THE FOUNDATIONS OF PSYCHOANALYSIS: AN ESSAY ON A PHILOSOPHICAL BOOK BY ADOLF GRUNBAUM
VANN SPRUIELL
Introduction
Psychoanalysis is a singular profession. Whether as a body of knowledge, method of investigation, or therapeutic modality, it resembles, in this are that way, an assortment of other disciplines. Yet, in each case of partial resemblance, psychoanalysis is in other important ways not similar.
Psychoanalysis superficially resembles some non-analytic forms of psychotherapy. But, unlike such psychotherapies, psychoanalysis eliminates, or at least radically reduces, the hallmark's of the operations of psychotherapy. Even the most psychoanalytic of psychotherapies, to some extent or another, practice manipulation through the use of the relationship between the patient or client and the therapist, are much more freed in giving advice, education, conversion via exhortation or other means, and especially the exploitation of suggestion. There is nothing wrong with the skilled use of such methods, of course. They may be quite necessary if it is not feasible or wise to rely upon a standard form of psychoanalysis.
Psychoanalysis is superficially like the professions of history and biography -- but not very much. Like historians, psychoanalysts attempt to reconstruct aspects of the past, but analysts as therapists are interested mainly in the past as it is repeated in the present: as it is the present in unfortunate ways. Analysis does not attempt to eliminate the past -- far from it, it often works to restore it. But it tries to make sure the past is past. Of course in our work we would like to come as close as possible to understanding "what really happened" in past external reality, even if we care more about "really is going on" in present external reality. But in contrast to most historians and sociologists, analysts are more interested in the world of the internal reality of the individual (McLaughlin, 1981; Thoma & Kachele, 1975). Similarly, cultural anthropologists do some of the same things analysts do, but their concerns are much broader -- the past and present conventions and sanctions of given societies and how these compare to other societies.
In another way psychoanalysis differs from the worthy fields of history, sociology and anthropology. A psychoanalyst is confronted with and aims to facilitate, change, and reorganize within his domain; he is interested in structural changes which take place within this patient while he and the analysand work mutually within the same field of forces. Such changes imply alterations in the field that both patient and analyst are observing -- in both its conscious and its unconscious aspects. Thus, while an analyst tries to do the best he can to be a rational, logical, consistent observer of his field, he puts to work his own mind as a fundamental part of the same field, including its potentially non-rational aspects.
It is possible to find parallels with psychoanalysis in such diverse fields as embryology, paleontology -- even with the fields of immunology and epidemiology. But in every respect the disciplinary frameworks (shared subjects of interest, the physiological concerns, basic assumptions, favored conceptual models, aims and techniques (Kuhn, 1977) differ.
Each respectable framework assembles and configures the data and knowledge accumulated within it in unique ways. Psychoanalysis, like other respectable intellectual endeavors, aims to be self-consistent. At the same time it looks for at least potential congruence of its findings with comparable findings from others. If one set of findings absolutely contradicts the other, attention is hopefully paid to the conflict -- from both points of view.
Analysts since Freud have had great, if cautious, respect for opinions made from vantage points of competence in other professions. Often, the central body of psychoanalytic knowledge has been influenced by them. Obvious examples include some of the controlled observations of the development of infants, studies of experiments' of nature (such as congenital blindness), the influence on characterological tendencies by pre-natal endocrinological abnormalities (including changes following permanent corrections of some congenital defects -- it is beginning to be possible actually to alter parts of what Freud called 'constitution').
A problem exists, however, when critics who are competent within their own fields presume also to have a competent, in-depth, understanding of human minds. Most often, they have (or lack) the intuitive understandings available to all lay people. Freud once said, wryly,
If you raise questions in physics or chemistry, anyone who knows he possesses no 'technical' knowledge will hold his tongue. But if you venture upon a psychological assertions you must be prepared to meet judgments and contradictions from every quarter ... Everyone has a mental life, so everyone regards himself as a psychologist. (1926, p.192).
At this time, only 'mainstream' psychoanalysis comes close to having an armature on which a rational general theory of mind might be built, a frame within which it can operate which can demarcate it from other working general theories of mind, along with a sufficiently cohesive body of professionals to construct it, put it to use, and alter it, if necessary. It is an open system of thought and within it, as is desirable, challenges to theories and specific practices are common.
Psychoanalysis has no rivals yet because, so far, alternative theories have inevitably left out, or prematurely tried to refute, one or more of the sets of empirical findings, not conjectures or primary assumptions, that have been discovered in practice, for examples, the operations of the dynamic unconscious, the ubiquity of intrapsychic conflict, the complex compromise of multiple motivations, the operations of transferences in human relations, the complex epigenetical patterns of development. These functions go beyond usual intuitive understandings, and there is an understandable wish to avoid some of them. With the omissions, the theoretical results, naive or sophisticated, tend to become closed systems, belief systems, not open systems of evolving thought (Rangell, 1983; Guttman, 1985). So far, not one rival has had enough lasting success to attract and hold the generations of professionals capable of creating either a cohesive scientific are a cold easy than non-scientific working disciplines.
THE INTERFACE
Like it or not, psychoanalysis has parallels with philosophy too, and in a peculiar way, straddles what once were called natural and moral philosophies. It has epistemological and ontological interests, of course, and in particular a concern with the ancient and still unsatisfactorily solved conundrum, the relations of mind to body. In this, it shares a position with neurobiology. Mind, we presume, cannot exist without a brain that functions brain; brain is only a puzzle of narrow interest without mind. Ultimately, literal or logical topological similarities will be found between neural and psychological sciences--and are beginning to be found (Changeux, 1985; Kandel, 1983; Reiser, 1984). But we are still in the dawn of the hopes Freud entertained more than ninety years ago.
But kinship does not always mean amity. Marshall Edelson (1984) says there have been three important philosophical challenges to the status of psychoanalysis as science. The first two, logical positivism and the claim by Popper that analytic propositions cannot in any way be falsified (and thus cannot meet the crucial requirements he believed demarcated science from non-scientific activities), seem in retrospect 'hardly worth the bother of a response'. But, says Edelson, "I cannot think that will ever be said about the third" (p. 1). This third, presumably more daunting philosophical challenge, has recently been mounted by Adolf Grunbaum is his The Foundations of Psychoanalysis (1984).
