11
Gadamer's Hermeneutics

,

A Philosophy for the Embedded Analyst

So few analysts are willing any longer to defend the blank-screen metaphor of the analyst's role, at least in its stark form, sans qualifications, that to attack it has become a gratuitous act, useful only as a kind of chest beating. And yet every time one is ready to consign the issue to oblivion, one reads some new article, or hears about someone's defense of neutrality, and the tired old argument comes to life. So perhaps the distinction between those who adhere to the orthodox clinical view and those who don't retains meaning after all.

Be that as it may, it has become much less interesting to bash the blank-screen metaphor than to delineate differences within the group of analysts who already accept some kind of modification of the most conservative view. Work in this area has only just begun. One of the few comparisons of this sort is by Greenberg and Mitchell (1983), who describe three solutions to the problem of how to represent interpersonal relations in psychoanalytic theory. One can, like Freud himself, and along with Hartmann, Mahler, Jacobson, and Kernberg, try to change drive theory so that relational matters are accommodated, but in a way that is ultimately instinctual. Or, like Sullivan and Fairbairn, one can dispense with drive theory altogether and begin anew with interpersonal relations as the cornerstone. The third alternative is a mixed model in which relational and drive phenomena are explained separately, the kind of solution adopted, according to Greenberg and Mitchell, by Kohut and Sandler.

Another set of categories has been proposed by Hoffman (1983), whose work has particular appeal to the working psychoanalyst because of its exclusive focus on clinical theory. Hoffman discerns two ways clinical writers have taken issue with the blank-screen metaphor: the conservative critique and the radical critique. In the conservative critique, as in Greenberg and Mitchell's strategy of accommodation, the problematic aspects of the old idea may seem to be addressed, but actually the essence of the problem remains. The conservative critic emphasizes some aspect of the analyst's "real" influence on the patient (as opposed to influence "distorted" by the patient). The judgment that the patient has seen the "real" meaning of the analyst's behavior is based on how the patient reacts, the criterion being the patient's correspondingly "appropriate" or "undistorted" response. For example, to the analyst's genuine helpfulness and capability, the undistorting patient responds by joining the "real relationship" or the working alliance. Similarly, the "appropriate" response of the undistorting patient to the analyst's neurotic countertransference is the patient's accurate observation of the analyst's inappropriateness. The patient's capacity to assess accurately these influences from the analyst is therefore a social phenomenon; it is an effect of the interaction, "uncontaminated" by the nonrational inner world. But this capacity on the patient's part exists separately from the rest of the patient's experience of the analyst, which is conceived by the conservative critic to be distinctly nonsocial, nonrational, and transferential. In this sense, conservative critics continue to employ the blank-screen model and the associated asocial paradigm: "According to these conceptions, there is a stream of experience going on in the patient which is divorced to a significant extent from the immediate impact of the therapist's personal presence" (Hoffman, 1983, pp. 390-391). That is, some part of the patient's experience is continuously fantasy-based, an ongoing eruption of the inner life, perhaps catalyzed by current external events, but only indirectly related to them. Transference is routinely viewed as distortion.

The radical critic, on the other hand, who may be identified with interpersonal psychoanalysis, the object relations school, self psychology, or even classical Freudian psychoanalysis, espouses a truly social conception of the analytic situation. The therapist is inevitably involved in the interaction, usually in ways he does not suspect. Hence there is no single truth, and transference and countertransference are unformulated experience, to be articulated by the selection of an interpretive perspective. The analyst is in no position to claim that transference is distortion. Rather, writes Hoffman, "the perspective that the patient brings to bear in interpreting the therapist's inner attitudes is regarded as one among many perspectives that are relevant, each of which highlights different facets of the analyst's involvement" (p. 394). The patient may actually be able to observe aspects of the analyst's involvement which the analyst has not yet seen. "More than challenging the blank screen fallacy," writes Hoffman, "the radical critic challenges what might be termed the naive patient fallacyt, the notion that the patient, insofar as he is rational, takes the analyst's behavior at face value even while his own is continually scrutinized for the most subtle indications of unspoken or unconscious meanings" (p. 395).

It is only by learning about himself that the radical critic believes the analyst can learn about the patient. Only in the act of disembedding himself from previously unseen aspects of the interaction is he able to experience—and perhaps say—something new. "At the very moment that he interprets, the analyst often extricates himself as much as he extricates the patient from transference-countertransference enactment" (p. 415). The radical critic assumes he is embedded in the interaction, that in certain important respects he is always blind to it. He continually queries himself about his participation, trying to create the possibility of a question, a foot in the door.

Issues of Knowing and Understanding

In the United States, we are in the midst of heightened psychoanalytic interest in the nature of knowing and understanding, led by the increasingly frequent citation of hermeneutic views.¹ It is no longer unusual for analysts to find clinical relevance in questions that only recently would have been deemed entirely metaphysical: What does it mean to know something about the patient? About one's experience of the patient? What is the nature of the material we are trying to understand? What is an interpretation? What qualifies as truth, and how can we recognize it? Who determines when we have reached it?

These new interests are natural and even necessary for radical critics, because as soon as one defines the basic unit as the interpersonal field and not the individual, one must replace the comforting assumption that the truth is already there, buried, awaiting discovery. The archaeological metaphor has outlived its usefulness. Like the blank screen, it is an image of stasis unsuited to represent the inevitable participation of both parties. For the radical critic, the "truth" about the patient is a mutual construction, an outcome of the interaction, always interpretive and only partially predetermined. And thus the definition of what it is for something to be unconscious is in a state of flux. For many analysts, lack of awareness now means the absence of explicit reflection, not the inaccessibility of a hidden reality. Besides unformulated experience, recent psychoanalytic concepts reflecting these trends are Stolorow's "prereflective unconscious" (Atwood and Stolorow, 1984; Stolorow, Brandchaft, and Atwood, 1987; Stolorow, 1988) and Bollas's (1987) "unthought known."

The Asymmetry of the Analytic Relationship

The acceptance that the process of constructing truth is mutual does not mean that the analytic relationship should be conceptualized as symmetrical. Egalitarian, yes; symmetrical, no.

To make the decision to enter treatment means to accept that a certain set of differential roles is to be assigned to oneself and the analyst. Basically, these roles are interpretive perspectives. The patient's role is to say everything about his experience that he can, including his experience of the analyst. The analyst's x ole, though, is hardly limited to an appreciation of what the patient says—a stance that would amount to an asymmetry of a different sort. As a matter of fact, many or most patients would be disappointed if that is what the analyst did. It is the analyst's responsibility not only to develop an empathic appreciation of the patient's perspective, but to have a perspective of his own. If the analyst did not work that way, little would happen, because there would seldom be any reason for the patient to think in new ways about himself. It might seem that the empathic stance of the self psychologist or the containing function of the object relations analyst contradict this formulation, but they do not. Those analysts, too, would agree that their analytic functioning requires them to have an independent perspective about what they are doing with their patients. The reason they treat their patients as they do is hardly that they accept what patients say at face value, but precisely that they do not. Like all analysts, they have their own ideas about what is called for, ideas that their patients might very well disagree with in the most strenuous way if they knew about them, and if they understood what these prescriptions meant about the way their analysts privately characterized them.

