Round Table: Discussion by Editors of Major Psychoanalytic Journals on their Perspectives of Current and Future Trends in Psychoanalysis
Theory and Practice: An Intimate Partnership or a False Connection?

Henry F. Smith

[Henry F. Smith sent this paper, which is the Editor's Introduction to no. 1/2003 of The Psychoanalytic Quarterly, as a background for his discussion at the Round Table of Editors of Major Psychoanalytic Journals held on June 15, 2003, at the Annual Meeting of the Rapaport-Klein Study Group]


[In response to Dr. Migone’s invitation to post something pertinent to our panel on "Current and Future Trends in Psychoanalysis," I hope the following will introduce some ideas that will stimulate further discussion. It is an Introduction to a recent special issue of The Psychoanalytic Quarterly on "How Theory Affects Practice."]

Eugen Bleuler (1912), the Swiss psychiatrist who gave schizophrenia its name, once said of schizophrenic thought that "it thinks" in the patient (p. 23, quoted in Forrest, 1965). So, too, does psychoanalytic theory "think" in the analyst.

Freud (1915) put it about as well as one can in discussing the origin of scientific theory. "Even at the stage of description," he wrote, "it is not possible to avoid applying certain abstract ideas to the material in hand, ideas derived from somewhere or other but certainly not from the new observations alone." He continued, "We come to an understanding of [the] meaning [of these ideas] by making repeated references to the material of observation from which they appear to have been derived, but upon which, in fact, they have been imposed" (p. 117, quoted in Smith, 1992, 1999).

Nor is the patient immune from his or her own personal indwelling theory. As Cooper (1985) has said, "Without a theory we are unable to select data from the massive jumble of impressions that constantly assail us. Neither psychoanalysts nor naïve psychologists -- the man in the street -- are able to function without a theory" (p. 5). And Friedman (1988) puts it bluntly: "We cannot think without theory" (p. 7).

A key player in the discourse on the relationship between theory and practice, Friedman has devoted much of his work to the elucidation of how theory "sits (or hides) in the therapist’s mind" (p. 9), where it illuminates opportunities, reassures the therapist that there is something to grasp, indicates how to grasp it, and steadies the therapist with a "cognitive brace" (p. 76), as it privileges and disciplines the therapist’s listening. In Friedman’s view, theory soothes the therapist’s inevitable and multiple discomforts, and particular theories or part-theories have evolved to address those discomforts in particular ways.

Friedman makes the presence of theory so immediate that we might at times be beguiled into thinking we need not distinguish one form of theory from another, as if we do not need to ask, Theory of What? Mind? Technique? What levels of theory are we talking about? Do they all have a function in clinical work?


Perhaps the clearest description of the different levels of psychoanalytic theory was Waelder’s (1962) summary of a symposium on scientific method. Taking his cue from a presentation at the symposium by Hartmann (1959), who had commented on the levels of abstraction in psychoanalytic theorizing, Waelder wrote:

In speaking of psychoanalysis or Freudian doctrine, one can distinguish between different parts which have different degrees of relevance. First, there are the data of observation. The psychoanalyst learns many facts about his patient which other people, as a rule, will not get to know. Among them are facts of conscious life which people are not eager to relate to others, not even to psychological interviewers, or about which they do not care to tell the truth, or the whole truth, or of which they do not usually think but which will occur to them and which they will relate in the psychoanalytic interview because of its peculiar climate mixed of relaxation and discipline, of intimacy and personal aloofness. To this, one must add the things which are not conscious or preconscious but can send derivatives into consciousness under the conditions of the psychoanalytic situation. The psychoanalyst learns not only about all such data but also about the configurations in which they appear. All these form what may be called the level of observation.
These data are then made the subject of interpretation regarding their interconnections and their relationships with other behavior or conscious content. This is the level of clinical interpretation.
From groups of data and their interpretations, generalizations have been made, leading to statements regarding a particular type such as, e.g., a sex, an age group, a psychopathological symptom, a mental or emotional disease, a character type, the impact of a particular family constellation, or of any particular experience, and the like. This is the level of clinical generalizations.
The clinical interpretations permit the formulation of certain theoretical concepts which are either implicit in the interpretations or to which the interpretations may lead, such as repression, defense, return of the repressed, regression, etc. This is the level of clinical theory.
Beyond the clinical concepts there is, without sharp boundaries, a more abstract kind of concept such as cathexis, psychic energy, Eros, death instinct. Here we reach the level of metapsychology.
Finally, Freud, like other thinkers, had his own philosophy, his way of looking at the world, and he was more articulate than many in expressing it. His philosophy was, in the main, the philosophy of positivism, and a faith in the possibility of human betterment through Reason—a faith which in his later life, in consequence of his psychoanalytic experience, became greatly qualified though not altogether abandoned. This may be called the level of Freud's philosophy.
These levels are not of equal importance for psychoanalysis. The first two, the data of observation and the clinical interpretations, are entirely indispensable, not only for the practice of psychoanalysis but for any degree of understanding of it. Clinical generalizations follow at close range. Clinical theory is necessary too, though perhaps not in the same degree. A person may understand a situation, symptom, or dream with little knowledge of clinical theory, and while this would certainly not be enough for a practicing analyst, one would yet have to recognize that such a person has a considerable measure of understanding of psychoanalysis.
Metapsychology, however, is far less necessary, and some of the best analysts I have known knew next to nothing about it. These are the kinds of hypotheses about which Freud (1914) said that they are "not the bottom but the top of the whole structure [of science], and they can be replaced and discarded without damaging it" (p. 77).
Freud's philosophy is largely a matter of his time and has little bearing on psychoanalysis. [Waelder, 1962, pp. 619-620]

Given how much play the role of philosophy has had in recent debates about both theory and practice, Waelder’s last comment is at odds with much of contemporary rhetoric, as I shall discuss below.

Clear as Waelder’s levels of theory are, his notion that some levels are closer to the data than others, or less relevant than others, has come under sharp criticism. Brenner (1980), for example, has disputed the idea that higher levels of abstraction lie further from the data of observation. Echoing Freud’s comment above, he writes, "In every branch of science even the simplest observations involve ideas of the highest order of abstraction," adding "what makes a theory useful" is not the level of abstraction but "the degree to which a given theory is supported by the relevant data" (p. 200). Even if Waelder simplifies the mix of theoretical levels with which we work clinically, as a descriptive exercise, his hierarchy remains, in my view, a useful tool for analyzing the way levels of theory become muddled in our discourse (Slap & Levine, 1978; Smith, 1997). Further, even though the data of observation are shaped by theory to an extent Waelder did not take into account, his levels of abstraction appear to be descriptively accurate across a wide range of psychoanalytic schools of thought.

Friedman’s inclusion of all levels of theory in his considerations of practice becomes clearer in the light of Brenner’s comment above. As Friedman (1988) himself puts it, "A ‘clinical theory’ purified of metapsychology is probably a self-contradiction" (p. 86). I would conclude, therefore, that in one sense, the distinction between a theory of mind and a theory of technique is an artificial one. All theory that derives from and participates in the clinical situation is ultimately clinical theory, and all efforts to organize our observations, experience, and interventions in the analytic situation can be considered exercises in the use of theory.


If indeed theory "thinks" in the analyst, as Freud, Waelder, Brenner, and Friedman all testify, each in his own way, it would seem essential for us to know not only what our theory is thinking, but also how and where it thinks. While all analysts must seek help from their theories more deliberately and explicitly at some moments than at others, for some analysts, theory seems to regularly inhabit the front of the mind, whereas for others, it sits somewhere to the rear or to the side, a presence but a less insistent one.

Bearing in mind that when analysts write about their work, there is a complex relationship between their writing and their analyzing, when some analysts write about their work, we think we see clearly and cleanly the operation of their theory. Gray (1986), for example, gives us detailed clinical observations that demonstrate his use of structural theory as a kind of optical filter through which he can observe moments of conflictual interference on the surface of the patient’s mind. Although they share a common commitment to the structural model, in Gardner’s (1983) clinical descriptions, theory itself is more elusive. Gardner (1995) writes, "I prefer a theory that guides my attention, but does so gently. I prefer a theory that’s simple enough to remember and complex enough to ‘forget’" (p. 90). If Gray’s approach appears to be theory-near, Gardner’s might be said to be theory-distant -- or near at some moments and distant at others. Similarly, whereas Gray (1982) suggests that technique lags developmentally behind theory, Gardner (1995), while agreeing, nevertheless counters that "some aspects of theory lag behind practice," adding "art, in the main, goes far ahead of theory, and it’s theory’s job to try persistently to close the gap" (p. 79).