The present essay will seek to demonstrate that, to the contrary, Grunbaum's challenge, judged on its own, is without much intrinsic merit. Why then write such an essay? The reasons have to do neither with the respectability of Grunbaum's work nor that it has spurred sundry laymen to argue in popular magazines that psychoanalysis is dead or soon will be. Rather, my reasons have to do with the fact that number of favorable responses to Grunbaum's work have come from psychoanalytic writers.
For example, an editorial in the Journal of the American Psychoanalytic Association (Shapiro, 1985) praises Grunbaum for being 'substantially more sympathetic toward psychoanalysis' than some of the more virulent lay-critics. The editorial goes on to say that 'the general tone of Grunbaum's text is even-handed, thorough, devoted to an exploration of whether we can find a scientific case for psychoanalytic interest ...' (p. 7). Other analysts have apparently been even more favorably impressed. Among them: Philip Holzman (whose paper, 'Psychoanalysis: Is the Therapy Destroying the Science?', was announced as a winner of the American Psychoanalytic Association's Prize for 1984 and whose paper appeared in 1985. Holt (1984) and Eagle (1984) are similarly impressed. Even Edelson (1984), who contests Grunbaum's claim that single-subject clinical research is without scientific validity, still agrees with his basic philosophical position.
Understandably, most scientific practitioners do not care much about philosophers' opinions of their work. After all, if chemists had heeded British academic philosophers, they never would have dared invent atomic theory or found the periodicity of the periodic table. On the other hand, a growing number of psychoanalysts do not believe we can afford to disdain questions about the underlying philosophical assumptions in our field (and we do have such assumptions), or questions about the relative value of our investigative techniques and the truth-values of the evidence generated by them. And interest also grows, inside and outside our field, in the nature of scientific disciplines: what are the impersonal ideas and personal activities which allow us to sort out that which allow us to sort out that which is scientific from that which is not? For a variety of reasons, psychoanalysis lends itself as a convenient test case for such questions (Kuhn, 1977).
It is true that some analysts, rather simply ignoring difficult epistemological matters, have wished to finesse them by proposing that scientific ambitions for psychoanalysis be abandoned entirely. Unfortunately, these colleagues have been dubbed 'hermeneuticists' (thus blurring the important distinctions among them and blurring the meaning of the word, hermeneutic). But however named, and whatever ways they differ, they agree in urging psychoanalysis to join the company of humanistic disciplines like history or literary criticism. If wished, psychoanalysis could be called a 'hermeneutic science' (Gill, 1983). And that supposedly would be that. Others declare that psychoanalysis is in fact a natural science, and that's that. But perhaps the majority of those who care one way or the other disagree with these extreme views -- as they disagree among themselves. There is no consensus about alternatives.
Similarly, philosophers, philosophers of science and historians of science have been intrigued by the same epistemological and ontological puzzles. Some of them have made themselves familiar with our field, and a few have become practicing analysts as well as philosophers. One hopes, in a community of individuals who respect each other, that the resulting interactions will be constructive. But other philosophers with outstanding credentials are not only less knowledgeable, but unaware of their ignorance. They doubt that psychoanalysis has any valid reason for existence. Psychoanalysis nevertheless has survived; nothing much has come of these latter interchanges but alienation.
GRUNBAUM'S BOOK
The title of a book, announcing that it is a philosophical critique of the foundations of psychoanalysis, is apt to make the analyst-critic wary. What does its author mean, psychoanalysis? And what foundations is he talking about? Can the reader get some notion about what is supposed to be built on those foundations? Has the author had some measure of experience doing psychoanalysis, or at least some formal training? Or is he an autodidact who has merely read about the subject?
Adolf Grunbaum's book (1984) comes along at just the right time to be a test case of its own. While he does not specifically proclaim the future demise of psychoanalysis, he argues every reason to believe that such an ending not only can but should happen. At first, his book, constructed as it is out of a series of papers, seems ungainly, poorly organized. The organization becomes more apparent once examined closely. The first section takes a 'play-like' stance; that is, it first provisionally accepts the scientific probity of clinical observations. From this platform, it presents plausible arguments designed to destroy non-scientific approaches to psychoanalysis, fashionably labeled, as mentioned above, as 'hermeneutic' in nature. The second part of the strategy is to withdraw the provisional acceptance and present every conceivable reason to doubt that clinical evidence has any scientific value at all. On the basis of these arguments, Grunbaum can announce that there is no reason to believe that psychoanalysis has anything more to offer scientifically than heuristic questions, no reason to believe that it has any specific benefit at all therapeutically. In fact there are reasons to think it is worthless. As for the heuristic questions, if testable at all, they will be answered by non-psychoanalytical investigators, not by psychoanalysts.
FOUNDATIONS
Grunbaum's book presumes to examine foundations. Foundations might be taken to be the fundamental, core concepts, the tenets of a discipline. Freud defined the tenets of psychoanalysis variously, sometimes in recognition of the principal of psychic determinants and, the existence of a dynamic unconscious, and infantile sexuality, etc... At other times he pointed to the phenomena of transference and resistance (Guttman, 1965, 1985).
Or the foundations may not be the theories at all. Rather, Freud said in 1914, they are the primary, disciplined observations made by experience and trained professionals. The foundations of psychoanalysis, he said, are identical with empirical observations -- clinical observations. But, he added a year later in 1915, these observations of the yield of phenomena are not made randomly: they are guided by processes of pattern recognitions and selections, the nature of which are only dimly understood. These statements hold among neural scientists to this day. Freud had highly sophisticated conceptions of science -- although not many contemporary writers recognize this fact. Theoretical ideas, he said, are not the foundation of science but part of its superstructure -- a superstructure which can be altered or even discarded in favor of superior formulations. The foundation is observation alone, ordered at the outset by the human capacity to recognize gestalten, even if dimly seen in ways that later prove to be erroneous.
Freud did not find it necessary to rebut smooth and seemingly unassailable negative constructions made out of radical doubts. If any and every observation made by any and every observer is to be doubted, the only consequence can be nihilism. Skepticism of this sort amounts to an something impregnably hopeless, as a Waelder (1962) and Wisdom (1967), among many others, have pointed out. It would be impossible to live in ordinary life, much less be a scientist, if one seriously committed oneself to a babble of mindless doubt.
When doubt is supplied tendentiously as it is in this book, the result is neither an exercise in idealistic purity nor nihilism -- it is simply something that is misleading. Grunbaum attacks selectively. On the one hand he challenges psychoanalysis relentlessly, whether as 'Hermeneutic' or 'scientific theory.' On the other hand, he accepts, without voicing any criticism whatever, the validity of attacks on psychoanalysis from the most dubious sources. In particular, he shows on questioning faith in the veridicality of 'scientific, empirical studies' made by behavioral scientists - though one could not presume him to be so naive as to be less cautious in such faith than behavioral scientists are themselves. But in whatever way Grunbaum means foundations, he finds those of psychoanalysis 'irretrievably corrupt' -- 'contaminated'.
The book, then, touches upon:
1. Contemporary philosophical positions having to do with in epistemology and ontology.
2. The various attitudes among philosophers, philosophers of science, historians of science, and scientists themselves about the definition of science, including demarcations between what can and what cannot be thought to be scientific.
3. The definition and nature of psychoanalysis; the veridicality of its studies of individual minds.
4. The veridicality of empirical studies of psychoanalytic theories made by outside observers.
5. The nature and presentation of the author's own arguments.
In a review-essay, it is not possible to cover so much territory. Inasmuch as it seems to me to be a mistake to abandon scientific ambitions for psychoanalysis, the points of view lumped together in his book as "hermeneutic" will be bypassed. Some of Grunbaum's criticisms of the work of Ricoeur and that of Habermas work seem apt. However, it is hard to agree with a harshnesa of his approach to their ideas; his indictments do not seek to contest but to bury. Similarly, Popper's general doctrine of falsifiability (that a claim must, to be considered as part of science, be capable of being falsified) has been challenged convincingly; too many examples of psychoanalytic hypotheses have been, or at least are capable of being falsified, to cite. In that respect, some of the criticisms leveled in Grunbaum's book are valid, if not original. But Popper's philosophy, as it relates to psychoanalysis, should not be dismissed in other respects, particularly in terms of the mind-body question or the question of multiple realities.
I will have a bit more to say about the philosophies of science, but will devote most attention to Grunbaum's conception of psychoanalysis, his notions about empirical scientific studies, the quality of the scholarship, and the nature of his presentation.
PHILOSOPHIES OF SCIENCE.
Reading Grunbaum's book, one would never guess that he represents a point of view held only by minority of philosophers. He identifies himself as an inductivists of a particular sort -- one who believes that repeated confirmations of theories cannot be probative, but the various theories explaining empirically observed events can be narrowed and some of them can even be eliminated by empirical testing. He follows the stern canons of science put forward by neo-Baconians and followers of John Stewart Mill. In addition he classifies himself among those philosophers who use hypothetico-deductive organizations of logic. Not a scientist, he approaches science with idealistic, pre-conceived, reductionistic standards. He does not seem to recognize that on more abstract levels of theory, not even the exact sciences could stand up to such an approach.
Nevertheless, Grunbaum's claims are presented so dogmatically one might forget that Hume's challenge to believers in induction (to find any demonstrable 'necessary connection between a supposed causal action and two events, e.g., the collision of two billiard balls) has not been met. There have been many attempts, but a consensus of philosophers is still unsatisfied. And in the face of 20th-century scientific thought, even the 'eliminative inductivists' no longer claim to be able to reach absolute truths. Only probabilistic truths, they say, are achievable.
Not a philosopher, I am unable to evaluate the most important criticism made of his stance by some of Grunbaum's fellow philosophers: that the approach is particularly unsatisfactory because it postulates an infinite number of theoretical explanations for a given event, and then proposes to eliminate -- from the infinite number -- those theories which are less satisfactory on a probabilistic basis (Black, 1958; Agassi & Jarvie, 1979; Laorr, 1985). But as a psychoanalyst, I cannot imagine the actual use of such perfectopmistic standards
A serious layman may read as much as he can of the contemporary literature of philosophy, philosophy of science, and history of science -- and, acknowledging the limitations of his competence, still perceive something clearly: there is a cacophony of disagreements among all these judges. There is no consensus about the definition of science, or its demarcations from other fields of learning. Various criteria are set and debates occur about whether a given discipline should or should not be considered a science. There are controversies about the nature of reality itself, and the appropriate epistemological stances to explore and later represent it (McLaughlin, 1981).
There is no agreement even about how to scientist works and how given sciences develop. The erosion of certainties made by post-Neutonian physics, post-Darwinian biology, post-Jamesian psychology, Freudian psychoanalysis, and the still unchallenged destruction by Godel of the belief that it is even possible to construct an ideal mathematical world without internal contradictions, have been accompanied by more positive trends, among them the recognition by Kuhn (1962, 1977) and others of the social dimensions present in any scientific discipline. The social and psychological dimensions have been explored by psychoanalysts and others knowledgeable about psychoanalysis have actually worked as scientists in other fields (Ecker, 1985; Guttman, 1965, 1985; Kuhn, 1977; Polanyi, (1974), Waelder, 1962 -- and Freud).
GRUNBAUM CONTRA FREUD
To return to specifics, what is this 'psychoanalysis' Grunbaum writes about? It is like nothing colleagues of my acquaintance recognize. First of all, the reductionism is extreme. Out of Freud's sets of theories, held at differing times and having varying levels of abstraction, the author distills 'the' Freudian theory. It is essentially that of Freud before 1897, but includes selective statements -- but not their contexts -- made by him later. And what is Grunbaum's distillation? He begins by taking a few more or less casual comments of Freud -- and then translates them as if they were formal philosophical statements. For example, repression, Freud said, very early, was the fundamental concept of psychoanalysis. He obviously meant that the understanding of repression led to the understanding of the dynamic unconscious and the vindication of the principle of psychic determinism. And thus he was able to say -- in a way we would regard as a schematic over-simplification -- that repression, and the maintenance of repression, is the source or cause of the symptom. This came at a time when explicit symptoms were still major therapeutic preoccupations.
Grunbaum's system gives no sense that he is talking about one of Freud's earliest and most simple beliefs -- and of which was quickly disabused: that specific psychoneurotic symptoms had specific and single causes. Grunbaum gives no suggestion the Freud later developed a collection of more or less closely associated theories, or that during each epoch of Freud's professional career additions or modifications were made. No indication is given that the theory of repression at first covered all defenses, later that it was one of several, or that the role of anxiety was changed in a revolutionary way. In fact, each development of Freud's thought is actually a retreat from his youthful beliefs that he had found relatively specific causes for relatively specific illnesses. For example, Freud's (1925) 'Inhibition, symptoms and anxiety', is dismissed as 'his disappointingly fuzzy 1924 essay on anxiety...' (p. 165).
Nevertheless, repression -- for Grunbaum -- is still the linchpin of 'the' theory -- "the cornerstone of the psychoanalytic edifice." If repression is undone therapeutically, and the symptom disappears, then at least 'the' theory is not falsified. However, the disappearance of 'the symptom' supposedly must happen following a correct interpretations, else 'the' theory is invalidated. It is said (p. 185) that all subsequent analysts subscribe to this. They do not. In my opinion all subsequent analysts who thought about the matter would see Grunbaum's formulation as a naïve caricature. Contemporary thought about the mechanisms of therapeutic action is too complex for it to be feasible to discuss here. One can only sat that Freud's latter-day concepts, much less those of present-day analysts, bear no resemblance to Grunbaum's ideas.
Freud meant that an interpretation should connect with something within the patient that seems 'real' to him, plausible, and has consequences in terms of not only intellectual but emotional reverberations. But even if it does it will be regarded as a provisional interpretation, subject to revision as more insights accrue. But Grunbaum blew up the one casually used word, 'tally', into something he (not Freud) named the 'Tally Argument'.
Grunbaum believes he was the first to notice this 'cardinal epistemological defense', with 'its bold lawlike premise' (1984, p. 127). He (not Freud) then used this 'thesis' to set forth a number of 'claims'. They seem to the psychoanalyst peculiarly insistent and dogmatic, as if written in stone -- in no way resembling the spirit of Freud's statements, then or later. Freud -- according Grunbaum -- claims that suggestion does not create an "irremediable contamination' (Freud used no words like 'irremediable contamination'); that the analytic treatment operates differently from all rival therapies in 'not resting ultimately on suggestion' (he didn't say exactly that either); that the causal claims can be validated within the analytic situation through the resolution of conflict through interpretations, and do not 'need to encompass experimental controls and statistical comparison with untreated samples' (Freud said nothing on those pages about experimental controls; he did say that statistical studies of psychoanalysis, as of 1917, would be worthless because the cases assembled at that time were too diverse and the items to be examined too heterogeneous).
Grunbaum's renderings of Freud's arguments are 'hard-edged', as if Freud were some sort of professional logician. Freud's actual arguments were 'soft-edged', leavened by the knowledge that he was presenting psychoanalysis in a light hat might be understandable for beginners. He was neither producing a tight philosophical argument no his final statements on these matters. Grunbaum acknowledges, but bushes aside, Freud's wiser, more restrained, more cautious explanation of therapeutic action twenty years later (1937).
Instead, Grunbaum, after complimenting Freud for being a 'superior scientific methodologist' (p. 130), goes on to say,
Unless the methodologically damaging import of the patient's compliance with his doctor's expectations can somehow be neutralized, the doctor is on thin ice when purporting to mediate veridical insights to his client rather than only fanciful pseudo-insights persuasively endowed with the ring of verisimilitude. Indeed, if the probative value of the analysand's response is thus negated by brainwashing, then Freudian therapy might reasonably be held to function as an emotional corrective not because it enables the analysand to acquire bona Fide self-knowledge, but instead because he or she succumbs to proselytizing suggestion, which operates the more insidiously under the pretense that analysis is nondirective. (p. 130).
What destructive twisting and distortion of Freud's words on the subject! What invention of concretistic-seeming phrases that Freud never used! Grunbaum repeatedly refers to this 'defunct Tally Argument', that 'aborted Tally Argument'. He does the same thing with something he (not Freud) invents and terms the 'Necessary Causal Thesis", or "Freud's Master Proposition' (p. 140), which he abbreviates as 'NCT'. NCT is the supposed claim of Freud's that only the psychoanalytic method of interpretation and treatment can yield correct insight 'into the unconscious pathogens of his psychoneurosis', and the analysand's correct insight is 'causally necessary' for successful treatment. It must be remembered that Grunbaum argues that the scientific probity of clinical observations depends upon therapeutic results (just because Freud said so early in his career).
Having established what he means by psychoanalysis, Grunbaum proceeds to examine his creation. Obviously it depends upon a concept of the operations of a dynamic unconscious. It depends upon what evidence might be provided for such an unconscious primarily by the use of free association, but also by the interpretation of dreams, slips of the tongue, and artistic productions. Finally, it depends upon therapeutic results: to repeat, results here seen trivially as the undoing of the symptom by the undoing of repression.
The philosopher then examines in succession concepts having to do with dreams, parapraxes, and free association. He brings up examples of challenges of each concept -- without examining sources and with no criticism whatever of them. No critic of psychoanalysis, Eysenck, for example, is criticized. If the opponent of psychoanalysis is from within our field, adjectives like 'prominent' are inserted before his name. Analysts who are supposed to be 'hermeneuticists' are contemptuously vilified. The few analysts cited who defend the scientific bases of our field are ridiculed. Brenner is 'hollow and evasive'; others have arguments which are 'pathetic -- the examples could be vastly extended.
This is reinvention of a version of Freud as stupid or foolish academic philosopher, followed by his demolition. The evidence for the psychoanalytic theory of dreams, Grunbaum avers, is thoroughly flawed. And suggestion is the worst of all the devils in the ointment. Everything Freud did clinically, and we do, is 'irretrievably corrupted by suggestion and bad logic, bad claims. Our successes are explained by what Grunbaum calls 'The Placebo Effect'. Therefore, according to him, if psychoanalysis is to have any scientific validity, that has to be shown by empirical methods conducted by non-psychoanalysts.
I hope the flavor and style of Grunbaum's thinking, so alien to Freud's thought (not to speak of that of contemporary psychoanalysts), and so tendentiously, perfectionistically 'logical', has been illustrated. These matters are examined, in the course of a widely ranging critique of the work of philosophers interested in psychoanalysis, including Grunbaum's, by Wallerstein (1985). He remarks that the so-called Tally Argument and NCT showed that 'Grunbaum has simply pushed through an open door that has been widely open for more than half a century', and goes on to remark that the disappearance of such naïve theories of reality 'is not news to either psychoanalytic theorists or psychoanalytic resarchers. Nor should this on logical grounds occasion surprise'.
How Freud would have resented being presented as a philosopher, much less a bad one! He is known to have had mixed feelings about philosophy. To Fliess, he wrote, 'I secretly nurse the hope of arriving by the same route [of medicine] at my original objective, philosophy. For that was my original ambition ... " (Masson, 1985). His admiration of Nietzsche, who influenced probably every well-educated person of his generation, is well known. Yet, in his public writing, he took pains to distance psychoanalysis from philosophy:
Philosophy is not opposed to science, it behaves like a science and works in part by the same methods; it departs from it, however, by clinging to the illusion of being able to present a picture of the universe which is without gaps and is coherent, though one which is bound to collapse with every fresh advance in our knowledge. It goes astray in its method by over-estimating the epistemological value of our logical operations ... (1933, pp. 160-61).
Let us go more deeply into what Freud actually stressed. He was an omnivorous reader, and there is no doubt that he had an understanding of those philosophers who aimed then -- and sometimes still aim -- for the 'really-real' Truth. But clearly, he was skeptical about the idealization of logical operations. As an older man, he may not have become acquainted with such twentieth-century philsophers as Wittgenstein, on one end of the spectrum, or some neo-Kantians, like Cassirer, on the other. In the opinion of Stent (1975, 1978), a neuro-biologist, the philosophical underpinnings of psychoanalysis are more related to the latter group than to the nineteenth-century positivists. Wurmser (1978) called Freud 'a philosophical thinker of first rank'. However, Freud was a thinker about philosophical matters, not a philosopher. Nevertheless, no philosopher dealing with human affairs can afford to ignore him, if for no other reason than his demonstration that the ego is not master in its own house. Philosophers before his time, like Nietzsche, had asserted as much, but had been unable to demonstrate.
Freud's thought was conscious rooted in nineteenth-century positivism and determinism. But he had a healthy skepticism:
[Science's] endeavor is to arrive at correspondence with reality -- that is to say, with what exists outside and independently of us ... This correspondence with the real external world we call 'truth' (p. 170). But this 'correspondence' is only partial, he quickly adds. Science 'collects observations of uniformities in the course of events which it dignifies with the name of laws and submits to its risky interpretation. And consider the small degree of uncertainty which it attaches to its findings Everything it teaches is only provisionally true ...' (p. 172, my italics).
In Lecture 13 of his Introductory Lectures on Psychoanalysis, Freud (1915-17) tried to combat the common accusation that psychoanalysis was 'nothing more than a particularly well-disguised and particularly effective form of suggestive treatment and we should have to attach little weight to all that it tells us about what influences our lives, the dynamics of the mind or the unconscious' (p. 452). He showed from analytic experience how hard it is consistently to influence patients in any lasting way.
The doctor has no difficulty, of course, in making him a supporter of some particular theory and in this making him share some possible error of his own. In this respect the patient is behaving like anyone else -- like a pupil -- but this only affects intelligence, not his illness. After all, his conflicts will only be successfully solved and his resistances overcome if the anticipatory ideas he is given tally1 with what is real in him. Whatever in the doctor's conjectures is inaccurate drops out in the course of the analysis ... (p. 452).
Freud went on to show that provisional interpretations are constantly being altered, corrected, and put into context in the course of a long analysis. The analyst is skeptical of quick confirmations of interpretations precisely because they might be the result of compliance on the part of the analysand, and thus the products of suggestion. The analysis of transferences and resistance acts to counter suggestion, as does the gradually increasing comprehension of his function as a whole, along with the understanding of the structural changes which take place in the course of treatment. In these ways Freud sought to differentiate psychoanalysis from other psychotherapies which either unwittingly depend upon suggestion or deliberately exploit it.
The doctor has no difficulty, of course, in making [the patient] a supporter of some particular theory and this making him share some possible error of his own. In this respect the patient is behaving like anyone else -- like a pupil -- but this only affects his intelligence, not his illness. After all, his conflicts will only be successfully solved and his resistances overcome if the anticipatory ideas he is given tally1 with what is real in him. Whatever in the doctor's conjecture is inaccurate drops out in the course of the analysis ... (p. 452).
Freud went on to show that provisional interpretations are constantly being altered, corrected, and put into context in the course of a long analysis. The analyst is skeptical of quick confirmations of interpretations precisely because they might be the result of compliance on the part of the anlysand, and thus the products of suggestion. The analysis of transferences and resistances acts to counter suggestion, as does the gradually increasing comprehension of his function as a whole, along with the understanding of the structural changes which take place in the course of treatment. In these ways Freud sought to ifferentiate psychoanalysis from other psychotherapies which unwittingly depend upon suggestion or deliberately exploit it.
Compare these statements of Freud's with Grunbaum's 'systems' mentioned above. And compare them with an example of Grunbaum's perfervid prose:
Without the vindication of some other as yet unknown epistemic underpinning, not even the tortures of the thumbscrew or the rack should persuade a rational being that free associations can certify pathogens or other causes! For, without the stated therapeutic foundations this epistemic tribute to free associations so far rests on nothing but a glaring causal fallacy. Therefore, it is unavailing to extol the method of clinical investigation by free association as a trustworthy resource of etiologic inquiry, while issuing a modest disclaimer as to the therapeutic efficacy of psychoanalytic treatment. And one is dumbfounded to find that noted psychoanalysts such as George s. Klein have done just that (1976, pp. 36-38). Also, amazingly, the renowned analyst Judd Marmor conjectured, as we recall, that it was the accidental need to earn his livelihood as a psychiatric practitioner which drove Freud to utilize his investigative tool (of free association) simultaneously as a therapeutic instrument.
The shabbiest arguments of all are presented to derogate the therapeutic effects of analysis. It will be remembered that Grunbaum, following an early line of Freud's reasoning, made the therapeutic effects of analysis the centerpiece of his arguments. On page 161, the author says:
In recent decades, comparative studies of treatment outcome from rival therapies have failed to reveal any sort of superiority of psychoanalysis within the class of therapeutic modalities that exceed the spontaneous remission rate gleaned from the (quasi-) untreated controls (Smith, Glass, and Miller, 1980; Rachman and Wilson, 1980; Strupp, Hadley, and Gomes-Schwartz, 1977). But if analytic treatment is thus not superior to its rivals in the pertinent diagnostic categories, it becomes quite reasonable -- though not compelling -- to interpret its therapeutic achievements as placebo effects. And if so, the therapeutic successes of psychoanalysis are not wrought after all by the patient's acquisition of self-knowledge, much to Socrates' sorrow. In this vein, the psychiatrist Jerome Frank has contended that the analyst, no less than his competitor, heals neurotics by supportively counteracting their demoralization, not by excavating their repressions.
What Grunbaum knew, but did not say, was that the papers to which he referred do not in any adequate way psychoanalytic results in the comparative studies. The studies, in fact, failed to show anything of relevance to standard or classical psychoanalysis (Schachter, 1985). But Grunbaum 'covered himself' by saying, 'if analytic treatment is thus not superior ....' And, 'if so, then the therapeutic successes are not related to insight at all, but like other therapies to the supportive counteracting of patients' demoralization' (my italics). The unwary reader would think that there is 'scientific' (and thus 'reliable') evidence challenging the therapeutic results of psychoanalysis. There is not.
DISCUSSION
A reasonably careful reading of Grunbaum's book reveals it to misrepresent psychoanalytic thought and psychoanalytic claims in many more ways than I have been able to indicate. He invents a philosophy for it and then subjects it to perfectionistic epistemological standards of an impossible stringency. And he misrepresent the literature he cites.
Many psychoanalysts, who were trained according the epitemological standards of the nineteenth century, have been in a quandary. Of course they would like their work to be regarded a scientific, at least aiming toward scientific standards. But their positions as observers are different in certain ways from those of other scientists, and their knowledge of people is different. They know more about people, including knowing about themselves, than most scientists and philosophers. Bertrand Russell2 believed (1928, pp. 2942) that these insights might benefit science in general, rather than pose spurious threats.
PHILOSOPHY IN THE WORLD OF ACTUAL PSYCHOANALYTIC WORK
It would be fair to ask of philosophers who are interested in psychoanalysis that they acquire understanding in depth not only of psychoanalytic theory but also the nature of the work. This work creates an investigatory framework which generates empirical data of a particular set and arrangement. The discipline has its own methods, on different levels, to work as hard as it can to validate or throw into question concepts on the various levels of theoretical organization: clinical generalization, clinical theory, and more abstract theoretical formulations (Waelder, 1962). Philosophers then would be in position to be constructive in their criticisms of our epistemology -- rather than attempt to impose pre-conceived and idealistic criteria which are not even agreed upon by other philosophers or scientists themselves (Wilson, 1985). Hopefully the great scientific my of the twentieth century, that physics should be the model for all sciences, can be put to rest. Or, if not, at least we should be shown in what actual ways the criteria derived from physics can be applied to the biology and psychology of living creatures.
In my work, I would not even try to 'prove core tenets of psychoanalysis. For example, the belief in psychic determinism is not provable. But its practical, heuristic validity has shown its usefulness countless times. No satisfactory attack on its fundamental usefulness has been plausible, even if it can no more proved a law of the universe than physical determinism can. Psychic events do 'make sense'. Just as, for practical purposes, we take existence if not the understanding gravity for granted (awaiting a consensus among physicists about a unified field theory), so, also, we can use the fundamental assumptions of psychoanalysis as working and workable hypotheses.
On other levels, the concept of the dynamic unconscious has been validated not only through psychoanalytic experience, but through abundant scientific and non-scientific experiences. The existence and importance of infantile sexuality is not a theory but has the status of a regularly observed set of empirical facts. The same holds for the existence of transferences and defensive operations. Of course, attempts to disprove these tenets are welcome. There are ways they could be falsified if they are false. But no such efforts have been remotely successful so far.
If I were interested in a clinical investigation, I would work on smaller matters, just as another scientist would not try to duplicate each original observation found in his text-books. Suppose I am interested in some observations which suggest that specific character trait is regularly associated with (not simply caused by) existence of some centrally active, dynamically coherent, unconscious fantasy. And further, that as the previously unconscious fantasy becomes more available to the patient's consciousness, and he is motivated to take it into account in so far as it influences his life, the intensity of the investment in this character trait becomes lessened. It would not be difficult to demonstrate to fellow analysts that this fantasy contains some derivatives of infantile sexuality and aggression, along with plausible transferential references to me or other contemporary persons. On the other hand, it might be quite difficult to demonstrate these connections to professionals who do not possess the technical knowledge of psychoanalysis to which Freud referred.
But must I live long enough to consider an infinity of other explanations before my observation would have probative possibilities? I am assuming, of course, that my observation would have been put alongside other analogous observations by colleagues, and that it would be challenged over the course of time by alternative explanations offered by fellow analysts. I am also assuming that reliable findings made by colleagues in neighboring disciplines would not have cast a shadow on the plausibility of my findings at any important level.
More centrally, it is a fair assumption that all along during the analysis I would have been taking into account, with that combination of acceptance and skepticism characteristic of analysts, some possibly questionable aspects of my notion that this unconscious fantasy and that particular character trait entailed each other: I might not have heard my patient correctly, might be tendentiously drawn in certain theoretical directions by my own unconscious needs, might have not reduced, as far as possible, every element of suggestibility operating in patient and me, might be being 'conned' by patient's need to please me, might be missing something else being covered up by the overt revelations. No discipline is any more assiduous than psychoanalysis in attempting to identify and cope with elements that might potentially distort its evidence. But that does not mean that its practitioners can expect to be omniscient, or be expected to be, or be adjudged worthless because their best and most careful efforts are imperfect. In fact they are beginnings.
The example given, that two factors entail each other, a specific, important and persisting unconscious fantasy and the existence of a particular character trait, and the study of the alteration of both in a context of the analysis of transference and resistance, assumes only one level of theory building. The more abstract generalizations -- what Freud called the super-structure of theory -- also are considered in various ways by the discipline of psychoanalysis. Some of these theories can only be tested in the market-lace, so to speak. Others can be examined for internal consistency, the avoidance of tautologies, the lack of conflict with each other, perhaps better supported theories. Some of them can be tested by non-psychoanalytic techniques, and such external empirical evidence is very much welcomed. Most important of all, psychoanalytic research utilizing multiple investigators studying multiple analytic patients has been conducted and is being conducted, despite the current lack of adequate support by public and private funds.
It is worth reiterating that no other methodology or technology now exists to study the phenomena I chose earlier as an example. I challenge any investigator to try to devise a methodologically sound way to explore the clinical questions raised above without having to begin with a number of already well-documented psychoanalytic core concepts.
PSYCHOANALYTIC EPISTEMOLOGY
Books like Grunbaum's distract us from serious matters. No other discipline is expected to defend itself against perfectionistic philosophical 'canons' like those advanced in his books. Psychoanalysis is, in many ways, in the same boat with a number of other biological fields. A fascinating paper,3 by a leading epidemiologist, Alfred Evans (1978), examines the vicissitudes of the Koch-Henle Postulate -- the conditions necessary for proving a specific causation of disease by specific micro-organisms. The epistemological problems are startlingly like some of those of psychoanalysis. A comparatively simple theory (but one not held as invariable or law-like even by its originators -- despite being held so by many followers) gradually became altered by new information, e.g., the existence of viruses as pathogens, to become a much more complex set of conditions still known as the Koch-Henle Postulate. The understanding of the multiple and elaborate variables in the field of immunology continues to deepen. Some of the problems involved in immunology, particularly auto-immune responses, are analogous to psychoanalytic interests. So are questions brought up by other disciplines, for example, field research in cultural anthropology.
Psychoanalysis has real questions to face that are completely unlike those posed by Grunbaum. How do we advance our scientific investigations while maintaining therapeutic integrity? Can we develop better rules of evidence? Better methods of verification? Better, and more comparably organized, single case studies? Better methods of intra- and interdisciplinary communication? Study those investigations made by non-psychoanalysts that are relevant to our discipline?
Psychoanalysis, besides being one of the natural sciences, besides being in some ways like the discipline of history, is both simultaneously -- and it is something new (Wurmser, 1978). While it shares many specifically analogous investigative and theoretical puzzles with other fields, it is as a whole unique in the degree to which the observer is part of the field observed and in the degree of complexity in producing verifiable evidence in single case studies -- an area in which Edelson (1984) studied productively.
There is more than one logic of inquiry in science (Whitley, 1985). The organization of a given field develops differently organized kinds of knowledge. Psychoanalytic clinical work is an investigative method framed by certain quite unique sets of rules (Spruiell, 1983). Out of the unique mutuality and intersubjectivity (Southwood, 1976; Blomfield, 1983) that ensues in a successful analysis (by 'intersubjectivity', I mean only its ordinary meaning, not the generalized connotations and implications with which it has been invested by other recent authors), observations are made and theories devised which order and to some extent explain these observations. Accordingly, the discipline has had to develop its own techniques to test its own statements for truth values, especially the slowly accruing self-corrections available in a long psychoanalytic process. It does without saying that relative veridicality, whether concerning the understanding of the history of a given individual, or concerning the truth values of abstract theories, has to depend upon extensive periods of mutuality between analyst and analysand, with mutual corrective processes always in operation.
There is something important to be added. As any discipline becomes more scientific in nature, it becomes more capable of making impersonal truth statements. But even purely mathematical statements are communications and methods of operation utilized by people. Contrary to the beliefs of some critics of psychoanalysis, the tenets of the field do not rely upon collections of anecdotal reports. Rather, they are developed slowly, in fits and starts, by the collectivity of those who belong to the discipline. Not only are ideas and concepts examined, defined, and redefined, proposed hypotheses are repeatedly challenged, tested, provisionally accepted or rejected, they are put to the tests of clinical experience by numbers of working psychoanalysts. Many false starts, premature reformulations, supposedly new 'paradigms' are put forward. Over time, it becomes clear that some new ideas (paradigms to one side) simply don't work; others do; still others suggest new directions.
An important book by Whitley (1985) explores both the diversity of scientific fields and the complications that external factors raise. Whitley constructs a typology expressing such factors as social organization which determine reputation, access to the literature, dependence on funding, on public approbation, relations with adjacent disciplines, the degrees of task uncertainties, and ambiguities in concepts. To acknowledge these factors is not to imply that scientific knowledge is totally relative (and 'contaminated', to use Grunbaum's word), nor to urge a radical skepticism of solipsistic subjectivity. But it is to imply that the growth of all sciences is complex, variable, and not to be reduced to one model (such as that of the so-called 'exact sciences' Nor is it possible to make more than approximations to truth, though respectable people try.
SUMMARY
Adolf Grunbaum, a philosopher, has written The Foundations of Psychoanalysis, which has attracted attention by analysts and non-analysts. The book is a slashing attack first on hermeneutic approaches to psychoanalysis; once he believes he has demolished hermeneutic credibility, he turns his attack onto Freudian clinical approaches -- and finds them devoid of any scientific validity. Grunbaum believes data acquired within the analytic situation as 'irretrievably contaminated' by suggestion, incorrect assumptions about the nature of evidence, and bad logic. Clinical evidence is supposed to have only heuristic value; the various hypotheses can only be tested by non-psychoanalytic methods. This essay contests all this on the basis of Grunbaum's own lack of understanding of the nature of psychoanalysis, and his tendentious methods of scholarship, which implacably attack anything psychoanalytic, and, simultaneously accept without criticism various dubious 'researches' by non-psychoanalysts. However, the purpose of this essay is not to consider Grunbaum's work. The aim is to challenge psychoanalytic writers who accept his arguments either wholly or in part.
FOOTNOTES
1. The words 'tally' and 'tallies' were used a total of fifty times in the Standard Edition (Guttmann et al., 1980), always in casual, informal senses, as in the present instance. But on this occasion, it was used as the basis for Grunbaum's 'Tally Argument'.
3. Brought to my attention by Dr. Paul Ecker.
REFERENCES
Agassi, J. & Jarvie, I. (1979). The rationality of dogmatism. In Rationality Today, ed. T. Geroets. Ottowa: Univ. Ottowa Press.
Black, M. (1958). Induction and probability. In Philosophy in the Mid-Century, ed. Raymond Libansky. Florence.
Blomfield, O. (1983). Review of The Standing of Psychoanalysis, by B. Farrell. Int. J. Psychoanal., 64:112-115.
Changeux, J. (1985). Neuronal Man: The Biology of the Mind. New York: Random House.
Eagle, M. (1984). Recent Developments in Psychoanalysis: A Critical Evaluation. New York: McGraw-Hill.
Ecker, P. (1985). Discussion of works by Edelson and Grunbaum. Pittsburgh, PA., 12 April, 1985.
Edelson, M. Hypothesis and Evidence in Psychoanalysis. Chicago: Univ. Chicago Press.
Evans, A. (1978). Causation and disease: a chronological journey. Amer. J. Epidemiology 108: 249-258.
Freud, S. (1914). On narcissism: an introduction. S.E.14.
------------ (1915). Instincts and their vicissitudes. S.E. 14.
------------ (1915-1917). Introductory lectures on psychoanalysis. S.E. 15-16.
------------ (1925). Inhibitions, symptoms and anxiety. S.E. 20.
------------ (1926). The question of lay analysis. S.E. 20.
------------ (1933). New introductory lectures on psychoanalysis. S.E. 22.
------------ (1937). Analysis terminable and interminable. S.E. 23.
Gill, M. (1983). The point of view of psychoanalysis: energy discharge or person. Psychoanal. Contemp. Thought, 6:523-551.
Grunbaum, A. (1984). The Foundations of Psychoanalysis. A Philosophical Critique. Berkeley, Los Angeles, London: Univ. Calif. Press.
Guttman, S. (1965). Some aspects of scientific theory construction and psychoanalysis. Int. J. Psychoanal., 46: 129-136.
--------------- (1965). The psychoanalytic point of view: basic concepts and deviant theories. Int. J. Psychoanal., 66: 167-170.
--------------- Jones, R. & Parrish, S. (eds.) (1980). The Concordance to the Standard Edition of the Complete Psychoanalytic Works of Sigmund Freud. Boston: G.K.Hall.
Holt, R, (1984). The current status of psychoanalytic theory. Amer. Psychol. Assn. Meeting, Aug. 1984.
Holtzman, P. (1985). Psychoanalysis: is the therapy destroying the science? J. Amer. Psychoanal. Assn. 33:725-770.
Kandel, E. (1983). From metapsychology to molecular biology. Explorations into the nature of anxiety. Amer. J. Psychiatry, 140: 1277-1293.
Kuhn, T. (1962). The Structure of Scientific Revolutions. Chicago: Univ. Chicago Press.
----------- (1977). The Essential Tension. Chicago: Univ. Chicago Press.
Laor, N. (1985). Psychoanalysis as science: the inductivist's resistance revisited. J. Amer. Psychoanal. Assn. 33:149-166.
Masson, J. (Trans. & Ed.) (1985). The complete Letters of Sigmund Freud to Wilhelm Fliess. 2887-1904). Cambridge, MA: Harvard Univ. Press.
McLaughlin, J. (1981). Transference, psychic reality and countertransference. Psychoanal. Q. 50: 639-664.
Polanyi, M. (1974). Scientific Thought and Social Reality, ed. Fred Schwartz. Psychological Issues. Monograph 32. New York: Int. Univ. Press.
Rachman, S. & Wilson, G. (1980). The Effects of Psychological Therapy. New York: Pergamon Press.
Rangell, L. (1983). Defense and resistance in psychoanalysis and life. J. Amer. Psychoanal. Assn. 31: 147-174.
Reiser, M. (1984). Mind, Brain, Body. New York: Basic Books.
Russell, B. (1928). Sceptical Essays. London: Unwin.
Schachter, J. (1985). Discussion of works by Edelson and Grunbaum. Pittsburgh, PA. 12 April, 1985.
Shapiro, T. (1985). Editorial: psychoanalysis, philosophy and the Public. J. Amer. Psychoanal. Assn. 33: 5-9.
Smith, M., Glass, G. & Miller, T. (1980). The Benefits of Psychotherapy. Baltimore: Johns Hopkins Univ. Press.
Southwood, H. (1976). Analysis as experience. Int. J. Psychoanal., 57: 167-174.
Spruiell, V. (1983). The rules and frames of the psychoanalytic situation. Psychoanal. Q., 52:1-33.
Stent, G. (1975). Limits to the scientific understanding of man. Science, 187: 1052-1057.
------------ (1978). Paradoxes of Progress. San Francisco: W.H.Freeman.
Strupp, H. Hadley, S. & Gomes-Schwartz, B. (1977). Psychotherapy for Better or Worse: The Problem of Negative Effects. New York: Jason Aronson.
Thoma, H. & Kachele, H. (1975). Problems of metascience and methodology in clinical psychoanalytic research. Annual Psychoanal., 3: 49-119.
Waelder, R. (1962). Psychoanalysis, scientific method, and philosophy. J. Amer. Psychoanal. Assn. 10: 617-637.
Wallerstein, R. (1985). Psychoanalysis as a science: a response to the new challenges. The Freud Anniversary Lecture, New York Psychoanalytic Institute. 23 April, 1985.
Whitley, R. (1985). The Intellectual and Social organization of the Sciences. New York: Clarendon (Oxford Univ. Press).
Wilson, E. (1985). Personal communication.
Wisdom, J. (1967). Testing an interpretation within a session. Int. J. Psychoanal., 48: 44-52.
Wurmser, L. (1978), The quandary of research in psychoanalysis. A philosophical study. Presented at the Fall Meeting of the American Psychoanal. Assn., New York, NY.