For most analysts, though, we might say that analytic work is defined by a vacillation between the roles of empathizer and interpreter. In Racker's (1968) roughly equivalent terms, the analyst alternates between observing and acting on a modulated grasp of his own concordant and complementary identifications with the patient. This is what it is to be an analyst, and most patients who enter treatment thereby endorse their analysts to occupy that role. The patient is purchasing the analyst's specifically analytic services, and they include a set of interpretive perspectives that the patient appropriately considers whenever the analyst deems them useful to add to the conversation. This is a freely accepted asymmetry.

But there are also many cases in which the patient enters treatment without making such an endorsement, explicitly or implicitly. These people, who are more difficult for the analyst to deal with and are often characterized as borderline, narcissistic, and so on, would have reactions to knowing the analyst's formulations ranging from disorganizing disappointment and depression through utter rage. The analyst, under these circumstances, has no choice, of course, but to keep many or most of his own interpretive perspectives to himself, at least for the time being. But does that mean he does not develop them? Absolutely not. As a matter of fact, we are perfectly well aware that if he did not develop them, the treatment would blow up immediately (and it may, anyway, of course), because the only way the analyst imagines how he can effectively treat the patient is by reference to these private formulations. As I just mentioned, this is the situation of the self psychologist and of the analyst who works by containing the patient's projections; but it is also the situation of any analyst working with a patient who cannot tolerate a fully collaborative analytic relationship.

Where is the egalitarianism in this? Does this situation not sound a little like the relationship between an authority and a subject? The reason it is not is really very simple: besides the fact that the patient always has the power to terminate the relationship, of course, he also always has the right—even the responsibility, if he disagrees—to dispute what he knows of the analyst's understandings. In the work of the radical critic, what the patient has to say about himself or the analyst, or the two of them together, may be quite valuable; this depends on how convincing it is to the analyst, just as what the analyst says depends for its ultimate therapeutic value on the conviction it arouses in the patient (see chapter 9). The necessity for each participant to convince the other is the egalitarianism of the analytic relationship, and it exists despite the fact that the relationship is asymmetrical. In any particular case, the patient may have a more useful or coherent perspective on himself (or on the analyst) than the analyst does. Asymmetry does not imply the analyst's infallibility, after all, only the ongoing attempt of two people to occupy certain more or less cooperative and complementary stances relative to one another's participation. (For a view closely related to this one, see Aron, 1996.)

With all that said, I must return to the point that the analyst is always embedded in the analytic situation in such a way that the ideal distribution of roles, the ideal therapeutic collaboration, is utterly impossible, even for the analyst, almost all the time. But the ideal is there, and we try to meet it as best we can. Thus, in the following discussion of Gadamer's hermeneutics, wherein Gadamer implicitly portrays the conversation in which understanding takes place as not only egalitarian but symmetrical, the analytic reader must remember that Gadamer was a philosopher and not a psychoanalyst. He did not grapple with conversations in which, from the outset, it is recognized that one person has special reasons not to understand (at least we hope that the analyst's reasons not to know are usually less compelling than the patient's). The patient's special reasons are one of the factors that make the psychoanalytic conversation unique in all of life, and that build asymmetry into the analytic relationship's very foundation.

Gadamer: A Philosopher Relevant to Clinical Practice

For Hans-Georg Gadamer, as for most modern hermeneutic philosophers, understanding (in the general sense, now, not the psychoanalytic one) is inevitably a matter of selecting and formulating one interpretation from the multiple plausible alternatives in any given situation. Reflective meaning cannot be said to have existed prior to its revelation in language, at least not in an apprehensible form. All of this, of course, is entirely consistent with the conception I have presented as the articulation of unformulated experience in language.

Gadamer claims that understanding requires us to disembed ourselves from our preconceptions—from what he calls "prejudice" or "prejudgment." New understanding emerges when we are able to make visible the expectations that have such a large role in the formulation of experience, and this can occur only in dialogue with the other. "In the last analysis," writes Gadamer (1962), echoing the radical critic, "all understanding is self understanding" (p. 55).

But the unique emphasis that makes Gadamer intriguing and useful to the embedded analyst is his view that prejudice is not only the bane of new experience, but the source of it. Gadamer looks to prejudice just as the radical critic looks to countertransference. It is not merely error. It is not noise in the system. Rather, it is both what makes possible new understanding and what stands in its way. Without prejudices we would be free to formulate anything at all and unable to make a single thought. Thus Gadamer (1975) can say, "Every experience worthy of the name runs counter to our expectation. . . . Insight is more than the knowledge of this or that situation. It always involves an escape from something that had deceived us and held us captive" (pp. 319-320). But he can state with equal conviction, "Prejudices are biases of our openness to the world. They are simply conditions whereby we experience something—whereby what we encounter says something to us" (1966, p. 8). Gadamer's philosophy is an exegesis of this paradox.

Reading Gadamer stimulates down-to-earth questions about clinical practice and clinical thinking. Because the aim of this chapter is to draw connections between Gadamer's thought, unformulated experience, and clinical psychoanalysis, much of what Gadamer has to say has been left out. His ideas are not discussed for their own sake. At the same time, though, a brief exposition of Gadamer's overall perspective and its historical context is unavoidable. The exposition follows in two sections, "From Schleiermacher to Gadamer" and "The Hermeneutic Circle," and to a lesser extent in a third, "Sharing a Tradition." The remainder of the chapter is focused on the implications for psychoanalysis practiced in the social paradigm. Chapter 12 takes up these views in a specifically clinical context.

From Schleiermacher to Gadamer: Rejecting Empathic Knowing in Favor of Mutual Influence

Hermeneutics is the study of the process of understanding. Prior to the work of Schleiermacher, the father of modern hermeneutics (he wrote in the beginning and middle of the 19th century), it had been assumed that ease of understanding was the norm. The meaning of the text was treated as if it should be self-evident to the competent reader. Understanding could be taken for granted; misunderstanding was the exceptional event that required explanation. Hermeneutics at that early time was really just the study of the practical problems involved in interpreting different kinds of texts. Schleiermacher, however, impressed with how difficult it is to grasp what an author really intended, reversed the original formula and originated the modern view, which still holds sway: Misunderstanding is the natural state of affairs; it is successful understanding that requires explanation. In this way, hermeneutics metamorphosed from the clarification of isolated difficulties in grasping meaning—the correcting of errors in comprehension—to the much more general study of what it is to understand (Palmer, 1969).

It is more reasonable to assume misunderstanding than understanding, said Schleiermacher, because intervening between the text and the reader are time, history, custom, changes of language, and so on. The reader is faced with a document encrusted and obscured by the debris of change. Thus, in the Romantic tradition from which Freud also eventually emerged, Schleiermacher came to believe that the truth lay behind one's reading of the text, and had to be reconstructed. One needed to put together in one's own experience the historical context in which the text was written. If that could be accomplished, one could read the text from within the vantage point of the writer, and it would then be possible to grasp its original meaning. Palmer (1969) describes the process: "For Schleiermacher, understanding ... is the reexperiencing of the mental processes of the text's author. It is the reverse of composition, for it starts with the fixed and finished expression and goes back to the mental life from which it arose" (p. 86). To the extent that the empathic reconstruction was accurate, the recreated meaning was accurate. There was a single truth in the text, and it preexisted the reader's encounter with it. The influence of the reader's present-day involvement in the world, to the degree that it interfered with a direct encounter of the writer's original intention, was a source of error. This is the most familiar model of bias, the traditional understanding of error forged in the Enlightenment: preconceptions veil and distort the truth; truth appears when error is removed. Thus, for Schleiermacher, hermeneutics was a set of procedures used to eliminate distortions. Though there was an undeniable role for intuition, the principles of understanding could be taught (Palmer, 1969).

