that has supposedly made a good beginning toward this goal.
After nine decades of sectarianism and self-isolation from the broader scientific community, however, psychoanalysis would hardly seem to be “proto” anything. The record to date suggests that when analytic theorists undertake to replace one of Freud’s implausible ideas, they do so without benefit of methodological safeguards against further error. Unlike demonstrably empirical discoveries, furthermore, psychoanalytic reforms rarely bring the warring Freudian factions together; nearly every change has meant a further sundering, a new occasion for dogmatic belief within a subset of the faithful. This fission has been caused, not by the exuberant youthfulness that Freudians ascribe to their discipline, but by the fact that it is not, strictly speaking, a discipline at all. That is to say, psychoanalysis as usually elaborated eschews the means of discriminating warrantedly between rival hypotheses about mental functioning and paths to symptom removal. As for the “hard-won insights” that we should hesitate to dismiss, most of those that may actually possess merit were current before Freud’s time and will presumably remain accessible in the future.15
Two factors condemn innovative Freudian metapsychologies to the subjectivism that yields one unresolvable quarrel after another. The first is that psychoanalytic reformers have only their own taste to guide them in deciding which metapsychological concepts to perpetuate and which to challenge. Notions like “id” and “oedipus complex” and “pleasure principle” take their meaning from a network of postulates that generate no straightforward behavioral consequences. Thus the presence or absence of such consequences in a given instance cannot serve as a test of theoretical adequacy. On the contrary, the elasticity of Freudian interpretive rules, whereby any phenomenon can bear either its manifest meaning or an opposite meaning or some third, “displaced,” meaning, enables every interpreter to convince himself that his concepts are backed by evidence. This absence of constraints encourages the riding of hobbyhorses instead of a principled adjudication of differences.
And second, most psychoanalytic theoreticians have relied on “findings” that will be forever unavailable to outside parties. The Freudian community retains its self-respect by assuming that the author of a paper, because he has been analyzed and officially trained, has acquired an objectivity and scrupulousness rarely found among the laity. But with the best will in the world, a Freudian innovator meets no methodological barrier against the temptation to misinterpret, embroider, or censor his essentially secret case histories. Scientific responsibility is thus lodged precariously not in the watchdog process whereby investigators check the replicability of one another’s announced results, but in individuals telling self-serving anecdotes about anonymous patients. In a community operating by such rules, metapsychological innovation comes cheaply—and is prized no less cheaply by guardians of established views.
Let us suppose, however, that reformers could persuade their colleagues to do without high-level metapsychology and restrict themselves to those concepts that provide a quasi-phenomenological account of therapist-client relations—concepts such as resistance, transference, repression, and regression. Could psychoanalysis get by on that minimal basis? Some revisionists think so, but they overlook the fact that even these terms are shot through with implications about infantile sexuality, the alleged stages of libidinal development, the early-traumatic etiology of neurosis, and the analyst’s privileged insight into that etiology. Psychoanalysis as a drawn-out regimen presupposes those ideas; without them, there would be no conceivable justification for the laborious and costly task of anamnesis.
If psychoanalysts wish to continue practicing under the old rules, therefore, they cannot afford to go very far toward admitting how little they really know about the mind. To most present-day analysts, Freud’s system is at once reassuring in its apparent conferral of authority and embarrassing in many of its details. The analysts’ best hope of preserving their elevated standing in a volatile society is to patch over the more obvious defects in classic psychoanalytic thought while continuing to maintain that they understand what causes and cures mental instability, what the basic instincts are, how sexual identity is formed, and so forth. We will see, however, that such ad hoc repair not only fails to go far enough; in some cases it makes matters worse by taxing the theory with overcomplication and by drawing attention to the flaws it is intended to cover.
To examine the liabilities of psychoanalytic reform at close range, let us consider one of the most respected of contemporary theorists, Heinz Kohut. Though Kohut is less well known to laymen than a Fromm or an Erik-son, among the cognoscenti he comes nearer to fitting the image of a properly Freudian innovator. His two major works, The Analysis of the Self (1971) and The Restoration of the Self (1977), have a large and grateful following, and recently he has acknowledged his eminence by issuing a two-volume collection of papers, graced by a 106-page introduction by Dr. Paul H. Ornstein extolling what is now said to be “the leading paradigm of psychoanalysis.”16 Needless to say, there is room for dispute of that honor; psychoanalytic theory is so easy to play variations on that anyone with a handful of followers could think he had revolutionized the field. But if Freud’s couch were passed along like the throne of St. Peter, it might well find itself today in Dr. Kohut’s office in Chicago.
It is not just Kohut’s stature but also his deference to empirical values that makes him an appropriate figure for assessing the future of psychoanalysis. In his papers he repeatedly admonishes his colleagues against hasty conclusions and unnecessarily rigid views. He is careful to reject biological explanations of psychologically described events, and he tries wherever possible to rely on tangible phenomena rather than on abstract reconstructions. One of his main concerns within the American Psychoanalytic Association, furthermore, has been the promotion of research. And unlike some analysts, he does not hesitate to point out areas in which psychoanalysis has shown itself deficient. By manner, at least, no one would inspire more confidence in the capacity of Freudians to set their house in order.
Sadly, however, Kohut’s empirical scruples turn out to be mostly cosmetic—a matter of scientistic rhetoric rather than true rigor. Psychoanalysis in its current state is already, for him, “this great new edifice of human thought” (695), “this new sun among the sciences of man” (684), “the science that reaches farthest into the breadth and depth of the human soul” (683-84), and the rightful candidate to become “man’s scientific leader” (682). Nothing in his writings suggests awareness of the key defect in his “science,” namely, the fact that we have typically been asked to take an analyst’s word both for the raw data he reports and for the correctness of his interpretations.
On the contrary, Kohut refers with assurance to “the laboratory of psychoanalytic treatment” (523), as if analysts in their fifty-minute hours were properly testing the debatable hypotheses of Freudianism instead of applying familiar and notably compliant rules of interpretation. And when he calls for more research, he has in view not controlled studies that might determine which concepts deserve to survive, but mere consolidation and publication of what analysts have already learned “from the clinical situation” (602). To be sure, Kohut worries about bias in research—but he does so along the quaint diagnostic lines established by Freud. One researcher, he tells us, may have reacted so strongly against his early sadistic impulses that he is too compassionate to be objective (607), and another may suffer from “a libidinal hypercathexis in the visual-cognitive area” (608) and so lack balanced judgment. In Kohut’s perspective, the problem of skewed results apparently vanishes as soon as the investigator has achieved “mastery and integration of infantile precursors of his research interest” (608).
I have mentioned as one of the handicaps of psychoanalysis its intellectually compromising subservience to Freud. Kohut, true to his role as an enlightened adviser to the movement, discusses this very problem at some length. Analysts in training, he says, encounter Freud as “the great father-figure and teacher of our science” (796), and thus they acquire “an attitude of firmly established identification with an idealized figure (or, in reaction formation, of rebelliousness against this identification)” (796). As Kohut sees it, this confusion of emotional dependency with the scientific posture is a warping force, inducing the individual psychoanalyst either to rubber-stamp Freud’s ideas or to undermine them by hitting on some gratuitous novelty to emphasize.
One might expect Kohut, therefore, to be especially wary of idolizing Freud. Such is not the case. Psychoanalysts, he announces, are lucky “because in this finite life we are able to participate in the work of one of the few geniuses of mankind, a participation which in our field appears to me to be more intimate and profound than in most other sciences” (393). As we will see, Kohut senses that his own ideas are in crucial respects opposed to Freud’s; like other revisionists, he faces the psychological and rhetorical problem of diverging from the “genius” while continuing to share in his glory. The solution is to take on faith great chunks of doctrine that are not immediately at issue and to appease Freud’s offended ghost with hyperbolical praise.
Thus, for example, Kohut asks why Freud thought that female sexuality is an outcome of thwarted phallic strivings rather than a natural biological development. Could it be that the master had a “circumscribed blindspot” (228) on this issue? Kohut raises the possibility only to wave it away without discussion. It must have been an admirable “reliance on clinical evidence” (228) that kept Freud from exercising common sense: “Penetrating beyond the feminine attitudes and feelings of his patients, he regularly found the struggle over phallic strivings, and, white he accepted biological bisexuality, he rejected the postulate of a preceding psychological phase of femininity without psychological evidence for it” (228). Kohut does not allow that Freud’s “penetrations” may have been tendentious, or that “evidence” supplied by indoctrinated adult patients may not necessarily be evidence of an infantile stage of development, or that Freud’s acceptance of “biological bisexuality” was just as arbitrary as his decision to make penis envy the raison d’être of femininity. Ordinary empirical prudence must be stifled in deference to “the great father-figure and teacher of our science.”17
More is involved in such kindness than filial homage, however. If Kohut were to grant the full extent of Freud’s interpretive whimsy, he would be disavowing his own right to base conclusions about early childhood on remarks made by supine grownups. Kohut must remain convinced that psychoanalytic theory is objectively dictated by facts that impress themselves upon the clinician who is schooled to listen empathically and ponder introspectively. Indeed, Kohut is so enamored of this idea that he thinks empathy “should become the guiding ideal of all the sciences . . . the scientist’s commitment to it should take the place of the pride in his methodological and technological expertness which he has felt up to now” (703). Without flinching, he tells us that “Freud’s attitude concerning the development of female sexuality is only one of many examples of his faithful adherence to the introspective and empathic method of observation” (228-29). Yet neither Freud nor Kohut would be able to explain how introspection and empathy generate adequate controls for error and idées fixes, much less how they enable the Freudian thinker to make inferences about psychic energy, stages of infancy, various kinds of libido, oedipal and pre-oedipal fixations, and the rest of the leaden conceptual baggage that psychoanalysis carries wherever it goes. Nor could they explain why analysts, if their tools of investigation are indeed superior to those of experimental natural science, remain divided within squabbling sects, each with its boast of having corrected the others’ faulty understanding of the mind.
Kohut himself is the founder of one such sect, purportedly the guardian of fundamental advances in therapy as well as theory. His contribution has been to rescue narcissism from its lowly place in Freudian thought and to endow it with its own special drive, its hitherto neglected vicissitudes (the “narcissistic disturbances”), its ideal line of normal development, and a happy ending, namely, the production of a wholeness that Kohut designates as “the self.” Kohut, Ornstein, and others now regard “the psychology of the self” as a second and superior half of the great theoretical edifice begun by Freud. Until he himself came along, Kohut reflects, psychoanalysis saw only “Guilty Man,” in conflict with his libidinal drives. Now we can also deal precisely with “Tragic Man,” whose goal is not gratification but self-realization. At last psychoanalysis can “attempt to make contributions—scientific contributions—to the understanding of some of the most important activities of man, such as his religions or his art, that do not dissolve these activities into their elemental constituents and, by doing so, become blind to their essential significance . . .” (923-24).18
Like other Freudian explorers, Kohut tells us little about the procedures he followed in arriving at his discoveries. From what he does say, however, we can discern the outlines of a characteristic pattern. In his case as in some others, the analyst’s starting point is apparently a perception of indifferent therapeutic success—a perception he keeps to himself until after the breakthrough, whereupon he becomes more or less candid about past failures, though not about present ones. Thus Kohut reveals that for some fifteen years before 1974, he had felt “increasingly stumped” (888) by as many as half of his cases. (One looks in vain for any acknowledgment of that fact in papers written during the period.) Goaded to try new ideas, the analyst decides to attach extra importance to some element of behavior or theory—in this case narcissism—that earlier investigators have overlooked or considered incidental. Then, still believing that whatever he learns from analytic sessions must be direct evidence of infantile development, he links the newly central factor to the “archaic experiences” that his patients have supposedly recovered from their repressed depths.
To the analyst’s delight, everything falls into place; the patient who is destined to become the latest Dora eventually accedes to his interpretive hints and feels, presumably, that the stalled analysis must be getting somewhere at last. The treatment then proceeds according to the customary Sisyphean technique, but with an altered vocabulary and a different set of assumptions about the nature of the patient’s woes, the root causes of trouble, and the signs of progress and backsliding. Certain forms of childishness on the patient’s part, for example, are now considered productive, for they indicate revival of the conflicts that have just been granted theoretical preeminence.
Needless to say, the new point of view is no less spuriously confirmed by “clinical results” than the one it has supplanted. In evaluating his apparent success, the analyst is unlikely to wait until he can learn whether the “cure,” if any, was only temporary. Nor will he inquire whether his own enthusiasm may have served as a placebo, or whether an extant rival theory might have yielded the same results, or whether Dora may have simply gotten older and more resigned to other people’s obsessions, or whether he should take into account those patients who have resisted his new line of coaxing and have thus failed to ratify his destiny as a medical pioneer. And in writing up his scientific advance for the edification of his colleagues, he of course provides no data to support the claim of sharply increased therapeutic success. To do so would be to break with Freud’s precedent and to imply that there may be something lax about the trusting atmosphere that enables each reporting analyst, whether or not he is a competent therapist, to have a turn at innovation.
Kohut is typical in another important respect as well: he does everything he can to minimize the extent of his departure from orthodoxy. As a holistic concept, Kohut’s idea of selfhood is contradictory in spirit to the mechanistic, trifurcated model of mind developed by Freud. Potentially, then, Kohut is as much a schismatic as Jung, who stood Freudian theory on its head. He tells us, however, that he built his new system keeping in mind that “the new psychology of the self must remain in an unbroken continuum with traditional psychoanalytic theory to preserve the sense of the historical continuity of the group self in the psychoanalytic community” (937). For a sentimental or political reason, in other words, Kohut was determined to insert his ideas into the preexisting framework of metapsychology, whether or not they belonged there.
That is precisely what he did, with results that defy succinct description. Dr. Ornstein’s labyrinthine introduction traces the stages of Kohut’s delicate progress toward conceiving of two selves—in effect, one for Freud and one for Kohut. The “narrower” self, as Ornstein notes, is “embedded (albeit somewhat loosely) in a mental-apparatus psychology and ego psychology,” while the broader or Kohutian one “is conceptually independent from and has moved beyond drive theory and ego psychology” (98). As a concordat with the Freudian establishment, this arrangement has much to recommend it. As a supposed psychological finding, it requires the faith of a Bernadette.
Kohut’s practice, to speak bluntly, is always to throw a bone to Freud while saving the choicest, most humanistic-looking morsels of theory for himself. Thus he fully accepts Freud’s account of the psychoneuroses, in which a repressed drive is seeking satisfaction, but charts as his own territory the more congenial narcissistic disorders, in which “an injured, narcissistic ego is seeking reassurance” (22). The mind, it seems, is a creaky contraption when working for Freud but a soulful being when working for Kohut. Again, instead of challenging Freud’s murky idea of libido, Kohut tactfully avers that there are two kinds of libido, one leading to the traditional Freudian terminus of “object love,” the other leading to self-love and onward to selfhood, a realm where the penurious economy of Freudian energy expenditure is nowhere to be seen. And if just two selves seem insufficient to placate the orthodox, Kohut is willing to supply more—provided everyone agrees that he holds the patent on the best one. “We recognize,” he says,
the simultaneous existence of contradictory selves: of different selves of various degrees of stability and of various degrees of importance. There are conscious, preconscious, and unconscious selves; there are selves in the ego, the id, and the superego; and we may discover in some of our patients contradictory selves, side by side, in the same psychic agency. Among these selves, however, there is one which is most centrally located in the psyche; one which is experienced as the basic one, and which is most resistant to change. I like to call this self the “nuclear self.” (96)
As this passage suggests, there are no limits to the complication that psychoanalytic reformers will add to already dubious postulates in order to avoid an open break with tradition. So long as analysts want simultaneously to free themselves from Freud’s biologism and to bask in its remaining prestige, they will continue ornamenting the stark branches of id psychology with “selves”—and with neutralized energies, and desexualized drives, and integrative functions, and conflict-free spheres of the ego—without noticing that they are creating, not a credible metapsychology, but a Rube Goldberg apparatus that bears no examinable relation to experience.
