A Critique of the Postmodern Turn in Psychoanalysis: 
Recent Work of Mitchell & Renik

Morris N. Eagle

[Paper presented on June 10, 2000, at the Annual Meeting of the Rapaport-Klein Study Group]


I present this paper as a work in progress rather than a completed one. In doing so, I return to our early tradition. In attempting to revise and update my Recent developments in psychoanalysis book (Eagle, 1984), it became clear to me that the basic divisions and schisms in psychoanalysis no longer have primarily to do with different ‘schools,’ but with what Irwin Hoffman refers to as different paradigms. Irwin identifies them as positivism versus constructivism. He writes that the fundamental change in psychoanalysis is not the shift from the drive to the relational model, but "from a positivist model for understanding the psychoanalytic situation to a constructivist model" (p. 135). I think this is a misleading description because, as Paul Meehl has noted, most psychologists and analysts know little or nothing about the history of positivism nor the range of its complex meanings, but instead use it as a buzz word to refer to the supposed scientistic and objectivist bad guys. However, I think I know what Irwin means and I agree with him that the primary divisions in psychoanalysis have to do with different world views. I think that the basic division can be best described in terms of what John Searle (1998) calls the "Enlightenment vision" versus postmodernist and other critiques of and attacks on that vision. In this regard, some recent psychoanalytic developments are but one expression of broad culture-wide challenges to and attacks on the "Enlightenment vision" and the default positions with which it is associated. One such world view is reflected in the recent work of Mitchell, Renik, Schafer, and Spence - I will refer to them as the ‘new view’ theorists and will focus on a recent paper by Mitchell (1998) and two recent papers by Renik (1996, 1998) which I take as paradigmatic of current postmodernist influence in psychoanalysis.

I have entitled my paper "The postmodern turn in psychoanalysis" because the writings of some influential contemporary psychoanalytic theorists reflect the influence of postmodern ideology. However, one can, if one prefers, ignore the term ‘postmodern’ and simply identify a number of recent trends in psychoanalysis that underlie and are reflected in the formulations of the ‘new view’ theorists.

Why devote such attention to the papers by Mitchell and Renik and to the ‘new view’ theorists? What is at stake beyond the analysis and critique of these papers and of the positions these authors take? My impression is that these theorists have been quite influential and attractive to many in the psychoanalytic community, particularly to new students. And it is my belief that adoption of some of the positions taken by the ‘new view’ theorists entails essentially dismantling the psychoanalytic enterprise, or certainly radically changing it. One of the reasons that the formulations of the ‘new view’ theorists have been attractive and influential is that they include legitimate and cogent criticisms of traditional theory. Many of these criticisms, although not entirely new, are justified. The problem, as I will try to show, is the conceptual and philosophical formulations in which they are embedded.

Another reason that the ideas of the ‘new view’ theorists have been attractive to many is that they appear to immunize psychoanalysis against the kind of cogent criticisms that Grünbaum (1984) has directed against classical theory by, in effect, responding: "These criticisms no longer apply because we don’t think that way anymore." Indeed, one wonders whether the seeking of such immunization is one of the implicit factors that motivates some of the positions taken by the ‘new view’ theorists. I will return to this issue later in the paper.

A third factor in understanding the context in which the ideas of the ‘new view’ theorists need to be placed is, as noted, the broad cultural turning away from and attack on what Searle (1998) calls the "Enlightenment vision" and the default positions that characterize that vision. As Searle notes, the challenges to the "Enlightenment vision" have been variously called "social constructionism, pragmatism, deconstructionism, relativism, postmodernism…" (p. 15) and "perspectivism" (p. 20). These trends are all well represented in contemporary psychoanalysis.

As Peter Gay points out, the clearest expression of the "Enlightenment vision" in psychoanalysis is found in classical theory. In criticizing classical theory, legitimate criticisms of certain aspects of that theory have been extended to and intermixed with the very dismantling of the "Enlightenment vision."

In introducing and linking the "cornerstone" concept of repression - the essence of which is the failure to acknowledge the truth about oneself - to pathology, and in linking the lifting of repression - the essence of which is increased self-knowledge - to cure, psychoanalysis became a clear expression of the "Enlightenment vision." Self-knowledge was now not only a Socratic virtue, but also a clinical necessity. Freud was not only interested in therapeutic success - other approaches could claim success - but in linking such success to the accurate uncovering what was going on in the patient’s mind. This basic idea remained a core one whether hypnosis was employed to uncover traumatic memories or interpretation to uncover unconscious wishes and defenses. Indeed, when Fliess accused Freud of suggesting ideas to his patients, Freud (1950) retorted "you take sides against me and tell me that ‘the thought-reader merely reads his own thoughts into other people,’ which deprives my work of all its value" (p. 336).

Thus, central to traditional psychoanalysis is the belief that psychoanalytic interpretations do not simply constitute suggestion, but really correspond to the inner reality of the patient, that is, they uncover truths. Furthermore, the uncovering of these truths and the self-knowledge they engender are held to be liberating and curative. This truly constitutes a marriage between the Enlightenment project and the clinically curative.