Gadamer claims just the opposite, and in so doing he grounds his work in the phenomenological description of understanding, specifically eschewing any attempt to influence its practice. Because of the very nature of understanding, says Gadamer, it cannot be taught. It happens, it is unbidden. Understanding is not reconstruction to Gadamer, but mediation between reader and text, or between two participants in a conversation. Understanding takes place in dialogue: it requires the involvement of two parties, it goes on in the present moment, and it depends on present attitudes, not on recreating the past—all of which are consistent with the social paradigm.

We cannot understand, according to Gadamer, without reference to the tradition in which the understanding is meaningful; but that tradition is nowhere spelled out fully. Tradition exists in the innumerable unarticulated prejudices with which we approach the world. Tradition is actually the sum of all prejudice. Each person is an embodiment of tradition, a living expression of the history of cultures. And thus, to refer to prejudice is to refer to the basis of human existence. Gadamer writes, "It is not so much our judgments as it is our prejudgments that constitute our being."² We are at one and the same time, then, our own tool for understanding and the very substance of what must be understood. Since we embody what we wish to know, we cannot very well perform a directed search for truth. One does not know what to look for; one cannot see one's own eye. We are necessarily embedded in tradition because it represents the only possibilities we have.

It is language, and language only, that defines reflective meaning, and the limits of possible reflective meanings are precisely the same as the limits of language. If a meaning cannot yet be spoken, it does not yet exist in a form that could be reflectively understood; if it is not within the capacity of language to represent, it can never be reflective meaning at all. The sum total of all those meanings that are capable of linguistic representation is a second way Gadamer defines culture, or tradition. That is, the possibilities of language, the sum of prejudice, and tradition are all ways of referring to the same thing.

What we understand is not separate from us. The truth cannot be located and uncovered. Rather, says Gadamer, we must recognize that we already are the truth, that the problem is formulating the relevant aspects of our being. All we can do is to make a dialogue with another person in which we engage our prejudices in such a way that they become clear to us, and understanding emerges. The participants in what Gadamer calls a "genuine conversation" try to do just this by sensing the constraints that keep their dialogue from opening further. Thus the one who wishes to understand cannot set out knowing he will accomplish the task, as one could in making the decision to take a walk or watch television. Rather, one is attuned, as prepared as possible to see and sacrifice prejudices in the interest of receiving the other's message.

In all these ways, Gadamer rejects Schleiermacher's belief that the meaning of the text is hidden and preexisting, available only through empathic communion with the writer. Reflective meaning does not exist prior to the interaction in which it occurs. It is an event, not a thing. Because of our embeddedness in tradition and our consequent reliance on prejudice, we cannot, says Gadamer, depend on being able to produce an empathic communion just because such a connection is desired.

This disagreement sounds remarkably similar to the issue regarding empathic knowing which separates interpersonalists and some object-relations writers from self psychologists (e.g., Bromberg, 1989; Stern, 1988, 1994). The analyst who believes he has no choice but to work from a position of embeddedness cannot trust his own intention to enter the patient's inner world, to learn through empathic immersion. Neither the patient's experience nor his own is transparent to his scrutiny.

There is another connection worthy of note. For a psychoanalyst, Schleiermacher's position that truth lies underneath or behind the text immediately brings to mind the concepts of manifest and latent content. Schleiermacher's characterization of the reader trying to see beyond the appearance to the writer's true intention is like the classical analyst trying to penetrate defensive obscurities and distortions to reveal the patient's true motivations beneath. These are varieties of the "hidden reality" view (Neisser, 1967). On the other hand, Gadamer's characterization of meaning as an articulation and selection of possibilities dovetails nicely with the radical critic's rejection of transference and countertransference as distortion and the affirmation of these phenomena as legitimate, albeit individual and analyzable, selections of possibilities.

Thus the field of hermeneutics, like psychoanalysis, seems to be moving from an older conception of meaning as a static phenomenon, a thing that preexists its interpretation, to a newer understanding that meaning is an activity, an event, and can only take place in interaction. Some of the differences between Schleiermacher and Gadamer resurface in the contemporary dispute between Gadamer and Habermas, though in more subtle form. I will return to this famous debate later in the chapter.

The Hermeneutic Circle

The concept of the hermeneutic circle is the framework of hermeneutics. When modern hermeneutic writers differ, their disagreements center on its proper interpretation.

The idea arises from a paradox: We can understand only those communications that we can locate in their proper contexts. Unless an utterance can be placed in the appropriate configuration of tradition, its meaning remains obscure. Imagine trying to understand certain passages in Freud without knowing that he was taking issue with Jung; or trying to understand a dream without knowing the events of the day preceding it, or the patient's associations to it; or trying to understand a patient's barely supportable characterization of a coworker as greedy without knowing that the patient himself is terribly afraid of being greedy.

But here is the rub: We cannot select the appropriate context without first having grasped what we claim we are trying to understand. How can one know, before understanding the dream, how to select the relevant events from the myriad possibilities of the previous day? How can one settle on the significant aspects of the patient's character without already having decided what the transference to the co-worker means? Palmer (1969) puts it this way: "How can a text be understood when the condition for its understanding is already to have understood what it is about?" (p. 25).

The hermeneutic answer is that we comprehend by means of continuously projecting complete understandings into communications from the other, communications we actually understand only partially. We extrapolate complete understandings on the basis of these partial understandings we already have. And where do the partial understandings come from? They are in turn stimulated by the complete understandings we project. Thus is the circle closed. But of course, something else must occur, or else comprehension would be nothing but self-reference. All comprehension is a process of projecting partial understandings into fully rounded ones, and then modifying these projections on the basis of what we actually come into contact with in conversation with the other person. In other words, when we understand, we have been able to treat our projections like hypotheses; and when we do not understand, it is because we have not managed to adopt this degree of uncertainty. This is the hermeneutical circle, and it is clear what the problem for the hermeneutic thinker is: One has to take a position about how we avoid seeing nothing more than we expect to see. Under what circumstances can projections be hypotheses and not givens?

At least this is the primary problem for Gadamer, because of the role of tradition in his conception of understanding. Whenever we explicitly understand something, what has happened is that we have made into expressive language some unexamined aspect of tradition. Productive prejudices are those that can be illuminated in such a way that new understandings come into view. They are capable of being disconfirmed.