For purposes of discussion, I want to make it clear that by theory-near and theory-distant, I am trying to suggest a number of variables in analyzing. First, how deliberately and consciously does an analyst use his or her theory, how near to awareness is it at any given moment? Second, how closely integrated with that theory are the observations the analyst makes? These two characteristics may coexist but are not synonymous. And finally, how fluidly does an analyst move between a more conscious and less conscious grip on his or her theory, or within any one preferred mode of analyzing? These variations in the use of theory are largely independent of the wide range of phenomena analysts may preferentially select for observation.

Along with these considerations of the location of theory, consider Sandler’s (1983) observation that while we all have a preferred, or what he calls a "public" theory, we maintain various preconscious, private theories that live in quite harmonious contradiction with our public or primary theory, as long as we do not bring them into the light of day. I would note that we all make observations that do not fit easily with our preferred theories. If we can catch them, such observations represent opportunities for a creative advancement in understanding, as long as we resist the temptation to enshrine them prematurely with new theoretical explanations. Given our natural and essential inclination to generalize from our observations, this is more easily said than done.


If the relationship between theory and practice is always an intimate one, with theory both derived from and imposed on observation, as Freud indicated, it would appear that what we observe in practice are clinical units in which theory and observation, while occupying distinct conceptual levels in Waelder’s terms, are thoroughly interwoven. It also seems clear that every theoretical commitment tends to push our habits of practice in one direction or another.

As I have indicated (Smith, 1997, 1999, 2001), however, in the contemporary literature, it has become commonplace for writers, deliberately or inadvertently, to represent theory and practice as much more tightly linked than is warranted in an effort to make their practices seem more lawful than they are. Under such circumstances, rather than theory and practice cooperating as intertwined but separate entities, the two are in fact conflated, and the intimate partnership between them becomes a false and misleading one.

Thus, we find analysts obeying a clinical theory, derived from the observations of practice, as if the practice were now "owned" by the theory, which, in turn, tells them what to do. Others deliberately buttress a new or preferred technique with a new theory of mind -- or even of brain (Smith, 1997) -- or justify their interventions according to their own philosophical or epistemological assumptions. We hear analysts retrospectively explaining their use of self-disclosure, for example, on the grounds that it is consistent with their intersubjective principles, rather than because it is clinically indicated with a particular patient. Note that in any of these examples, not only do the distinctions Waelder drew between one level of abstraction and another collapse on themselves, but the sequence that he described from observation to theory is inverted.

I want to emphasize that in my illustration, I am not taking a stand for or against either self-disclosure or intersubjectivity. Rather, I would suggest that the form of argument linking theory and technique in such a fashion ignores the fact that many aspects of practice -- including what we reveal about ourselves, deliberately or not, along with our tone of voice, gestures, affective engagement, and authenticity, not to mention the level of uncertainty we tolerate and our capacity to question our own assumptions -- are not the province of any one theoretical position or school of analysis. Conceptually, theory and practice exist in different domains, at different levels of abstraction, and are far more loosely coupled than these false connections imply.

In his important contribution to this issue of The Psychoanalytic Quarterly, Fonagy (2003) elaborates on several of these points from a somewhat different perspective. He argues that analytic practice developed historically by trial and error, and that analytic theory grew out of analytic practice. In that respect, it was, at least in part, an inductive process, from observation to generalization. But subsequently, Fonagy suggests, we have all tried to use theory deductively, as if it could tell us what we are seeing and what to do. It is this latter use of theory as a directive for practice that concerns me here.