In justifiably reacting against certain aspects of classical theory, the ‘new view’ theorists have taken positions that, I believe, create at least as many difficulties as the ones they were intended to resolve and that essentially entail a dismantling of the very core assumptions on which the psychoanalytic enterprise rests.

Let me begin with Mitchell’s 1998 paper entitled "The analyst’s knowledge and authority." It is clear that Mitchell wants to dispute what he believes is the typical classical analyst’s arrogant claim that as a neutral, objective observer and armed with psychoanalytic theory, he or she has virtually direct access to the Truth about the patient’s mind. So far so good. For certainly, such a claim would be an unwarranted and arrogant one. However, Mitchell goes from a justified rejection of this sort of claim to the arguments that (1) neither the analyst nor the patient uncovers or discovers anything in the patient’s mind; and (2) the only expertise the analyst has is in "meaning-making, self-reflection, and the organization and reorganization of experience" (p. 2).

Mitchell makes clear in his paper that at the core of his dispute with traditional views is his conception of the mind as constructed rather than uncovered. I say "rather than" because it is Mitchell who presents constructing and uncovering as mutually exclusive. Let me quote some passages from Mitchell’s paper that illustrate his point of view. As he notes in the Abstract of his article, his criticisms of traditional psychoanalysis rest on a particular conception of the nature of mind. According to Mitchell, "there are no clearly discernible processes corresponding to the phrase ‘in the patient’s mind’ for either the patient or the analyst to be right or wrong about" (p. 16). The analyst (or anyone else) does not discover or uncover something [in the mind] "that has a tangible existence" (p. 17). Rather, "mind is understood only through the process of interpretive construction. This is equally true for the first person who is the mind in question as well as for someone in the third person position who is trying to understand the mind of another. Further, this is true for both conscious and unconscious mental processes" (p. 16). Other expressions of Mitchell’s position include his rejection of the "traditional claims… that the central dynamics relevant to the analytic process are preorganized in the patient’s mind, and that the analyst is in a privileged position to gain access to them" (p. 18); and his belief that "in an important sense, consciousness comes into being through acts of construction either by others or, through self-reflection, by oneself" (p. 16).

Hence, the classical analyst’s purported arrogance, misplaced certainty, and unearned authority are only side issues. One can temper these attitudes and continue to believe that there is something in the patient’s mind to be uncovered and discovered. The main issue is that Mitchell appears to believe that the idea that one uncovers anything in the patient’s mind rather "interpretively constructing" it is a misguided and mistaken view. So intent is Mitchell in rejecting any trace of the idea that one uncovers anything preexisting that, as we have seen, he also rejects the traditional claim that the central dynamics relevant to the analytic process are preorganized in the patient’s mind. Mitchell’s position here has led Meissner (1998) to comment "It seems odd to me that one would think of the patient, as he enters the consulting room for the first time, as without a history entirely of his own, without a developmental background, without a psychology and personality that he has acquired and developed in the course of a lifetime, all accomplished before he had any contact with the analyst" (p. 422).

In viewing uncovering and constructing as mutually exclusive and in rejecting the former and opting for the latter, Mitchell does not allow the possibility that one both interpretively constructs and uncovers what is in the patient’s mind - in the sense that one infers what is going on in the patient’s mind. That Mitchell does not intend inference by the term "interpretive construction" is seen in a number of ways. Firstly, inferences are just the sort of thing one can be right or wrong about, even if probabilistically. Secondly, if he meant inference by "interpretive construction," it would not follow that there are no preorganized central dynamics in the patient’s mind. Thirdly, it would make little sense to say, as Mitchell does, that "interpretive construction" applies equally both to understanding one’s own mind as well as another’s mind, and to conscious as well as unconscious experiences. For surely, one does not infer one’s own conscious experiences as one does another’s. Understanding another’s mind is mediate and indirect whereas one’s own conscious experiences are immediate and direct. One does not infer them, one simply has them - even though there may be inaccessible complex constructional and computational processes that underlie these conscious experiences. All this suggests that what Mitchell means to say is that mind is literally constituted by "interpretive construction." He wants to sound the death knell to any version of the idea that the analyst uncovers or discovers what is in the patient’s mind. (As an aside, I wonder if this position is, in part, a reflection of Sullivan’s substitution of an interpersonal field for the commonsense idea of an individual personality.)

If mind is constituted by "interpretive construction," then, of course, it follows that "there are no clearly discernible processes corresponding to the phrase ‘in the patient’s mind’" for either the patient or analyst to be right or wrong about (p. 16) and that there is no singularly correct canonical interpretation of another’s mind. But it also follows that there are no multiple correct interpretations. And that is so because the issue of correct - or even plausible - interpretations only applies if one thinks that it is possible to evaluate the degree to which an interpretation corresponds to something independent of that interpretation - not when the interpretation constitutes what is being interpreted.

Similar to Mitchell, Renik (1993, 1996) too seems to dichotomize between uncovering and discovering on the one hand and providing "new perspectives" on the other - his version of "interpretive construction" - and then opts for the latter as the primary task of analysis. He overlooks the fact that quite often adopting a new perspective that makes a difference in one’s life does involve discovering a bit of self-knowledge that was hitherto inaccessible and unknown. And finding some important truth about oneself almost always shifts one’s perspective about oneself, at least to some degree. Consider Renik’s interpretation that his patient’s hostility and guilt toward her sister served as a defense against being critical towards her parents. On the one hand, this interpretation entails taking a new perspective toward the patient’s longstanding hostility and guilt towards her sister: something like, "why don’t you look at your anger and guilt towards your sister in this way?" On the other hand, however, the interpretation also speaks to the uncovering of the patient’s hitherto inaccessible, angry and critical feelings toward her parents. Thus the interpretation is not simply "why don’t you take this perspective toward your anger at your sister" but also "there is more to your anger and guilt toward your sister than meets the eye," more to be uncovered.

A question that comes up at this point is the nature of a psychoanalytic project in which one gives up the idea of uncovering or discovering anything. As Larry Friedman (1998) notes, if one accepts this point of view, then analysis "is, indeed an adventure of a vastly different sort than we have imagined" (p. 260). He goes on to express the concern that "it is hard to imagine how an analyst would work who no longer believes in hunting for something that is already there to be discovered" (p. 261). If all we offer are coherent and serviceable meaning systems and free ourselves from the questions of whether they tally with something real in the patient, how do we differ from religious and cult proselytizers of all stripes? I will come back to this question. Let me turn now to another related issue.

Before I do, I want to note here that when I speak of uncovering and discovering, I am not necessarily limiting that concept to the uncovering of specific unconscious wishes. For one can also uncover fantasies, defenses, unconscious pathogenic beliefs, one’s internal working model, the procedural memories and ‘rules’ that characterize aspects of one’s behavior, and so on. It seems to me that when Mitchell rejects the role of uncovering in analysis, he implicitly has in mind the idea associated with Freudian theory that one uncovers fully formed and timeless unconscious wishes that are ‘waiting’ to be exposed. He writes: "to understand unconscious processes… is not to expose something that has a tangible existence as one does in lifting a rock and exposing insects beneath" (pp. 17-18). He does not seem to allow for the possibility that in uncovering unconscious material, one can expose something less tangible than insects, but nevertheless just as real. I will come back to this issue later in the paper.

How does one select and evaluate different "interpretive constructions"?

A critical question regarding "interpretive constructions" of another’s mind is how one selects among different possible interpretations. Mitchell is aware that talk about many possible good interpretations prompts the question of the grounds for this or that interpretation. In an attempt to demonstrate his rejection of "irresponsible relativism" and to show that he is not espousing an ‘anything goes’ position, Mitchell observes that "believing that there is no correct canonical version of the patient’s mind does not suggest that all versions are equally valid or compelling" (p. 9) and then goes to assert that "compelling and generative meaning systems do not work well if they are contradicted by known facts" (p. 10). (It should be noted that Mitchell’s concern that the analyst’s interpretations and meaning systems should not be contradicted by "known facts" is not because such contradictions indicate that the interpretations are false, but rather because they "do not work well"). Mitchell’s other basis for selecting among different interpretations and meaning systems is their ability to make for a "richer and less self-sabotaging existence." So, put simply, Mitchell is suggesting that one evaluate and select among different possible interpretations and meaning systems on the basis of their therapeutic effectiveness.

It turns out that Mitchell limits "known facts" or "factual events" (p. 9) to delineated physical occurrences such as "your mother died when you were five; your father lost his job, became depressed, and was treated with ECT" (p. 9). Psychological events such as "your mother withdrew from you when your younger sister was born; your father gave up hope and became demoralized; or your father tended to act seductively with you" (p. 9) do not qualify as "factual events," but rather are classified by Mitchell as "interpretations of complex interpersonal relationships" (p. 9).

At least two questions are in order. First, how does Mitchell know that "compelling and generative meaning systems do not work well if they are contradicted by known facts?" We certainly know that the ‘big lie’ and the meaning systems that are derived from the ‘big lie’ can, unfortunately, organize people’s experiences and belief systems, and can work exceedingly well for certain purposes.

Factual Events versus Complex Interpersonal Relationships

Second, why does Mitchell count as a "factual event" something like "your mother died when you were five" and not as a "factual event" something like "your mother withdrew from you when your sister was born"? Of course, the former is an easily ascertainable historical fact and the latter is complex and certainly far more difficult to ascertain. But that does not make the former a "factual event" and the latter a thing that is non-factual in nature. For either mother withdrew (or withdrew to this or that degree) or she did not.

Different aspects of a patient’s history may be associated with different methods and ways of knowing. But when something happened and what someone thought or felt about it are both facts of the world. Neither has an inferior existential status.

It appears that for Mitchell, experience in general, particularly complex thoughts and feelings such as emotionally withdrawing or becoming demoralized, cannot be "factual events" because, as he repeatedly tells us, they are "interpretively constructed." Hence, whereas Mitchell entertains the idea that the meaning systems offered by the analyst can be checked against past "factual events," he does not allow for a similar checking against past mental events, such as withdrawing and feeling demoralized because such events are part of the meaning system the analyst has constructed.

That Mitchell seems to reserve the status of factuality to physical events, such as your mother died when you were five years of age, and withheld that status from psychological phenomena is also reflected in his statement noted earlier that "to understand unconscious mental processes in one’s mind and that of another is not to expose something that has a tangible existence as one does in lifting a rock and exposing insects beneath" (pp. 17-18). But what would it mean for any mental content or process, conscious or unconscious, have a "tangible existence"? Would one see it, touch it, hear it? In a sense, then, it is, of course, true that mental events of any kind, conscious and unconscious, have no concrete "tangible existence" in the same way that chairs and tables, overt behaviors, and neurons do. When I see a red ball or feel tired or have the thought that it is warm today, only my report of my experience or my mental state (or some other observable behavior, such as yawning or taking my sweater off) has a "tangible existence," that is, can be directly observed. However, this does not mean that my seeing a red ball or feeling tired or having a particular thought is any less real than an ordinary physical event. Nor does it mean that the former are to be relegated to an inferior or more questionable existential or ontological state.

That psychological phenomena such as consciousness and conscious experiences do not have a "tangible existence" was essentially the ground on which behaviorists denied them the status of "factual events" and excluded them from the subject matter of psychology. It is ironic to find Mitchell, who almost certainly would not view himself as a behaviorist, adopting what is, in some respects, a classical behaviorist position. Thus, unless Mitchell wants to take a classical behaviorist stance - which, I assume, he does not - there is no clear import to the observations that "to understand unconscious processes… is not to expose something that has a tangible existence as one does in lifting a rock and exposing insects beneath" (pp. 17-18). One can just as readily and legitimately make the same statement with regard to conscious processes. The absence of a concrete and literal "tangible existence" in regard to mental events does not make them any less ‘existent,’ any less part of the world - something Searle (e.g., 1998) has repeatedly pointed out with regard to consciousness and conscious experience.

I hope I am making clear what appear to be Mitchell’s implicit views regarding the difference between physical and psychological events. Put very succinctly, for Mitchell, only physical events have factuality and "tangible existence," whereas mental events, possessing no independent factuality and reality, are constituted by "interpretive construction." In a certain sense, Mitchell adopts a variety of philosophical physicalism in which only ordinary physical events are real and have factuality, whereas mental events are formless epiphenomena which have no intrinsic character but rather can be "interpretively constructed" in many different ways. Hence, whereas meaning systems and interpretations need to be constrained by "factual events," they need not be similarly constrained by mental events since mental events are "interpretive constructions" that can presumably be tailored to fit any meaning system.

However, as Searle (e.g., 1998) has repeatedly pointed out, subjectivity and mental states are as much part of the ontology of the world as tables and chairs. Thus, whereas, say, being thirsty is irreducibly subjective, that I am thirsty is as much an objective fact of the world as the fact that there are tables and chairs. Of courses, the kinds of mental states in which analysts are interested are far more complex than the state of being thirsty and they present far more complex epistemological problems. However, epistemological issues should not be confused with ontological ones. Subjective mental states are no less objective facts of the world that concrete physical objects. As Larry Friedman (1998) has noted, the issue for many of what I call the ‘new view’ theorists is not uncertainty, but doubts about "factuality itself" (p. xvi).

Uncovering Truths versus New Meaning Systems as the Goal of Psychoanalysis

It is not clear that Mitchell needs to concern himself with "known facts" as constraints on interpretations. For, given the logic of his position, the only constraints on interpretations, the only way they need to be evaluated, is their therapeutic effectiveness.

It seems to me that, at rock bottom, Mitchell (as well as many other contemporary psychoanalytic writers) are essentially arguing that in contrast to the traditional psychoanalytic idea that learning the truth about oneself is a critical means of bringing about therapeutic change, they believe that new meaning systems, alternative perspectives, and related factors are the real agents of change. Hence, according to them, a central project of psychoanalysis is not uncovering truths about the patient’s mind but developing or constructing new meaning systems, perspectives, and narratives that presumably make for a better life. This is a relatively simple idea that does not, in itself, require philosophical discourses on the nature of the mind and of the possibility of discovering truths about the mind.

In an important sense, then Mitchell’s entire discussion of the nature of the mind is largely beside the point. In taking the position that he does, Mitchell echoes Renik (1996) and Schafer (1992) - simply substitute for new meaning systems, "alternative perspectives" and "retellings" or "constructed narratives."

It would be much cleaner, clearer, and consistent, if Mitchell, as well as Renik, Schafer, and as we shall see, Spence, dropped all talk about truth and knowledge. If the name of the game is, so to speak, the provision of meaning systems or alternative perspectives that make for a better life, then the central operative questions are whether patients do, in fact, adopt new meaning systems and new perspectives, whether this or that meaning system or alternative perspective does, indeed, make for a better life, what one means by a better life, and how one goes about determining or ‘measuring’ the degree to which such an outcome has or has not occurred.

The Nature and Goals of Psychoanalytic Treatment

At the heart of Mitchell’s position, as well as of much of contemporary psychoanalytic writing, is a disillusionment with the core traditional idea, linked to the "Enlightenment vision," that insight, awareness, and self-knowledge - that is, gaining access to one’s unconscious mental contents and modes of defense - constitutes an important route to liberation and therapeutic change. The disillusionment with the role of insight has, I believe, at least two sources. One is the familiar clinical experience that insight and awareness do not necessarily lead to significant therapeutic change. The other source for the disillusionment with the role of insight is a more epistemological one, namely, the chronic suggestion problem, that is, the question of whether interpretations that appear to generate insight really correspond to what is going on in the patient’s mind, that is, "tally with something real in the patient."

By adopting the framework offered by Mitchell (as well, as we shall see, by Renik), the analyst is freed of the burden that accompanies the claim that his or her interpretations tally with what is real in the patient, that is, that he or she uncovers or discovers anything.

Thus, in one conceptual fell swoop analysts can unburden themselves of a dogged epistemological problem that they have been unable to deal with effectively. By relinquishing the idea that analysts uncover anything in the patient’s mind and adopting a thoroughly constructivist position analysts can avoid the discomfort of what Mitchell (1998) calls the "Grünbaum syndrome," and can assure themselves of the "almost total irrelevance to contemporary clinicians" (p. 5) of Grünbaum’s critique. However, as Friedman (1998) points out, if one accepts this view, then analysis "is, indeed, an adventure of a vastly different sort than we have imagined" (p. 260). That it is hard for even Mitchell and Renik to imagine how an analyst can work who no longer believes in hunting for something that is already there is suggested by two pieces of evidence. One is the disjunction between the conceptual stance taken by Mitchell (and, as we shall see, Renik) and the clinical material presented. And the other piece of evidence is the need to preserve the idea that, after all, psychoanalysis does deal in the currency of truth and objectively even if that requires some rather odd redefinitions of these concepts. I will discuss each of these issues in turn.

Disjunction between Conceptual Stance and Clinical Material

Although maintaining that there are no mental processes corresponding to the phrase "in the patient’s mind" about which one can be right or wrong and after arguing that there are no "preorganized" central dynamics in the patient’s mind, Mitchell nevertheless proceeds to interpret his patient Robert’s dream in terms of "his struggles with his son [that] were in some measure reflective of struggles with a part of himself that had been buried" (p. 23). He also notes that the patient’s father "was internalized by him in a complex fashion" (p. 23) and notes the patient’s struggle "with the sense that he has tragically mutilated his own inner resources and potentiality" (p. 24).

If these struggles of the patient do not qualify as preorganized central dynamics, it is difficult to know what would qualify. Similarly, if these struggles and internalizations that Mitchell identifies do not refer to mental processes "in the patient’s mind" about which one could be right or wrong, it would be difficult to know what would qualify as "patient’s mental processes."

Another example of a disjunction between the conceptual stance taken and the clinical material presented is also seen in Renik’s (1996) paper "The Perils of Neutrality." Similarly to Mitchell, he talks about providing new perspectives rather than uncovering. He then describes an interpretation that he makes to a patient, noted earlier, that her hostility and guilt toward her sister served as a defense against being critical towards her parents. On the one hand, this interpretation entails taking a new perspective toward the patient’s longstanding hostility and guilt towards her sister: something like, "why don’t you look at your anger and guilt toward your sister in this way?" On the other hand, however, the interpretation also speaks to the uncovering of the patient’s hitherto inaccessible, angry, and critical feeling toward her parents. Thus the interpretation is not simply "why don’t you take this perspective toward your anger at your sister" but also "there is more to your anger and guilt toward your sister than meets the eye," more to be uncovered.

Without the assumption that the patient unconsciously harbors critical feelings toward her parents, that is, without the assumption that such feelings are "in the patient’s mind," Renik’s interpretation becomes, in essence, nothing more than "look at your hostility and guilt toward your sister this new way - it will help you feel better." Unless Renik’s interpretation has something to do with what the patient actually thinks and feels - in this case, harboring critical feelings toward her parents - it is totally arbitrary. Without any concern that one’s interpretation correspond in some way to what is ‘there’ in the patient’s mind, a whole range of interpretations or new perspectives or new meaning systems could conceivably help the patient feel better. How does this view differ from suggesting, say, becoming a born-again Christian or an orthodox Jew in order to feel better?

The other way that ‘new view’ theorists exhibit their discomfort with a complete dismissal of the question of whether analytic interpretations tally with something real in the patient, that is, possess truth and objectivity, is to redefine these concepts. Thus, Renik’s (1998) solution to the uncomfortable possibility that new perspectives that are (purportedly) effective in analysis may have little or nothing to do with the truth is simply to redefine truth so that "what is true is what works" (p. 492) - when it is clear that what he means by "what works" is that which increases the patient’s happiness.

As for the issue of objectivity, his answer to the question of "how can an analyst be objective if an analyst is irreducibly subjective" (p. 491) simply "lies in recognizing that in analysis, as everywhere in life, observations of reality are constructs, formed in relation to specific subjective interests" (p. 491).

Renik then provides an example of observations being determined by subjective interest by contrasting a hiker’s observations of the sun rising in the east and setting in the west with the astronomer’s observations which "show that the sun does not move through the sky at all" (p. 492). Renik’s conclusion is that "[t]hough it could be argued that the hiker’s and the astronomer’s observations are in contradiction, we can also say that they are both objective, each in relation to a different subjective interest" (p. 492). Note, by the way, the parallel between Renik’s blurring of the distinction between subjective and objective and the postmodern insistence that supposed truths are no less subjective values than any other beliefs or opinions. Renik’s reasoning here perverts the ordinary and basic difference between a subjective phenomenal experience and a proposition about the world. Apply Renik’s reasoning to a hiker who experiences the earth as flat - which, of course, is the way we all do experience the earth in our daily interactions with the world. According to Renik’s logic, the hiker or anyone else who experiences the world as flat and the astronomer whose observations show that the world is round "are both objective, each in relation to a different subjective interest." But this conclusion is, of course, mistaken - the earth is round, not flat and many a hiker walking on his or her flat path knows that as well as the astronomer. Perhaps what Renik wants to say is that for the purpose of the "subjective interest" of negotiating our ordinary everyday activities, the experience of the earth as flat, or even the belief that it is, is perfectly adequate - it works for our ordinary purposes. However, it turns out that our subjective experience of the earth as flat is not an adequate basis for determining the objective nature of the shape of the earth as a whole. These two examples neatly demonstrate the difference between a heuristic that ‘works’ and a proposition that is true. Thus, although the heuristic that the earth can be considered as flat works for many purposes, it is literally false.

In an apparent recognition that there might be some difficulties with his equation of what works (i.e., what helps the patient feel better) with what is true and objective, Renik raises the question of whether he would say that a patient is objective if he has a grandiose delusion that makes him happy. His answer is "my experience is that delusional grandiosity does not, in the long run, make people happy… if the patient’s happy self-satisfaction did, in fact, seem to last, however, I would feel obliged to review my perception of the patient as delusionally grandiose" (p. 494). So intent is Renik on clinging to his idiosyncratic conception of objectivity that if the (now only apparent) delusion made the patient happy for a sustained period of time (how long a period?), he would feel obliged to believe that the patient was being objective rather than delusional. What if someone has a blatant delusion, say, that he is the King of England, and what if this delusion makes him happy over a long period of time? Would Renik feel obliged to review his perception of the person as delusional? Would he not want, instead, to review his assumption that delusional grandiosity cannot ever, in the long run, make anyone happy? Otherwise, he is forced to say that what appeared to be a delusion was not, after all, really delusional. Would he then conclude that the (formerly deluded) patient was the King of England? These are the kinds of absurdities to which Renik’s reasoning leads.

The reluctance to relinquish the idea that psychoanalysis has at least something to do with the ‘truth of the matter’ is also reflected in the writings of Spence (1982) and Schafer (1992). Thus, Spence introduces the concept of "narrative truth" which, it turns out, essentially refers to the persuasiveness of a narrative. Since we know that many things can be persuasive without being true, it is not clear why the term "narrative truth" is employed. But the concept caught on, in part, I suspect, because the use of the term "truth" disguised somewhat, what is, after all. Spence’s main idea, namely, that the currency of psychoanalytic narratives lies primarily or entirely in their persuasiveness. By using the term "narrative truth" one could convince oneself that one was in the business of some kind of truth rather than mere rhetoric. But this is not just a harmless matter of choice of words. For, whether or not intended by Spence, the concept of "narrative truth," reflects the postmodern insistence that "all distinctions between truth and falsehood must be dissolved" and that supposed truths are really based on rhetoric and "persuasive power." (Vattimo, 1985).

One finds a similar reconceptualization of terms in Schafer’s (1992) idiosyncratic definition of insight as referring to "those retellings that make a beneficial difference in a person’s construction and reconstruction of experience and adaptive conduct of life" (p. v). But what if a blatantly false or fictitious "retelling" made a "beneficial difference" in a patient’s life? Schafer’s redefinition of insight is a way of saying that insight does not matter - what matters is any "retelling that makes(s) a beneficial difference." Thus, Schafer’s redefinition of insight is a variant of Spence’s "narrative truth" and Renik’s "what is true is what works."

Whatever other differences exist among Mitchell, Renik, Schafer, and Spence, they share in common the view that psychoanalysis is in the business of generating meaning systems, perspectives, interpretations, constructions, narratives - whichever term one prefers - that will make a "beneficial difference" or that will lead to a richer and less self-sabotaging existence. All the other talk, including labored definitions or redefinitions of the nature of mind, of truth, of insight, and of objectivity, is really beside the point. They can dispense with all that. Although the logic of their position suggests that the truth of the narrative - that is whether it corresponds to something real in the patient - is largely irrelevant and although they come close to saying just that, they cannot seem to take the final plunge into postmodernism and instead engage in conceptual gymnastics.

I suspect that in addition to not wanting to appear "irresponsible relativists," another reason that the ‘new view’ theorists cling to the idea that they are concerned with truth, objectivity, and so on, is that, when they enter treatment, virtually all patients assume that they will gain self-knowledge and learn more of the truth about themselves in psychoanalysis. If this assumption is simply an illusion, why are patients not disabused of this notion? As Sass (1992) asks, is the presumed impossibility of determining the veridicality of interpretations a secret being kept from patients? Do patients really believe that there is nothing corresponding to the phrase "in my mind" about which they or their analysts can be right or wrong? Do patients accept the goal of becoming more "relativistic historians" of their lives (Schafer, 1980)? When Renik makes the interpretation to his patient that her anger at her sister is partly a defense against feeling critical and angry towards her parents, does she interpret this intervention as merely a ‘new perspective’ that might ‘work’ for her or does she understand Renik to be saying that she actually and in fact harbors angry and critical feelings towards her parents? One wonders how long patients would remain in treatment if they were told that the analyst is not interested in uncovering and discovering truths, but rather in offering "aesthetic fictions" (Geha, 1984), coherent narratives, or the like. One also wonders how well analysts could conduct treatment if their general philosophical ideas that appear in journal articles and books fully infiltrated their day-to-day clinical work.

The reluctance of the ‘new view’ theorists to relinquish any concern with truth may be based on their intuition that it is difficult to understand how the psychoanalytic project can survive when the "interpretive constructions," meaning systems, narratives, perspectives, and so on, that are formulated in psychoanalysis, are freed from any links to something real in the patient. It seems to me that the tortured redefinitions of truth and objectivity are the result of an attempt to preserve the psychoanalytic project while undermining its basic assumptions.

Summary and Conclusions

I have tried to show that in reacting against some inadequacies of the traditional view of the analytic situation, some contemporary writers have taken positions that are untenable. These positions do not constitute an adequate basis for an understanding of the analytic situation, for an understanding of the patient’s mind, and for the further development of psychoanalytic theory.

I have also tried to show that in this work, inspired by postmodernism, the sharp dichotomy between uncovering and discovering on the one hand and developing new perspectives and meaning systems on the other is unwarranted. Rather, the two go hand in hand. There should be a way to deal with the deficiencies of traditional theory that does not entail the kind of radical swing of the pendulum that generates at least as serious difficulties as are present in the traditional view. Finally, there is no evidence at all that the position advocated by the ‘new view’ theorists leads to more effective treatment, or to needed types of research on psychoanalytic treatment.

If "external realism" is characterized by the assertion that there is an external world that exists independently of us, then, I would suggest, one might define ‘internal realism’ as asserting that mental states in other exist independently of us. That social influences (both developmentally and contemporaneously) may shape mental states and that it may be impossible to develop mental states, as we understand them, without social interaction, does not alter the assertion that they exist independently of us and our "interpretive constructions." Indeed, to argue that social factors influence and shape our mental states is to acknowledge that the latter have an independent reality. Similarly, that it may be difficult, sometimes impossible, to ascertain another’s mental states and that all we can do is infer them, does not alter the fact that they exist independently of our "interpretive constructions." Indeed, the thrust of this paper has been to argue that if they are not simply arbitrary or intended as therapeutic fictions, then our "interpretive constructions" are evaluated in terms of whether they tally with something real in the other. This seems to me the heart of the matter. It is difficult for me to see how any psychoanalytic program, whatever the particular psychoanalytic ‘school,’ can operate that does not accept ‘inner realism,’ that is, the idea that the patient’s mental states exist independently of the analyst’s "interpretive constructions" and that a central task of that program is to discover and understand the actual nature of these mental states.


Cavell, M. (1998) In response to Owen Renik’s ‘The analyst’s subjectivity and the analyst’s objectivity’. International Journal of Psychoanalysis,79, 1195-1202.

Eagle, M.N. (1984). Recent Developments in Psychoanalysis. A Critical Evaluation. New York: McGraw-Hill (reprint: Cambridge, MA: Harvard Univ. Press, 1987). Internet edition of chapter 12:

Eagle, M. N. (1987) The psychoanalytic and the cognitive unconscious. In, R. Stern (Ed.) Theories of the unconscious and theories of the self. Hillsdale, N.J.: The Analytic Press, pp. 155-189.

Friedman, L. (1998) Overview. In O. Renik (Ed.) Knowledge and authority in the psychoanalytic relationship. Northvale, N.J.: Jason Aronson, pp. VIII-XXII.

Freud S. (1887-1902 [1950]). The Origins of Psychoanalysis. Letters to Wilhelm Fliess, Drafts and Notes: 1887-1902 (Bonaparte M., Freud A., Kris E., editors). New York: Basic Books, 1950 (London: Imago Publishing Co., 1954).

Freud, S. (1916-1917) Introductory lectures on psychoanalysis, Part 3. Standard Edition, 11, 243-496. London: Hogarth Press (1961).

Geha, R.E. (1984) On psychoanalytic history and the "real" story of fictitious lives. International Forum of Psychoanalysis, 1, 221-229

Gill, M.M. (1982) The analysis of transference, Vol. 1: Theory and technique. New York: International Universities Press.

Gill, M.M. (1994) Psychoanalysis in transition. Hillsdale, N.J. : The Analytic Press

Green, A. (1986) The private madness. London: Hogarth Press.

Grünbaum A. (1984). The Foundations of Psyhoanalysis. A Philosophical Critique. Berkeley, CA: Univ. of California Press.

Holzman, P.S. (1985) Psychoanalysis: Is the therapy destroying the science? Journal of the American Psychoanalytic Association, 33, 4, 725-770.

Kirshner, L.A. (1999) Toward a postmodern realism for psychoanalysis. Journal of the American Psychoanalytic Association, 47, 2, 445-463.

Meissner, W.W. (1998) Review of S.A. Mitchell’s "Influence and autonomy in psychoanalysis". Psychoanalytic Books, 94, 419-423.

Mitchell, S.A. (1998) The analyst’s knowledge and authority. Psychoanalytic Quarterly, 67, 1, 1-31.

Mitchell, S.A. (2000) Reply to Silverman (2000). Psychoanalytic Psychology, 17, 1, 153-159.

Neisser, U. (1967) Cognitive psychology. Englewood Cliffs, N.J.: Prentice-Hall.

Neisser, U. (1976) Cognition and reality. San Francisco, CA: W.H. Freeman & Co.

Ogden, T. (1997) Reverie and interpretation: Sensing something human. Northvale, N.J.: Jason Aronson.

Piaget, J. (1954) The construction of reality in the child. New York: Basic Books.

Renik, O. (1993) Analytic interaction: Conceptualizing technique in light of the analyst’s irreducible subjectivity. Psychoanalytic Quarterly, 62, 553-571.

Renik, O. (1996) The perils of neutrality. Psychoanalytic Quarterly, 65, 495-517.

Renik, O. (1998) The analyst’s subjectivity and the analyst’s objectivity. International Journal of Psycho-Analysis, 79, 3, 487-498.

Sass, L.A. (1992) The epic of disbelief: The postmodernist turn in contemporary psychoanalysis. In, S. Kvala (Ed.) Psychology and postmodernism. London: Sago, pp. 166-182.

Schafer R. (1980). Narrative Actions in Psychoanalysis (Heinz Werner Lecture Series, 14). Worcester, MA: Clark Univ. Press, 1981.

Schafer, R. (1992) Retelling a life. New York: Basic Books.

Searle J. (1998) Mind, language, and society: Philosophy in the real world. New York: Basic Books.

Silverman, D.K. (2000) An interrogation of the relational turn: A discussion with Stephen Mitchell. Psychoanalytic Psychology, 17, 1, 146-152.

Smith, H.F. (1999) Subejctivity and objectivity in analytic listening. Journal of the American Psychoanalytic Association, 47, 2, 465-484.

Spence, D.P. (1982) Narrative truth and historical truth. New York: Norton

Stern, D.B. (1999) Unformulated experience: From familiar chaos to creative disorder. In, S.A. Mitchell & L. Aron (Eds.) Relational psychoanalysis: the emergence of a tradition. Hillsdale, N.J.: The Analytic Press, pp. 77-107.

Stern, D.B. (1989) The analyst’s unformulated experience of the patient. Contemporary Psychoanalysis, 25, 1-33.

Stone, L. (1981) Noninterpretive elements in the psychoanalytic situation. Journal of the American Psychoanalytic Association, 29, 69-118.

Vattimo, G. (1985) The end of modernity. Baltimore, MD: John Hopkins University Press.

Weiss, J. and Sampson, H. (1986) The psychoanalytic process: Theory, clinical observations, and empirical research. New York: Guilford Press.

[Note: An Italian version of this paper was presented on May 13, 2000, at the Associazione Studi Psicoanalitici (ASP) of Milan and on May 20, 2000, at the University of Florence; an Italian translation, titled "La svolta postmoderna in psicoanalisi", appeared in Psicoterapia e scienze umane, 2000, XXXIV, 4: 5-44. A later version of this paper, authored by Eagle M.N., Wolitzky D.L., & Wakefield J.C., and titled "The analyst's knowledge and authority: a critique of the 'new view' in psychoanalysis", appeared in the Journal of the American Psychoanalytic Association, 2001, 49, 2: 457-89. Another version will appear in Psychoanalytic Psychology]

Morris N. Eagle, Ph.D.
Derner Institute, Adelphi University
Garden City, New York 11530
Tel. (00) (1) (516) 877-4809
440 Riverside Drive, # 8
New York, NY 10027-6829
Tel. (00) (1) (212) 865-3574
Fax (00) (1) (516) 749-4525
E-Mail <>

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