Sharing a Tradition: Cultural Differences and the Special Case of Psychoanalysis

What is understood gains its meaning as much from what remains unarticulated as it does from the words themselves. Each piece of understanding is surrounded and given meaning by "the infinity of the unsaid." Even around each word, giving context and relation, is a "circle of the unexpressed."³ Over and over again, Gadamer emphasizes that understanding is as much a process of "concealment" as "revealment." In bringing one possibility into the light, experience is organized in such a way that other possibilities are "darkened." But those possibilities thus kept from the light are still significant in the making of meaning. To understand is not only to grasp what is said, but also to be so much a part of the tradition of the speaker that one also grasps the relation of what is said to all that is not said.

For example, consider a Chinese teacher of American literature, living in a rural area of China, who could talk with what seemed a respectable degree of familiarity about certain modern American novelists, but who could so little conceive of the tradition within which these writers work that, when the topic of conversation somehow turned to New York (he had heard of it, but knew no more than that), he could not understand how Manhattan skyscrapers could be taller than the single two-story structure in his own village, it was not so much that he refused to believe this fact, but rather that there was no way for him to make sense of it. It just could not be anchored in his own tradition. To paraphrase Gadamer, there were no conditions in this man's world whereby this encounter could speak to him. Needless to say, the friend who related this story, when she finally grasped that her interlocutor's attitude was due to bafflement and not to obstinacy, immediately reassessed the preceding conversation about novels, which she suddenly saw had been more a matter of verbal mimicry than real substance. She had unknowingly supplied the relevant context for whatever he had said.

Thus the process of understanding can go on only between people who share a tradition—or who are at least aware that they do not share one. Some years ago I offered to help a Japanese man with his English. Finding that we got along well, we decided to combine work and play by having lunch together once a week and going over cartoons in the New Yorker. I taught him vocabulary by naming the pictured objects, and I introduced grammatical points and matters of convention by explaining the thoughts the cartoons conveyed. But no matter how careful my explanations, most of the time my friend didn't get the jokes. I would fish around, trying to figure out what assumptions I brought to the situation that he did not. Trying to imagine how he saw, or didn't see, the most obvious parts of my world (how do you explain why it is funny for a man to bite a dog to someone for whom the very idea of a pet is uproarious?) led to hilariously concrete explanations of things I had never even thought to say. Eventually, at some moment that inevitably surprised me, I would learn that we had reached some meeting of the minds, because he would burst into laughter, more often than not right in the middle of my sober explanation of slippers or some such thing. He was frequently no more able than I to make sense out of why his understanding had come together in that particular moment.

Since analyst and patient generally share more or less the same set of cultural prejudices, the situation between them is often less baffling than it was between my Japanese friend and myself—often, but not always. My friend and I had the advantage of knowing how unfamiliar we were with each other's prejudices. We looked for misunderstanding in every sentence that passed between us. In psychoanalysis, on the other hand, even though the participants may share a culture, the relationship can seldom be as explicitly based on the mutual recognition of each other's prejudices. The most dedicated investigators of transference and countertransference are nevertheless inevitably involved in many conventional and unquestioned (and therefore masked) aspects of relatedness. They do not know what to look for. And even when something alerts them to the possibility of specific questions, the task of making visible the interweaving of the very tailored prejudices two people develop between them is more difficult than recognizing cultural differences. The clash of cultures, while it may be hard to understand, is at least unmistakable.

Genuine Conversation and the Creation of the Analytic Field

Gadamer differs from his predecessors in rejecting the possibility of prescribing a method of understanding. There is no way to describe how being comes into language—which is to say, there is no way to describe how one ought to approach the world to gain the truth. The concept of an exclusive method—a set of rules, such as the scientific method, prescribing how knowledge is to be determined—is legitimate only so long as what is learned is separate from the one who learns it. If the truth is separate, the question of the best way to approach it is a logical one. But if truth is the outcome of the grasp of being in language, there cannot be a superior way to accomplish it, because one cannot escape being part of the event one wants to capture. One method cannot be recommended over others. As a matter of fact, according to Gadamer (1962) the conscious application of a method can never result in revelation. He writes, "The real event of understanding goes beyond what we can bring to the understanding of the other person's words through methodical effort and critical self-control. Indeed, it goes far beyond what we ourselves can become aware of. Through every dialogue something different comes to be" (p. 58). Method to Gadamer is reification, the concretization of unexamined assumptions. Belief in method subverts the discovery of truth. Method should be questioned routinely, its assumptions understood and evaluated, never merely followed. And thus Gadamer's approach does not have specific technical implications for psychoanalytic practice. One could even make the case, on the basis of these ideas, that the concept of technique in psychoanalysis is specious.

Technique is not specious, however; we will always need it. Our recognition that it is a social construction, not an essence, hardly means we should wash our hands of it. What is specious (and here Gadamer would undoubtedly agree) is the claim that one kind of technique is the true one, or that the analyst's participation is in error if it is not organized by technique. Hoffman (1994) suggests that whatever technique we use structures our participation, not by leading us in the paths of righteousness, but by setting up a dialectic between technique and spontaneous personal participation. Without technique, psychoanalysis would be no different than any other conversation; and without the analyst's spontaneous participation, psychoanalysis would be emotionally arid. Each of these kinds of participation on the part of the analyst can exist, however, only against the background of the other; technique and personal participation are both necessary if either is to be meaningful. The patient only cares about our spontaneity because of our usual demonstration of a commitment to a disciplined way of working; and our way of working is meaningful only because the patient knows this technique often puts a rein on our spontaneity.

As one interacts with a text or another person, the hermeneutic circle goes around and around. First, expectations develop, based on what one has brought to the encounter and on the history of the encounter itself. Each time (moment to moment) an understanding is called for, one projects into the future a preconception of what the other person or the text will disclose. This projection is based on the assumption that the other person or the text is internally consistent, and that the past is therefore a relevant basis on which to construct a possible future. One completes the partial meanings one has gleaned. To the extent that the expectations we form on this basis are explicitly known, one can see whether or not the situation with which one is faced confirms or disconfirms them. On the other hand, to the extent that one is not aware of one's expectations, one does not learn. "Understanding," says Gadamer, "is primarily agreement," by which he intends to say that to understand is to see what the other means. Yet this is an insufficient expression. Here is Warnke (1987), who has contributed a book-length study of Gadamer.

Genuine conversation sounds like psychoanalysis. But doesn't the patient consult the analyst precisely because he cannot converse, cannot disconfirm his expectations, because he is caught in the endless stalemate that is transference? Well, yes, but this brings up something very interesting, an elusive point that happens to be especially clear from Gadamer's perspective. In treatment, the construction of transference and countertransference, even during the time when these interactions are still being blindly enacted by patient and analyst, is at least as significant as their eventual understanding in words. The relationship must be made; it must come into being. The participants in an analysis must develop a shared "tradition" to which to refer. Transference and countertransference need to feel so familiar to the participants, even if invisible and vaguely unwelcome, that when the right words emerge, they are recognized. It is only the familiarity of the relatedness, the way it fits and feels, that eventually makes it possible, in Gadamer's words, for something in the rush of events to "stand," to "come to a halt."

Psychoanalysis, no less than any other discipline, is limited by the culture at large in the possibilities its participants can envision, and in that sense tradition means the same thing between analyst and patient as it does between reader and text. The tradition that matters most in psychoanalysis, however, is that miniscule (relatively speaking) subset of invisible prejudices that analyst and patient create between them, the interpersonal field, what we might call the "being" of the analytic situation. It takes time (unless something dramatic and unfortunate occurs, usually only in the treatment of very disturbed patients) for analyst and patient to influence each other deeply and subtly enough to "have a relationship," to be in each other's presence in such a characteristic way that something can be said about it. The field must be created before it can be understood; it must be put together before it can be taken apart. Out of all the unformulated possibilities, two people come to a way of being together in which certain behaviors and experiences are allowed and others are prohibited, and in which the degree of recognition of, or enforced blindness to, various aspects of experience is prescribed. They make a history, complete with prejudices, and then they try to disconfirm them. And thus, Gadamer's scheme highlights the way in which, in psychoanalysis, prejudices are both one's blinders and the ground against which one is eventually able to discover a figure. This point has been recognized by those analysts who have seen that the analyst must regularly be able to play (and of course that means genuinely play, i.e., without awareness) a reciprocal role to the patient's transference—at least for a time—if the treatment is to be alive and helpful (e.g., Feiner, 1982; Jacobs, 1991; Levenson, 1972; Sandler, 1976; Racker, 1968; Searles, 1979; Tauber and Green, 1959).

Because it takes time to establish the pattern of what will be seen and what must not be seen—the distribution of light and darkness in the relationship—the analyst may be clearer about certain aspects of the patient's character in the first few sessions than he will be again until much later in the treatment. (This observation, the accuracy of which is unmistakable as soon as it is stated, was a personal communication from Earl Witenberg, reported by Bromberg, 1982.) At the beginning, analyst and patient have yet to establish their characteristic relatedness, and there is little reason for the analyst not to use all his observational powers. As time passes, however, he is caught in the grip of the interpersonal field constructed by the patient and himself, and it is only when the relevant issues can be analyzed that the analyst is again free to see. Of course, from the point of view of the radical critic, there is every reason to suspect that the same thing goes in the reverse direction as well. The patient, too, is probably freer to see the analyst at the beginning than when the pattern of prejudice is set in place.

If indeed we create a shared tradition and then discover it, coming to see the other person clearly in psychoanalysis is a more complex matter than understanding a text—more complex, too, than what Gadamer means by understanding the other in a conversation. The added complexity arises from the fact that in psychoanalysis, to see what the other means requires that one unveil the other's preconceptions of oneself, the preconceptions that interlock with one's own and contribute to maintaining one's blindness. As far as I know, Gadamer does not address this intriguing problem, which in fact might be expected to arise between any two people, not just participants in the analytic situation.

Open Questions, Commitment, and Countertransference Involvement

Prejudice is not necessarily inaccurate, only invisible. The questioning of prejudice may lead to disconfirmation, or it may lead to the conclusion that one's implicit assumption is in fact worthy of being consciously adopted. But prior to the decision regarding what to do with prejudice, there is that moment when one is suddenly capable of seeing it—and an alternative to it. It becomes possible to reflect on the prejudice one had been buried in until that moment, and one also sees the other's perspective. A comparison is available. The creation of such a comparison is what Gadamer calls finding an "open question." In an open question, a clear choice between alternatives appears, and answering the question therefore means learning something. Here is a brief illustration.

Recently, I wondered to myself whether the suddenness of a patient's decision to leave her job might have to do with a recent conflict with a supervisor (which she had discussed with me the week before), or whether, as in the patient's presentation of the issue, the new job itself was the sole motivating factor. I wondered whether the patient might have had more second thoughts if the conflict had not occurred. The new job paid better, but the old job carried more prestige. Notably, in discussing her decision with me, the patient did not mention the conflict with the supervisor at all. I brought up the question with the patient and, at least partly because my uncertainty about the answer was real, the patient accepted the question and was able to be curious about it. (She found it especially interesting that she had not even considered the problem with the supervisor.) In this case I formulated an open question in my own mind, and in conveying it to the patient, helped formulate the same open question in hers.

The questioner must mean the question if it is to be open; that is, the question may actually be an interpretation. For a question to be open, the one who asks must believe that both alternatives are possible, and this can only happen when the analytic work is going on right at the edge of what the participants know. The uncertainty must be genuine. So, for example, it would not have been an open question for me merely to have asked the patient whether there was anything more to the decision than met the eye. This question does not derive from genuine uncertainty, because it does not refer to a specific contrast between alternatives, only to the possibility of one. Genuine uncertainty for Gadamer requires particularity; one must know precisely what one is uncertain about.

Most open questions are not made explicit by questioners, not even to themselves. My example of the question asked by the analyst about the job change is cumbersome by comparison to the kinds of fleeting, subtle open questions that are more the rule. The kind of experience Gadamer refers to as an open question, as a matter of fact, is seldom actually spoken or even reflected upon. Gadamer means to describe a ubiquitous kind of experience, which is a precursor of every thought that qualifies as new. And here the process has been traced as far back toward its origins as it can be. This is the mystery at the heart of knowing. It cannot be said where open questions come from. And if we do notice them, they appear—we do not have the sensation of creating them.

Sometimes open questions are visible in hindsight. For instance, there was a particular moment I recall in the second or third hour I spent with a new patient when it occurred to me for the first time that this person spent very little time with friends and seemed not to care. How did this thought arise? The patient was referring to some upsetting news she had received the prior evening. She had been alone before, during, and after the event. I imagined her feeling lonely, but then I realized she was not conveying this feeling. Alternatives became available: perhaps she had the feeling, but was keeping it to herself; perhaps she was capable of the feeling, but was avoiding it (would it have made her too vulnerable in front of me, someone she did not yet really know?); or perhaps she characteristically did not experience loneliness. My expectation that she would have been lonely became visible to me at the same moment that I saw that she had not been. It took inquiry, of course, to establish which of the possibilities was most accurate, but the essential observation had emerged. The open question that preceded it can be reconstructed easily enough in this case. It would go something like, "Was the absence of companions painful to her, or did she feel no need for them?"

In what conditions do open questions arise? What can be done to encourage the perception of alternatives? (This, of course, is the radical critic's primary problem.) Gadamer says that any understanding requires commitment. "If there is any practical consequence of the present investigation," he says in the introduction to Truth and Method (1975)

Understanding requires commitment to a preconception, which means the willingness to give oneself over to a way of seeing. Before a prejudice can be known, before one can even know one has it—in fact, before it could even be said that a particular prejudice exists—one must be committed to a way of understanding. And this commitment must be deep enough to be implicit (in Polanyi's, 1958, similar conception, tacit), so that in coming into awareness in language, it broadens one's horizon. What this means in psychoanalytic terms is that the analyst must be involved with the patient, willing to feel toward the patient as anyone else would, though of course not only as anyone else would (cf. Hoffman, 1994). It has long been taken for granted that the patient's emotional involvement in the analysis is crucial. But the shared tradition needed by the patient and analyst cannot be created without this same commitment from both participants. The analyst must be unafraid to commit himself to a wholehearted experience of the countertransference. This point is hardly novel, but Gadamer's perspective offers us a new route of access to it.

The Fusion of Horizons: Any Understanding Reflects a Change in the Field

Gadamer calls that moment in which a preconception and its alternative can be differentiated a "fusion of horizons." "Horizon," writes Weinsheimer (1985), paraphrasing Gadamer, "is another way of describing context. It includes everything of which one is not immediately aware and of which one must in fact remain unaware if there is to be a focus of attention; but one's horizon is also the context in terms of which the object of attention is understood" (p. 157). When horizons fuse, it is because the visibility of one's own preconceptions makes it possible to grasp the context in which the other's meaning exists. In seeing one's own expectations one falls into the other's. One simply finds oneself there, and a new experience is suddenly available; one has stepped into the configuration of prejudices the other occupies, the same distribution of light and darkness, the same relation of articulated possibility to unformulated being. This event is often described in psychoanalysis as empathy, though it is clear from Gadamer's stress on the analysis of preconceptions that he could not take such a view. At the same time, however, one can recognize in the fusion of horizons Gadamer's continuity with Schleiermacher. Although he may have rejected empathic knowing and reconstruction in favor of mediation, Gadamer must tackle the same problem. Somehow he must explain how one senses the context in which the other's statement is meaningful. The horizons fused are those of the person with the prejudice and the person who has provoked the possibility of alternative—for the present purpose, the analyst and the patient. And of course it happens in both directions. That is, analyst and patient at different times are each in the position of grasping preconceptions, as well as providing the conditions that makes the other's grasp possible. When the analysis is moving well, each is receptive to questions from the other, just as a reader, to understand, must not only put questions to the text, but must also allow himself to be questioned by the text. Gadamer's understanding of the hermeneutic task might double as a definition of the best moments of most analyses, at least those conducted by analysts for whom the equality of Gadamer's conversants does not represent a violation of the therapeutic situation. Prior to the following passage, Gadamer (1966) has just complained about the "leveling tendencies" of our industrial age, which influence language to become nothing more than a "technical sign-system."

Thus Gadamer implies that the process of understanding requires a successful interpersonal integration. Nothing can be grasped that the relationship between the parties does not allow. Here is another (1975) description of that integration. Once again, read it as if it applies to psychoanalysis.

Of course, this kind of relatedness, while it may occur with some frequency in the interaction between a reader and a text, is the ideal in psychoanalysis, one analysts are always glad to achieve but do not routinely expect. But in addition to drawing attention to this fact, these passages carry a significant implication for psychoanalysis: Any nezu understanding is a crystallization of some aspect of the interpersonal field. That is, because understanding requires a genuine exchange, and genuine exchange is precisely what is missing when patient and analyst are caught in the grip of the field, it follows that whenever some new thought is available to either patient or analyst, a shift in the relationship must have occurred to allow it. This is true whether or not the new thought is manifestly about the relationship. Any new thought on the part of either analyst or patient is an occasion for the analysis of transference and countertransference, because its very existence means that the participants have stumbled into some kind of freedom between them that was not there before. The successful pursuit of psychoanalytic inquiry, no matter what the content under discussion, is the same thing as expanding the possibilities of the therapeutic relationship.

Along with the radical critic, one must conclude, on the basis of Gadamer's view, that no analytic examination of the patient's history, of a current relationship, or of anything else can occur outside the influence of the field. Here the analyst finds himself on familiar ground. There can be no such solitary entity as "the patient's past," for instance. It is not possible to tease out a person's history in the abstract, "uncontaminated" by the context in which it becomes relevant. And neither can the interpersonal field be defined in a way that separates it from anything that goes on in it. Every relationship shapes the content that occurs in the conversations of its participants. For "his past," one must substitute "his-past-in-our-present." But does this mean that there is no reality outside ourselves to which we can refer. Does anything go?

Definitely not, no more than in the work of Sullivan or any other theorist whose work is inherently social, because Gadamer's criterion of truth is consensual validation. Valid meaning is a construction accepted by the community of interpreters, and it emerges in language—once again reminiscent of Sullivan. A meaning is not valid, that is, unless it can be understood by someone else. For Gadamer, writes Linge (1976), as for the radical critic, "interpersonal communication . . . [is] the locus for the real determination of meaning" (p. xxxiii).

It is probably even more important to emphasize that, for Gadamer, this consensual validation is rooted in its capacity to represent reality. The world cannot simply be understood in any undisciplined way we feel like, because reality does exist. It is really there for Gadamer, though our routes of access to it are inevitably shaped and selected by history and culture. But as I have tried to say already, in Gadamer's thought, unlike much of postmodernism, our situatedness in history and culture is the key to the possibilities for understanding. Our embeddedness in our time and place is, if we use it creatively, our means of revelation, not merely a barrier to our knowing. It is our participation in the life of our cultures—whether in contentment or dissent—that underwrites our pursuit of the truth. Sass (1988) writes, in reference to Gadamer's perspective, "For, though it may be impossible to discover a single meaning, this does not mean that anything goes, that listeners can legitimately ascribe any meaning to any discourse. The hermeneutic approach is a sort of 'middle way' between objectivism and relativism" (p. 254).

The Priority of Prejudice: Gadamer and Habermas

The hermeneuticists most cited in psychoanalytic literature are Ricoeur and Habermas, not Gadamer. And for good reason. Ricoeur (1970, 1977) published a book-length study of Freud's work and an influential essay, which appeared in a psychoanalytic journal, on the question of proof in Freud's writings. Habermas (1971, chapters 10-12) makes clinical psychoanalysis his model for what hermeneutics should be. For Gadamer, psychoanalysis is merely one variety of hermeneutic inquiry. Why, then, should the radical critic turn to Gadamer?

Answering this question requires a review of the debate between Gadamer and Habermas.⁴ In Habermas's ideological analysis, Gadamer's acceptance of the value of tradition encourages a kind of conservatism. Habermas is concerned that when tradition is structured according to ideology oppressive to humankind, Gadamer's approach would never expose it, and would in fact capitulate to it, treating it as inescapable. Oppressive ideology, in other words, because for Gadamer it would be part of tradition or prejudice, would never be noted to be oppressive, but would be accepted—perhaps even honored—as part of the ground against which the figure of new understanding appears.

This problem with Gadamer's thought, says Habermas, is the result of Gadamer's belief that hermeneutics—that is, the act of interpretation—is primary, that it precludes preexisting objectivity. It is the result of Gadamer's position that we are always caught in tradition, that we are the product of our prejudices and our attempts to illuminate them. But Habermas argues that, if we agree on the primary and prehermeneutical value of emancipation, we do have a basis for objectivity. For if we can identify ways in which tradition acts to oppress us, we can change the conditions of our living. Therefore, objectivity comes before hermeneutics; an insistence on liberation, the objective "good," should shape our attempts to interpret and understand. Social science, to Habermas, is defined by its commitment to discover the invisible sources of oppression in the traditions we inherit. Social science is therefore based in objectivity, and psychoanalysis is the prime example. In this way, hermeneutics loses its primary status, and becomes instead the study of how the social sciences—which now become the primary discipline—make visible the restrictions on freedom and equality that are embedded in our history.

Gadamer's reply to this argument is that we cannot legitimately exempt anything from examination, including our prejudices about freedom and equality. Why should we believe that such an exemption is not itself a prejudice in need of examination? Is Habermas not claiming that by believing in the absolute value of emancipation, and in our absolute ability to perceive it and to know how to accomplish it, we can transcend history and simply decide not to be involved in the event of understanding? Can we believe that it is possible to step outside history in this way? Can we ever conduct understanding without participating in it? The best way to ensure that we know as much as possible—about everything, oppressive ideology included—is not to make exceptions, but to attempt to bring into the light all our prejudices. In this way, freedom and equality will ultimately be best served. For Gadamer, hermeneutic reflection devises the questions that are then directed to social science, not the other way around. We should not simply resist authority, as Gadamer believes Habermas encourages, but try to examine the prejudices that lead us to accept it. If we find that these prejudices are wrong, we should resist; but it is perfectly conceivable that they are acceptable and accurate in any particular case, and in that instance we have no reason to resist. It is hermeneutical reflection we must insist upon, not unreasoning resistance.

Yet we must grant that the explicit attention Habermas gives to the problems of power and oppression is welcome, and we must recognize that Gadamer's work falls short here. Gadamer's work, at least when it is applied to psychoanalysis, should be supplemented by the views of writers who, in the words of Jane Flax (1996, p. 577), attend to "the historical contingency, radical incompleteness, and intrinsic insufficiency" of the narratives through which we understand race, gender, and other constitutive aspects of our identities.

But when the problem of power is taken up in relation to Gadamer's work, as Philip Cushman (1995) has already started to do within psychoanalysis, it should not be taken up as the objective fact it is considered to be by Habermas. Conceptions of power do not have to preexist hermeneutic analysis to participate in shaping it.

Then how shall we bring our increasing awareness of these problems to bear on our clinical work? As far as I am concerned, the task is best accomplished in the same way I have described the patient and analyst dealing with unconscious resistance; that is, resistance is not dealt with as an objective fact, but as an aspect of subjectivity that has to be made phenomenologically appreciable (chapter 9). In an analogous way, patient and analyst need to converse about the likelihood of power relations until the presence of such influences becomes convincing to both analyst and patient, and until specific, contextually relevant shapes of power become clear and compelling. We need a phenomenology of power in psychoanalysis no less than a phenomenology of resistance. Once we are educated to the ubiquity of power relations, there is no reason that they cannot be approached hermeneutically, in dialogue.

There are unmistakable parallels between this dispute and the schism between the social and asocial paradigms of psychoanalysis, and it is probably now clear why the radical critic would side with Gadamer, agreeing that we cannot escape participation in the event of understanding—that we "cannot not interact" (Levenson, 1972). But it should also be pointed out that in the debate between Gadamer and Habermas the field of psychoanalysis is limited to the works of Freud, which means that a good deal of the argument ends up being irrelevant to the radical critic. If attention is restricted to Freud's asocial paradigm, there is no particular reason to challenge Habermas's argument. The conservative classical analyst, after all, does believe that interpretive schemes can be objectively applied to the patient's productions. The events to be understood occur inside the patient, separate from the analyst, so that the analyst does indeed stand outside the event of understanding. Far from being expectable, involvement with the patient (i.e., countertransference) outside the analyst's technically defined role is (at least more often than not) avoidable and problematic.

Schafer and the Interpersonal Field

The prejudices between analyst and patient—the interpersonal field—is not where hermeneutic attention in psychoanalysis has been focused to date. The most extensive, sophisticated, and widely known writings on hermeneutics in clinical psychoanalysis, those by Roy Schafer (1983, 1988, 1992), focus on how hermeneutics aids analysts in their attempts to emancipate the patient from oppression by the internal, intrapsychically theorized unconscious. For Schafer (1983), a theory is an interpretation, not science: "it shapes the phenomena to be analyzed, and it selectively accentuates and organizes them" (p. 31). It is Schafer's theme that clinical conduct and understanding is inescapably theory-laden, and he encourages the comparative study of psychoanalytic theories as a means of deepening our appreciation of theory's role. He also accepts that comparative study can reach only so far, because "each analytic approach tends to be self-confirming, and its results cannot easily be compared to those obtained differently" (p. 31). But Schafer leaves unquestioned the assumption that competent analysts generally shape their work with patients according to a rational application of the theory they prefer. He leaves unexamined, in other words, the analyst's embeddedness in the field. Instead, for Schafer, the competent analyst can be trusted to make careful, (relatively) emotionally removed observations. The analyst's observations may be theory-laden, and we may need to choose our theories very carefully for that reason; but within the "self-confirming" realm of its own operation, each theory remains in this view a rationally applied mode of observation.

But our experience with our patients is laden with more than theory. I would argue that theory is not even the most significant of the factors that influence the analyst's understanding. After all, an analyst is already armed with theoretical commitments when he meets a patient for the first time. Though the particular way the theory is applied is a function of the analyst's unfolding experience with the patient, the theoretical commitments themselves are not developed in interaction. But consider personal prejudices: starting with the very first meeting, these begin to form and shape the experience analyst and patient have in one another's presence. At the outset, the effect of tailored personal prejudices is small; the field in these initial sessions is largely a reflection of the analyst's and patient's pasts, composed of one another's relatively global attributions about what "kind" of person the other is. Each participant has plenty of life experience to use in these early diagnostic assessments, and the analyst also has a theory. But these attributions, like the analyst's theoretical commitments, are not developed in interaction; they are imposed on the basis of preexisting points of view. As time passes, the influence on the field of the participants' initial diagnoses of one another declines, to be replaced by an intricately woven pattern of implicit personal prejudices. I have tried to say that it is prejudice that constitutes the field—the transference-countertransference—and that it is therefore prejudice that eventually must be analyzed. Of course, in this view, the analyst by this time is deeply involved, incapable of anything like the rational, emotionally detached application of a theory.

At some point along the way, some combination of the analyst's perspicacity, intuition, and prejudices begins to inform his theoretical grasp of the patient and the relationship; this changed use of theory then feeds back on the web of the analyst's prejudices, which influences the analyst to participate differently in the relationship. Eventually, theory and prejudice begin to bear a dialectical relation to one another in the analyst's mind. Each becomes so much the child of the other that it is difficult to know which was the original parent. Was it theory that shaped in its own image our eventual interpretation of the prejudice, or was it the unconscious influence of the prejudice that led us to think about the patient, ourselves, or the relationship in the terms of a particular bit of theory?

We usually cannot give neat answers to this question. But we can use this brief account to emphasize the dialectical nature of the process. The analyst's use of theory occurs within a social web that becomes so ubiquitous and immensely complex that it eventually weaves theory, too, into itself; in turn, theory continually plays a crucial role in shaping our perception of prejudice.

Schafer could not be called a radical critic, nor would he wish to be; accordingly, it should not be surprising that he approaches this problem in a way more consistent with Habermas than with Gadamer. As a matter of fact, Schafer (1983, p. 234) cites Habermas as his primary hermeneutic influence.⁵ Like Habermas, Schafer seems to accept that there are certain important ways in which the observer can remain separate from the observation. The analyst may be embedded in his theoretical commitment, but he is not necessarily embedded in the field. As a contrast to the view of the radical critic, consider one of Schafer's (1977) descriptions of transference, which conveys his confidence in the capacity of the well-functioning analyst to remain clear of unconscious entanglement in the field, and therefore to know the meaning (at least within the domain of the analyst's preferred theory) of the patient's experience and conduct. "The major transference phenomena represent the achievement of such simplified, focused ways of defining and acting within the analytic relationship that there can be no mistaking their meanings or avoiding their emotional manifestations or implications" (p. 124; my italics).

From the perspective of a radical critic armed with social constructionism, Schafer, while he alerts us to one half of the dialectic of theory and participation, never even considers the other, which lies in the web of personal prejudices. I must disagree with Schafer in this sense. But let me ignore for a moment the fact that Schafer does not structure the issue as a dialectic. In order to compare Schafer's position and my own, let us reformulate Schafer's as an emphasis on the theoretical pole of the dialectic and my own as an emphasis on the participatory one. If we put the issue this way, we must admit that neither position is simply right or wrong—though each position's advocates would certainly claim that their position better suits what we know. It is a matter of vantage point. A Gadamerian view encourages us to focus not only on the objects of understanding, but on the inevitable shaping influence of the dialogue itself, whereas Habermas encourages us to focus less on the dialogue and to step back to gain a more critical view of the things to be understood.

The Triple Hermeneutic

Gadamer's views are a particularly good way to deal with a related, knotty problem in the logic of social-scientific knowledge that has been noted by a number of philosophers (see Taylor, 1971). There are two parts to this problem. The first part, which applies to both social science (Taylor, 1971) and natural science (Kuhn, 1970, 1977), has to do with seeing theory as a cultural product. The argument goes that the understanding of science is "situated": it is a "language game" (Wittgenstein) or a "semantic field" (Taylor), which means that "the direction of research within a given field, the standards by which research hypotheses are confirmed and rejected, and the criteria for the relevance of data all refer to a normative framework of assumptions, conventions, and purposes" (Warnke, 1987, p. 109). The knowledge of science is therefore not objective, and it is certainly not free of values. On the contrary, in this view, as in Foucault's (1980), such knowledge actually represents the institutionalization of values. Schafer recognizes this point in his effort to establish a comparative psychoanalysis.

The second part of the argument, however, applies only to the social (or human) sciences. For these areas, it is not only fields of study that constitute language games. The subject matter itself—human living—is a language game. That is, the way we live every day—the common meanings our behavior and linguistic expressions have for us, the content of our everyday concepts, and all the rest of those taken-for-granted underpinnings that make life coherent and meaningful—requires its own hermeneutic exegesis. Just in order to have a subject matter, social sciences must assume this preexisting, socially constructed organization. We refer to this aspect of living by means of concepts such as as Mead's (1934) "generalized other," or Berger and Luckmann's (1967) "everyday life," which are socially constructed and yet experienced as inevitable and objectively real. Thus, any attempt to study human life inevitably requires what Anthony Giddens (1976) calls a "double hermeneutic," and any understanding in human science cannot be accomplished simply by applying a language to a content, but requires the convergence and mediation of two different languages.

The analyst comes to the analytic situation, then, with two hermeneutic tasks. Since each analytic pair's experience with one another is built from the same vocabulary of culturally defined elements available to us all, the analyst must grasp implicitly the common, everyday meanings of the interpersonal field; then the analyst must understand the way these meanings can be recombined in the realm of theory. This is the double hermeneutic. But soon another task arises, the one to which I have just drawn attention. Any interpersonal field is unique in its particular details, and it accrues more and more detail over time. As its particularity increases, to the first two tasks we must add the job of understanding the very particular selection of common meanings that have become habitual and invisible in the analytic situation. These are the very tiny and specific, socially constructed sets of organizing principles to which I have been referring as the interpersonal field. The field demands its own exegesis, an exegesis that requires, but is not accomplished by, a grasp of the social fabric the participants have in common. Any study of the events that take place in one particular relationship, then, requires something more of the researcher than the double hermeneutic refers to. Here we have a third level of the mediation of meaning, what Protter (1996), who makes the same point, calls a "triple hermeneutic." What this means for psychoanalysis is that, if we accept that any event in the treatment is embedded in the very particular set of prejudices that the analyst and patient have established in the field, then the act of understanding requires the convergence and mediation of not just two, but three different languages.

Because Gadamer believes understanding is consensus through dialogue, his work is particularly well suited to the conditions of the double hermeneutic. Consensus, after all, is defined as the understanding of what the other person means within the other's frame of reference—and that frame of reference, of course, is the everyday world. Think how much more necessary this kind of understanding becomes when the other's frame of reference is not just everyday life, but that tiny portion of it specific to a particular interpersonal field. Thus, Gadamer's work, essential when the hermeneutic is doubled, becomes even more crucial when it is tripled.

But what should the Analyst Actually do?

Gadamer does not offer a practical answer to what is, after all, the question of most import to psychoanalysts: How can prejudice be disconfirmed? How do we come to be able to ask open questions? To tackle these questions directly Gadamer would have had to contradict himself, because any answer he would give would amount to a recommended method. His final word on the matter is that we must develop and maintain what has been variously translated as "effective-historical consciousness," "consciousness of effective history," and "the consciousness in which history is ever at work." This unwieldy phrase simply means that one must remain receptive to the new and unfamiliar by being open to the effects of history (i.e., tradition) on one's understanding. History must be allowed to live in us, to direct our explicit awareness, to arrive at a knowledge of itself through our consciousness of prejudice. We must be a conduit for the conversion of being into language. I have tried to emphasize that in psychoanalysis the most crucial history to which we must be alive is the prejudices we develop in interaction with the patient.

That Gadamer offers nothing more concrete than this will be unsatisfying to anyone attached to the notion that psychoanalysis should be practiced according to a specifiable technique. It may seem that Gadamer's view encourages nihilism, or perhaps laziness, since there is no way to search out the truth. It may even seem to imply that psychoanalytic education is unnecessary. Why don't we just "converse?" But Gadamer's view does nothing to weaken the case for analytic discipline and education, because to adopt this perspective means placing great stress on the necessity for the analyst to question himself about whatever he takes for granted, to find a way into "seeing what is questionable" (Gadamer, 1966, p. 13). And seeing what is questionable requires a consistency of self-reflection that would be next to impossible without careful supervision and a training analysis that touches the analyst in the way he hopes his patients will be touched. To understand someone else requires innocence and openness, which are not only gifts, but accomplishments of education and experience.

To work from the view of understanding that Gadamer and the radical critic have in common, the analyst faces a high degree of uncertainty. All he can do is try to identify good questions (Levenson, 1988; Stern, 1992b), never being sure he is successfully pursuing even that limited aim. Whatever its difficulties, though, the perspective an analyst finds in the work of Gadamer will be familiar to the many analysts who already believe that in clinical practice one has no choice but to become part of something beyond one's ken, to be inhabited by the unbidden, to surrender to the mystery by which we formulate what we know. Gadamer's critique of method would suit Masud Khan (1969), for instance, in whose oft-quoted words, "We are all the servants of the patient's process" (p. xxxi). Levenson (1982), who cites the same remark of Khan's, also comes to mind: "some process is going on that they [patient and therapist] have not initiated or energized. There is the remarkable experience of being carried along by something larger than both therapist and patient: A true sense of an interpersonal field results. The therapist learns to ride the process rather than to carry the patient" (p. 13).