I would suggest that the format of much of our literature confirms and recapitulates the very sequence Fonagy is outlining. A writer may begin with a review of the literature, followed by a clinical vignette, and then a theoretical comment that seems to derive from the clinical material. This is then accompanied, not infrequently, by another vignette, and soon the clinical material seems to follow from the theory, rather than the other way around. In some cases, this appears to be a rhetorical device to grant authority to the practice in question; in others, it seems a deliberate attempt to ground both observation and technique in a general theory. Even with no such conscious motivation on the part of the writer, but simply as a byproduct of the narrative sequence, the effect tends to be the same on the reader: both theory and practice feel more persuasive because of their mutual reinforcement. Sometimes we observe this sequence in a single article; at other times we see it unfold over a series of articles or in the work of multiple authors, as schools of analysis evolve and become established.

It is not difficult to see how these trends might take root even without artful intent. When patients oblige us by getting better under our care, we take this as confirmation not only of our practices, but also of the theories we have evolved to explain them. Moreover, as readers, we seek such reassurances. That is one of the reasons we read. The point is that sooner or later, theory begins to function as law, and practice follows behind.

One can see the effect of these false connections not just in the use of theory as a technical directive, or in the deliberate building of new theory, but also in the doubts that begin to gather around the old. As I indicate elsewhere in this issue (Smith, 2003, pp. 80-89), a master clinician like Philip Bromberg, for example, anchors his clinical insight and technical agility to a theory of mind based on dissociation. It is the dissociative mind that underwrites, and hence "necessitates," his attention to the multiple states of the patient. This is a radically different view of the mind than the one on which many analysts were raised. It may in fact be correct, but I have heard clinicians argue that in order to be flexibly attuned in the clinical hour, as Bromberg is, or to adopt the practices of any number of other compelling contemporary writers, they must discard the old concepts they learned, including the view of a mind in conflict. In other words, to adopt the practice they have to adopt the theory with which it has been packaged, an assumption that I would suggest is no more warranted than that the two needed to be so tightly linked in the first place.

We might note in passing that it is not uncommon in psychoanalytic education to see courses on comparative psychoanalysis taught along the lines of just such a one-theory/one-practice model, with no disclaimer offered on the limits of theory and its role in practice. What results is a kind of idealization of theory that may be particularly problematic in those programs with a specific theoretical agenda. In my experience, such idealizations, if unchecked, may leave students feeling demoralized by their own uncertainties, grasping at one theory or another to guide them in the heat of the clinical moment, and foreclosing the development of their own clinical voice with premature theoretical commitments.

I am not singling out anyone as responsible for these false connections between theory and practice. We all inevitably try to use theory to simplify our work, rather than to reveal its complexities, and we find solace in the general rule as a protection from the exigencies of the particular situation. Moreover, we all try to give our preferred techniques a theoretical legitimacy they cannot claim, rather than depending on clinical evidence to prove their efficacy. We do this because we do not know what else to do. But as a result, over time, our literature comes to resemble a kind of patchwork quilt of theory and practice, in which each appears to support the other with scarcely any evidential connection, and almost every apparently useful intervention can be explained after the fact by a wide variety of contemporary theories.


This special issue of the Quarterly was conceived by Sander Abend, who, together with Owen Renik, chose the contributions and did much of the editorial work before I became the Quarterly’s Editor. I am very grateful to them both.

You will note that, while all the papers explore different aspects of how theory affects practice, some of them lean more toward theory and others more toward practice. In the former group, in addition to Fonagy’s exploration of the scientific status of psychoanalytic theory, Reed assumes the nearly impossible task of trying to observe how theory functions in her mind as she analyzes, Rey de Castro illustrates his experience of "free-floating theory" in the clinical hour, as he explores our inability to pin down clinical theory in the first place, and I differentiate several contemporary views of conflict and their influences on clinical technique. The other authors, Aisenstein, Blevis and Feher-Gurewich, Busch, Ferro, Hirsch, and the Ornsteins, have all contributed generous samples of their own clinical work to illustrate the function and effect of their own preferred theories. The volume concludes with Michels’s typically masterful discussion.