As everyone knows, Freud explained intellectual resistance to psychoanalysis in terms of injured human pride. His discovery that our minds are enthralled by repressed wishes, he announced without undue humility on his own part, was the third great blow to anthropocentrism following those administered by Copernicus and Darwin. Yet it is curious how readily many of us absorbed the putative insult and defended it as what it distinctly is not, a scientifically confirmed truth. The real question to be resolved is not why people resisted a doctrine that found in every physician a deflected sadist, in every artist a former dabbler in his own feces, in every infant a murderous and incestuous schemer, in every decent act the sublimation of a barbaric impulse. The question is rather why so many people fell cheerfully into line with these and equally lurid ideas, expounded with no more proof than the say-so of a compelling stylist.
Any answers are bound to be partial and conjectural. Yet I think we can recognize certain nonrational appeals within psychoanalysis, even if we cannot assign them relative degrees of importance. It would seem, for example, that for many intellectuals psychoanalysis has been, not a blow to human pride in general, but a means of elevating pride among a corps of privileged knowers who, by subscribing to the Freudian movement, rescue themselves from doubt and insignificance. It is as a cathartic and redemptive science that psychoanalysis has claimed our loyalty. “Normal science” is dry and impersonal, narrow in focus, and increasingly incomprehensible to the envious humanist. Psychoanalysis, by contrast, offers each of its believers a total vision that spans the entire history of our species, links biology and psychology, and unveils the innermost scandalous wishes animating heroes and ordinary folk, great works of art, and whole systems of law, philosophy, mythology, and religion. What is so humbling about that? Could Faust have asked Mephistopheles to show him much more?19
In part, then, we can suppose that psychoanalysis survives because it feeds extravagant intellectual hopes. It is energized not by the skeptical spirit but, as Freud knew in calling himself a conquistador, by a wish to overleap barriers and arrive at a comprehensive, countertraditional system of insight. Perhaps that is why armchair radicals, at a time when psychoanalysis as a therapy and science is in full retreat, can still base whole books on the most extreme elements of Freudian theory, such as polymorphous perversity and the death instinct. The reason they are not dissuaded by empirical criticism may be that their aim is precisely to have done with the restraining effects of empiricism. If I read them correctly, they wish to wield terms of discourse implying a realm of essence beneath and opposed to the visible world—a realm which, if we could only set it free, would work the overthrow of every tyranny of custom, tradition, and entrenched privilege. So, too, the “hermeneutic” school, which conceives of psychoanalysis only as a means of performing abstruse manipulations of texts, appears to have distilled from Freudianism its will to explanatory power while openly proclaiming the irrelevance of verification.
That sense of special power may be felt even by circumspect Freudians who eschew deep theory and try to confine themselves to such relatively accessible concepts as denial, projection, and identification. I was myself such a Freudian for a decade or so—excited yet also made wary by the prospect of holding a master key to interpretation. I remember taking comfort from the analysts’ rhetoric of meticulous devotion to fact, even while I deplored some of the shortcomings I have mentioned here. To be a cautious Freudian was to have the best of both worlds: allegiance to transpersonal standards of knowledge and a head start toward certainty. Possessing that head start, or the illusion of it, a partisan of psychoanalysis will understandably hesitate to entertain more than selective doubts about his doctrine.
Finally, many veterans of therapy have their own reasons for clinging to the faith. They have had what they call confirming experiences—crystallizations of emotion, flashes of self-recognition, and enhanced insight into people or problems that had been troubling them. We need not doubt the genuineness of these catharses; the prolonged strain intrinsic to a Freudian patient’s situation, combined with the analyst’s disapproval of conventional, self-protective responses, probably sharpens the analysand’s perceptions, however temporarily. What needs doubting is whether such experiences, which can be approximated in any number of other therapies or even in outright brainwashing, really ought to count as “confirming.” Memorable though they may be, they prove nothing at all about the correctness of Freudian theory or the long-run usefulness of the therapy. Try telling that, however, to someone who thanks his analysis for his sanity and who now sees the work of id and ego everywhere, much as his counterpart a few centuries ago detected the equally confirming influence of angels and devils.
For several reasons, then, a rush of defections from the Freudian ranks seems unlikely. Yet psychoanalysis, I would expect, will fade away just as mesmerism and phrenology did, and for the same reason: its exploded pretensions will deprive it of recruits. If most people cannot readily forgo an entire world view that has sustained them, they find it easy enough to avoid the strange-looking intellectual fashions of another era. Psychoanalysis, once so familiar that, in Erich Heller’s words, “it comes close to being the systematic consciousness that a certain epoch has of the nature and character of its soul,”20 is in the process of becoming strange. There is no reason to believe, much less to hope, that the process will be reversed.
Whether or not my prediction is accurate, there is an immediate moral to be drawn: a concern for empirical standards should make us reconsider the habit of borrowing bits and pieces of psychoanalytic thought for ad hoc explanatory purposes. Perhaps some Freudian ideas, pried loose from their context of presumptions about mental hydraulics and libidinal development, may eventually prove to have some merit. If so, it will be because they have won a test against concepts that take less for granted and fit better with inductive sense. In the meanwhile, we would do well to struggle along without recourse to instant “depth.” When we do, we will find that we have sacrificed, not complexity of understanding, but something nearly opposite: a pretended intimacy with realities that cannot be captured in the crude and monotonous language of Freudian explanation.
1 Critics have pointed out that the third edition of DSM (1980), like its predecessors, reflects political as well as medical and scientific opinion. Quite true: when smoking replaces homosexuality as a mental aberration, more of the credit must go to caucuses than to new findings. For that very reason, however, we can safely regard the DSM's demotion of “neurosis” as a sign of waning psychoanalytic influence.
2 In Anna Freud's words, “In competition with the pyschotherapies [analysts] are justified to maintain that what they have to offer is unique, i.e., thoroughgoing personality changes as compared with more superficial symptomatic cures.” Difficulties in the Path of Psychoanalysis (International Universities Press, 1969), p. 17.
3 See Lester Luborsky, Barton Singer, and Lise Luborsky, “Comparative Studies of Psychotherapies: Is It True That ‘Everyone Has Won and All Must Have Prizes’?” Archives of General Psychiatry, 32 (1975), pp. 995—1008; Leo Goldberger, Roger Reuben, and George Silberschatz, “Symptom Removal in Psychotherapy: A Review of the Literature,” Psychoanalysis and Contemporary Science, 5 (1976), pp. 513—536; and Allen E. Bergin and Michael Lambert, “The Evaluation of Therapeutic Outcomes,” in Sol L. Garfield and Allen E. Bergin, Handbook of Psychotherapy and Behavior Change: An Empirical Analysis, and ed. (Wiley, 1978), pp. 139—189.
4 Freud, according to two sympathetic observers, “never presented any data, in statistical or case form, that demonstrated that his treatment was of benefit to a significant number of patients he himself saw,” and he “chose to demonstrate the utility of psychoanalysis through descriptions of largely unsuccessful cases.” See Seymour Fisher and Roger P. Greenberg, The Scientific Credibility of Freud's Theories and Therapy (Basic Books, 1977), pp. 285, 281.
5 Why does the contest between psychotherapies appear to end in a massive dead heat? “The most potent explanatory factor is that different forms of psychotherapy have major common elements—a helping relationship with a therapist is present in all of them, along with other related, nonspecific effects such as suggestion and abreaction” (Luborsky et al. p. 1006).
6 See Jerome Kagan, Change and Continuity in Infancy (Wiley, 1971), and “The Baby's Elastic Mind,” Human Nature, January 1978, pp. 66—73.
7 Fisher and Greenberg, p. 364.
8 See Richard E. Nisbett and Timothy Wilson, “Telling More Than We Can Know: Verbal Reports on Mental Processes,” Psychological Review, 84 (1977), pp. 231—259.
9 Fisher and Greenberg, p. 303.
10 Benjamin B. Rubinstein, “On the Clinical Psychoanalytic Theory and Its Role in the Inference and Confirmation of Particular Clinical Hypotheses,” Psychoanalysis and Contemporary Science, 4 (1975), p. 31.
11 Observers who believe that at least a modest handful of Freudian hypotheses have been experimentally supported include Fisher and Greenberg (footnote 4 above); Paul Kline, Fact and Fantasy in Freudian Theory (Methuen, 1972); and Lloyd H. Silverman, “‘the Reports of My Death Are Greatly Exaggerated,’” in Science and Psychotherapy, ed. Raphael Stern et al. (Haven, 1977), pp. 255-282. Although the claims advanced by these Freudians are open to challenge, their acceptance in toto would lend only marginal credence to general psychoanalytic theory.
12 The locus classicus of Freudian investigations that rule out commonsense alternatives is a study purporting to show that, because a number of female college students dreamed of penises, the concept of penis envy has received significant support. The same study confirmed that women tend to dream about weddings and babies, i.e., “displaced penis envy.” For an authoritative discussion of this and other poorly conceived investigations, see Hans J. Eysenck and Glenn D. Wilson. The Experimental Study of Freudian Theories (Methuen, 1973).
13 See especially “Epistemological Liabilities of the Clinical Appraisal of Psychoanalytic Theory,” Psychoanalysis and Contemporary Thought, 2 (1979), pp. 451-526. I am indebted to Professor Grünbaum not only for several ideas in the present essay, but also for searching criticisms of my various drafts.
14 “Victims of Psychiatry,” New York Review of Books, January 23, 1975, p. 17.
15 See, e.g., Lancelot Law Whyte, The Unconscious Before Freud (Basic Books, 1960).
16 The Search for the Self: Selected Writings of Heinz Kohut: 1950-1978, 2 vols., ed. Paul H. Ornstein (International Universities Press, 1978), Introduction, p. 91. Subsequent parenthetical references will cite these continuously paginated volumes; all quoted italics appear in the original text.
17 Kohut does worry that Freud's theory strains credulity by asking us to believe that every little girl in the history of the world has responded with “envy, shame, rage, and denial” (785) to her genital incompleteness. Natural selection, however, comes to the rescue. Perhaps, Kohut suggests, “in the prehistorical past of the human race those females of the species who reacted with greater sensitivity to the experience that they had no penis had a higher survival rate” (785).
18 In practice, however, Kohut treats works of art just as previous analysts have, mining them for points of likeness to his favorite ideas. He asserts, for example, that paranoia arises genetically from the incapacity of a child's “self-objects” (parents) “to mirror the child's total self,” and he adds: “Kafka described this situation poignantly in Metamorphosis: Gregory Samsa experiences himself as non-human while his parents in the next room speak about him in the third person singular” (743n.). Had they spoken about “Gregory” in some other person or number, they might have been beyond even Dr. Kohut's help.
19 The replies of several pro-Freudian acquaintances to a draft of this essay have reinforced my guess about the correlation between psychoanalytic allegiance and animus against “normal science.” My respondents uniformly declared that since their interest in Freudianism was not a scientific one, my animadversions about the dubious scientific status of psychoanalysis were wasted on them. They did not care to recognize that I am here criticizing psychoanalysis, not for the technicality of failing to qualify as a science, but for being so conceptually muddled and empirically dubious that it does not warrant our belief. “Science,” which many Freudians conveniently confuse with materialism or positivism or behaviorism, is their straw man—an antithesis to the humanistic spirit that binds them, however irrationally, to psychoanalysis, for both Freud and humanism strike them as championing the beleaguered imagination.
20 “Observations on Psychoanalysis and Modern Literature,” in Psychiatry and the Humanities, ed. Joseph H. Smith, Vol. 1 (Yale University Press, 1976), p. 35.