Influence and Autonomy in Psychoanalysis, By Stephen A. Mitchell.
Hillsdale, NJ: The Analytic Press, 1997.

Reviewed by Philip Ringstrom, Ph.D., Psy.D.

In this landmark volume Stephen Mitchell draws into his crosshairs the central target of psychoanalytic discourse since Freud's first attempts to develop a clinical methodology free of the direct suggestion of hypnosis.  The question Freud undertook to answer is how the analyst can become as little influential as possible in the development of the patient's autonomous sense of self.  This is a noble part of psychoanalytic heritage: no other theory or practice of mental health so occupies itself with this ultimate quest for untainted liberation.  Nonetheless, the history of answers to this question has been problematic.  Freud's greatest discovery, of the patient's unconscious, has subsequently become recognized not as a "bug under a rock" waiting to be discovered, but as a mental process intricately intertwined in the complexity of the minds of both analyst and analysand.  From this vantage point we see the interplay of two unconscious processes that often results less in discovery than in a collaborative investigation that creates simultaneously what is being discovered.  In the very act of "symbolizing" the "symbol," process and product, intimately entwined, are codetermined.  On the positive side of this humbling revelation is the fortuitous result that in a successful analysis not one but both parties' minds must develop beyond their constraints.

What has been missing heretofore, however, is a language that truly captures the complexity of the analytic interaction.  It is in this vein that Mitchell, in his latest book, seeks to make a contribution.  What does it really mean, for instance, that the analyst is a participant observer?  What does it mean that every analytic coupling is a unique dyad?  What does it mean when the analyst's subjectivity, and therefore its influence, are seen as "irreducible"?  Or when it is said that transference and countertransference are two faces of the same dynamic?

Mitchell wants us to see that influence is something not to be eschewed, but rather to be embraced and seriously studied.  For him, the best way to do that is to free psychoanalysis from the encumbrance of what he argues have come to be recognized as ideals impossible to achieve--anonymity, abstinence, and neutrality, to mention but a few--and to place in their stead a methodology of self-reflection.  To get to this point, Mitchell first builds his argument by examining some of the efforts already made in this direction, efforts that by virtue of their partial success, warrant the further effort of this book.

Mitchell begins by pointing out that there has been a problem in the way we have conceptualized therapeutic action, and that interpretations have failed so often because they have tended to repeat in the analytic relationship the very qualities being interpreted--as, for example, when the interpretation of a patient's shame produces the effect of shaming him, irrespective of the analyst's intent.  Mitchell notes that this situation may sometimes be corrected by invoking the contemporary Kleinian solution of interpreting how the patient processes the interpretation.  But as helpful as this tactic can be, it, like most, does not fit every occasion.  Indeed, Mitchell argues that there is no platform on which we can stand to gain a clear vision of what is occurring, one that would consistently allow us to escape this dilemma.  "Interpretations fail," he writes, "because the patient experiences them as old and familiar modes of interaction.  The reason interpretations work, when they do, is that the patient experiences them as something new and different, something not encountered before. " (p. 52)

In the interpersonalist movement of Sullivan, Fromm, and Thompson, Mitchell finds a real attempt to recognize, in both "ecological" and "participatory" principles, that "the environment plays a crucial role in creating, shaping, and maintaining personality and psychopathology" and that "the psychoanalyst is never simply an observer of the data that the patient provides but is always also a participant in co-creating that data" (p. 64).  But even the early interpersonalists, Mitchell points out, failed to recognize that participation and observation are never separable.  Sullivan's own notion of the "parataxic distortion" holds sacrosanct the idea that even though the psychoanalytic process is a profoundly interpersonal one, the analyst remains the arbiter of what is real and what is distorted.  As such, the analyst is seen as needing to be firmly in control of the analysis and, except for "inevitable periodic departures," should never be surprised by what emerges.

While contemporary interpersonalists such as Levenson have challenged their forebears, the tendency remains for them to believe that the analyst, despite participating in what unfolds, can step back sufficiently from his or her influence on the analytic relationship to see the same patterns of interaction that any analyst would view as plaguing the patient's life.  Even in Ehrenberg's innovative work on engaging the "intimate edge," a technique whereby the analyst repeatedly confronts patients about their interpersonal style of engagement with the analyst, Mitchell raises the question of who defines the edge.  "Edges," he writes, "can be defined only within perspectives, and interactions, because they involve at least two parties, contain, by definition, more than one perspective" (p. 93).  In the end, while Mitchell recognizes the interpersonalists' attempts to assiduously study, account for, and mitigate analyst influence, he questions whether talking about the analyst's influence can ever really eliminate or even reduce it.

The contemporary Kleinians are still another group that has made important inroads in exploring the influence of the analyst on the analytic process.  Coming out of a tradition that once hewed to an exclusively one-person psychology-- for example projective identification was seen by Klein as an exclusively one-person fantasy-- the contemporary Kleinians ingeniously spelled out the interpersonal influences under which the analyst may in fact come to identify with the analysand's projections.  Indeed, in identifying with them, the analyst comes to experientially understand aspects of the patient's primitive organization, thus furthering a capacity for containment of disruptive affects, as well as a deepening of the patient's empathic engagement.  In addition to this evolution in their ideas, the contemporary Kleinians extricated their methodology from its so-called "wild analytic" past, in which interpretations of unconscious material were presumably made without respect to the timing that was so much a consideration of the ego psychologist.  The Kleinians came to recognize that rapid and early interpretations were likely signifiers of "the analyst's anxiety and difficulty in containing the patient's terror and fantasies" (p. 125).  This truly signified the maturation of the Kleinian model along relational lines.

Yet Mitchell observes that despite these radical shifts within their tradition, the contemporary Kleinians maintain the analyst's position of final authority over the veridicality of the patient's narrative.  Mitchell speculates that this is because the Kleinian loyalist believes that the relationship of analyst and patient as breast to baby is not metaphor at all, but that "at the deepest unconscious levels, the analyst is the breast for the patient and the analyst's interpretations are good or bad milk or both"(p. 135).  This traditional position holds that even prior to contact the analyst knows that because of the patient's "anxieties over his dependency on that primary object, the patient will do anything and everything to destroy its goodness" (p. 135).

Hence the Kleinian is admonished to "stand firm" against such undermining.   Interaction is automatically interpreted (if at first only privately to oneself) under a "policy of suspicion" (p. 138).  To spoil the analytic function, upon which the analysand loathes feeling dependent, attempts will be made, it is thought, to "draw the analyst into noninterpretive interactions" (p. 136).  Mitchell argues that to pursue such a policy places the contemporary Kleinian at odds with the emergence of recent theories of constructivism, perspectivism, and hermeneutics.  The irony in all of this, he believes, is that adherents of such a position are hardly equipped to claim freedom from influence, since they view every aspect of analytic interaction as having a particular valence even prior to its emergence.  Thereafter, they hold, the analyst's job is to track it and point it out to the patient, even if these days this is done with considerably greater sensitivity to tact and timing.

It is in this context that Mitchell asserts that there is no issue today more important for contemporary psychoanalysis than the definition of knowledge and authority.  He warns that psychoanalysts have abused their respect for science by converting it into "scientism."  The former he says, refers to "the accumulation of certain kinds of knowledge by way of certain methods":  the latter to "the belief that such knowledge will tell us all we need to know about human experience, meaning and values" (p. 205).  In our overzealous embracing of the latter, we have at times developed an inflated estimation of our capacity for objectivity.  But Mitchell cautions also against a reactionary tendency in contemporary psychoanalysis to celebrate total subjectivism and facile relativism.

Instead Mitchell admonishes us that self-discipline and clinical responsibility grow out of a respect for the intersubjective nature of psychoanalysis.  "The patient's autonomy," he writes, "is more honestly and meaningfully protected through the acknowledgment of the analyst's influence than through claims to illusory objectivity" (p. 221).  In chapter 5, "Varieties of Interaction, " he supports this argument by reviewing the uniquely different ways in which the "unbidden qualities of countertransference are used by Theodore Jacobs, Darlene Ehrenberg, and Thomas Ogden, three practitioners coming, respectively, out of Freudian, interpersonalist, and object relations traditions.  As for the first of these figures, Mitchell sees him as summoning (though not actively) "ghosts of Jacobs past," using memories of episodes with significant others to guide him in deciphering what the patient is unconsciously communicating.  By contrast, Ehrenberg is much more grounded in the present.  She is stirred "because she wants something from or is threatened by something with this patient right now" (p. 148).  Finally, Ogden's version of countertransference explores neither Jacob's resonating world of "ghosts" nor Ehrenberg's two separate subjects, "but rather the generation of a uniquely combined subjectivity" (p. 151) wherein the therapist is a recipient and processor who evokes, contains, and gradually symbolizes the "deepest, most primitive features of the analyst's own internal object world.  The kind of analytic presence that is possible with each patient in any particular hour is an extension of the analyst's relationship to his or her internal world" (p. 154).

Not surprisingly, among these three there are dramatic differences regarding self-disclosure of countertransference reactions.  Because Jacob's associations are ghosts from the past, there is seldom rationale for introducing them.  Similarly, Ogden would find his own disclosure "likely to serve as a distraction and an escape from the silent and often lonely work of struggling with the demons that have been stirred up" (p. 154).  For Ehrenberg, however, since her "upset" is in the present, not to share the patient's impact on her, or what it is like to be in the patient's presence, would be tantamount to "a contrived withholding" (p. 150).

Mitchell's invocation of three master clinicians he greatly respects, however, is intended not simply to demonstrate three variants of the relational approach, but to indicate that in each, as in many more, there are styles from which others may borrow.   No one of them should be considered veridical in determining the truth, but all should be available for the analyst to use during the course of an analysis.  Though Mitchell eschews "eclecticism or even model-mixing," he does argue that clinicians should mold their own "distinctly personal model" (p. 195), taking advantage of the "conceptual cornucopia" contemporary psychoanalysis offers.

This line of argumentation lays bare what is perhaps the core of the book, which is presented most succinctly in chapter 6, "The Analyst's Intentions."  There Mitchell argues that the historical ideals of psychoanalytic practice have fallen in consequence of questions regarding "objectivity," the proliferation of competing psychoanalytic theories, the gradual admission by all analytic tendencies that countertransference is less aberrant than normal, and the general recognition that anonymity can not be preserved, nor all forms of gratification avoided, despite all efforts to do so.  In short, Mitchell argues, abstinence, anonymity, neutrality, flawless empathy, containment, listening without memory or desire, and authenticity are no longer ideals to be pursued.  Mitchell acknowledges that some mainstream analysts, such as Kernberg, have come a long way in agreeing that anonymity is not only impossible, but that it "actually does great mischief because it pressures the analysand to agree that the analyst operates as an interpretive function, a faceless oracle" (p. 180).   Still, like Lawrence Friedman, Kernberg holds that neutrality, though not achievable, remains an ideal worth striving for.  In strong protest, Mitchell, in agreement with Owen Renik and Irwin Hoffman, declares that trying to do something that is impossible is not the most useful framework for the analyst's efforts.  Indeed, it can erroneously lend weight to the assumption that the analyst can retain a perch independent of the interaction of which he or she is a part, and from that position assume veridical authority.

One may rightly ask Mitchell what, if we cast off our ideals, we are to strive for.   What prevents psychoanalysis from becoming a directionless, "anything goes"  modality?  In this Mitchell would answer that while we should avoid such ideals, we cannot avoid intention.  Indeed, being intentionally intentionless is a paradox.  Mitchell reminds us that Bion, to practice "without memory or desire," had to both remember and want to do so.  Mitchell writes, "I would describe the intention that shapes my methodology as a self-reflective responsiveness of a particular (psychoanalytic) sort.  In putting it this way, I am suggesting that my way of working entails not a striving for a particular state of mind, but an engagement in a process" (p. 193).  Mitchell's elaboration of this idea is worth quoting at length:

This kind of analytic participation is neither simple nor naive; self-reflective responsiveness to the patient is a highly cultivated skill.  It assumes that the mind of the analyst, like that of the patient, is characterized by shifting, discontinuous self-states and self-organizations;  it presumes that mind is generated into interpersonal fields of reciprocal influence; it presumes that self-reflection is itself always, necessarily, perspectival and highly selective.  It takes a long time to learn to experience and use oneself in this fashion.  Doing so involves hearing and following different levels of meaning at the same time, something like the way a simultaneous translator learns to hear one language and speak another or the conductor of an orchestra can hear separate lines that generate polyphony.  The analyst, in this view of analytic process, learns to track and engage in simultaneously, different lines of thought, affective responsive, self-organization [p. 194].

Here Mitchell underscores his personal recommendation for psychoanalysis, which resurfaces throughout the book.  It is that each of us entertain multiple self-organizations, no singular one.  Thus, in the course of his work with a patient, Mitchell allows various of these organizations to arise in interaction with the patient and to "inform" him of something about both the patient and himself.   "In actuality," Mitchell writes, "these choices go by too fast to allow prior conscious reflection on all of them.  But I try, retrospectively, to articulate the implicit, preconscious nature of these choices" (p. 196).  On the same page, Mitchell poses seven self-reflective questions, too long to list here, that seem to ask, Is what we just did here today opening or closing, vitalizing or deadening, connecting or avoiding, focusing or confusing, liberating or constraining, playfully exploring or just fooling around?  "Thus," he concludes, "one of my fundamental concerns is to shape my participation in a fashion that minimizes my constraints on the patient's range of experience and to help her to minimize her own constraints" (p. 197).

The multiple self-organization model appears again in chapter 8, "Gender and Sexual Orientation in the Age of Postmodernism."  For Mitchell, "theorizing about gender is the perpetual dialectic between biological/essentialist accounts and constructivist accounts," the former rooted in nature and the latter "purely a socially constructed category"(p. 24).  Like language in Chomsky's account, gender and sexuality are biological, insofar as all cultures exhibit them; but, again like language, they are decidedly shaped by the culture in which they emerge.  Thus, one need not feel compelled to choose between biological and constructivist models of gender, but may instead regard them as "a helpful tension that perpetually generates new forms of organizing experience, a kind of potential space that is particularly well suited to the analytic process, in its continual reworking of past and present fantasy and actuality, internal and external, to generate new meanings" (p. 245).  In this manner, the question of gender orientation for both analyst and analysand is not foreclosed, but to develop a truly personal sense of the meaning of each one's sense of gender and sexuality, each must ask what constraints anatomy, temperament, and developmental history pose in forming the identifications and choices that create a gendered self-experience.

These days, psychoanalysis seems occupied with two central questions.  First, does the clinical material emanating from the patient's unconscious seek primarily a reparative, developmentally constituting response from the analyst, or is it more in the service of defensively repetitive styles of relating that inhibit the very growth the patient ostensibly seeks?  Second, to what extent is the analyst a significant participant in the shaping of the answer to the first question?  Readers for whom these questions resonate should consider Stephen Mitchell's latest a "must read."  Mitchell captures, in a manner both congenial and erudite, the extraordinary complexities involved in discerning whether the patient at any given time is better engaged by developmental arrest models of psychoanalysis or by relational conflict models.  Often, he argues, we just don't know, and the analyst must be willing to wade into the mix to discover the answer.  Psychoanalysis, he believes, is not a province susceptible to explanation in terms of any single theory: rather, it must be open to analysts' engagement with many, with the aim of finding those that work best in given moments in a specific analytic relationship.

Philip A. Ringstrom
5004 Haskell Avenue
Encino, CA 91436
Fax 818-906-3269
E-mail: Ringsite@aol.com

Ringstrom, P. (1998). Review of Influence and autonomy in psychoanalysis., J. Amer. Psychoanal. Assn., 46:619-627.

Return to DSPP's Home Page