Rather than settling any of the many questions that lie before us, we hope with this project to unsettle some of your own theoretical predispositions, both as to the content of your theories and to your use of them in practice. You will note that we have included a range of psychoanalytic approaches in this volume, some of which you may find frankly unsettling in themselves. It is my belief that only as we develop a capacity to study the work of others more dispassionately can we begin to determine what separates "us" from "them," and what might lead toward a degree of reconciliation. It is also my belief that if we were each to report what we observe ourselves doing in a form as close as possible to the way we both observe and do it, we might challenge familiar theoretical affiliations, and in so doing, discover a few strange new bedfellows. We might even put to rest some of our seemingly endless, and most assuredly fruitless, political battles.

Henry F. Smith, M.D., Editor of The Psychoanalytic Quarterly


Bleuler, E. (1912). The Theory of Schizophrenic Negativism, trans. W. A. White. New York: J. Nervous & Mental Disease Publishing.

Brenner, C. (1980). Metapsychology and psychoanalytic theory. Psychoanal. Q., 49: 189-214.

Cooper, A. (1985). A historical review of psychoanalytic paradigms. In Models of the Mind: Their Relationships to Clinical Work, ed. A. Rothstein. New York: Int. Univ. Press, pp. 5-20.

Fonagy, P. (2003). Some complexities in the relationship of psychoanalytic theory to technique. Psychoanal. Q., 72: 13-47.

Forrest, D. V. (1965). Poiesis and the language of schizophrenia. Psychiatry: J. Study of Interpersonal Processes, 28: 1-18.

Freud, S. (1915). Instincts and their vicissitudes. S. E., 14: 117-140.

Friedman, L. (1988). The Anatomy of Psychotherapy. Hillsdale, NJ: Analytic Press.

Friedman, L. (1997). Ferrum, ignis, and medicina: return to the crucible. J. Amer. Psychoanal. Assn., 45: 21-36.

Gardner, M. R. (1983). Self Inquiry. Hillsdale, NJ: Analytic Press, 1988.

Gardner, M. R. (1995). Recollections: sexuality, neurosis, and analysis. In Hidden Questions: Clinical Musings. Hillsdale, NJ: Analytic Press, pp. 76-99.

Gray, P. (1982). "Developmental lag" in the evolution of technique for psychoanalysis of neurotic conflict. J. Amer. Psychoanal. Assn., 30: 621-655.

Gray, P. (1986). On helping analysands observe intrapsychic activity. In Psychoanalysis: The Science of Mental Conflict: Essays in Honor of Charles Brenner, ed. A. D. Richards & M. S. Willick. Hillsdale, NJ: Analytic Press, pp. 245-262.

Hartmann, H. (1959). Psychoanalysis as a scientific theory. In Essays on Ego Psychology: Selected Problems in Psychoanalytic Theory. New York: Int. Univ. Press., 1964, pp. 318-350.

Sandler, J. (1983). Reflections on some relations between psychoanalytic concepts and psychoanalytic practice. Int. J. Psychoanal., 64: 35-45.

Slap, J. & Levine, F. (1978). On hybrid concepts in psychoanalysis. Psychoanal Q., 47: 499-523

Smith, H. F. (1992). Screen language and developmental metaphor. Ann. Psychoanal., 20: 95-113.

Smith, H. F. (1997). Creative misreading: why we talk past each other. J. Amer. Psychoanal. Assn., 45: 335-357.

Smith, H. F. (1999). Subjectivity and objectivity in analytic listening. J. Amer. Psychoanal. Assn., 47: 465-484.

Smith, H. F. (2001). Obstacles to integration: another look at why we talk past each other. Psychoanal. Psychol., 18: 485-514.

Smith, H. F. (2003). Conceptions of conflict in psychoanalytic theory and practice. Psychoanal. Q., 72: 49-96.

Waelder, R. (1962). Psychoanalysis, scientific method, and philosophy. J. Amer. Psychoanal. Assn., 10: 617-637.

[Note: This paper is the Editor's Introduction to no. 1, 2003, vol. LXXII, pp. 1-12, of The Psychoanalytic Quarterly. We thank for the permission]

Henry F. Smith, M.D.
17 Hammond Street
Cambridge, MA 02138-1915
E-Mail <>

Web Editor: Paolo Migone. For suggestions or corrections, please e-mail to: