Stern (2010) Partners in Thought - Working With Unformulated Experience, Dissociation, and Enactment

Partners in
Thought
Psychoanalysis in a New Key Book Series

Volume 12
PSYCHOANALYSIS IN A NEW KEY BOOK SERIES
Donnel B. Stern
Series Editor
When music is played in a new key, the melody does not change, but the
notes that make up the composition do: change in the context of conti-
nuity, continuity that perseveres through change. “Psychoanalysis in a
New Key” publishes books that share the aims psychoanalysts have always
had, but that approach them diἀerently. The books in the series are not
expected to advance any particular theoretical agenda, although to this
date most have been written by analysts from the interpersonal and rela-
tional orientations.
The most important contribution of a psychoanalytic book is the com-
munication of something that nudges the reader’s grasp of clinical theory
and practice in an unexpected direction. “Psychoanalysis in a New Key”
creates a deliberate focus on innovative and unsettling clinical thinking.
Because that kind of thinking is encouraged by exploration of the some-
times surprising contributions to psychoanalysis of ideas and findings
from other fields, “Psychoanalysis in a New Key” particularly encourages
interdisciplinary studies. Books in the series have married psychoanalysis
with dissociation, trauma theory, sociology, and criminology. The series is
open to the consideration of studies examining the relationship between
psychoanalysis and any other field—for instance, biology, literary and art
criticism, philosophy, systems theory, anthropology, and political theory.
But innovation also takes place within the boundaries of psychoanaly-
sis, and “Psychoanalysis in a New Key” therefore also presents work that
reformulates thought and practice without leaving the precincts of the
field. Books in the series focus, for example, on the significance of per-
sonal values in psychoanalytic practice, on the complex interrelationship
between the analyst’s clinical work and personal life, on the consequences
for the clinical situation when patient and analyst are from diἀerent cul-
tures, and on the need for psychoanalysts to accept the degree to which
they knowingly satisfy their own wishes during treatment hours, often to
the patient’s detriment.
PSYCHOANALYSIS IN A NEW KEY BOOK SERIES
Donnel B. Stern
Series Editor
Vol. 12 Vol. 6
Partners in ἀ ought: Wounded by Reality:
Working With Unformulated Experience, Understanding and Treating
Dissociation, and Enactment Adult Onset Trauma
Donnel B. Stern Ghislaine Boulanger
Vol. 11 Vol. 5
Heterosexual Masculinities: Prologue to Violence:
Contemporary Perspectives From Child Abuse, Dissociation,
Psychoanalytic Gender ἀ eory and Crime
Bruce Reis & Robert Grossmark (eds.) Abby Stein
Vol. 10 Vol. 4
Sex Changes: Prelogical Experience:
Transformations in Society An Inquiry Into Dreams
and Psychoanalysis and Other Creative Processes
Mark J. Blechner Edward S. Tauber & Maurice R. Green
Vol. 9 Vol. 3
ἀ e Consulting Room and Beyond: ἀ e Fallacy of Understanding
Psychoanalytic Work and Its and the Ambiguity of Change
Reverberations Edgar A. Levenson
in the Analyst’s Life
Therese Ragen Vol. 2
What Do Mothers Want? Contemporary
Vol. 8 Perspectives in Psychoanalysis
Making a Difference in Patients’ Lives: and Related Disciplines
Emotional Experience Sheila F. Brown (ed.)
in the ἀ erapeutic Setting
Sandra Buechler Vol. 1
Clinical Values:
Vol. 7 Emotions ἀ at Guide
Coasting in the Countertransference: Psychoanalytic Treatment
Conflicts of Self Interest Sandra Buechler
Between Analyst and Patient
Irwin Hirsch
Partners in
Thought
Working with Unformulated
Experience, Dissociation,
and Enactment
Donnel B. Stern
New York London

Chapter 3 originally appeared in Psychoanalytic Dialogues, 13, 843–873, 2003.
Chapter 4 originally appeared in Contemporary Psychoanalysis, 40, 197–237, 2004.
Chapter 5 originally appeared in Psychoanalytic Quarterly, 78(3), 701–731, 2009.
An earlier version of Chapter 6 appeared in Psychoanalytic Inquiry, 29(1), 79–90, 2009.
An earlier version of Chapter 7 appeared in Psychoanalytic Dialogues, 16, 747–761, 2006.
An earlier version of Chapter 8 appeared in Mind to Mind: Infant Research, Neuroscience, and Psycho-
analysis, Slade, Berger & Jurist (eds.), © 2008 by Other Press, LLC.
An earlier version of Chapter 9 appeared in Psychoanalytic Dialogues, 18(2), 168–196, 2008.
Reprinted with permission.
Routledge Routledge
Taylor & Francis Group Taylor & Francis Group
270 Madison Avenue 27 Church Road
New York, NY 10016 Hove, East Sussex BN3 2FA
© 2010 by Taylor and Francis Group, LLC

Routledge is an imprint of Taylor & Francis Group, an Informa business
Printed in the United States of America on acid-free paper

10 9 8 7 6 5 4 3 2 1
International Standard Book Number: 978-0-415-99969-4 (Hardback) 978-0-415-99970-0 (Paperback)
For permission to photocopy or use material electronically from this work, please access www.copyright.
com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood
Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and
registration for a variety of users. For organizations that have been granted a photocopy license by the
CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
Library of Congress Cataloging‑in‑Publication Data
Stern, Donnel B.
Partners in thought : working with unformulated experience, dissociation, and
enactment / Donnel B. Stern.
p. cm. -- (Psychoanalysis in a new key ; v. 12)
Includes bibliographical references and index.
ISBN 978-0-415-99969-4 (hardcover) -- ISBN 978-0-415-99970-0 (pbk.) -- ISBN
978-0-203-88038-8 (e-book)
1. Schemas (Psychology) 2. Experiential learning. 3. Experience. 4. Cognitive
dissonance I. Title.
BF313.S74 2009
150.19’5--dc22 2009007473
Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
and the Routledge Web site at

http://www.routledgementalhealth.com
To Kathe, Lexi, Babette, and Mikey
Contents
Acknowledgments xi
Introduction xiii
1 The Embodiment of Meaning in Relatedness 1
2 Conversation and Its Interruptions 25
3 The Fusion of Horizons: Dissociation, Enactment, and Understanding 43
4 The Eye Sees Itself: Dissociation, Enactment, and the Achievement of

Conflict 71
5 Partners in Thought: A Clinical Process Theory of Narrative 107
6 Shall the Twain Meet? Metaphor, Dissociation, and Co-occurrence 131
7 Opening What Has Been Closed, Relaxing What Has Been Clenched:
Dissociation and Enactment Over Time in Committed Relationships 147
8 On Having to Find What You Don’t Know How to Look For: Two
Views of Reflective Function 161
9 “One Never Knows, Do One?” The Relation of the Work of the Boston
Change Process Study Group and Relational Dissociation Theory 183
References 207
Index 223
ix
Acknowledgments
For their readings and commentary on various drafts of the chapters of

this book, I thank Ghislaine Boulanger, Philip Bromberg, Rich Chefetz, Bill
Cornell, Phil Cushman, Muriel Dimen, Stewart Crane, Robert Grossmark,
Adrienne Harris, Betsy Hegeman, Irwin Hirsch, Irwin Hoἀman, Jill Howard,
Ruth Imber, Hazel Ipp, Shelly Itzkowitz, Lawrence Jacobson, Manny Kaftal,
Peter Lessem, Howard B. Levine, Sharon Mariner, Russell Meares, Robert
Millner, James Ogilvie, Lois Oppenheim, Jean Petrucelli, Barbara Pizer,
Stuart Pizer, Bruce Reis, Harry Smith, Steve Stern, Don Troise, Steve Tublin,
and Cleonie White. I also thank the International Association for Relational
Psychoanalysis and Psychotherapy for proposing an online symposium in
November, 2007 devoted to a discussion of Chapter 4, “The Eye Sees Itself.”
The symposium was organized, administered, and moderated by Peggy
Crastnopol and Bruce Reis, and Dan Shaw, Orna Guralnick, Susan Sands,
and Estelle Shane participated as panelists.
I am grateful, too, for the opportunities I have had to present chapters
of this book to professional audiences around the country, and especially
for the discussions after those presentations, which revealed to me points
about my own thinking that I would not have considered without the
stimulus of those interchanges. Ongoing dialogue with Philip Bromberg,
as good an editor as he is a friend, clarified and stimulated my thinking all
along the way. Another good friend, Phillip Blumberg, deserves special
thanks for his continuous support, interest, and enthusiasm, his ceaseless
willingness to read drafts, and his capacity to raise questions that forced
me to think through issues that I had not seen clearly prior to his observa-
tions. When you are the editor of the series in which your book appears,
as I am, you don’t have an editor yourself. It is a lot to ask of someone to
take on that job. I never asked Phil to do that, but that is, in eἀect, what he
did. Most of all, I thank my wife, Kathe Hift, a psychoanalyst herself, who
read and helped me with this book at every step, and whose encourage-
ment over the years made this book a reality.
xi
Introduction
I noticed that this book had been written out of the corner of my eye.
After the publication of Unformulated Experience in 1997, I found myself
drawn to the ideas of enactment and dissociation and the conception of
the self as multiple. These interests had been prominent in Unformulated
Experience. In fact, looking back, I now realize that I also made the link
in that book between dissociation and enactment that I am making in
this one, although I was then exclusively using the language of transfer-
ence and countertransference. In the earlier book, however, the link was
primarily implicit and therefore less fully theorized (see D. B. Stern, 1997,
Chapters 10–12).
In the years that followed the publication of Unformulated Experience,
the link between dissociation and enactment took center stage for me.
During those years I was also writing about other things, though, so that,
although I knew that the work on dissociation and enactment was accu-
mulating, it did not occur to me that what I was writing added up to a
book. I was brought up short by the realization that a book was already
nearly written.
And yet there it was. The papers had thematic unity, each of them com-
ing at the subject matter from a diἀerent perspective. For that reason, this
book is not a linear account of its subject, developing toward a final con-
clusion, but a collection of various approaches to the relationship of dis-
sociation and enactment.
In Chapter 1 of this book, I try to formulate the ideas that I believe
undergird the chapters that follow. The idea that most meaningfully links
them is a particular take on the nature of relationship and its place in
unconscious process. I conceptualize unconscious psychic life as unfor-
mulated experience, vague psychic material that, under the right cir-
cumstances, becomes articulated or realized meaning. Unformulated
experience is potential meaning, and in that way quite diἀerent from the
fully formed meanings that are assumed in theories based on the concept
of unconscious fantasy.
xiii
xiv Introduction
Chapter 1 presents the perspective that the meanings that arise from
unformulated experience are catalyzed by the states of relatedness in
which the meanings emerge. Patient and analyst are, in the most literal
sense, partners in thought because the interpersonal or relational field they
create between them, mostly without their awareness, and the changes
that take place within that field, have everything to do with the shape and
nature of what they experience together. This point holds, as we shall see
throughout this book, not only for the patient’s and the analyst’s expe-
rience of each other and the relationship between them, but for all that
they feel, think, and perceive in each other’s presence—and sometimes
what they experience elsewhere as well. The content of the session, in other
words, whether or not it expressly concerns the analytic relationship, is the
outgrowth of clinical process, conscious and unconscious.
Each time I make this point I feel compelled to add a proviso. Or
perhaps I should call what I want to add here my attempt to lend per-
spective to the problem. It is not the whole story to say that the content
of a clinical session is the outcome of its process. Process is also the
outcome of the sequences of its content. I will take the position in this
book, for instance, that the narratives we create in clinical work are the
unplanned expressions of clinical process (Chapter 5). But I am also
careful to say in that chapter that the narratives that come about in this
unwitting fashion then contribute to the shape and nature of the clini-
cal process that follows them. It may be an expression of nonconscious
clinical process, for instance, when the analyst has a new thought about
the analytic relatedness; but that new content (i.e., the new thought),
in the same spontaneous way, then aἀects the unfolding of the clinical
process that comes after it. It would distort the clinical situation to look
at the problem too exclusively from the perspective of either process or
content. Each plays a part.
But with that said, I must also say that this book is an attempt to correct
what I feel has been too great an emphasis in our field on the construc-
tion of clinical reality by the analyst’s consciously chosen interpretations,
and too little an emphasis on the spontaneous growth of clinical reality
from unplanned clinical process. Since clinical process is continuously
influenced by unconscious factors contributed by both the analyst and the
patient, my emphasis is on the analyst’s attempt to understand what we are
continuously in the process of unwittingly constructing. As I pursue this
point of view, and as a corrective, I lay more emphasis on the creation of
content by process than I do on the creation of new process by preexist-
ing content. I do want it to be clear, however, that I recognize that process
Introduction xv
and content have a mutual influence on each other, sometimes facilitative,

sometimes inhibitory.
Rethinking Unformulated Experience
Recently I have reconceptualized the theory of unformulated experience.

The change is recent enough to have taken place after most of the work on
this book had been finished. I claimed in Unformulated Experience that
only experience we can reflect on in verbal terms can be formulated at
all, and that only experience we have constructed in this way should be
described as formulated. In the reconceptualization I have just completed,
however, formulated meanings are not limited to the verbal register. I now
take the position that both verbal and nonverbal meanings can be formu-
lated. I refer to the formulation of verbal meaning as articulation and that
of nonverbal meaning as realization.
These new ideas do not appear explicitly in this book, however; they are
instead the core of another book that is now in preparation. But the changes
I will propose in the theory of unformulated experience are nevertheless
present in implicit ways throughout these pages because the writing of
these chapters was what made it inescapable to me that changes in the
theory of unformulated experience were necessary. What most gripped
me was the nonverbal nature of the clinical events that eventually break
up enactments and often make their dynamics possible to experience in
conscious, reflective ways.
Let me explain just a bit about what I mean. I observe repeatedly
in this book that what eventually interrupts enactments is not verbal
insight but a new perception of the other and oneself (see Chapters 3–6,
but especially Chapter 4). Although in the chapters of this book I do
not make an issue of the fact that this new perception is nonverbal, the
point has always been unmistakable to me, and once I formulated it for
myself I could not stop thinking about it. How could I maintain that the
process of formulation must be verbal if I also took the position that the
event to which I am pegging therapeutic action—the new perception—
is nonverbal?
Even before dissociation and enactment took center stage for me, I had
felt for many years that I assigned too exclusive a role to verbal language
in the formulation of experience. Furthermore, I had always known, of
course, as all of us do, that many significant clinical events take place apart
from verbal language. Now, with the “new perception” serving as the straw
xvi Introduction
that broke the camel’s back, it was clearly time to redefine the process of
formulating meaning in a way that would make it possible to conceptual-
ize the formulation of nonverbal meanings.
But the exploration of that subject, and others related to it, must await
the book that is now in preparation. In the book you now hold, I must be
satisfied to announce the change in view that is to come and to prepare
you to encounter the implicit expression of this change in the chapters
that follow.
The Chapters
Chapter 1, in addition to what I have already described about it, contains

the first of many clinical examples of dissociation and enactment. In the
course of reading that first clinical vignette (which, unlike those that fol-
low, is fictional), most readers will think about projective identification.
And so, following the introductory example, I compare projective iden-
tification and the modern Kleinian/Bionian view with the Interpersonal/
Relational perspective on enactment and dissociation that is the subject of
this book.
Chapter 2 presents the hermeneutic ideal of true or authentic conver-
sation, which is the dialogic creation of understanding, also referred to
by Hans-Georg Gadamer (2004) as “the fusion of horizons.” Authentic
conversation is one way to describe the ideal of the psychoanalytic situa-
tion; in fact psychoanalysis does work that way much of the time. But true
conversation breaks down during enactment. Enactment can be defined
as the interruption of true conversation. And so Chapter 2 constitutes a
hermeneutic portrayal of the dialogic context that enactment destroys and
to which, once we are involved in enactments, we try so hard to return.
Following this theme, Chapter 3 examines the link between dissocia-
tion and enactment in hermeneutic terms. Starting from the position that
all understanding is context dependent, I claim that one of the most sig-
nificant contexts for clinical purposes is the self-state. How we understand
the other, and ourselves, depends on the state(s) we occupy. Dissociations
between the analyst’s states of self can therefore limit or impede under-
standing of the analysand by depriving the analyst of a fitting context
within which to grasp what the analysand says and does. Such impedi-
ments can become what we understand as mutual unconscious enact-
ments. Clinical understanding under these circumstances requires the
breach of the dissociations, which ends the enactment. I lay out some of
Introduction xvii
the implications of thinking about transference and countertransference

along these lines, with detailed examples illustrating the consequences of
both the analyst’s dissociations and their eventual resolutions. Among the
advantages of this way of thinking is that it amounts to a psychoanalytic
account of the hermeneutic circle.
Chapter 4 is probably the heart of my ideas about dissociation and
enactment. The chapter addresses the question: If being part of an enact-
ment means being blind to the nature of our involvement with the patient,
as I claimed in Chapter 3, how can we ever understand the countertrans-
ference? How can the eye see itself? I expand the portrayal of enactment
oἀered in Chapter 3, where I introduce the conception that mutual enact-
ment is the outcome of the interlocking dissociations of analyst and patient.
Chapter 4 presents the view that, in cases of mutual enactment, neither the
analyst nor the patient can experience alternate, or conflicting, views of
the other; each is limited to understanding the other through a single lens.
These reciprocal dissociations are breached only when either the patient
or the analyst can transcend this “single-mindedness” and allow the expe-
rience of multiple, conflicting perceptions of the other. Until the conflict
can be contained within one mind, it remains split into the pieces rigidly
experienced by the patient and the analyst as enactment. I take the posi-
tion that conflict must be viewed as an achievement. Enactment is not the
result of too much conflict in the personality, but of too little.
Chapter 5 focuses on how narrative is created in psychoanalysis, oἀer-
ing an alternative to the views of Schafer (1983, 1992) and Spence (1982),
an alternative based in the concept of unformulated experience that shows
how enactment derails narrative formation and makes psychotherapy
essential. The chapter’s title, “Partners in Thought,” from which the title of
the book is drawn, refers to the kind of productive clinical process, based
on what I describe as “witnessing,” that becomes impossible in the pres-
ence of dissociation and enactment and that takes place once again when
dissociation and enactment are resolved. Even in the absence of other
people, we learn about ourselves by listening to our own thoughts about
ourselves through the ears of the other. We need an emotionally respon-
sive witness, even if that witness is imaginary, in order to know and feel
our experience. We need a witness, that is, if experience is to fall into a
satisfying narrative order. The origins of the witness lie in earliest infancy,
in the roles played by responsive parents in reflecting back to us a version
of ourselves that eventually becomes the self. We need a witness to become
a self; and later in life, in similar fashion, we need a witness to heal our-
selves. Patients listen to themselves as they imagine their analysts hear
xviii Introduction
them and in this way create new narrative freedom. But only what feels
like “me” can be witnessed. “Not-me” can only be enacted. The resolution
of enactments is crucial in psychoanalytic treatment not only because it
expands the boundaries of the self, but also because it reinstitutes and
broadens the range within which patient and analyst can witness each
other’s experience. Narrative is not the outcome of the analyst’s objective
interpretations, but an emergent, co-constructed, unbidden outcome of
clinical process.
Chapter 6 approaches dissociation and enactment through the lens of
trauma and memory. Even when trauma can be remembered, the memory
often does not infuse the present with vitality or emotionality, as other
memories do. To become a vital part of experience, trauma must be linked
with other current experiences. Such links are metaphorical, in the sense
meant by Lakoἀ and Johnson (1999). I use their work to suggest that, in
metaphor, the meaning of a memory is “carried over” or “transferred”
from the past to a present experience. When such transfer takes place,
trauma can be reflected on because it can now be seen against the back-
ground of other experience. Transfer is made possible by “co-occurrence,”
or the simultaneous presence in one’s mind of a memory and a present
experience. I describe such co-occurrences as potential, unformulated
metaphors; they can be either actualized or refused. Modell (2003) tells us
that when metaphors are unconsciously refused, a common event among
trauma suἀerers, traumatic experience is prevented from becoming part
of “emotional categories.” The resulting isolation of the experience is what
I mean by dissociation. I therefore refer to the unconscious refusal to tap
the potential of co-occurrence, that is, the unconscious, defensively moti-
vated refusal to create metaphor, as dissociation. As in any other instance,
dissociated states are liable to be enacted. And so the unconscious refusal
to create a metaphor is one source of unconscious enactment.
Chapter 7 extends the concepts of dissociation and enactment beyond
the bounds of the clinical situation into the kinds of interpersonal pat-
terns seen in long-term romantic relationships. The chapter explores the
way that certain kinds of mutual dissociations and enactments can come
to define relatedness over long periods of time.
Chapters 8 and 9 compare dissociation/enactment theory to two prom-
inent contemporary approaches to clinical understanding. In Chapter 8
I compare my thinking to the mentalization theory of Fonagy and his
collaborators (2002), a body of work that constitutes a profound explora-
tion of the nature and significance of reflective awareness. I ask whether
the view of reflective function held by Fonagy and his collaborators is the
Introduction xix
same as my own. It seems that therapeutic action in mentalization theory

requires the analyst to be capable of mentalizing, which is exactly what I
believe cannot be done during an enactment; and so it also seems that the
therapist must be responsible for avoiding enactments whenever possible.
To me, such avoidance is not really possible. Whatever wiggle room we
have around enactments exists more often in how we respond to finding
ourselves already part of them than in the choice about whether or not to
enter them in the first place. And even if it were possible to avoid enact-
ments, it would usually not be desirable, because dissociated material (not-
me) is not symbolized, but unformulated, and is therefore only available
via the experience of enactment. Finally, it is my position that the most
clinically significant instances of reflective function take place unbidden.
Although the analyst can try to create conditions under which reflection
can flourish, I believe she cannot control its appearance in her mind. And
so, despite the overlap in our approaches, and despite holding the opinion
that mentalization theory is one of the most significant and clinically use-
ful bodies of work in contemporary psychoanalysis, I conclude that the
architects of mentalization theory and I probably think at least a bit diἀer-
ently about the place of reflective function in therapeutic work.
Chapter 9 compares dissociation/enactment theory with the work
of the Boston Change Process Study Group (BCPSG; 2002, 2005, 2007,
2008; D. N. Stern et al., 1998), another highly influential body of work
that has deeply influenced my perception of clinical events, especially in
its brilliant conceptualization of the nonverbal and nonsymbolic aspects
of what transpires in the consulting room. After presenting these sub-
stantial areas of overlap in our work (it is really only the similarities of
our thinking that make our diἀerences interesting), I focus on three
areas. First, I take the position that the work of BCPSG, to this point,
does not address the kinds of emotionally intense enactments that have
been presented in the Relational literature, including throughout this
book, and I oἀer a speculative explanation for this diἀerence. I then turn
to BCPSG’s application of nonlinear dynamic systems theory, observ-
ing that that theory is used by BCPSG in a literal sense and suggest-
ing certain problems this leads to for psychoanalysis. I take the position
that nonlinear dynamic systems theory should probably be used meta-
phorically in psychoanalysis, and I oἀer my explanation why. Finally, I
address BCPSG’s objectivist, scientific approach to psychoanalysis and
psychotherapy and compare it to the constructivist or hermeneutic one
I have adopted.
1
The Embodiment of Meaning in Relatedness
One cannot include everything about one’s ideas every time one writes
so I must refer readers to my earlier book (D.B. Stern, 1997) for a descrip-
tion of the theory of unformulated experience. However, there are points
about those views that should be reiterated at the outset of this book, at
least briefly, because they are crucial and frequently misunderstood. First,
the question of relativism. I conceive unformulated experience as poten-
tial experience, and I intend by that point to suggest that the shape of the
next moment’s formulated meanings is not entirely predetermined, that
there is always some ambiguity to be resolved in experience, some for-
mulation of the unformulated that remains to take place, some emergent
quality in the creation of whatever is to come next. At times this position
about the inevitable ambiguity of the next moment has been mistakenly
interpreted to imply that the process of formulation is unconstrained, as
if unformulated experience can become any meaning one pleases to give
it. That is not what I believe (D.B. Stern, 1997, pp. 28-32, 203-233). I want
to avoid altogether the implication of relativism and unconstrained sub-
jectivism. The idea of unformulated experience is a hermeneutic view well
described by Sass (1988): “For, though it may be impossible to discover a
single meaning, this does not mean that anything goes, that listeners can
legitimately ascribe any meaning to any discourse. The hermeneutic view
is a sort of ‘middle way’ between objectivism and relativism” (p. 254).1
I have taken pains elsewhere in this volume to describe the formula-
tion of meaning as a dialectical process, but I want to emphasize dialectic
1 Among hermeneutic writers, Hans-Georg Gadamer has had the most influence on my thinking,
and therefore on the ideas in this book. For introductions to Gadamer’s philosophical herme-
neutics, see Palmer (1969), Bernstein (1983), Warnke (1987), Grondin (1997), and Dostal (2002).
Gadamer’s magnum opus is Truth and Method, published in a new translation in 2004. A collec-
tion of informative but accessible essays (Gadamer, 1966) is a good point of entry to Gadamer’s
thought. For the intersection of Gadamer’s hermeneutics and psychoanalysis, see Orange (1995),
Messer, Sass, and Woolfolk (1988), Sass (1988), and Stern (1997).
1
2 Partners in Thought
here, too, in the beginning, because another implication I want to avoid

is that unformulated experience has no structure and that the process of
formulation is therefore unidirectional or one-dimensional, i.e., “nothing
but” emergent, uninfluenced by the continuity provided by reality and the
more structural aspects of personality. That is not the position I take.
The meanings that can be validly created from any unformulated expe-
rience are a joint outcome of pre-existing structural meanings and the
emergent influence of the present moment. The role of structure, from this
perspective, is played by constraints on what unformulated experience
can become. From the hermeneutic perspective, reality cannot be directly
apprehended; it can be perceived only through the lenses of tradition, his-
tory, and culture. But reality is there, and it shows itself in the continuous
constraint it exercises on our freedom to create experience. In Gadamer’s
(2004) hermeneutic view, all experience is interpretation. It is reality that
provides the limits within which we are free to create valid experience,
and beyond which we recognize experience as lie or distortion.1
The degree of constraint on unformulated experience ranges from high
to low, and this degree of constraint diἀers with context, over time, and
from one kind of experience to another. In the case of tight constraints,
there is little “wiggle room,” which is to say that in these cases the range
of formulated meanings that can validly be made from unformulated
experience is narrow. In other cases, constraints are looser, and the range
of meanings that can validly be formulated is therefore correspondingly
wider. Consider, for example, a painting consisting of two fields of color, one
pink and the other orange, the two fields seeming to float on an oἀ-white
background. Let us say that I simply attend to the impact the painting has
on me, trying to formulate my reaction, or the kind of aἀective experience
I have while looking at it, or the place of this painting in the tradition of
art from which it arises. In the cases of experiences and thoughts such as
these, the interpretive quality is undeniable, and the range of formulations
that can be made without violating the constraints of reality is very wide,
indeed. But if I ask myself instead what are the colors of the painting’s two
fields, the experiences I can formulate without violating reality are so few
1 This point is too simple in one very important respect that I cannot pursue here, but that needs
to be recognized. Uunconscious ideology, the aspect of culture that serves our minds not as the
potential for new meaning but as a straitjacket for it, is just as eἀective as reality itself in setting
limits on our freedom to experience. Because reality can only be apprehended via the tools made
available by culture, the distinction between limits that are based in reality (and therefore intrin-
sic to human experience), and those that are ideologically based (and therefore imposed on us for
reasons having to do with the invisible workings of power [see Foucault, 1980]) is always at issue
and painfully difficult to accomplish.
The Embodiment of Meaning in Relatedness 3
that the interpretive aspect, while it remains real enough, is constrained

enough to seem trivial.
One routine eἀect of formulating meaning is to provide a constraint
on what future meanings can become. The impact of the past, and the
presence of continuity in the personality, are thereby assured, because
the most significant constraints on what unformulated experience can
become in its next incarnation, or in the next moment, are the meanings
that have come before. The creation of meaning is once again dialectical:
formerly created meanings influence the future to take their shape, while
the unique influences of the present and the future encourage the refor-
mulation of past meanings. I made this point in Unformulated Experience
(D.B. Stern, 1997) in words to which I continue to subscribe:
The given and the made are a dialectic, neither ever excluding the other and
both constituting every meaning and moment. Without the opportunity to
change previously structured experience, and without that previous structure
to feel and think against, new experience would be impossible. We would be
trapped in an evanescent subjectivism. But, on the other hand, without our
capacity for an imagination that goes beyond experiential regularities, without
the animation of spontaneous expression and the continuous reworking that
represents our ceaseless eἀort to understand, we would never be able to redeem
our experience from the stasis of dead convention. It is reflection that saves the
unconscious from being nothing more than a set of strictures, and makes it a
precious resource instead; and it is the unconscious that oἀers reflection on
the fecund and ever-changing materials with which to carry out its life-giving
mission (p. 30).
The Location of Unformulated Experience
Where is unformulated experience? Does the question even make sense?

Should we really say that it exists anywhere? Certainly we would not want
to say that it is “located” in the kind of hypothetical netherworld that
most of us are no longer even willing to call “the” unconscious. The era
of psychic geography is dead. Unformulated experience is possibility, the
various potential meanings that might expand from the present moment.
Only one, or some, of these potentials are ever realized. Unformulated
experience is the source of what experience can become. And so, because
it does not yet exist, it cannot really be said to be located anywhere, not
even in the brain.
We can say that the present moment actually exists in our minds (wher-
ever that is), but the present is probably the only kind of experience about
which we remain confident enough to ascribe to it literal existence. Take

the example of memory. We once thought memory had a literal existence.
When we think of experience that perseveres over time, we think imme-
diately of memory, and so it was natural for us to think that each memory
was represented in stable, physical fashion somewhere in the brain. That
natural conclusion is what motivated the futile search for the engram
(Lashley, 1950).
But the way we see memory has changed, and in particular, our view
of the literalness of what we remember has been qualified. We accept
today that we reconstruct memory each time we consult it. We constantly
remake the past in the service of present purposes. Citations of this point
go back at least to Bartlett (1932), and in fact, cognitive and developmental
psychologists have recently come to a broad and striking consensus that
remembering is a constructive process (e.g., Tulving & Craik, 2000). In an
important sense, then, even the past is unformulated. We continuously
reconstitute the events of our lives, even if those reconstitutions often bear
a significant similarity to the ones that have come before. And of course,
if the past is substantially unformulated, whatever we imagine about the
future must be in even greater flux. As soon as we really consider these
things, it is apparent that we live as if life were much more stable and pre-
dictable than it is. And so, while some psychoanalysts today seem enthu-
siastic about nailing down the origins of experience in cerebral structure
and process, many of the rest of us are skeptical of theories that “source”
experience too specifically or concretely, whether the location of that
source is said to be the unconscious or the brain.
But there is one “place” in which it may be meaningful to give unfor-
mulated experience a home that is at least metaphorical: relatedness.
Relatedness is a category of living in which I mean to include not only
the relevant conduct that transpires between people, but all the relevant
aἀective, cognitive, and conative experiential processes. In imagining
these intrapersonal processes, I am thinking, for example, of interaction
between states of self, or for those who prefer a diἀerent language, internal
object relations.
Relatedness is the nexus from which experience emerges. It is in the cru-
cible of each moment’s relatedness that unformulated experience is articu-
lated into knowable, usable, experience-able living: meaning is potentiated
and prohibited by what transpires between us. Since the events of related-
ness are only partially under our control, we are not the only authors of the
meanings that we formulate in what we usually feel is the privacy of our
own minds; but at the same time, and paradoxically, because the process
of authorship goes on outside awareness, we are unaware, more often than

not, of the extent to which we do create the experience that emerges from
relatedness.
Mind and Relatedness: Hand and Glove
Contemporary interest in the relationship between dissociation and enact-

ment, each phenomenon significant in its own right, each the inspiration
for a literature of its own,1 is part of the broad interest in psychoanalysis in
process theories of treatment. I hardly know whether to call this interest
explosive or longstanding, because it is both: process theories have now
held the floor in psychoanalytic clinical debate for decades. Across theo-
retical schools whose adherents otherwise usually claim to diἀer from one
another in significant respects, I believe there has developed an emphasis
on the close examination of clinical process, a move toward the study of
the events of the treatment for their own sake.
What do I mean by “for their own sake”? I mean to refer to the social
aspects of therapeutic relatedness. I mean that today there are analysts of
all persuasions who tend to understand the transference and the coun-
tertransference as “interlocked” with one another, as Wolstein (1959)
so presciently put it, each playing a role in holding the other in place.
Now, of course, I hasten to add that I do not mean to equate “social” and
“sociological.” I am hardly setting aside unconscious influence. In fact, it
could reasonably be claimed that I, along with most analysts who iden-
tify themselves as Interpersonal or Relational, actually privilege mutual
unconscious influence between patient and analyst. At the very least, we
emphasize it heavily.
Even those analysts who continue to maintain that the unconscious
fantasies of the individual mind lie at the heart of life and of treatment
(the majority of analysts may very well hold this view; see Bornstein, 2008)
would agree that their interest in process theories of therapeutic action is
today extensive. Yet it does seem to me that anyone who accepts the latent/
manifest distinction, and who imagines each mind to be housed within
1 Prime examples of the Relational literature on enactment include Aron, 2003a, b; Bass, 2003;
Black, 2003; Benjamin, 1998; Bromberg, 1998, 2006; Davies, 1997, 1999, 2003, 2004; Davies &
Frawley, 1994; B. Pizer, 2003; S. A. Pizer, 1998; D. B. Stern, 2003, 2004. For the contemporary
psychological and psychiatric literature on dissociation, see Dell & O’Neil, 2009. For the contem-
porary clinical psychoanalytic literature on dissociation, see Bromberg, 1998, 2006; Howell, 2005;
D.B. Stern, 1997.
a single skull, has to hold a diἀerent view of analytic relatedness than I

do. I know that even those analysts who take the most conservative posi-
tion about transference—i.e., that it is a distortion based on the demands
of the inner world—would readily accept that diἀerent transference con-
figurations are uppermost at diἀerent times. In taking this position, one
might say, these analysts are indicating their acceptance that context plays
a significant role in analytic relatedness. But most of these same analysts
would also take the position that each one of these transference manifes-
tations is attributable to the activation/expression of a particular endur-
ing, underlying unconscious fantasy and the defenses against it. The latent
meaning of the transference continues to be, for this group of analysts,
the truth of transference. The truth exists in the unconscious and awaits
discovery. This book is based in a diἀerent view.
The traditional assumption that unconscious meaning is fully formed
also has a significant influence on analysts’ understanding of their own
participation in the treatment. It is acknowledged by adherents of the
traditional view, of course, that analysts, like any other human beings,
are themselves capable of transference; but many of these analysts also
believe that they are supposed to be capable, because of their own analy-
ses, of observing the development of their own transferences in such a
way that they avoid being caught up in them. Everyone these days accepts
that countertransference can be informative, but it is still common for
some analysts to believe that they should be able to stand aside from these
aspects of their experience and analyze them before they appear in ana-
lytic conduct.
In this orientation toward transference and countertransference, which
was dominant until at least the mid-1980s in most of the field, and still
exists in some quarters, the analyst’s authority is at least partly defined by
knowing what is true, and the most important thing the analyst should
know, or come to know, is the contents of the patient’s mind. It is pos-
sible in this frame of reference, in a fairly straightforward way that can be
embraced without disrespecting anyone, to believe that the analyst knows
the truth about the patient.
But now consider what happens if we think in the frame of reference
provided by the concept of unformulated experience. From this vantage
point, experience is not uncovered but constructed, and the construction
of experience involves the selective formulation of some portion of the
available possibilities. Experience ceases to be a given at any level. It is
not that we “have” the experience and then interpret it. Rather, we are
continuously involved, without our awareness, in crafting the very shape
our experience takes. It is very seldom that we know exactly how and why
we make experience as we do. Experience, instead, arrives in our minds
unbidden; but its unbiddenness makes us no less responsible for its shape
than we are in any psychoanalytic frame of reference. Unconscious mean-
ing is not fully formed and waiting to be uncovered; it is instead potential
meaning, meaning that might become actualized under the right circum-
stances—and “the right circumstances” means “the right interpersonal
context.”
That was the message of Unformulated Experience (D.B.Stern, 1997).
In these terms, there can be no single reality underlying any transfer-
ence, no unitary truth to be uncovered. Transference and countertrans-
ference can no longer be viewed as distortions, but must be understood
instead as unconsciously selected perceptions that are quite real, but that
represent only one, or a few, of the possibilities that might be actualized.
Transference and countertransference, in other words, are created in just
the same way as any other interpersonal perceptions. What sets them
oἀ is not the means of their creation, or their sources, but their rigidity.
Transference and countertransference are rigid selections from among the
possibilities for relatedness. Unlike other interpersonal perceptions, they
do not necessarily shift easily and freely with changes in circumstances;
the person in the grip of transference turns away from the possibilities
for feeling, thinking, and being that might make a new experience of the
other available. The experience that would change the transference (that
would be a change in the transference) is simply not constructed. For the
moment, for simplicity’s sake, I will not explore the fact that transference
cannot be maintained this way without the collaboration of the other. Let
me just make the familiar claim that, even (or maybe especially) in the case
of unconscious processes, it takes two to tango. Paraphrasing Winnicott:
there is no such thing as transference.
As soon as you take this kind of constructivist view, in which meaning
is not predetermined but created in dialogue, the idea of a single, objective
reality becomes unsupportable. I hasten to add, however, as I always do
when I make this point, that the claim that reality is multiple is not at all
the same thing as taking the position that reality is relative. The multiple
possibilities for meaning all have a real existence, even if it is not sensible
to refer to those potential existences as objectively verifiable.
I must also make sure to answer another objection I have often heard
made to my work and the work of other interpersonal and relational writ-
ers. It is claimed by some that in interpersonal and relational views indi-
vidual subjectivity has been eclipsed—that is, that the individual mind has
disappeared, leaving only the dyad. That simply is not the case. To recog-
nize the role of the other in the selection of the conscious contents of one’s
own mind is hardly synonymous with suggesting that one’s mind does not
have its own, separate existence. Even the idea that the mind is distributed,
a claim I make later in this chapter, is perfectly consistent with individual
subjectivity. Let me be as clear as possible about this point: the recognition
of the influence of the other, even the recognition of the influence of the
other on the contents of one’s own mind, does not imply for one moment
the rejection of the individual mind.
Of course, as soon as we reject a single objective reality, the analyst’s
old status as the arbiter of such a reality also becomes impossible to sus-
tain (e.g., Benjamin, 1988, 1995, 1998; Cushman, 1995; Aron, 1996; Elliott
& Spezzano, 1996; Hoἀman, 1996; McLaughlin, 1996; Mitchell, 1997,
Chapter 7; Greenberg, 1999; Zeddies & Richardson, 1999). A training
analysis can no longer be considered an inoculation against countertrans-
ference enactment, because countertransference and its enactment are
not diἀerent in kind from everyday interpersonal life. We can no longer
define countertransference as irrational and then contrast it with a ratio-
nal non-countertransference. All experience is subjective, the analyst’s as
thoroughly as the patient’s. (This point has now been made so often that I
would not know how to limit citations of it.) We must now understand that
we all continuously, necessarily, and without awareness apply ourselves to
the task of selecting one, or several, particular views of another person
from among a much larger set of possibilities. That is the way we all deal
with one another, all the time. We can no longer specify psychopathology
as a certain kind of mental content, or even, as some writers have rede-
fined unconscious fantasy (e.g., Sugarman, 2008), as a mental process. All
we can really specify is the degree of freedom or rigidity that character-
izes a person’s approach to experience, and even that specification must be
understood as a judgment, vulnerable to unconscious influence. The only
clinical question about transference and countertransference that really
matters becomes how thoroughly willing and able each person is, under
some specified set of circumstances, to consider alternative meanings.
Before I go any further, I think I should reinforce a point I made in
beginning this chapter: my perspective does not require the dismissal of
the concept of character, or any of our other ways of grasping the conti-
nuity of the personality. Those ideas are far too useful to be dismissed.
But our notions about continuity, or what are commonly referred to as
structural aspects of personality, do have to become multiple. We can no
longer do with a single character structure, for instance. That idea needs
to be replaced with a more flexible conception in which diἀerent aspects

of character appear under diἀerent circumstances, circumstances that we
then need to specify. Even those of our ideas that refer to the mind’s conti-
nuity, in other words, must become more responsive to context.
Now, when I say that the only clinical question about transference and
countertransference that really matters is how thoroughly willing and
able each person is to consider alternative meanings, I mean by “willing”
and “able” to refer to phenomena that transcend consciousness. This kind
of will and ability should probably be called intentionality, a commit-
ment thorough enough to extend into unconscious life. In Chapter 3, for
instance, in the case of Hannah, the patient’s commitment to her treat-
ment, and her willingness to become vulnerable to me at a frightening,
risky moment that one might have expected to provoke an unconscious
defensive maneuver, were powerful enough to create a collaborative dream
instead—a dream that proved to be what she and I needed to end a lengthy
and rather dire enactment.
To hold out for meanings alternative to the ones that pop up with least
eἀort in our minds is an admirable kind of choice. At the same time,
though, because freedom of this kind is rooted in intentions that reach
so deep in us, these are “choices” that, in an important sense, can hardly
be made otherwise. They define us; they are what make us ourselves. And
they are the motives that create the possibility of psychoanalytic treat-
ment. Hannah, the patient in Chapter 3 to whom I just referred, did not
“choose” her dream, of course, not in the sense in which we usually use
that word; but in another sense she did, because the dream was her inven-
tion, however unconscious the creative process may have been.
It has been several decades since the revelation of the mind’s contents,
or their manipulation in the service of creating a diἀerent kind of content,
has been the core of anyone’s theories of therapeutic action. We have come
to believe that the way the mind works is far more important than what
it holds. For many of us, the way the mind works is inseparable from the
mind’s context; and the mind’s context is relatedness. Mind and its con-
text are a ceaselessly changing unity. It has been many years since most of
us thought of the analyst as a decoder of the patient’s unconscious secrets,
those congealed nuggets of meaning that were supposed to exist in a hid-
den, timeless realm in which context was only relevant to the extent that it
provided the canvas on which disguised expressions of those secrets were
projected. Those of us who have made the shift from models of content to
models of context and process think and write about psychoanalytic work
only infrequently as the interpretation of a greater truth that lies behind
or underneath appearances. Instead, we believe that whatever we can have

of emotional truth is most often embodied in the nature of our relatedness
with one another and what we do with it.
Ongoing Interaction and the Formulation of Meaning
In any particular moment, with any particular person, some of what one
“is” with the other stays unformulated, while other parts of one’s being-
with-the-other are available to us as perceptions. What we either perceive
or leave unformulated runs the gamut: we feel our own emotions and
infer the emotions of the other; we attribute intentions to ourselves and
the other; we perceive conduct, to which we ascribe meaning, and so on.
All of these phenomena may be either perceived or left unperceived, their
potential untapped. When they are left undeveloped, we are making the
“choice” to leave them—in a word—unformulated. But if they are formu-
lated, perceptions of oneself or the other may take either a verbally sym-
bolized form that allows reflective consciousness, or a nonverbal form that
then becomes part of the intuitively organized relating to come.
Our perception of interaction is always several steps behind our partici-
pation in it. Perceptions of self and other, including our perception of our
own and the other’s aἀects and ways of being, arrive in our minds after the
relatedness that embodies them, not before it. And that observation then
leads us to the conclusion, perhaps counter-intuitive, that our perceptions
of self and other are at least as deeply shaped by interactions as interac-
tions are shaped by pre-existing perceptions (cf. Renik, 1993a).
For the most part, we take it for granted in everyday life that things are
the other way around—i.e., that it is perceptions that shape interactions,
not interactions that shape perceptions. We generally take it for granted,
in other words, that it is our views of others that determine how we treat
them. There is truth in that view, of course, because once we absorb the
self/other perceptions embodied in one moment of interaction, that new
awareness participates in shaping the interaction that comes next. And so,
to attain a balanced view we must recognize that the relationship between
perception and action is, like the relationship between content and process
that I discussed in the Introduction of this book, reciprocal: each plays a
role in creating the other.
I mean “perception” to include what it seems to us that we and the other
person are like, and what it is like to be involved with each other. It is
aἀect, the “feel” of the situation, that gives an interpersonal perception

its boundaries or wholeness, that gives us a meaningful way of demarcat-
ing one perception from another; it is aἀect, in other words, that makes it
meaningful to refer to what something is like. When we say that percep-
tion is what it is like to be with the other, that is, and what the other is like,
what matters most to us is what it feels like to be with the other and how it
feels like the other is being with us.
If we are to perceive what something or someone is like, we must
experience the possibility of alternative (D.B. Stern, 1990). To take
advantage of the potential meanings embodied in interactive experience
is to be open to multiple interpretations of oneself and the other. This is
the mature form of the capacity to reflect on one’s own experience and
that of others, or mentalization (e.g., Fonagy et al, 2002), in which we
are not necessarily attached to any particular interpretation of our own
and/or the other’s feelings and motives, but remain open to whatever we
come across. To absorb the potential meaning of interpersonal events is
to be curious, to allow oneself, with a willingness that derives not from
moral force but from desire, to imagine as freely as possible the ways of
grasping and feeling one’s own and the other’s conduct and experience.
Unfettered curiosity is an ideal, never actually created but worthy of our
aspirations toward it.
And that is to say that, even after the lag between the event and its per-
ception, we never absorb all the potential meanings embodied in the inter-
actions in which we participate. We develop only some of the potential
meanings of relatedness, leaving the others unformulated. This selective
process goes on continuously. We could hardly develop every potential
relational experience; if we did, we would be so preoccupied that we could
do nothing else. For this reason, much of relatedness is on “automatic
pilot,” and it is this automaticity that is responsible for a large portion of
what we never perceive. The necessity for automatic relating also allows us
to maintain in our daily lives highly selective, problematic, and meaning-
ful patterns of perception, patterns of perception that make up a substan-
tial portion of what we call transference and countertransference, without
ever facing the challenge to notice them.
We are able to symbolize in words some of our perceptions of the mean-
ings embodied in relatedness—that is, we can name and think about some
ways of being. Other ways of being, while not available in words, never-
theless are perceived either nonverbally (e.g., in the form of imagery) or
subsymbolically (i.e., in experience that is not represented in conventional
verbal or nonverbal symbols, but in aἀect-laden procedural forms closely
related to the original experience).1 Especially significant is the subsym-

bolic, which, Bucci (1997) tells us, is the primary means by which aἀect
can be represented. The Boston Change Process Study Group (BCPSG;
2002, 2005, 2006, 2008; Stern, D.N. et al., 1998) has brought a key aspect
of the subsymbolic to our attention under the rubric of implicit relational
knowing (IRK). IRK can be sensed or felt, but usually it cannot be symbol-
ized in verbal language. Conceptions like Bucci’s subsymbolic mode and
BCPSG’s IRK show us why it is so often crucial to ask patients, “What was
(is) it like?” The question invites a kind of groping toward explicit meaning
without insisting on a precise verbal formulation. Verbal language can-
not specify or articulate IRK or the subsymbolic; but there are moments
in which it can be used in a loose, relaxed way that draws our attention
further into places that are hard to illuminate, perhaps making possible
our nonverbal or subsymbolic appreciation of phenomena that might oth-
erwise remain wholly in the dark.
We are seldom explicitly aware of this part of what we perceive, but
when we do notice it, it is comfortable enough; it does not feel alien or dis-
turbing. Most of the time, IRK, the nonverbal, and the subsymbolic simply
unfold in a “natural” way that we never explicitly notice at all. Perception
in these modes is the primary language, as BCPSG suggests, in which the
moment-to-moment, unnoticed negotiation of relatedness, or what we
might call nonreflective living, takes place. If we can comfortably “be that
way” with the other person, or can comfortably perceive and tolerate the
other “being that way” with us, then, we can say that that way of being is
available to us for use in nonreflective living.
Conceptualizing Enactment
To this point I have been discussing parts of our experience of related-

ness that, while sometimes inconsistent with explicit symbolization, are
unquestionably tolerable. But some of the potential meaning embodied
in relatedness is not tolerable at all. These unformulated parts of related-
ness seldom, if ever, become meanings we use in living. That is, even if
we otherwise could formulate them in reflective consciousness—even if
their modes of representation were consistent with registration in reflec-
tive awareness—they remain unformulated for unconscious dynamic
1 See Bucci (1997) for a description of what she calls the verbal, nonverbal, and subsymbolic modes
of information processing.
reasons. They are, that is, dissociated. In the Interpersonal/Relational

model I introduced in Unformulated Experience and expand in the present
book, dissociation is the primary unconscious defensive process, replac-
ing repression. Dissociation and unformulated experience are the build-
ing blocks of a psychoanalytic model of mind (D.B. Stern, 1997, 2009).
I mean dissociation to refer to the process by which, for unconscious
defensive reasons, either the patient or the analyst, or both, fails to poten-
tiate some portion of the verbal, nonverbal, or subsymbolic meaning
available in the interaction of which they are part. Some of this selective
perception takes place as a result of dynamically-controlled inattention
to certain possibilities for social meaning, which I (D.B. Stern, 1997) have
described as “passive dissociation,” or “narrative rigidity” (an attachment
to certain narratives so strong that other alternative narratives are never
imagined, and the relevant observations are never formulated). In other
instances, selective perception is the outcome of more typical unconscious
defensive processes: the specific rejection of certain meanings, the uncon-
sciously motivated turn away from the possibilities for formulation. This
is what I have referred to as “dissociation in the strong sense” (D.B. Stern,
1997).
The unconscious motive for dissociation is conceptualized diἀerently
from the unconscious motive for repression. That is, the motive for dis-
sociation is not the unconscious intention to avoid certain kinds of expe-
rience, i.e., certain aversive or anxiety-provoking aἀects, meanings, or
memories. Instead, what moves us to unconscious defense in the theory of
dissociation is the need to avoid assuming a certain kind of identity. The
unconscious purpose is to avoid the creation of a certain state of being, or
self-state. The unconscious necessity is that we must not be a certain per-
son; and we suddenly would be that person if we were to formulate certain
key parts of an interaction in which we are involved. I give illustrations
of what I mean in most of this book’s chapters. For the time being, let me
just say that the person one must not be—the self-state one must not find
oneself inhabiting—is someone who felt disappointed, bereft, frightened,
humiliated, shamed, or otherwise badly hurt or threatened. One must not
be the person to whom that thing happened, the person who has the feel-
ings, memories, and experiences that come with being that person. What
we defend against is not any single feeling, fantasy, thought, or memory,
but a state of identity, a way of being. I refer to this dreaded state of being,
after Harry Stack Sullivan (1954) and, more recently, Philip Bromberg
(1998, 2006), as not-me. Quite literally, this is the part of subjectivity that
must not be me.
But of course life does not necessarily cooperate with our intentions,
this intention included. Circumstances—interactions with others—some-
times conspire in such a way that the dissociated way of being threatens to
erupt into awareness. Interaction, for example, might seem to be moving
in a way that we unconsciously perceive as a threat to wrest us into a posi-
tion in which a certain kind of humiliation cannot be avoided. That must
not happen. The only course of action left to the dissociator who needs
to protect himself from such imminent danger is the externalization of
the way of being that one must not take on oneself—the interpersonaliza-
tion of the dissociation. Enactment is the last-ditch unconscious defensive
eἀort to avoid being the person one must not be, accomplished by trying
to force onto the other what defines the intolerable identity (see especially
Chapters 3–8).
A Fictional Illustration
One manages relatedness most of the time in such a way that the possibili-
ties for the eruption of not-me are minimized. Let us imagine a simpli-
fied illustration, a man who, on the basis of his relationship with a parent
whom he feels experienced his needs as selfish and burdensome, is in
danger of being overwhelmed by shame when he feels a certain degree of
self-interest. Awareness of self-interest, especially self-interest that might
conflict with someone else’s interests, is, for him, not-me. These aspects of
his character are observed by certain others in the man’s life; but he him-
self has never formulated them. Nevertheless, he manages to construct
characteristic states of relatedness with others in which his self-interest
almost always takes a back seat to theirs. He does this on the basis of the
process I have described as passive dissociation, dissociation in the weak
sense, or narrative rigidity. That is, he is so attached to living out a narra-
tive of self-sacrifice that he is never faced with circumstances that would
make him feel greedy. And generally, this way of living works for him: he
is known as generous by some people, masochistic by others. In any case,
he is generally able to avoid having to be keenly aware of his self-interest,
and this is especially true when his self-interest would conflict with the
interests of the person with whom he is interacting.
I restrict the use of the phrase mutual enactment to interactions in
which both the patient and the analyst are, individually, sufficiently threat-
ened by awareness of their own dissociated material to unconsciously
interpersonalize it. This happens when the patient, enacting his own dis-
sociated material, accuses the analyst of something that the analyst cannot
tolerate being, or treats the analyst in a way that threatens to arouse the
same intolerable identity—that is, the patient awakens a dissociation in the
analyst. The analyst, who is no more able than the patient to use his mind
eἀectively in a situation of this kind of imminent threat, in some fashion—
perhaps subtly, perhaps openly—turns around the patient’s accusation or
the way the patient is treating him, thereby interpersonalizing his own
(the analyst’s) dissociation. In mutual enactment, in other words, both the
analyst and the patient externalize dissociated material; both analyst and
patient are caught in an unconsciously motivated way.
Let us say that the fictional person I have just introduced becomes a
psychoanalytic or psychotherapy patient. And let us say that paying his
analyst’s fee has been a bit of a strain, and might even mean that he will
have to forego something he wants, such as a vacation to a place he has
been thinking about visiting for years. At this point the analyst brings up
raising the fee. The patient is taken by surprise. He was not prepared for
this, and he is attached enough to his vacation plans that he cannot adopt
his usual, automatic attitude of self-sacrifice. He is threatened with feeling
a powerful surge of self-interest, one that will conflict with the interests of
his analyst. If this happens, the patient will be not-me: he will feel intoler-
ably greedy, ashamed, and self-hateful. Instead of trying to negotiate the
issue, which is just too risky a course, the patient conveys, in words or in
conduct, something like this to the analyst, thereby interpersonalizing the
dissociation: “You’re some kind of greedy bastard, aren’t you, charging
such a high fee? Aren’t you already getting plenty out of me?”
If the analyst is also someone who cannot tolerate a certain degree of
self-interest (and many such people do become psychoanalysts, after all),
he responds in a way that interpersonalizes his own dissociation, his own
version of not-me, conveying in one way or another that (for example) the
patient is wildly unreasonable, that he (the analyst) has perfectly justifi-
able expenses that require that he collect thus-and-such a fee, and so on.
This is not the only thing that could happen, of course, and not even the
most common. Let us say that the analyst can tolerate his own self-interest
without feeling too greedy and ashamed. In that case, the clinical interac-
tion still qualifies as enactment, but not as mutual enactment. This kind of
enactment is what takes place when, in the face of one person’s enactment,
the other person in the interaction manages to respond in a way that rec-
ognizes the dissociation without responding from within a dissociation of
his own. The analyst who is not caught in a reciprocal enactment is instead
able to mentalize (Fonagy, et al, 2002) his own experience, saying to him-
self something like, “OK, I’m bent out of shape here. This person is getting
under my skin. I hate feeling this way. I’m angry at the patient over the
accusation, and maybe a little guilty about my fee. But exactly what is hap-
pening to me? Why am I angry? Why guilty? What exactly is the patient
setting oἀ in me? What can I learn about the patient from my reaction?”
Let us go back for a moment to the observation that our perception of
events lags behind the events themselves. The perceptions that would allow
you to know or sense what is happening with the other are not immediately
available, under even under optimal conditions. And so we must conclude
that there is always a brief period after every interpersonal event during
which we literally don’t know what we are doing. If we think of enactment
as unconscious participation in relatedness, the time lag might lead us
to take the position that all interactions are enactments, or have enactive
qualities, even those interactions in which both participants’ perceptions
of ongoing interaction are relatively unobstructed. I have not adopted that
way of defining enactment, however, because the phenomenon would then
have to be understood as omnipresent. We would never cease enactment;
and in this way the term would lose any specific meaning. And so, despite
my acceptance that all interaction does indeed have unconscious aspects,
I prefer to restrict the use of the term “enactment” to the interpersonaliza-
tion of dissociation. I oἀer a fuller exploration of these issues in the chap-
ters that follow, especially in Chapter 4.1
1 It goes without saying that all conduct, both the patient’s and the analyst’s, is partially the outcome
of unconscious motivation. I want to be clear that I am restricting only the use I am making of the
term “enactment,” not the range of unconscious motivation that is involved in ongoing interaction
between analyst and patient. A great deal of nonconscious motivation and other kinds of influ-
ence do not qualify as dissociated, i.e., are not unconscious for defensive reasons. Think of all the
mutual regulatory influences we pass back and forth in the clinical situation without ever being
aware of doing so (e.g., Beebe & Lachmann, 1998, 2002), or consider the procedurally represented
experience that the Boston Change Process Study Group (1998, 2002, 2005, 2007, 2008) refers to
as “implicit relational knowing.” In Bucci’s (1997) multiple code theory, much of the experience
encoded nonverbally, and especially subsymbolically, is nonconscious (at least in verbal-reflective
terms), although not necessarily for any kind of defensive reason. Bucci suggests that aἀect, in
particular, is encoded subsymbolically.
This point is hardly unique to the dissociation model, of course. The same point can be made in
classical models, in which much nonconscious influence on conduct and experience comes from
sources that are descriptively but not dynamically unconscious—i.e., not repressed.
Dissociative Enactment and Projective Identification
No doubt my fictional example provokes some readers to think about pro-

jective identification. It does seem that the two conceptions are similar in
certain respects. Most meaningfully, both are ways of thinking about how
unconscious influence is conveyed from one person to another, and a suc-
cessful outcome in both cases has to do with transformation of experience
from a “raw” form that cannot be used in making meaning (not-me in the
terms of dissociation theory; beta elements in Bion’s [1962, 1963] frame of
reference) to a more benign and useful form (me; alpha elements).
The similarities, but particularly the diἀerences, between these two
views deserve much more careful study and exclusive focus than I can
give them here. The issues are complex, both because of the concepts
themselves and because diἀerent authors employ them diἀerently. Neither
enactment nor projective identification means the same thing to all those
who use the terms.
And yet, because I imagine that the clinical examples that appear in
most of the chapters of this book will continue to make readers think
about projective identification, I will hazard the beginning of an analy-
sis of the diἀerence between the concepts. I do so, though, as an aid to
the reader rather than as an attempt to stake out a position. That larger
goal requires more scholarly study and documentation than I give the
hypothesis here.
The diἀerences between the ideas of enactment and projective iden-
tification that are most relevant to the clinical problems I am discuss-
ing in this chapter revolve around the ways the analytic relationship is
conceived. Among those analysts for whom projective identification is
central, analyst and patient occupy complementary roles, container and
contained, and the most aἀectively charged exchanges between the two
participants are understood to take place mostly within psychic life, less
frequently in the actual conduct of analyst and patient. If the exchange
spills over into conduct in a notable way, especially if the conduct is
the analyst’s, and especially if it is the kind of distressing conduct that
is often the heart of the matter in what are considered enactments by
Interpersonal and Relational analysts, the containing function is under-
stood to have failed.
For Interpersonal and Relational analysts, on the other hand, for
whom the roles of patient and analyst are more mutual and reciprocal
than complementary, the stage on which enactments are played out has a
more complex geography. It encompasses psychic life, of course; but the

unconscious parts of both the analyst’s and the patient’s psychic lives are
also expected to play out on the field of actual conduct. There is a greater
emphasis than in the work of those for whom projective identification is
central on the way analyst and patient treat one another. The examination
of what is happening in the room between the two people—the details of
what is taking place, and the meanings each participant (i.e., not only the
patient, but the analyst as well) ascribes to those things—is a routine part
of Interpersonal and Relational practice. This way of working derives from
the emphasis on relatedness that is the subject of this chapter.
Of course, psychic life is hardly the exclusive precinct of those who
think in the terms of projective identification, nor are Interpersonal and
Relational clinicians the only analysts who investigate clinical relatedness.
Analysts from both orientations see unconscious material reflected in both
the minds and the conduct of the participants. It is a matter of emphasis:
Interpersonal and Relational analysts, I believe, simply look more rou-
tinely than their colleagues for reflections of enactment in their own con-
duct and that of the patient. It is an important aspect of what is distinctive
about Interpersonal or Relational practice. The kind of enactment that
is often felt to be a failure of containment by those who use the concept
of projective identification is often the very event, for Interpersonal and
Relational analysts, that sets in motion the most important work of the
treatment. For these analysts, it is frequently only via aἀectively charged
enactments that not-me, because it cannot be symbolized, can appear in
the treatment at all.1
1 Mitchell (1997, Chapter 4) argues that interaction in the Kleinian frame of reference is not fully
mutual, that there remains in that body of work the (unstated) assumption that there are impor-
tant limits on the vulnerability of the analyst to the patient’s unconscious influence. On that view,
because the Interpersonal/Relational perspective is fully social, one might argue that projective
identification and dissociative enactment diἀer on the basis of the degree of their acknowledg-
ment of the analyst’s unconscious involvement with the patient. That may be the case. But the
problem is complicated by the fact that it is primarily Bion’s revision of the concept of projective
identification that is used today; and those who use Bion’s revision do not necessarily describe
either Bion or themselves as Kleinians (e.g., Ferro, 2005a), and in fact feel that Kleinians did not
know what to do with Bion’s revisions of Klein’s ideas. Ferro (2005a) writes that Bion, “while he
derives from a definitely Kleinian matrix,” developed his thinking to the point where “it was no
longer a linear expansion of the previous [model], but one which brings about a quantitative leap”
(p. 1535). And so analyses like Mitchell’s of Kleinian writers do not necessarily apply to the under-
standing of writers such as Ferro, who is one of the contemporary writers best known for his use of
projective identification. All of this complication is the reason that the comparison of dissociative
enactment theory and contemporary models of projective identification requires careful study
and documentation.
Not-Me Does Not Require Trauma
Sometimes, perhaps because of my use of dissociation, readers have been

left with the impression that not-me is routinely the outcome of trauma. I
want to be careful to avoid that implication. Trauma is indeed often cen-
tral in the creation of not-me; but not-me can also be the outcome of much
less easily observed psychic processes.
The fictional patient I just discussed, for instance, might have grown up
worrying that, if he were demanding about the satisfaction of his wishes,
he would have hurt or overwhelmed a parent whom he believed to be weak
and emotionally inept. To demand too much from a loved person who is
nevertheless perceived in this way could provoke the sense that one’s own
demands are hurtful enough to others to be intolerable to oneself. Such
a person might come to feel, too, that his angry reaction toward the par-
ent’s perceived weakness was also not-me, since it might very well seem
to the patient that the parent would be helpless in the face of the patient’s
anger, resulting in guilt and shame on the patient’s part. Under the right
circumstances, such guilt and shame might come to seem intolerable. The
patient might also worry that he would enjoy causing the parent pain, for
example, and might be unable to bear the guilt provoked by the formula-
tion of this recognition.
The point I want to emphasize about this elaboration of my example
is this: it would be possible for the fictional patient I have described to
develop and hold his perceptions of the parent for any number of reasons,
and some of these would be unformulated reasons with dynamic signifi-
cance. That is, the relational origins of not-me are not limited to observ-
able, traumatic events. To put it another way, the origins of not-me are not
limited to events that have actually taken place in the external world.1 No
trauma would necessarily have had to take place in order for the fictional
patient I have described to grow up feeling that to be demanding or angry
is not-me.
Note that in this example, the patient’s “worries” about the parent’s
emotional capability would never have been formulated. These “worries”
can be conceived as something other than unconscious fantasies, as that
concept is usually used; they can be understood, that is, as possibilities for
meaning that are dealt with by never being fully known, felt, or sensed.
1 Any psychoanalytic understanding of trauma, of course, includes more than external events. The
case has also been made, though, that psychoanalytic views of trauma have given short shrift to
the actual traumatic events, especially in the case of adult onset trauma (Boulanger, 2007).
These possibilities for experience are dissociated for unconscious rea-

sons without ever having been formulated, and are then maintained in
an unformulated state; that is, they remain potential experience that has
never been actualized, not fully formed experience that has been repressed.
Unless or until the interpersonal field allows enough safety to formulate
the experience (and that outcome requires the successful resolution of
enactment), not-me remains unformulated indefinitely, and enactment of
not-me comes and goes as the interpersonal context shifts. The way the
concept of dissociation is used in this book, then, is not specific to trauma.
Instead, as I have already claimed, dissociation and unformulated experi-
ence comprise a model of mind (D.B. Stern, 1997, 2009).
Unformulated Experience/Unmentalized Experience
Links have frequently been drawn between the idea of unformulated expe-
rience and certain other conceptions of unconscious experience, particu-
larly Bollas’s (1987) “unthought known,” Stolorow’s (Atwood & Stolorow,
1984; Stolorow, Brandchaft & Atwood, 1987; Stolorow, 1988) “prereflective
unconscious,” the “prelogical experience” of Tauber and Green (1959), and
Wolstein’s (1982) “transconscious experience.” But there is another link
that is at least as relevant to emphasize, the one between unformulated
experience and the various ways of understanding unmentalized expe-
rience. There is by now a substantial literature having to do with what
Lecours and Bouchard (1997), in a review of that body of thought, describe
as “the necessity of the mental transformation of raw, concrete, ‘unmenta-
lised’ experiences” (p. 855). Unmentalized experience is “raw” in the sense
that it has not been symbolized. Mentalization is symbolization of one
kind or another, and it makes thought, feeling, and the life of the mind
possible.
I have already cited Fonagy et al. (2002) in the context of unmentalized
experience. I add to that the work of Mitrani (1995) and, most recently,
Botella and Botella (2005). Earlier influential contributors to this literature
are Bion (1962, 1963), Segal (1957), and the Psychosomatic School of Paris
(Fain & David, 1963; Fain, David & Marty, 1964; Green, 1975; McDougall,
1985; Chasseguet-Smirgel, 1990; Marty, 1990, 1991). The task of mental-
ization, broadly defined, is one of the greatest challenges of infancy and,
in intrapsychic views, pre-exists the development of repression, which can
only come about once experience of a certain degree of structure has been
created. The wish, too (again, from within the classical perspective) is pos-
sible only after it becomes possible to think (e.g., Green, 1975). In mod-
els in which repression is central, experience can only be expelled from
consciousness—repressed—once it attains the kind of symbolic form that
awareness requires. Experience that is not symbolized cannot be thought,
nor can it be repressed, because such experience can never have been con-
scious in the first place.
That is the position I have always taken about unformulated experience,
although I reached it from a very diἀerent direction, one in which, among
other diἀerences, the primary defense is dissociation, not repression (D.
B. Stern, 1997, 2009). Here is an early presentation of the point (D.B. Stern,
1983):
Unformulated material is experience which has never been articulated clearly

enough to allow application of the traditional defensive operations. One can
forget or distort only those experiences which are formed with a certain degree
of clarity in the first place. The unformulated has not yet reached the level of dif-
ferentiation at which terms like memory and distortion are meaningful (p. 73).
Not-me, and all experience that has remained unformulated for uncon-
scious defensive reasons, therefore conveys something important in com-
mon with unmentalized experience: it has not been formulated and then
expelled from consciousness; rather, it has never been symbolized at all.
And therefore, as in theories of mentalization, the task of treatment as con-
ceived in this book is to make formulation possible where it was impossible
before, and in that way to expand the limits and the capacities of the self
or mind. Furthermore, accounts of mentalization and the theory of unfor-
mulated experience share the position that the rendering of symbolically
mediated experience is a continuous and vulnerable process that can be
interrupted, temporarily or for longer periods, by the ongoing conscious
and unconscious events of psychic and relational life. Both kinds of theory
conceive the creation of experience that can be used in making meaning
as a continuous process.
Relatedness and the Future of Meaning
Those who theorize the roots of transference to lie in unconscious fan-

tasy—even if, like many Freudian analysts today, they think about
unconscious fantasies not as configurations of drive and defense, but as
repressed narratives—are at the very least taking the perspective that
clinical relatedness does not itself constitute the object of clinical interest.
Relatedness is instead conceived as a means of access to that object—an
expression of the mind or something outside the mind, not part of the
mind itself. For those who depend on the concept of unconscious fantasy,
clinical relatedness is like the image projected on a movie screen: it con-
tains what you want to know, but if you want to aἀect the image in any
permanent way you had better ignore the screen and go find the film.
I believe, on the other hand, that the possibilities that, if actualized,
open experience in just the ways intended in psychoanalytic treatment,
are quite literally the possibilities of clinical relatedness. We do not need
to try to see “through” clinical events to the real substance, the fantasies
lying behind or beyond them. The future of any treatment instead exists
in the unseen, unformulated possibilities embodied in the conscious, and
especially in the unconscious, parts of clinical process, right there in how
the patient and analyst grasp one another’s personalities and intentions,
what they do with one another and how they feel about it. As the possibili-
ties for this jointly created interpersonal field unfold—some of them dis-
appearing without ever having been realized, others becoming actualized
as they emerge into a moment of relational relevance—they continuously
set the changing parameters for what patient and analyst can experience
in one another’s presence. In the simplest terms, what I want to say is this:
The experience that it is possible for the analyst and the patient to have in
one another’s presence is shaped by the nature of what takes place between
them. The possibilities between them are not contributed by a separate
source of meaning—i.e., fantasy—that shapes and influences relatedness
to conform to its image. The possibilities between analyst and patient are
instead the unformulated possibilities of relatedness itself.
This is what I mean by when I claim that meaning, at least the most psy-
choanalytically relevant meaning, is embodied in relatedness. Relatedness
is mind in action, and so what you see is, in an important sense, what
you get. It is often reasonable to take the position, and especially often
reasonable in clinical work, that what mind does is what it is. Relatedness
is simultaneously the site of the mind’s potential and a very significant
part of what it does. Relatedness does not, therefore, bear the relation to
mind that the image on the screen bears to the film. That is, relatedness
is not merely an expression of mind—relatedness is part of mind. This is
one of the most important areas in which it makes sense not to conceive
mind as unitary, not to conceive it as contained in the brain or the head, or
even as located somewhere inside the person, but as distributed. ἀ e future
of meaning is embodied in relatedness. That is, to repeat the point of this
chapter, the nature of the interactions in which we participate, and that

we are partially responsible for creating, has a great deal to do with which
versions of our unformulated experience we articulate or realize.
Psychoanalysts do not know in advance which new emotional/relational
contexts will transform old meanings into new ones, and it is unlikely
that they could set out to create those contexts even if they knew before-
hand which ones they were. Spontaneity is required. We must feel our
way toward the fulfillment of our intentions, allowing those intentions
to create and recreate themselves along the way. That is one of the lessons
of Alexander and French’s (1946) now long-ago experiment with correc-
tive emotional experience: we are all aware now that, for any number of
reasons, it just does not work to try to suss out the transference and then
calculate the correct way to avoid confirming it. The problem is not only
that we cannot predict the meaning to the patient of our interventions.
We certainly can’t predict such things—that is true as far as it goes. But
our inability to predict is not the primary problem. We must be both more
authentic and less calculated than the Alexander and French technique
allows, however well intentioned it was at the time. It is not enough to
know how to do our work, although knowing how to do it is essential. We
have to mean what we do.
We do not necessarily have any idea when we and our patients will
find ourselves in a transformative context. It just happens. I (D.B. Stern,
1983, 1997) have referred to this event, after Bruner (1979), as “eἀective
surprise,” and have described the process by which we try to encourage
these events (we cannot simply cause them or will them into being) as
“courting surprise” (D.B. Stern, 1990, 1997). Under favorable conditions,
that is, under conditions that allow therapeutic action, we notice, after
the fact, that the emotional/relational context has changed, and we see its
impact on us. In the chapters that follow this one, those moments come
about over and over again. The heart of our work takes place beyond our
capacity to grasp the events that make it up. That does not mean that it is
unlearnable; it means, rather, that it is a form of praxis. Like art and like
craft (e.g., Sennett, 2008), and like the mind itself, psychoanalysis is what
it does.
Enactment is a kind of extreme selective attention, a set of perceptions
of the other, and of oneself in relation to that other, that are so rigid that
no other possibility can be imagined, at least temporarily. Because any
perception that grows from need inevitably has a constricting eἀect on the
range of interpretive understandings we can allow ourselves to have, any
perception of the other person that represents greater freedom or curiosity
is inevitably broader than the transference and countertransference that

preceded it (D.B. Stern, 1997, Chapter 7). Need stifles curiosity by shining
light only into certain corners; it requires continuous eἀort, and the kind
of intentionality that I have claimed goes beyond consciousness, not to
succumb to this selectivity, to stay open to our capacity to allow alterna-
tive perceptions to form in our minds. Over and over in this book, it will
happen in clinical illustrations that enactments end in the appearance of a
new perception of the other.
And so, here at the beginning of this book I reaffirm and extend the
theme of Unformulated Experience: we best serve those with whom we
work not by attempting the revelation of pre-existing psychic content,
but by working toward a greater freedom to experience, an unlocking of
the aἀective/relational problems, large and small, that arise between our
patients and ourselves. As we unlock those problems, the possibilities for
the analytic relatedness expand; and as they expand, the possibilities for
the experience analyst and patient can have in one another’s presence also
expand.
Because both the analyst and the patient are involved in the related-
ness that grows up between them, oἀering help of this kind requires that
not only the patient, but the analyst, find a way to accept a greater free-
dom to experience. Patient and analyst must become partners: partners in
thought, yes, but also partners in feeling and every other aspect of expe-
rience. We work toward a wider range of aἀects, thoughts, and percep-
tions that allows us to feel and sometimes formulate meanings that have
remained outside our capacity to live. We work toward a resumption of
the therapeutic conversation that has been interrupted by enactment, a
resumption made possible by our acceptance of some small part of what
has been not-me. The interruptions are as important as what they inter-
rupt. The aim is less to know something diἀerent than to be something
diἀerent. I hope that the chapters of this book contribute to the articula-
tion of such a vision of our field.
2
Conversation and Its Interruptions
We say that we “conduct” a conversation, but the more genuine a conversation

is, the less its conduct lies within the will of either partner. Thus a genuine con-
versation is never the one that we wanted to conduct. Rather, it is generally more
correct to say that we fall into conversation, or even that we become involved
in it. The way one word follows another, with the conversation taking its own
twists and reaching its own conclusion, may well be conducted in some way, but
the partners conversing are far less the leaders of it than the led. No one knows
in advance what will “come out” of a conversation. Understanding or its failure
is like an event that happens to us. Thus we can say that something was a good
conversation or that it was ill fated. All this shows that conversation has a spirit
of its own, and that the language in which it is conducted bears its own truth
within it—i.e., that it allows something to “emerge” which henceforth exists.
—Hans-Georg Gadamer (1965/2004, p. 385)
What Is Conversation?
In the chapter following this one, I discuss dissociation and enactment

from within a hermeneutic frame of reference. In this chapter, I contex-
tualize that chapter, and the ones that follow it on dissociation and enact-
ment, by revisiting the way that the hermeneutic philosopher Hans-Georg
Gadamer (1965/2004), who has influenced me for many years, thinks about
the matter of true conversation, or collaborative, authentic dialogue.1 I
have been drawn to think of dissociation and enactment as the uncontrol-
lable, unconsciously motivated interruption of that kind of dialogue (e.g.,
D. B. Stern, 1990, 1991, 1997; see also Chapter 3). And so understanding
1 Gadamer’s magnum opus, Truth and Method, was published in German in 1960, but it was not
published in English translation until 1975. In 1965 a second edition of the book had appeared,
and it is this edition that was translated into English. Then, in 2004, a revised translation of the
second edition was published. I have used the revised translation throughout this book. To repre-
sent this complex bibliographic history, I cite the book thusly: Gadamer, 1965/2004.
25
the nature of true conversation in its uninterrupted form should set us on

the path for the chapters to come.
Hermeneutics is rooted in historicity, the perspective that what we
consider good and bad in life—in fact, what it is possible for us to find
meaningful at all—is the outcome of our embeddedness in history. Our
time and place simultaneously define the range within which we can cre-
ate our possibilities for meaning and limit our access to possibilities out-
side that range. By participating in what Gadamer calls true or genuine
conversation, we can understand the limitations imposed by history to
some degree, thus making it at least somewhat more feasible for us to take
them into account and sometimes even to transcend them. Gains of this
kind come, first and foremost, from the conversation itself, of course; but
they also arise from the assumption of historicism we start with, which
encourages us to assume that we are blind to possibilities that, if given the
opportunity, we might rather know about.
For Gadamer (1965/2004), understanding emerges in dialogue; and
dialogue has a very particular meaning to him, a meaning that lies at the
heart of what most psychoanalysts believe is the ideal of clinical process.
Conversation is a process of coming to an understanding. Thus it belongs to

every true conversation that each person opens himself to the other, truly
accepts his point of view as valid and transposes himself to such an extent that
he understands not the particular individual but what he says. What is to be
grasped is the substantive rightness of his opinion, so that we can be at one with
each other on the subject. Thus we do not relate the other’s opinion to him, but
to our own opinions and views. (p. 387)
The idea that we should try to understand “not the particular individual
but what he says” may seem to contradict the aims we pursue in clini-
cal practice. But I do not believe it does; Gadamer is merely trying to say
that any conversation that matters must be about something, and that the
dialogue is successful to the extent that the conversants find their way to
common ground about whatever that is. Furthermore, in consonance with
contemporary thinking about the analyst’s use of her own experience,
Gadamer believes that we understand the other’s position by understand-
ing ourselves. It would be better, actually, to make this point reciprocally:
we reveal ourselves in our own eyes by understanding the other, and we
understand the other via what we learn about ourselves. In any case, the
route to an understanding of what the other is trying to say lies through
our own experience. We must change our own experience of the other to
understand her.
Conversation and Its Interruptions 27
Gadamer’s reference in this passage to “substantive rightness” may be

jarring to those of us who think of psychoanalytic work as a matter of sub-
jectivity and intersubjectivity. But Gadamer is not speaking epistemologi-
cally; that is, he is not embracing objectivism. Rather, he is emphasizing
that the partners in a conversation are not acting directly on each other,
as they would be, for instance, if they were yelling at each other or kissing
each other. That is what would constitute subjectivity in his frame of refer-
ence: some kind of statement or action that conveys, “This is what I want
from you; or this is what I am going to do to you.” The substantive right-
ness Gadamer is describing is defined by an attempt by each conversant
to address what he understands the other to be thinking and feeling. The
focus, in other words, is on developing a sense of the other’s meanings, not
in getting something from the other or making something happen.1
The process of conversation is therefore inevitably and continuously
interpretive; and the interpretations are motivated, to whatever extent
each participant can manage, by the attempt to see the object of the dia-
logue the way the other is seeing it. Along the way, of course, this extended
interpretive process also features other kinds of interpersonal processes,
such as disagreement, paradox, prescription, cajoling, and exaggeration.
And sometimes, of course, the therapist will think she is simply attempt-
ing to understand the patient, but unconsciously she is attempting to make
something happen.
Each partner in the conversation begins the process of understanding
from within his own horizon of understanding. When the conversation
is successful, the interpretations of each by the other come closer and
closer to each other, each participant repeatedly modifying her interpre-
tation to conform to what she now understands the other to mean. There
occurs what Gadamer repeatedly refers to as “agreement,” a word that is
not meant to imply that the conversants believe the same thing, but that
they understand the same thing. Belief is an entirely separate matter, in
the same way that it is in clinical process, where the question of whether
we believe the patient is right or wrong is virtually never a primary con-
sideration, or even a relevant one, in guiding our listening. Gadamer’s
1 Of course, if we are to be true to clinical reality here, we also have to add that, in the very act
of understanding the other, we participate in creating the meanings to be understood. We must
acknowledge that our attempt to understand is hardly “pure,” in the sense that we are doing noth-
ing else with the other than pursuing understanding. We must acknowledge, in other words, that
we are always also acting on the other in ways that we may or may not be aware of. We are, in other
words, always mixing our attempt to understand with other intentions that actually do say “This
is what I want from you” or “this is what I am going to do to you.” More about this later in the
chapter.
famous expression of this idea of agreement is “the fusion of horizons”

(see Chapter 3).
And so, for Gadamer, the conversation that works its way toward truth
focuses on what emerges between two people, not on what one of them
finds within the other. Perhaps most immediately, this part of Gadamer’s
thinking reminds psychoanalytic readers of the ideas contributed by
Benjamin (e.g., 1999, 2002, 2004) and Ogden (e.g., 1994, 2004) that two
subjectivities working together in the analytic space co-create an “analytic
third,” an intersubjective relationship between them that has an indepen-
dent existence. In a more general sense, Gadamer’s description of a true
conversation is what all clinicians understand as a smoothly progressing
analytic inquiry.
Gadamer himself had little to say about psychotherapeutic conversa-
tions, except (oddly enough) to indicate his belief that his work was prob-
ably not applicable in that setting. For example, just after the passage I
quoted above, Gadamer goes on to say this, contrasting a focus on what he
calls “individuality” with the more hermeneutic emphasis on what is said:
“Where a person is concerned with the other as individuality—e.g., in a
therapeutic conversation or the interrogation of a man accused of a crime—
this is not really a situation in which two people are trying to come to an
understanding” (p. 387). It is clear from Gadamer’s (1965/2004) equation
of “therapeutic conversation” with “the interrogation of a man accused
of a crime” that he does not view psychotherapy as we do. Apparently,
Gadamer thought that psychotherapy was limited to some kind of attempt
to influence, examine, blame, or change the other—and that, of course,
would not be what he meant by a true conversation. He does not seem to
grasp that individuality could itself become the object of true conversa-
tion. What those of us who practice it understand as psychotherapy, of
course, is consistent with both individuality and true conversation. The
ideal of psychotherapy and psychoanalysis, like the ideal of true conversa-
tion, is the relatedness of I and Thou.
Gadamer’s work is, in part, a response to the early 19th-century
Romantic hermeneutics of Schleiermacher, one of the great hermeneutic
innovators. A review of Gadamer’s divergence from Schleiermacher illu-
minates the source of Gadamer’s antipathy to the view that understand-
ing is a grasp of the other’s mind. For Schleiermacher, says Gadamer
(1965/2004), “the author can really only be understood by going back to
the origin of the thought” (p. 186). What does that require? It requires that
one reconstruct the very process of creation—from within the mind of the
creator. It requires that one “identify” with the creator. In the end, because
the reconstruction of the creation reveals many aspects of the process

that were unconscious to the original creator, the one who understands
what an author has written, according to Schleiermacher, aims “to under-
stand a writer better than he understood himself ” (p. 191; emphasis in the
original). Keep in mind that this reconstructive understanding does not
require the contribution of the other; the other has nothing to do with
it, apart from, of course, being himself the object of the understanding.
That is, Schleiermacher developed the Romantic belief that understanding
requires neither mutual understanding nor conversation or dialogue, but
rather an empathic understanding of the individuality of the creator of
the thought, an understanding that is created in the monadic or isolated
mind of the one who understands. Gadamer believes that Schleiermacher’s
hermeneutics has been handed down in an uninterrupted way from then
to now; the understanding of the other through identification is “a formula
that has been repeated ever since” (p. 191). “It is ultimately a divinatory
process,” writes Gadamer (1965/2004):
a placing of oneself within the whole framework of the author, an apprehension

of the “inner origin” of the composition of the work, a re-creation of the creative
act. Thus understanding is [for Schleiermacher] a reproduction of an original
production, a knowing of what has been known …, a reconstruction that starts
from the vital moment of conception, the “germinal decision” as the composi-
tion’s organizing center. (p. 186)
Contrast this view with Gadamer’s claim that understanding cannot

be monadic, but must be mutual and must concern not the source of what
must be understood, but its actual nature. We understand not by recon-
structing the mind of the other, but by doing more and more to remove
the diἀerences between his view and our own. From the other side of the
dialogue, the partner in conversation is doing the same thing. There is a
process of mutual accommodation. Each of us begins by identifying the
other’s meaning with what it most reminds us of. Gadamer calls the ini-
tial meaning we bring to bear a “prejudice” or a “preconception,” words
that are used without opprobrium in this context. In true conversation,
we deal with the other via preconception only at the outset, because as
soon as possible we allow the other’s contribution to the dialogue to query
us about what we mean. Sometimes the query from the other is literal.
Imagine that my preconception of the other’s meaning is that he is angry
with his spouse. Perhaps I say, “So you must have been very angry.” But the
other was not angry, or at least he did not think he was, and he responds
by saying, “Well, no, I wasn’t really angry. I was more baffled than angry.”
My impression of what he intends to say now changes, and whatever I say

next reflects that change. The situation continues to develop.
A similar sequence of questions and answers might occur on a more
implicit or nonverbal level. Let us say that, in response to my own impres-
sion that the patient is angry with his spouse, I make a sympathetic noise;
or maybe I convey something by the very absence of a response. The patient
understands, in any case, that I have the wrong impression and says, “No,
no, I was actually more baffled than angry.” And in addition, of course, we
could construct a scenario in which both sides of the conversation went on
without explicit symbolization.
The kind of mutual questioning that furthers a true conversation can
take many forms. We often hold psychotherapeutic conversations, for
instance, in which, despite one’s acceptance that the other person believes
what he says, one still thinks it is possible that a diἀerent meaning is sub-
merged, or potentially present, in that content. In other words, true con-
versation can include the attempt to discuss meanings that one member
of the conversation believes are either unconscious or implicit in what the
other says. In my example, perhaps I think that the other patient really is
angry with his spouse, but does not want to know it; that is, I think that his
bafflement, while sincere, is defensive. Gadamer’s model can be expanded
to include this kind of psychoanalytic example because the fusion of hori-
zons is not necessarily defined by an agreement between partners in dia-
logue that an unconscious meaning is present between them; the fusion
requires only that the partners both grasp either that the unconscious
meaning is a reasonable implication or that it is not. Of course, in the end,
the implication of an unconscious meaning by the analyst helps, when
it does, because it organizes experience of the patient that had not been
organized before (Loewald, 1960), and so the fusion of horizons usually
is comprised by an actual agreement about the nature of the meaning.
What I mean to emphasize is that Gadamer, like psychoanalysts, rejects
the authoritarian imposition of meaning; and like an increasingly large
group of analysts, he insists that meaning is jointly created.
Oἀering an understanding, provoking a question, modifying one’s
understanding—this is the stuἀ of conversation for Gadamer. We do not
reconstruct the literal history of the meaning that we find in the other, as
Schleiermacher believed. We do not directly construct the other’s mental
process. Each conversational partner interprets the meaning of the other,
absorbs the answering reaction, and then interprets again, until a fusion
of horizons comes about.
In this way, Gadamer’s portrayal of understanding is intersubjec-

tive. It requires two people; it is inevitably and continuously dialogic.
Schleiermacher’s is not; the object of understanding is more passive. He
need only make his meaning known. He need not simultaneously par-
ticipate in the more active way necessary for the one who understands in
Gadamer’s frame of reference.
Gadamer’s hermeneutics holds an implication for our theories of empa-
thy. As long as we mean by empathy that two people come into mutual
understanding, we do not contradict the hermeneutic perspective. But a
true conversation cannot take place between two people if only one of
them is trying to understand the other; nor can it take place if the under-
standing that is sought is felt to be grasped directly from the experience
of the other, as Schleiermacher portrayed it, as if, that is, the one who
understands somehow looks into the mind of the other and absorbs it
through identification. Empathy is an interpretive process, not a direct
apprehension of meaning (e.g., D. B. Stern, 1994), and it must be recip-
rocal. Of course, the psychotherapist or analyst is usually more focused
on the patient’s experience than the patient is on the therapist’s; but the
ongoing process needs to be reciprocal in at least the sense that therapist
and patient intend to understand what the other means by what she says
and by her conduct. Each needs to understand his or her task as the cease-
less contribution of meaning, the querying of the meaning oἀered by the
other, and acceptance of the other’s queries of one’s own meaning. When
empathy is reciprocal, it issues in mutuality, not in the grasp of one person
by the other. Racker (1968) puts this point beautifully: “To understand is to
overcome the division into two. … To understand, to unite with another,
and hence also to love, prove to be basically one and the same” (p. 174).
One of the points in Gadamer’s thought that has always fascinated me
and that I think links it most closely to clinical process is the fact that
conversation, if it is going on between two people who are doing their
best to understand each other, takes its own course, and that that course
is unpredictable. (See the epigraph at the beginning of this chapter.) The
course of an analysis can never be planned, even from one day to the next,
from one moment to the next. We do not know what will happen, and we
cannot sense where we will end up.
Why is this the case? In Gadamer’s (1965/2004) terms, it is because
of “the priority of the question in all knowledge and discourse that really
reveals something of an object. Discourse that is intended to reveal some-
thing requires that that thing be broken open by the question” (p. 357,
emphasis in the original). Gadamer goes on:
To ask a question means to bring into the open. The openness of what is in ques-
tion consists in the fact that the answer is not settled. It must still be undeter-
mined, awaiting a decisive answer. The significance of questioning consists in
revealing the questionability of what is questioned. It has to be brought into this
state of indeterminacy, so that there is an equilibrium between pro and contra.
The sense of every question is realized in passing through this state of indeter-
minacy, in which it becomes an open question. Every true question requires
this openness. (p. 357)
And so true conversation moves in mysterious ways. Seeing what is ques-

tionable is its heart. I return to my simple illustration, which I will now
embellish a bit for my present purpose. Let us say that I think my patient
may actually be angrier with his spouse than he seems to think he is. How
do I bring up this possibility? I could just ask him about it, of course. But
I will be much more eἀective in bringing something new into the open if I
can say something that conveys what suggests to me the questionability of
his conclusion. I need to make his denial of anger toward his spouse seem
possible within the terms of his own experience. And to do that, I must
try to grasp enough about what he means by what he says to make his own
prior, denying interpretation seem questionable in even his own eyes. So
after I say that he must have been angry, and he responds by saying that
no, he was really just baffled, maybe I say something like: “Well, a person
certainly could be baffled by what she said to you. But it seems to me that
a person could also be annoyed. Especially you. After all, your wife was
questioning your judgment about something that you pride yourself in
knowing about. It wouldn’t be surprising if you had felt annoyed, and as
a matter of fact, there was something in the way you said you were baffled
that made me wonder if you were. Do you know the tone of voice I’m refer-
ring to? Do you think that’s possible?”
Now, whether the patient responds in an accepting way to this remark
of mine has to do with many things. But if he feels reasonably safe with me,
and we have managed to develop at least the beginnings of a collaboration
about this aspect of his relationship with his wife, he may very well be able
to see, on the basis of my questioning, new possibilities in his own previ-
ous utterances. Something new has been brought into the open. He may
respond by filling out what I said, adding detail and emotional resonance
to it, until it once again begins to become something more than it was;
and at that point, perhaps I have another comment or question that opens
it even further. The outcome is unexpected and unpredicted because it
depends on the autonomous, emergent development of the subject matter
that exists in the between, not on the separate, individual experiences of

the participants. Something is, as Gadamer says, broken open.
And of course the same process could take place in reverse: The patient
might see possibilities to query something I said or did, and it might then
be me who, seeing the patient’s take on what is questionable in what I have
said or done, expands the material in an unexpected direction. No doubt I
also do not need to add that this kind of inquiry, in psychoanalysis, often
concerns the nature of the ongoing relatedness between the two conversa-
tional partners. In such instances it is the analytic relationship itself that
is brought into the open.
It is no accident, of course, that good psychoanalytic sessions often go by
quickly and eἀortlessly, and that bad sessions usually require backbreaking
work and can seem nearly interminable. Gadamer says that the forgetting
of ourselves is what happens when conversation is authentic. The subject
matter, or the play (a term that Gadamer shares with Winnicott, and uses
in a way that Winnicott would recognize), is all that exists; the players dis-
appear, at least in their own minds (see Gadamer, 1965/2004, pp. 110–118).
We are familiar with this phenomenon in psychoanalysis, where we write
things such as this about it: “There is the remarkable experience of being
carried along by something larger than both the therapist and patient: A
true sense of an interpersonal field results. ἀ e therapist learns to ride the
process rather than to carry the patient” (Levenson, 1984, p. 122, emphasis
is in the original).
At these times, analyst and patient are not living as much within their
individual consciousnesses of themselves as they are within the boundar-
ies of a joint event that is happening between them. Of course, this is a
complicated thing to say, because this joint event has everything to do
with the individuals who are part of it, and it requires from each partici-
pant a continuous and nuanced appreciation of his or her own experience.
I think I will have to leave it at this: Anyone who has been a psychoanalytic
clinician or a psychoanalytic patient knows what I am referring to.
But perhaps that is not all there is to say about it, after all, because, actually,
even those who have been neither patient nor analyst probably know what I
am referring to. As Gadamer tells us, the immersion of one’s self to the point
of feeling its loss is something that happens to everyone in true conversation.
The phrase of Gadamer’s that is translated as “true conversation” can
just as well be translated as “authentic conversation.” This authenticity is
key both to hermeneutics and psychoanalysis. But there is a seeming para-
dox here. Authenticity, Guignon (1965/2004) points out, is defined in two
ways that can appear to contradict each other. In one sense, the ideal of
authenticity is to own oneself, to achieve self-possession. We long to be all

that we can be, to self-actualize, to become what we are. In another pic-
ture of the good life, though, the ideal is what Guignon calls “self-loss” or
“releasement,” the giving over of one’s life to something greater than one’s
own feelings or needs. Do these two ideals not conflict? How can we own
ourselves if we give ourselves away?
The ideals actually do not conflict, for either Gadamer or psychoanalysis.
We have already seen that good sessions fulfill both ideals: In good sessions
we become more ourselves through a process in which we are, in certain
important respects, released from awareness of ourselves. Here is what
Guignon (2004) says about the matter. It applies as well to psychoanalysis
as it does to Gadamer’s thought, which is what Guignon is addressing.
Gadamer shows how the participants in the [authentic] conversation can leave
behind their self-preoccupations as they give themselves over to the to-and-fro
of the discussion. What becomes central in a dialogical situation of this sort
is not the opinion of this person or that person, but rather the subject matter
under discussion. We all have experienced conversations in which we get so
involved in the topic being discussed that we seem to become totally absorbed
in the discussion. The center and focus of an intense conversation is defined by
the ongoing play of ideas as they carry the matter at hand forward. The locus
of the activity as we experience it is not my mind and yours, but rather the
“between” made concrete in the issue of the truth of the matter we are discuss-
ing. In vital, intense discussions, egos fall away and are replaced by something
much more important: the matter that matters. (pp. 164–165)1
There are diἀerences over the question of what implications this view of
hermeneutics holds for psychoanalysis. I am going to develop what I want
to say on this point against the foil of a paper by Zeddies (2002), a paper
that I admire in every other respect than the one I am about to comment
on here (see D. B. Stern, 2002a, in which I make use of the points from
which I have learned). Like Guignon, Zeddies emphasizes that Gadamer’s
true conversation is not focused on the individual participants but on
the subject matter of the conversation: “In order to be carried along and
transformed by the subject matter, no participant can regard his or her
1 Philip Cushman (personal communication, 2009) comments on this quotation: “Perhaps we could
say that we are most ourselves when we are in the process of dialogue, of giving and receiving,
immersed in what Gadamer meant by the great circle of life. One way of thinking about this is to
draw from both Gadamer’s and Heidegger’s ideas about historicity. Through an encounter with
diἀerence, we come face to face with the enormous influence of history on us, and its limitations
on us as well. Through our social world, history constitutes us, and yet by encountering other
social worlds we come to realize our world is only one of many worlds, the truths of our world are
not the only truths. This is what prevents hermeneutics from being a historical determinism.”
own opinion or perspective as inherently superior, nor attempt to move

or coerce the conversation in any predetermined direction or toward any
particular goal” (p. 19).
So far, so good. But Zeddies develops this point in a direction it is
hard for me to follow, and that will be equally difficult, I think, for most
psychoanalysts. Zeddies starts out in a familiar enough vein. He writes
that, in transference/countertransference work, it may be necessary “to
understand what each individual contributes to the dialogue.” But then
he writes that this point applies mostly to “those who are in therapy to
have better relationships in life,” suggesting that this is a subgroup. For
me, having better relationships in life is no doubt the most important or
frequent reason for psychoanalytic treatment. It is over what Zeddies says
next, however, that I most disagree with him:
However, in contrast to much recent psychoanalytic writing, an overfo-

cus of [sic] subjectivity or intersubjectivity may actually prevent or obscure
Gadamerian true conversation. There must, Gadamer would say, be a third ele-
ment in the dialogue about which or through which the individual and collec-
tive subjectivities are concerned. Conversation and relationship cannot simply
be about us, lest we risk becoming disconnected from vitalizing traditions and
customs and consumed by narrow individual purposes and incentives. From
a hermeneutic perspective, it is through mutual concern with the subject or
topic of conversation that the individual subjectivities are enhanced and trans-
formed, and not through an open-ended, exhaustive exploration of individual
subjectivity itself. (p. 20)
Zeddies (2002) goes on to say that unless psychoanalysis looks “beyond

itself for moral and dialogical inspiration and grounding, its promises
will ring a bit hollow and its identity will be lacking in purpose and
meaning” (p. 20).
If I understand properly, psychoanalysis, with which Zeddies is gener-
ally quite sympathetic, is here being accused of navel gazing—or, in more
serious language, of being unwittingly hospitable to the expression of ide-
ology in the guise of therapeutics. I think that Zeddies is saying that an
ongoing focus on transference/countertransference can be excessive and
self-indulgent and can thereby lead to self-preoccupation and a turning
away from the things that most matter in life, probably social issues and
the ideal of community. If that is what he means, he seems to be suggest-
ing that to make the study of subjectivity the central preoccupation of psy-
chotherapy and psychoanalysis is to court self-centeredness, a perspective
with which I wholeheartedly disagree.
The point of psychoanalysis, it seems to me, is the mutual exploration

of subjectivity and intersubjectivity. In mutual study, these phenomena
become, in Gadamerian fashion, the matters that matter; they become the
objects of true conversation; they are precisely the “third element in the
dialogue” to which Zeddies refers. As a matter of fact, as far as I am con-
cerned, the ideal of “better relationships in life,” which Zeddies seems to
find wanting, is not only not self-indulgent; it is actually one way to repre-
sent the ideal of community. To have better relationships in life is to accept
one’s own narcissism fully enough to choose to live otherwise, is it not? To
have better relationships is not only to find in relationships more of what
one wants out of life, but also to be of more use to those one cares about—
more loving, authentic, courageous, generous. I do think that most psy-
choanalytic clinicians share this body of ideals.1
Zeddies has been among those who have made the point that psycho-
therapists must examine their practices with an eye toward revealing the
incursion of ideology.2 All these writers urge us to use the historicity of
hermeneutics to see how psychoanalysis is embedded in the larger cul-
tures to which it belongs. In particular, they have warned that psycho-
therapy can have the eἀect of valorizing self-contained individualism and
self-interest while drawing attention away from the ideal of community,
and that such an eἀect upholds the worst aspects of capitalism and the
rampant commercialism of our time.
The question is how best to address this concern. For me, there
is a danger in any recommendation about psychoanalytic treatment
that represents the straightforward imposition of a moral program.
To follow such a recommendation would be to fashion psychoanalytic
conduct along lines that do not grow out of whatever we can sense of
ongoing clinical relatedness. That does not mean that psychoanaly-
sis is not closely related to morality, of course. I have elsewhere taken
the position that our field is actually rooted in morality (D. B. Stern,
1996). We do need to find a way to apply hermeneutic principles to
psychoanalysis and psychoanalytic psychotherapy, but we need to do
1 I want to make sure not to be misunderstood about the question of ideals. What I am not propos-
ing is that this list of characteristics (“loving, authentic, courageous, generous”) comprises some
kind of humanistic catalog of the characteristics of human nature, or some essentialist vision of
morality. I am not suggesting that humans are “naturally good” or anything else in such a vein.
This is not the place for me to lay out in any detail what I do believe about the origin of human val-
ues. Suffice it to say, for the present purpose, that I take a hermeneutic view: We derive our values
from the traditions that comprise our possibilities for meaning.
2 See also Cushman (1990, 1995, 2005a, 2005b, 2007); Layton (1998, 2002, 2004a, 2004b, 2005);
Richardson and Zeddies (2001, 2004); Zeddies & Richardson (1999).
so by asking how we can use the insights of hermeneutics within our

existing convictions about the centrality of conscious and unconscious
clinical process.
If I read Zeddies correctly (now between the lines), he encourages us to
make a point of discussing moral and political issues with patients, or, if
not moral and political issues, then at least matters with moral and politi-
cal heft that fall outside the exploration of subjectivity and intersubjectiv-
ity. Now, it is true that we discuss with patients matters outside the analytic
relationship, and we do it all the time; but we do it from within the context
of what we sense is happening in the room.1 In other words, we do not
limit our clinical interest to an understanding of subject matter the patient
brings up; we also ceaselessly try to imagine that other subject matter, often
unexpressed, that always contextualizes the subject under discussion: the
place of the patient’s concern in the unfolding of the analytic relatedness.
Because we are always doing both these things, we never disregard clinical
process, and we always seek to embed our interventions in what we can
grasp and feel of that process. If we were to replace our focus on clinical
process with a diἀerent kind of focus, no matter how well intentioned—
some kind of concern with raising consciousness, for example—the treat-
ment could very easily become clumsy, inauthentic, argumentative, and
intellectualized. To structure psychotherapy according to any rule or tech-
nical prescription, without embedding the technique within what can be
sensed of the emotional environment being created in the here-and-now
with the patient, is to leave a psychoanalytic orientation behind.
What is it that is so significant about the value of clinical process as a
guide? Here I think we return to the matter of authenticity. Focusing our-
selves on what we can sense about the ongoing emotional situation in the
room, and then listening and participating from within the context of that
situation, is perhaps the most we can do to ensure the authenticity of our
work. Isn’t it the absence of a sense of authenticity, and especially of any
sense of releasement or self-loss (Guignon, 2004), that makes bad sessions
so hard and unproductive? And isn’t our most significant and challeng-
ing analytic task to find ways of participating that are both analytically
meaningful and personally authentic? Very often we feel that we are most
eἀectively satisfying our clinical demands of ourselves when we are least
aware of having to think about how to do the work; and we are least aware
1 Philip Cushman (personal communication, 2009) writes that this view amounts to “the encour-
agement to notice moral and political issues, but to do so when they emerge in the conversa-
tion, not to force them, impose them, or invent them.” In this connection, Cushman cites Gerber
(1990).
of thinking about what to do next when we are so thoroughly immersed

in the emotional sense of the session that we do not need to ask ourselves
what to do. We just know.
And yet, of course, we also have to admit that clinical process, and our
habits in understanding it, are just as thoroughly the products of history,
and thus just as completely moral and political, as any other social phe-
nomenon. Relying on clinical process as a guide hardly gives us a pass on
the necessity for developing a moral/political awareness of our work. So
what are we to do? It seems that we must do something (talk about social
issues, for instance), on the one hand, that we must not do, on the other.
Cushman (1995, 2005b) says that the task of writing about how to recog-
nize and discuss the political in psychotherapy is difficult, but not impos-
sible, and seldom attempted. He follows this proviso, though, by plunging
in, oἀering clinical vignettes that illustrate his contention that discuss-
ing the political in the clinical situation requires fashioning some kind
of compromise between Heidegger and Foucault. That is, it is not enough
to recognize the traditions that comprise our very being (Heidegger), nor
is it enough to deconstruct those traditions (Foucault). We must do both
(see also D. B. Stern, 2002b). Cushman intends his vignettes to illustrate
how he has managed this compromise, and the vignettes do convey what
he intends. But I find them more notable for something Cushman does
not comment on and probably did not set out consciously to convey: their
rootedness in the emotional atmosphere of the sessions in which they take
place. (Cushman 2005b) manages to consider the political and the clinical
simultaneously. Take as an example the following vignette:
Toward the end of a long therapy, [the patient] was struggling with a problem
at work that had profound implications for his life: when to accept jobs from
customers and when to turn them down. He came to realize that the moral
understandings about the good in his family of origin fit well with those of
the unscrupulous work world with which he was involved. He often felt placed
in the same moral predicament from which he suἀered in childhood. He is in
the position of being both victim and perpetrator: He is a victim because he is
forced to live in a framework he never agreed to and that he hates, and a perpe-
trator because in order to emotionally survive he must not care even though he
has promised to care, and because he has promised to do, properly, a job that
ultimately cannot be done properly, given his standards. “In other words,” I
responded, “you are in the middle of a moral conflict. According to the moral
code of your family and the society, you should accept the double bind: detach
and yet pretend to care, work conscientiously and yet know that the work is
impossible to accomplish, and all the while pretend not to notice that you are
feeling enraged, unsatisfied, and deadened.”
We came to realize that he is in the process of … [shaping] a new understand-

ing of what is moral. He has a sense that if he takes the job, he will be cheating
the client, living a life of detachment, denial, and falseness. And now he thinks
that is the wrong thing to do. But to refuse is to break the old family rules [and
the rules dominant in today’s society], to act in ways that they would think
immoral. He is caught in a dilemma
… Hoping to find others to encourage him to refuse the job, my patient
began asking his colleagues what they thought he should do. Contrary to
his hopes, each of them, without hesitation, advised him to take the job. The
money, they thought, was too good to turn down, no matter how damaging
it might be to him emotionally. …“It was like they stood up and spoke with
one voice! The voice was my mother’s voice, and it said, ‘Hey, your feelings
don’t count.’”
“Well,” I said, “it sounds like your mother isn’t the only one who thinks that
money is more important than emotional well-being.”
My patient smirked. … “Maybe it’s everyone—hell, it’s the system. That’s
the voice.”
“I think the voice you heard speaks through all these people, because it’s
sedimented in each of us. …”
My patient was silent for a moment. … “It’s speaking through everyone, he
said angrily, “my buddies, my competitors, everyone. It’s not just my mother,
as much as I’d like to blame it on her. It’s the whole system.” He thought for
a while, and then he looked up and smiled. “This is going to be more difficult
than I thought. How are we going to get the whole damn system into the room?”
We laughed together and I said, “You know, if it’s all the same to you, I’d rather
you take your ideas out there, instead of bringing … [everyone] in here.” (pp.
440–441)
There is nothing arbitrary about what Cushman (2005b) is doing here; if

he does have an agenda to raise the patient’s consciousness, you have the
sense that Cushman is subordinating it to his grasp of clinical process.
The words he contributes grow from his emotional sense of the session
and are responsive to the patient. Nothing is directly spoken about the
transference/countertransference, and yet everything that is said seems to
grow from it. One senses that Cushman is important to the patient, and
that the patient feels a kind of safety in his presence. I especially like the
laugh they share at the end, which conveys the comfort between them. In
my imagination, this comfort could accommodate an exploration of feel-
ings about Cushman that are less comfortable than those we see in this
example. It does not matter whether the patient actually has that degree
of freedom with Cushman—the example does its work by demonstrating
the viability of that possibility. This session does not sound intellectual-
ized or defensive; it strikes us as an authentic conversation, one in which
something new emerges.
Gadamer’s description of authentic conversation is a good fit for the

parts of psychotherapy and psychoanalysis that proceed with relative
smoothness, such as the basically friendly and collaborative interchange
between Cushman and his patient. A great deal of the time, actually, our
work does proceed that way. But that kind of work is not the most diffi-
cult part, and in the end, it may not be the most important. Much of the
remainder of this book concerns interruptions of true conversation, how
we understand such interruptions, and what we do about them. I want to
situate the following chapters in the general characterization of conversa-
tion I have just oἀered; but I also want to make sure to convey that true
conversation is precisely what dissociation and enactment make impos-
sible. Gadamer’s subject was the ideal of conversation; this book, by con-
trast, concerns the problematics of conversation. On the other hand, we
can also say that true conversation, as will become clear in the clinical
examples I oἀer in the following chapters, is what is restored once disso-
ciation is breached and enactment dissolves.
But exactly how does enactment interrupt authentic dialogue? The
answer to that question is actually fairly simple. True conversation requires
that the partners to the conversation not work at cross-purposes. I have
already emphasized, for instance, that true conversation cannot be about
who is right or whose views are superior. There can be no attempt to prove
the other wrong.1 As soon as the partners try to act on each other, rather
than understand each other, true conversation breaks down. But of course
enactment is precisely the attempt to act on the other, not to understand
him or her. Usually there is a sadomasochistic element in enactments, an
attempt to deal by domination and submission with what is unconsciously
felt to be an irresolvable diἀerence.
Gadamer does not consider the nuances of this issue, but we can. He
was a philosopher, not a clinician, and so he had the luxury of taking the
position that collaboration was simply present or absent. Psychoanalysts
are used to working with a much messier situation than that, because we
take for granted that all people are acting on one another all the time, even
during episodes of collaboration. That is the meaning of unconscious pro-
cess. Without our own awareness, we are all continuously attempting to
influence the other, and we are just as continuously absorbing the other’s
unconscious influence on us. And so, for a psychoanalyst, collaboration is
1 “There is, of course, a time and a place for those conversations, but dialogue isn’t one of them.
In fact, dialogue might make those other kinds of conversations go more smoothly and/or eἀec-
tively. Perhaps, indeed, those other kinds of conversations might be impossible without dialogue”
(Cushman, personal communication, 2009).
always a relative matter. It is impossible to believe that any conversation is

nothing but collaborative.
But with that proviso, we can still define true conversation, in psycho-
analysis, as relatedness that is relatively collaborative, relatedness in which
the partners’ primary intention is to understand the other, even if there
are other ways in which they are undercutting the fullness of that collabo-
ration by their unconscious attempts to act on each other.
Sadomasochism obviously interrupts the open, nondefensive, often
implicit, questioning of one another that makes up true conversation. We
have no choice, though, in our attempts to deal with enactments, but to
continue to try to understand, to restore true conversation. Often enough
we may as well try to put out an inferno with a garden hose; and when
the enactment is mutual, as it often is, we do not even have the hose. The
following chapters are an attempt to comprehend our difficulty in under-
standing at these times, the failure of our capacity to see what is question-
able in the other’s experience and in our own.
I do not take up the political aspects of enactments in the chapters
to come, but that does not mean that such a study would not be pro-
ductive. It is certainly true that the kind of conversation Cushman had
with his patient becomes quite impossible in the heat of an enactment.
But the fact that such a dialogue is impossible at such times does not
mean that the events in question do not have a political or moral side.
As a matter of fact, it seems to me that the very intensity of the happen-
ings and the fact that they so often end up being expressed through the
operations of power mean that enactments probably carry even more
expressly political and moral meanings than other clinical events. It
sometimes behooves us to think about these possibilities during the
course of an enactment and to bring them up with our patients once
they end. Who is attempting to dominate whom? Is a man angry with
a woman? A woman angry with a man? What might the reason for the
anger have to do with the genders of the participants? Whenever the
race, class, gender, sexual preference, or ethnicity of patient and ana-
lyst diἀer—meaning that there exists some kind of power diἀerential,
in addition to the diἀerential built into the distribution of the roles of
patient and analyst—the same questions can and should be asked once
the atmosphere cools enough to allow it. And the question of power
hardly ceases to be meaningful just because two people have the same
identity position. I can testify, having been there, that power relations
between two middle-aged, upper-middle-class, straight white men are
hardly absent!
3
The Fusion of Horizons
Dissociation, Enactment, and Understanding
Context and the Fusion of Horizons
We understand our patients when we locate their speech and conduct

within suitable or fitting contexts. We understand ourselves the same way,
by feeling our way into the contexts within which we feel that our own
experience and behavior are most convincingly situated. Contexts, for this
purpose, are enveloping atmospheres, at least as much matters of mood or
aἀect as of content, within which some kinds of experience and interac-
tion can take shape and others cannot. A fitting context is one that allows
the meaning of the other’s speech and conduct, or our own experience,
to unfurl: it enables the expansion of meaning, and it unlocks some sig-
nificant portion of the capacity of experience to signify. A poorly suited
context does not enable or potentiate in this way, but instead prevents,
stunts, or inhibits the meanings that might emerge. A poorly fitting con-
text ensures that new meaning remains locked away; it forces understand-
ing down paths that lead only to familiar destinations.
Once one is able to supply a suitable context, a meaning emerges; one
has understood. But the advent of new meaning not only marks the union
of conduct with a fitting context; new meaning actually constitutes this
union. In recognition of this view, the hermeneutic philosopher Hans-
Georg Gadamer (1965/2004) describes the birth of new understanding
as the “fusion of horizons.” The explicit new meaning that arises in the
fusion is only one of the possibilities that might have come into being; it
did not exist previously in the mind, at least not in finished form, but was
given its shape in the moment of the fusion. The potential meanings that
preexist the fusion of horizons, only some of which will come to explicit
43
fruition, therefore are less aptly described as unconscious in the usual

sense (that is, as hidden but fully formed) than as unformulated (D. B.
Stern, 1983, 1997). Thus the task of making the unconscious conscious is
not best described as allowing oneself to accept what is already there, but
rather as freeing oneself to articulate or construct what one has refused to
think about. Unformulated experience is not “there” to be revealed.1
Situating conduct in a fitting context, then, is not like finding the right
spot for a piece of a jigsaw puzzle. More often than not, our grasp of both
conduct and context is at least partially constructed in the act of under-
standing, a mutual accommodation that the phrase “fusion of horizons”
is meant to convey. Conduct and context modify each other until, in the
fusion, one finds that one can sense the goodness of fit from the inside of
what one is trying to understand. Uncertainty vanishes, at least for the
time being; one knows that this is right. And yet, as right as the new under-
standing may feel, what the next moment demands us to understand is
liable to be no less ambiguous than what we needed to understand the
moment before.
Once the fusion of horizons has taken place, one can speak to the other
person from within that other’s frame of reference. But the process by
which the analyst arrives at such a clinical interpretation, especially when
that interpretation concerns matters of some emotional significance in
the analytic relationship, often requires that a step take place prior to the
fusion between the horizons of analyst and analysand. Frequently a fusion
must first occur inside the analyst’s experience, a fusion between the hori-
zon of what-feels-like-me and the horizon of part-of-me-that-feels-alien,
the other in myself.
This fusion of the horizons of two or more parts of the mind allows the
analyst fuller conscious and preconscious access to his own subjectivity (a
term I use to signify the entire range of a person’s psychic life, including
its unconscious aspects), a freedom that then informs the clinical process,
sometimes allowing understanding where none was possible before. In
everyday psychoanalytic guise, just as the fusion of horizons (in the analyst’s
1 The claim that all experience is continuously constructed does not contradict our everyday rec-
ognition that some meanings are remarkably enduring. It is entirely consistent with the idea that
we continuously create our experience anew and suggests that in some cases we construct the
same meaning, or the same pattern in experience, over and over again. To put the point in con-
ventional psychoanalytic language: Unformulated experience can be highly structured, though
never so structured that multiple interpretations are excluded. Even those structured meanings
remain processes. Even the most highly organized unformulated meanings are therefore not static
objects or ruts worn in the brain, and never absolute, but predispositions toward certain kinds of
meaning-making and away from others.
The Fusion of Horizons 45
mind) between analyst and analysand is the most significant aspect of

what we refer to as clinical interpretation or the attainment of empathy, the
fusion of horizons within the analyst’s subjectivity is familiar to us as the
successful analysis of countertransference. (It would take me too far afield
at the moment to explore the point in detail, but the same two kinds of
fusion take place in the analysand’s experience, that is, in the analysand’s
understanding of the analyst’s experience, a process corresponding to the
growth in the analyst’s capacity to interpret and provide empathy and con-
tainment; and in the analysand’s understanding of himself, corresponding
to the analyst’s understanding of the countertransference.)
One of my intentions in this chapter is to show how productively the
hermeneutic portrayal of understanding dovetails with the psychoana-
lytic conception of the multiple self. As I have already hinted in referring
to the fusion of horizons within one’s own subjectivity, it seems to me
that it is productive to think of self-states as some of the most significant
of the contexts we employ every day in the clinical situation. One under-
stands conduct when one is able to imagine the self-state from which it
arose and then grasp it from within that context. I argue later that notable
episodes of misunderstanding in the analytic situation—that is, circum-
stances in which the possibility of a fusion of horizons between analyst
and analysand seems remote—often can be attributed to a dynamically
enforced separation, or dissociation, of the analyst’s relevant self-states.
The analyst unconsciously denies himself access to the context, to the state
of his own self—to the other within himself—from within which it would
be possible to construct the experience of the analysand. In more familiar
terms, the analyst does not understand himself well enough to understand
the patient.
In hermeneutic philosophy, the process by which conduct and context
are fitted to each other—the creation of understanding, the fusion of hori-
zons—is called the hermeneutic circle. The scope of application of this
idea is very wide: It includes our grasp not only of thoughts but of feelings,
and our understanding not only of other people and ourselves, but also of
art, literature, drama, and even science.
Not surprisingly, and as has already become clear in Chapter 2, the
hermeneutic literature focuses much less on unconscious motivation than
psychoanalysis does. Gadamer, in particular, whose work centers on the
phenomenology of understanding—the moment-to-moment episodes of
interpretation and understanding that constitute human experience—has
little to say on the subject. Other hermeneutic philosophers, especially
Habermas (1971) and Ricoeur (1970, 1977, 1981), have had closer relations
with psychoanalysis, but their accounts do not really address the moment-
to-moment construction of experience Gadamer (1965/2004) deals with
so well. And so, while preserving Gadamer’s picture of understanding, I
set out here to present what Gadamer has not: a psychoanalytic account of
the hermeneutic circle.
How does the hermeneutic circle work? Whenever we are trying to
understand, we are working with part–whole relations. We try to compre-
hend something new by grasping it partially, just enough to identify it as
an instance of something familiar, a meaning we already know. Then we
project this “whole” meaning onto the “partial” one we have constructed,
completing the partial meaning—and the circle. (I am not adapting psy-
choanalytic vocabulary by using “projection” here; the word is employed
in hermeneutics in a more general sense than in psychoanalysis and
implies nothing about defense. It signifies, rather, our imposition of
meaning on what faces us, based on what we are already capable of under-
standing.) The “whole” meaning is what I have been calling “context” and
what Gadamer refers to as “prejudice.” Gadamer does not use the word
prejudice in its most common, negatively valenced sense; he employs it
instead to signify that the contexts we project are expectations or biases
based on past experience, both our own personal history and the cultural
traditions sedimented in us. And, therefore, although prejudices, if they
remain unexamined, may obstruct understanding, they are also what
make understanding possible: Novelty can be grasped only from within
what is already familiar to us.
The key problem for any kind of understanding, clinical work included,
arises when the prejudice or context we project is not a good fit for what
we are trying to understand. In fortunate instances, the working psycho-
analyst sees that the context within which he is trying to understand the
analysand is inadequate; he sees that the meaning he has projected is inap-
propriate. Of course, this understanding requires that the analyst be able
to treat the context or prejudice as a hypothesis, that is, that he be able to
question the validity, for this purpose, of the familiar understanding he
is projecting. He must be sensitive to those moments when the detail that
faces him challenges the familiar thesis he intends to project: He must
maintain enough uncertainty to be in a position to reject what he may
otherwise be tempted to believe.
When everything is working well, the process of understanding, this
hermeneutic circle, is what anthropologist Cliἀord Geertz (1974) calls “a
continuous dialectical tacking between the most local of local detail and
the most global of global structure in such a way as to bring them into
simultaneous view” (p. 69). We use the parts to grasp the whole, and the
whole to grasp the parts. The reciprocal aspect of the process is the source
of its description as a circle. As Geertz puts it: “Hopping back and forth
between the whole conceived through the parts that actualize it and the
parts conceived through the whole that motivates them, we seek to turn
them, by a sort of intellectual perpetual motion, into explications of one
another” (p. 69).
This is an engaged, interactive, dialogic view of understanding. We need
the collaboration of the other. We cannot understand him by ourselves; we
cannot somehow choose, on our own, to experience exactly what the other
experiences. The value of empathy is not at issue; it is just that in a hermeneu-
tic view empathy is not some kind of monadic immersion in the experience
of the other. It is a process that requires the involvement of the one who is to
be understood. This point holds both for our understanding of other people
and for our understanding of the otherness within us: Understanding is
dialogue, even if it is a dialogue we have within our own skins.
But of course it is usually more comfortable to think along familiar
paths than to strike out in a new direction. It is easier to gloss over what is
diἀerent about the other than to keep ourselves open to possibility. How
do we maintain our curiosity and open ourselves to the other? What do
we do? The analyst begins by trying to keep in mind the strong likelihood
that he does not experience what the analysand experiences (and that he
does not fully comprehend what he himself experiences).
In other words, the analyst ceaselessly imagines and reimagines the
limits of his own grasp. When the analysand speaks, or the analyst notices
something about what the analysand has done, the analyst notes the inter-
pretation that occurs to him and then asks himself what else the analysand
might mean. What else could this speech or conduct signify? This is one of
those paradoxical, recursive, look-at-your-own-eye-seeing operations: The
analyst tries to query himself about what he might not be thinking. What
could he see diἀerently? What might he be missing? He must find a way to
allow the other’s experience to have its voice, to “speak,” perhaps even to
speak in ways the analysand did not know he intended. The analyst must
allow the other’s experience to disconfirm assumptions the analyst did not
even know he was making.
But now consider what happens when the analyst has no idea that he is
selecting an inadequate context for what he is trying to understand: That
is, the analyst does not understand, but believes he does. There is no dia-
logue here, no reciprocity, no fusion of horizons. What may appear to the
analyst to be an interchange is no more than the unwitting and unilateral
imposition of an interpretive frame on the patient’s experience, monologue

disguised as dialogue. Under these circumstances, the analyst is liable to
feel critical of the patient or baffled by him. The analyst does not see that
he is looking for evidence of what he already knows and, as a consequence,
is liable to feel (however guiltily) that the patient is recalcitrant.
Here we have what we might call the “vicious” circle, not the hermeneu-
tic one: It does not open into anything new, but leads only to the conclu-
sion that set it in motion in the first place. We are inside the vicious circle
when we know the answer before we ask the question, when, in fact, we are
not really asking a question at all. But when we are in its grip, the vicious
circle may not feel vicious: Until some later time, we may remain utterly
convinced of our openness to the other. The vicious circle sustains our
continuing projection of poorly suited contexts. It is what happens when
we cannot (or do not, or will not) see enough to sacrifice our preconcep-
tions: We stifle or stunt the other person’s meaning and, as a result, inevi-
tably find ourselves in some kind of relational bind. We shall see that the
vicious circle, in psychoanalysis, is the outcome of the dissociation of the
analyst’s self-states from one another.
The Multiple Self as Context in Flux
As a result of the work of a number of writers,1 it is today a familiar idea

in interpersonal and relational psychoanalysis that the self is not simple
and unitary but a more or less cohesive collection of self-states. These dif-
ferent self-states may be simultaneously knowable; there is no implication
that a person need be uncomfortable about knowing one self-state while
he is “in” another. The multiple self as the expectable, everyday condition
of identity is, among other things, a very helpful addition to our concep-
tualization of the role of context in understanding because our frequent
shifts from one self-state to another emphasize the continuous change in
the context of all understanding.
The idea of the multiple self in psychoanalysis owes much to Sullivan
(1950), who suggested that the interpersonal field determines the contents
of consciousness. The kind of relationship we establish with another person
has everything to do with what we can be aware of in that person’s presence
1 Most notably, Bromberg (1998, 2006); Chefetz (2003); Chefetz and Bromberg (2004); Davies (1996,
1997, 1998, 1999, 2003, 2004); Davies and Frawley (1991, 1994); Howell (2005); Pizer (1998); Slavin
(1996); and Slavin and Kriegman (1992, 1998). See also Flax (1996); Harris (1996); Mitchell (1991,
1993).
and what the other can be aware of in ours. (I use “presence” now both in
its physical sense and in the more metaphorical, intrapsychic sense as the
symbolic presences of the inner world.) It was Sullivan (1950), after all, who
said over half a century ago, “For all I know we have as many personalities
as we have interpersonal relations” (p. 221).
Admittedly, Sullivan was not proposing a multiple-self theory at the
time. He was criticizing the reification in the traditional idea that each of
us has something that we are justified in referring to as a unique, unitary
self. But add to this critique Sullivan’s emphases on trauma and on disso-
ciation as a defensive process, and you have the seeds of what has become
the multiple self in the hands of interpersonal and relational writers.
Sullivan’s great insight that what we can experience is defined by the field
becomes, in multiple self theory, the idea that each self-state is defined by
the experience that we are capable of creating, feeling, and formulating
from within it. When, for defensive reasons, we cordon oἀ certain self-
states from contact with the others, and then restrict ourselves only to
some of these and not others, we also restrict our access to the fullness
and depth of the experience it is possible for us to have. We thus reduce
the imagination, precision, and aἀective nuance of the formulations of the
experience we are capable of articulating.
In each life, the multiple self arises in the first place because each of us is
diἀerent in the presence of diἀerent others. Our selves are originally called
out by others, although soon enough our own states of being begin to exert
their influence on the lives of those same others, so that each of us is, in
some sense, called out over generations. The various selves of the multiple
self are the patterns of experience and conduct, especially emotionally
salient experience and conduct, that coalesce around the diἀerences in
our conscious and unconscious interactions with diἀerent people. Such
coalescences form around the patterns of others’ direct impact on us and
our reactions to them; but just as significant in the genesis of our states
of being is the process of identification. (Identification and the “direct
impact” of others are not necessarily easy to tease apart, though. I make
separate reference to them to emphasize that self-states have their sources
in both the intrapsychic and the interpersonal. But the dividing lines are
hazy, since we have no reason to identify with someone with whom we
have not been deeply involved in the interpersonal world.)
One good vocabulary here is Racker’s (1968): We might say that our self-
states are sometimes shaped around our identifications with the way those
with whom we interact experience themselves, what Racker calls concor-
dant identifications, or identifications with the other’s ego; and sometimes
our identifications arise from the way other people have experienced those
with whom we interact, what Racker terms complementary identification,
or identification with the other’s internal objects.
In any case, the upshot is that very early in life we begin to be diἀerent
with diἀerent people, and often enough diἀerent with the same person
under diἀerent circumstances, each significant change in our world call-
ing out particular patterns of feeling, thinking, behaving, and being. We
are familiar with the process by which, over time, patterns of being such
as these are internalized, so that even those self-states that originated in
direct interpersonal experience may come to be intrapsychic, thereafter
to be elaborated and reconstructed according to the purposes of the inner
world as well as the outer one.
The existence of multiple self-states, to the extent that they overlap
more or less freely (that is, to the extent that we can tolerate remaining
aware of one self-state while also existing at that moment within another,
what Bromberg [1998] calls “standing in the spaces”) is the definition of
“normal” or “expectable” dissociation, a relatively permeable kind of sepa-
ration between parts of the self that occurs simply as a consequence of
human relatedness. Under these circumstances, the boundaries between
self-states present little or no obstacle to the mind, traffic across the lines
proceeding with no more notice or eἀort than it takes to cross the demar-
cation of one municipality from another.
But often enough traffic across the boundaries is choked and obstructed,
and our theories reflect this. Dynamics and defense are also part of every-
day life, and in the conception of the multiple self, defense is the other
meaning of dissociation. In this second sense, dissociation is the uncon-
sciously enforced separation or disconnection of the kinds of experience
associated with diἀerent selves, a separation motivated by an unconscious
discomfort, or even a sense of doom or dread, about certain kinds of expe-
rience being simultaneously known, sensed, or felt. The more separate the
experience of two self-states must remain, the more dissociated, in this
second, defensive sense, we describe them to be.
We define defensively motivated dissociation, then, as the uncon-
sciously enforced disconnection between experience and its most fitting
context(s), a disconnection that prevents some of the understandings and
new meanings that would otherwise be possible. Under these conditions,
boundaries between states of self that, in less troubled circumstances, can
be crossed simply by focusing attention on experience and making a mini-
mal eἀort (like crossing that border from one town to the next) become
less like boundaries and more like fractures or chasms in a glacier: The
eἀort required to cross them is considerable, and there may be danger.
In the clinical setting, dissociation, because it deprives the analytic par-
ticipants of the emotional and intellectual contexts they would need to be
able to apply in order to understand the ongoing clinical interaction, is, in
a sense, really just another way to describe the origin and maintenance of
transference and countertransference. Transference/countertransference
enactments, in other words, are instances of mutual dissociation; and they
may represent the breakdown of the dialogue critical to the hermeneutic
circle. Enactments resolve only when one or the other member of the ana-
lytic couple reestablishes dialogue by gaining explicit awareness of how, at
that particular moment, the context he is supplying is inappropriate to the
other (or the “other” within himself); but at those times, by definition, it is
precisely that crucial, explicit awareness that is unavailable to either one.
What I am saying may make it seem that I am implying that mutual
dissociation occurs only in the most dramatic kinds of enactments—
impasses, breakdowns, or deadlocks; but that is not at all what I want to
say. Many instances of understanding ourselves or one another must over-
come a mutual dissociation; the dramatic enactments are only the most
obvious of these. All instances of successful understanding begin with
the absence of understanding; if they did not, there would be nothing to
be understood. The absence of understanding between two people often
means that they are occupying self-states that do not allow a fusion of
their horizons, at least for the time being. In any particular instance, new
understanding may arise with virtually no resistance to the participants’
grasp of one another’s meaning; but that ease of understanding merely
masks the fact that the process is roughly the same (though less intense) in
those facile moments as it is when understanding is much more difficult
to accomplish. In both kinds of moments—the unremarkable episode of
comprehension and the breaking of an impasse or deadlock—the two par-
ticipants must find their way to understanding and accepting that the con-
text they have been applying to the meaning of the other is inadequate to
the meaning the other wants to convey, or, for that matter, to the meaning
the other conveys unwittingly. In the unremarkable moment, the event of
grasping and sacrificing the old context and adopting the new one occurs
so instantaneously that it hardly seems worthy of note; in the deadlock, it
may be so difficult that it never happens at all.
We are grateful for the times when the analysand is able to see the rigid-
ity in his own selection of self-state and is able to amend it in such a way
that the deadlock is broken. And that probably happens much more often
than we write about. But we do feel, after all, and rightly so, that it is our
responsibility not to depend on the patient, but to find our own way to
an explicit awareness of what is wrong with the context within which we
understand the analysand. Often, at these times, we manage to learn some-
thing important about the analysand by observing our own experience.
Sometimes, for instance, as I describe in Chapter 4, we sense the eἀects
of the analysand’s self-states in the way the clinical interaction provokes a
kind of aἀective “chafing” in our experience. It is especially changes in the
analysand’s self-state that we sense this way. Something feels inconsistent
to us; something counters an aἀective expectation we did not even know
we had until that moment, something feels subtly “wrong” or contradic-
tory or just uncomfortable.
Therapeutic work is often the psychic equivalent of what we do when
an otherwise unremarkable walk in the woods is suddenly interrupted by
finding that our sweater has snagged on a twig or branch: We stop to inves-
tigate the source of the difficulty. Now bring this snag into the consulting
room and make it metaphorical. Perhaps that snagged feeling makes us
attend to a subtle change in our mood or even makes us note the presence
of a diἀerent feeling. We ask ourselves about that change, and we find that
we are responding to something about the analysand that we have not yet
explicitly noticed.
Such hints, which amount to our recognition of our own, previously
unformulated state of self, are often the first things that alert us to a shift in
the analysand’s self-state. A change in the analysand’s self-state, in other
words, is likely to provoke a change in our own state of self. The reverse is
also true, of course. Neither the analyst’s nor the analysand’s state shifts are
useful to the task of explicit understanding unless the change in state that
is provoked is noted and accepted by the one who wishes to understand.
We recognize the analysand’s self-state from within a part of our own
being that is responsive to the analysand’s; and, in its turn, our adoption
of a responsive state of being has occurred because we have found a way
to accept the impact of what the analysand is saying and doing. We can-
not observe the other’s self-state directly, but only through its eἀect on
us; the other’s self-states are visible only as a consequence of their impact.
Similarly, we understand our own self-states, and shifts in them, from
what we can sense about changes in our ongoing, conscious experience.
The analyst’s direct awareness of his state of self, of course, is the
exception and not the rule. It is crucial, but only for the task of explicit
understanding. Unnoticed changes in self-state ceaselessly oscillate
between analyst and analysand and constitute over time the spontaneous,
unanalyzed relationship that is the background against which the more

purposeful analytic work becomes meaningful and without which it could
not proceed.
Reaching an Understanding
Consider the following incident: Daniel, a very bright and articulate

middle-aged professional man, was in the early stages of his analysis.
We had already established a warm, collaborative relationship and were
both well aware, even at this early time in the work, that my reaction to
him carried particular importance. If he felt I was warm and affirmative
in a session, the time between sessions tended to go well for him; but if
he left with the nagging feeling that I was critical, disapproving, or cool,
he could become rather depressed (“deflated” was the word that seemed
to him to fit best), often about matters that, on their face, had nothing
to do with the analysis. In neither instance (that is, when he felt good or
when he felt deflated) was he necessarily aware, between the sessions,
that these feelings had to do with our relationship; but he was very open
to understanding them in this way when he returned for his next ses-
sion, and he felt that we were on to something significant. We both felt
that way.
On the day in question, Daniel heard me rustling around behind him
(he was on the couch). He turned around to face me and snapped, “What
are you doing?” He has very good peripheral vision (this had already been
established), and so, because my chair is not positioned directly behind
the couch but slightly to the side, he had been able to see that I was leaning
forward to look at my appointment book. I asked him what he imagined I
was doing. In the same reproachful tone, he said that I must be thinking
about some other patient’s appointment.
In fact, I had been thinking about him. It was significant, of course, that
he chose his particular interpretation of my behavior; and, with many or
most patients, I would have made the conventional move here and sug-
gested that we explore the sources of that interpretation. But I had a dif-
ferent feeling with Daniel. From what had taken place in the treatment
to that point, I knew perfectly well that he was capable of conventional
analytic exploration and that he even thrived on it. But I also had the sense
that this particular episode, because it concerned his belief that, emotion-
ally, I had disappeared from the room, was painful to him in a way that
exploration would exacerbate.
Unpacking Daniel’s reaction to the feeling that I had deserted him, of

course, might have been useful, but he and I had done this before: We both
already knew that he worried about whether I had a sustained interest in
him. And so I had some doubts that the pain the exploration would cause
would be worth whatever we would learn from it. Those doubts were espe-
cially acute given the fact that I actually had been thinking about him and
so, in that sense, had not disappeared at all. I imagined that we might get
further by exploring his sensitivity to the possibility of my disappearance
at a moment when he knew I had not left him. All this led me to feel that
carrying out a conventional inquiry at that moment would have amounted
to following the rules for the wrong reasons; I would not have been choos-
ing my course on the basis of the experience growing out of this particular
analytic engagement.
And so I pointed out to Daniel that, just before I looked at my sched-
ule book, he had been telling me about having made plans on a particu-
lar afternoon to go out with a woman he was excited about. I reminded
him that we had rescheduled an appointment for an afternoon in the near
future, and I told him that I had thought the appointment and his date
with the woman might be on the same day. (They were a day apart, in fact.)
I had been checking the date and had no idea that he would hear me doing
so, and certainly no idea that he could see what I was doing.
I imagine that my curiosity about the date of the make-up appointment
would be as objectionable to some patients as it would be if I were thinking
about another analysand. In both cases, after all, I would not have been
paying attention only to what the patient was talking about. I have no
doubt that my concern with scheduling would be interpreted as self-serv-
ing by some patients, while others would be sure I was looking to see if I
would get an hour to myself or away from them, or (in fewer cases) to see if
I would have to be disappointed by their absence. For myself, once Daniel
drew my attention to the direction of my own thoughts, I wondered if I
was checking to see if I was to be replaced by his woman friend.
Was this an instance of unconscious communication? Was I experi-
encing my own version of what Daniel was feeling—he being replaced by
another patient, I being replaced by his woman friend? To him, however, as
you might expect from what I have already said, my explanation was unal-
loyedly good. I was thinking about him, after all. He wove his associations
to this event into the rest of what he said that day. He was not as interested
in exploring his sensitivity to feeling that I was emotionally absent as he
was in the good feeling he had that I had cared enough about his presence
to wonder if his date conflicted with a session. He fretted, though, about
how I felt about his having snapped at me, since my revelation indicated
that I did not “deserve” it.
When I saw Daniel next, after the weekend, I learned that during the
intervening couple of days he had found it difficult to maintain hope that
his career plans would come to fruition. He had been mildly depressed. He
began to talk about circumstances that might have led to these feelings.
After a few minutes I asked him about how he had felt about the previous
session and whether his feelings over the weekend might be an instance
of the “deflation” with which we had become familiar. I made reference to
the interactions around “What are you doing?” He recognized immedi-
ately that there was something to this and plunged in. He talked about the
“snap” for a few minutes, and I eventually somehow got the drift (it was
implicit, unformulated in his mind, though somehow available to me in
what he said) that, in fact, there had been a kind of sly twinkle about his
delivery that I had missed. When he asked “What are you doing?” it had
been with the tone of an annoyed but just slightly amused parent catching
his four-year-old snitching from the cookie jar.
Make no mistake, he did mean to snap at me. But he also meant to
do it in a way that had a certain intimacy about it, with a kind of wink
accompanying it, and I had missed that completely. He had not been able
to articulate that I had misunderstood this subtlety (he never did have
the words for it until it somehow surfaced in what I could glean from
what he said on this second day), so he had just suἀered with it. He had
felt, between sessions, that I must hate somebody like him who caused so
much trouble.
I had the sense, as a matter of fact, that although he had not explicitly
formulated the aἀection in his rebuke, his “deflation” would have been
less painful to him, and perhaps would not have occurred at all, if I had
been able to respond to the intimacy he was oἀering and not just to the
hostility. It seemed even clearer to me now that it had not been so much
my reaction to the irritation in Daniel’s snap that had worried him, but
the fear that I would emotionally disconnect. He was less worried about
the damage I would do if I retaliated than he was about my turning away
from him. If I had better understood the warmth in his rebuke, perhaps I
would have accepted the hostility in a way that would have reassured him
about the connection.
Enactments, which are always a variety of misunderstandings, are not
necessarily deadlocks. Plenty of them, this one being an example, are situ-
ations in which, although the analyst participates unconsciously (my mis-
taking the afternoon of the rescheduled appointment for the afternoon of
the date; my missing the irony in his complaint about what I was doing),
the problematic aspects of this unconscious participation can be brought
to the analyst’s attention by relatively minor events. (The qualification
“problematic” is necessary because much of what facilitates the analysis is
equally unconscious, but never formulated.) In the terms I am using, this
is an example in which the boundaries between my relevant states of self
were obstructed but could be cleared easily enough that the treatment was
not badly snagged or detoured.
Take one detail: In one of my states of self or mind—let’s say, a state
that may have originated in my response to my mother’s characteristically
veiled criticism—I respond to criticism, in a prickly and defensive way, and
laughter doesn’t come as easily to me as it does otherwise. If I had heard
Daniel’s “snap” at me in that vein and did not have available to me, simul-
taneously, countervailing states in which criticism is a little less freighted,
it is not surprising that I would have missed the irony in his complaint, an
irony so subtle, after all, that Daniel himself had not explicitly formulated
it prior to my query about it in the next session.
And yet, in that next session, after the intervention of a period of living,
I was diἀerent; somehow that subtle irony was available to me, although it
was now only an echo. In workable enactments, the relevant dissociations
are unstable; an intervening period of living is usually enough to destabi-
lize the relevant dissociations between the analyst’s states of self, and that
destabilization, in turn, by freeing the analyst to use his mind, allows the
reframing of the situation and suggests a line of inquiry or interpretation.
Or it just allows the analyst a diἀerent state of being-with-the-other.
The vignette of Daniel illustrates our everyday experience that the con-
tours of our own self-states are seldom explicitly visible except by contrast
to those that come just afterward. Shifts in our states of mind reveal the
presence of prior contexts we usually had little choice but to live through
unwittingly. In this case, I could see neither my (apparently) defensive
portrayal of Daniel’s hostility nor the state from within which I had that
picture until I noticed my new portrayal of him on the following day. I did
not and probably could not know the “me” who reacted defensively until I
had also become the “me” who constructed the fuller portrait.
I might add here, to complete the picture, that despite my conviction
that I selected the best course in telling Daniel that I was actually think-
ing about him and not about another patient, I might think diἀerently
in the future. I can imagine my decision eventually appearing to me as
an unwitting installment in a deadlock, a much more problematic (and
potentially even more informative) kind of analytic interaction than the
easily destabilized one that took place between Daniel and me. Perhaps my
paternal countertransference to Daniel will seem stifling to him one day,
or even patronizing. Perhaps even now it feels that way to some parts of
him in ways neither he nor I yet appreciate.
Or perhaps (as I am inclined to think), if his feeling about my coun-
tertransference changes, my feelings will change as well, or perhaps, in
resonance with him, my feelings already will have changed by the time
the old ones become objectionable to him. Much of what is most mutative
about the analytic relationship, after all, happens in unnoticed reciproc-
ity, evolving without words, sometimes visible in retrospect, but probably
most often never noticed at all. At this point I believe that the loving parts
of my relationship with Daniel are good and useful and will remain so;
but I remain open to the possibility that they will also turn out to be more
complicated than that. In any case, to whatever degree it becomes possible
for us to attend to them, they will be analyzed as time passes.
Breaking the Grip of the Field
If the vignette of Daniel illustrates a hermeneutic circle that could be com-

pleted with only minor difficulty, the vignette I am about to discuss is an
example of a vicious circle, a deadlock, a closed system of stubborn and
stable dissociations on the part of both the analyst and the analysand that
prevents understanding for long periods, or even indefinitely.1
The crucial diἀerence between these two enactments is the kind of
dissociation that underlies them in both the patient’s mind and the ana-
lyst’s. My enactment with Daniel grew, on my part, from “weak dissocia-
tion” or “narrative rigidity” (D. B. Stern, 1997), defined as an involvement
with a particular storyline exclusive enough that alternative possibilities
1 The literature on enactment has become so immense that to cite it is impractical. There also exists,
though, a recent and smaller body of writing oἀering conceptions of impasse or deadlock. A repre-
sentative sample: Russell (2006a) on the “crunch,” an aspect of the repetition compulsion; Elkind
(1992) on the problematic interaction that takes place when the patient’s transference hits the
analyst squarely in his “primary vulnerability”; Ogden (1994) on the “intersubjective third” turn-
ing into “the subjugating third”; S. Pizer (1998) on the “nonnegotiable”; Ringstrom (1998) on a
variation on Bateson’s “double bind”; Benjamin (1990, 2000) on “breakdowns” of intersubjectivity
due to especially recalcitrant “doer-done to” complementarities; B. Pizer (2003) on the “relational
knot,” a form of Russell’s “crunch”; S. Pizer (2004) on impasse as “weak dissociation.” I thought
that the word deadlock had arisen spontaneously in my mind during the writing of this book until
I read Irwin Hirsch’s (2000) interview with Benjamin Wolstein and was reminded that Wolstein
coined the phrase “transference–countertransference interlock” in 1959 (!) to refer to enactments.
Perhaps my word is an unconscious appropriation of his.
are not noticed or articulated. Weak dissociation does not specifically

exclude certain possibilities; those possibilities just happen to be unno-
ticed alternatives to the story we are so focused on telling or experienc-
ing. Because of my own background, I was so “taken” with the storyline
in which Daniel’s snap was an expression of criticism that I missed its
other half. (S. Pizer [2004] also oἀers a moving and detailed illustration
of the use of the concept of weak dissociation to understand an enact-
ment.) “Strong dissociation,” on the other hand, is specifically defensive:
whatever is being defended against must not be experienced (D. B. Stern,
1997). Strong dissociation is deployed in the service of avoiding specific,
consciously accessible experience; it is simply not allowed to coexist with
the rest of the personality. Manageable enactments are those in which the
reciprocal dissociations are weak ones; in deadlocks, on the other hand,
the dissociations binding the experience that analyst and analysand can
have of each other are more likely to be the strong or defensive kind.
I had been working with Hannah, a middle-aged professional woman,
also very bright and articulate, for about six months when her some-
time complaints that I was understanding her inaccurately or narcis-
sistically began to become more frequent. When I asked her to explain
what I was missing, what she told me convinced me that I was miss-
ing nothing at all, that we were, in fact, saying exactly the same thing.
I could not grasp how she could think I was misunderstanding her.
When I would point that out and express bafflement about any diἀer-
ences between what we had said, she only felt that I was understanding
her less and less.
I tried over and over again to get it just right. Yet she still said that I
was hurting her more with each one of my mistakes. She was angry at me
about what she felt was my refusal to see things from her point of view;
but what I responded to more directly was her hurt. Despite my very clear
awareness that she and I had to be participating in some kind of enact-
ment, I began to feel guilty that I could not grasp what she was talking
about, guilty that I was hurting her and annoyed at what seemed to me to
be her influence on my taking these emotional paths.
I knew there had to be another way to see this situation, and I con-
tinued to ask her about our diἀerences in the greatest detail. I hoped to
be able to claw my way to a more empathic attitude; but more often than
not I did not find my way there. I came to feel that her complaints were
more significant as a kind of blaming operation than as attempts to convey
meaningful content. Increasingly, I found that my attitude toward her was
irritable. I would not have objected to making this problem between us
the focus of joint inquiry if I could have found my way to an open-minded

enough perspective, but I was not able to do that.
I felt that, if I were to try this kind of analysis from within my current
state, all I would accomplish was to blame her. I did not have to blame
her openly and explicitly to continue to hurt her, though, because I could
see plainly enough that my annoyance leaked through, and so she felt
increasingly blamed, anyway, and thus more hurt and angry than ever.
I felt quite alone, without a collaborator. I knew that there must be some
way to understand what was transpiring between us so that would release
us without requiring either of us to give in, but I just could not seem to
formulate that understanding. I could see that she felt the same way I did,
and I conveyed my understanding of this deadlock in an empathic way,
which helped, of course; at least she knew that I understood that she was
hurt, and I think she did believe, at least at times, that it mattered to me.
She knew that I imagined that there could be an understanding that would
help, but that I could not construct it. As useful as this empathic apprecia-
tion may have been, though, it did not help nearly enough, because it did
not change our basic positions vis-à-vis one another.
This was not a good situation. I found it increasingly painful. I felt more
and more guilty about my feelings and my seeming inability to stifle their
expression (which I hardly thought was a good solution, but I had no other
for the time being). And, of course, I also felt guilty about what seemed to
me to be my failure to be of use. I knew better than simply to blame myself,
but there is no inoculation for certain emotional impacts, even if you do
know better. I felt locked in.
This very painful period was more complicated than any single telling
can capture, as any treatment is, which is to say that it should probably
come as no great surprise that we also did good work during this time. The
relatedness never ceased being analytic: We continued to make the dead-
lock the focus of our curiosity and analytic concern. And so, during this
period, it seemed in keeping with the work we were doing that Hannah
continued to bring in dreams. She is an imaginative person who works
well with dreams. One day she brought in a dream that turned out to have
special import for both of us.
In the dream, Hannah was in my waiting room, except that, unlike
the waiting room in my real office, this one was in an apartment that she
knew was also my home. She was alone in the waiting room, but she could
hear people in the adjoining kitchen. She somehow could tell that they
were my family, laughing and enjoying one another’s company. On the
table in front of her was an open safety pin, the kind that used to be used
to fasten diapers. It was her impression in the dream that I had a baby and
that this pin was used for his diapers. She remembered (in the dream) that
during her own chaotic and disorganized childhood (her mother had been
deeply depressed, and Hannah had been lonely, frightened, and poorly
cared for), open safety pins had sometimes been lying about. During those
years Hannah had heard awful stories of babies and toddlers swallowing
pins that had been left open. The open pin on the table suggested to her
my carelessness about my baby and, worse, my murderous hostility. Yet it
was entirely clear to her in the dream that she was not about to close that
pin. She fully intended to leave it right where it was, wide open. She woke
up with the thought that we were both full of hostility. And that was what
she said to me in the session.
After everything I have said, it may be surprising that this was a revela-
tion to me. I had been focused on her hostility as a problem to be addressed
and on my own hostility primarily as a reaction to hers, which it may very
well have been to begin with. But even if that were true, we were far beyond
that now. I had not seen clearly that each of us was separately hostile, that
it was accurate to say that we were playing on a level field, and that each
of us was individually responsible. I had surely been aware of my hostility,
but my guilt about it and frustration over it had prevented me from think-
ing clearly about what part it played in my own subjectivity. I had not been
able to think about it eἀectively and so had been able only to suἀer over it
and wish to expunge it, once again, despite knowing better. Now suddenly
I could ask, why exactly was I angry?
Suddenly I had hypotheses about that. Suddenly I could use my mind
again. I was finally able to ask myself exactly who Hannah would have to
be, and how she would have to feel, to treat me as she did. In short order,
Hannah and I laid out a series of thoughts about the nature of our prob-
lematic interactions. We were able to stand back from them and think
about them eἀectively for the first time in months. My projections (I use
the word in its hermeneutic sense), which I had not been able to stop from
becoming certainties about her impossible treatment of me, shrank once
again to the manageable level of hypotheses.
Hannah and I have become involved again in this kind of exchange.
Understanding is not a magic bullet aimed at the future. As of this writing,
as a matter of fact, Hannah and I just faced another deadlock of this kind;
but, given that first experience, and our capacity to make individual and
mutual reference to it, I felt confident this time that we would find a way to
loosen the noose, and we did. It took much less time and eἀort than before
to reduce the vicious circle to the hermeneutic one. Perhaps we will have
to do this several times, or even many times, until it is worked through.

One of the simple and recurring lessons of psychoanalytic work is that
principles never substitute for experience. Each episode in the transfer-
ence/countertransference must be addressed and either resolved or not,
on its own terms.
There are any number of points that could be made about Hannah’s
dream, of course, and she and I explored many of them. For the present
purpose, what is important to me is that her interpretation of the dream
broke the grip the interpersonal field had on us both and allowed me to
occupy simultaneously the various states of self that I needed to occupy in
order once again to be free to think.
One might say that it was not so much the interpretation itself, but
Hannah’s sudden oἀer of collaboration that made the diἀerence. That
hypothesis may be right. Whether we focus on her collaboration or on
the insight she oἀered me, I take for granted that the episode could not
have taken place without the previous occurrence of other, perhaps silent
processes that must have been occurring over a significant period of time
between us. I feel sure that my dogged attempts to understand what was
going on, for instance, although they were unsuccessful, made a consider-
able impression on the nature of Hannah’s willingness to work with me.
Even in the worst of moments, I retained certain important aspects of my
capacity to be an analyst, as Hannah retained hers to be an analysand.
We continued to work successfully, however suἀused the hours often were
with the eἀects of the deadlock.
What happened in response to Hannah’s interpretation of her dream
was not as simple, then, as the resumption of my analytic capacity or
Hannah’s to be a collaborative analysand. Hannah’s oἀer of collaboration,
as surprising as it was to me, grew from both a conscious and an uncon-
scious context. Although I could not sense these contexts, no doubt I was
as central as she was in their development. That is the nature of many of
the surprises that occur in psychoanalysis: in fact, surprise is the feeling
we have when an event in the relationship takes place in a way that feels
out of the blue, devoid of context.
Hannah and I did not simply resume our analytic capacity. What we
resumed was our ability to think about this particular problem. It is partly
that specificity that draws me to the idea of the multiple self, with dissocia-
tions between various of its aspects, as an especially suitable way to think
about what happened. With any patient, we adopt many states of mind in
the course of the work. In my example, however, it was just one of those
states that was problematic—a key one, of course, and apparently isolated,
but just one nevertheless. It is equally important, though, that our even-
tual ability to deal with that one state was embedded, in ways that she and
I created but will probably never know, in all the others.
Where was my aἀective “chafing” in this illustration? That is not dif-
ficult to pick out. I knew that Hannah must be saying something that I
did not understand, because she was hurt and I was unable to oἀer an
eἀective analytic response. In a deadlock, it is often only the analyst’s
awareness of the inadequacy of his own response that prevents the situ-
ation from degenerating into a common argument and keeps hope alive
that the analyst (and the analysand, too) will find some solid ground on
which to stand and think.
In a deadlock, the analyst may be able to do no more than try to be open
to influences from outside the state of mind within which he is mired. A
deadlock means, after all, that the participants’ freedom is curtailed, which
suggests that neither can choose to pull the treatment up by its bootstraps.
In the case I have just described, it was the analysand’s generosity, courage
(she took the risk that I would not accept her observation), and powerful
motive to further the treatment (powerful enough that it infused a dream)
that made change possible. Both of us very much wanted something like
what she did to happen. As angry and hurt as she was, I think she was
aware that, in my way, I regretted the deadlock as much as she did, and I
think she must have had a pretty good idea that I would welcome her oἀer
of collaboration. So I cannot say that I had no freedom; better to say that
what little freedom I had took the form of my regret and sincere desire for
therapeutic movement. I may not have been able to shift my state of mind,
but I certainly wanted to, even if, outside my awareness, I must also have
wanted not to. Without the analysand’s belief in my desire to work and
to help, I suspect that it would have been just too dangerous, under the
circumstances, for her to make herself as vulnerable as the dream, and her
interpretation of it, made her.
In such a case, then, the analyst must wait, recognizing that he is in the
grip of a dissociation that he cannot control. He has no choice but to accept
that he cannot will a change. He can only remain open to the foreign ele-
ment that may reveal a path to freedom. Sometimes that foreign element
is an intervention by the analysand, as it was in this case; sometimes it is a
consultation sought by the analyst; sometimes it is a movie, a conversation
with a colleague, a play, a novel, or a psychoanalytic article. Sometimes
the analyst, in retrospect, can identify in the clinical interaction a variety
of that aἀective chafing, that unprovoked signal of the unexpected, that
alerted him to something about the patient, or about himself, that he had
not noticed before.
Perhaps most often, though, the foreign element is impossible to iden-
tify. In these less easily describable cases, at some point the interaction
between the analyst and the analysand seems to become infused with
some new quality that neither participant consciously set out to create.
Then, in response to these new meanings, the inner world (which may
actually have contributed the new shades of meaning in the first place)
shifts as well, making it possible for at least one of the participants to have
a deeply felt new experience of how the other feels. In the meantime, while
analyst and analysand both try to protect what they can of their collabora-
tion and curiosity through the storm, the analyst can do no more to melt
the deadlock than to register whatever he can of alternative ways of expe-
riencing and focus on the details of his experience of the interaction.
For the sake of perspective, leaving the case of Hannah, I return to the
case of Daniel. He and I both felt that I understood him rather well most
of the time. But remember my oversight: It turned out, you will recall,
that Daniel’s “snap” at me had an intimate, humorous edge that I com-
pletely missed at the moment it happened. Daniel had spoken to me in
this way frequently enough over the previous months of the treatment
that I might have been expected to be familiar with a certain wryness he
has. My oversight was clearly a kind of temporary blindness, a notable but
relatively unproblematic dissociation of the state(s) of self within which I
could have sensed and articulated the emotional nuance I missed. It is not
my own motivation that interests me so much here. The case of Daniel is a
reminder that dissociations and the enactments they inspire are seldom as
noisy, dramatic, and painful as the deadlock with Hannah was. The more
subtle dissociations and enactments, and our clinical responses to them,
are part of every analyst’s more or less unremarkable daily work.
A Note on Imagination
In this chapter I have described dissociation as an unconscious unwilling-

ness to experience certain states of being simultaneously. Elsewhere I have
placed much more emphasis on dissociation as the inability or uncon-
scious unwillingness to articulate certain aspects of one’s experience in
verbal language (D. B. Stern, 1997). These two meanings are really dif-
ferent ways of making the same point. To refuse unconsciously to enter
a certain state of being is (among other things) to refuse unconsciously
to create the particular explicit meanings that would be available from

within that state.
Dissociation is a constraint on the freedom of thought. It is sometimes
total, as in the absence of any shred of memory of early childhood abuse;
but more often the inability to articulate is not total at all. I may be able
to remember, for instance, or to formulate in the present moment, that
Daddy is angry at me; but depending on the history of the relationship and
the residue of that history in my mind, I may or may not have access to
the additional interpretation that Daddy is angry because he loves me and
does not want me to wander into the street. If I do not formulate the latter
experience, I come away feeling that he is unfair. Or perhaps, in a diἀerent
and more abusive situation, the field I put together with my angry father
allows only the interpretation that his anger is due to my badness, so that
I cannot articulate the perception that he is sadistic. Or I can articulate it
in some states of self and not in others. Or, like Hannah, I may be able to
see that my analyst is irritable, but I cannot formulate an interpretation
of my own behavior as provocative, so I feel only victimized. Dissociation
makes it difficult to reflect on certain aspects of experience; these things
just happen.
It is not only the freedom of thought that dissociation prevents. It is just
as significantly the freedom to feel. To stay with memory as an example:
We sometimes have a hazy recollection of traumatic events, or even a fairly
explicit one, but it is flat, dead, merely factual, lacking the aἀective nuance
we might expect to accompany it. Or perhaps a storm of aἀect is present,
but the emotional subtlety that would be needed to give the experience a
fuller meaning is absent.
In the largest sense, though, dissociation is not fully described as a fail-
ure of either thought, memory, or feeling. Dissociation is a failure to allow
one’s imagination free play. In many instances one can think of the fail-
ure of imagination as the collapse of transitional space (Winnicott, 1971)
into deadness or literalness. Merleau-Ponty (1964a, 1973), who captured
the sense of what I mean by imagination as well as anyone, writes that
“speech takes flight from where it rolls in the wave of speechless com-
munication” (1964a, p. 17). He tells us that creative speech, which is the
domain of imagination, “tears out or tears apart meanings in the undi-
vided whole of the nameable, as our gestures do in that of the perceptible”
(p. 17). Imagination is our capacity to allow language to work within us
as it will; dissociation is our inhibition of that capacity. If we wish to do
so, we can use Lacan’s terms here: Imagination takes place in the realm of
the Symbolic, dissociation in the Imaginary. Merleau-Ponty (1964a) put

it this way:
To make of language a code for thought is to break it. When we do so we pro-

hibit ourselves from understanding the depth to which words sound within
us—from understanding that we have a need, a passion, for speaking and must
(as soon as we think) speak to ourselves; that words have the power to arouse
thoughts and implant henceforth inalienable dimensions of thought; that they
put responses on our lips we did not know we were capable of, teaching us,
Sartre says, our own thought. (p. 17)
The absence of dissociation is not defined by the presence of some

particular experience that has been prevented from existing. That way of
thinking would be a simple-minded dualism, as if experience could be
only present or absent, formulated or unformulated. The absence of disso-
ciation is defined, instead, as (relatively) unfettered curiosity, a point that
immediately allows us to say that experience ranges from highly imagined
to highly dissociated, with all the implied variations in between. In the
case of Hannah, for instance, it certainly is not my intention to say that I
did not formulate my experience. Rather, I did not formulate it in certain
ways; I was not able to allow myself an open kind of curiosity, a freedom
of thought and feeling, an unconstrained imagination. I did formulate my
experience of the interaction with Hannah; it is just that the experiences
I formulated were too incompletely imagined to be useful. My experience
was constricted, highly and rigidly selective. I could not ask myself the
questions that would have opened up my own experience. I could not use
the possibilities my experience aἀorded me. I needed to be able to expand
my context—that is, my state of mind—to imagine more fully, to be able
to wrap my mind around a more inclusive and nuanced picture of what
Hannah and I were doing with each other.
Just as dissociation is a failure of imagination, enactment, too, is a
sign that imagination has failed. The failure is usually understandable,
informative, temporary, and relative; but it is a failure nevertheless,
because it is a kind of estrangement from within which we are locked
out of curiosity. Simultaneously, of course, enactment is also a kind of
unconscious communication, the only way that some experience can be
brought into the analytic situation. In that sense, enactment is a boon,
not a failure at all. Without it, in fact, imagination would have that much
less to transform. That these two interpretations of enactment seem to
contradict each other is merely one more evidence, if we needed one, of
how rich and manifold reality is: The world allows us interpretations that
logic might tell us cannot coexist; and yet they do. To complicate matters
further, many of us now think of the analyst’s unconscious involvement
as continuous.1
But, of course, imagination in some ways goes on continuously as well.
Experience is formulated and unformulated, free in some ways and con-
stricted in others. Each moment can be conceived in both ways, and prob-
ably should be.
When the Patient Reveals the Path to Freedom
We can describe the crucial element in a deadlock as each participant’s

loss of the capacity to see or value the other’s perspective; and so it fol-
lows that perhaps the chief means of relaxing a deadlock is the reawak-
ening of that capacity. Such a reawakening is easier to refer to than to
accomplish, of course, because people in deadlock with one another are,
by definition, trying to force one another into a certain set of percep-
tions. Seeing the other person’s perspective is precisely what neither
wants nor is able to do. Consequently, most direct expressions of one
participant’s feelings or perceptions about the other, or about the situ-
ation between the two, contain some variety of blaming; and blaming,
obviously, resolves nothing.
But there are certain kinds of direct expressions that, even from within
a deadlock, can and do change the nature of the analytic field. If the ana-
lyst makes a nonblaming observation of the analysand that is precisely
descriptive and heartfelt (but that may nevertheless convey anger or other
aspects of the deadlock’s aἀective charge), there is the chance that the
analysand’s capacity for empathy about the circumstance in question may
be awakened, or reawakened (and I might add here that the analysand’s
empathy for the analyst is something we depend on more often than we
acknowledge). Once the analysand’s empathy for the analyst’s position is
aroused, the analyst finds it much easier to occupy the states of mind in
which the analysand’s self-state can be formulated. Of course, this may
also happen the other way around, which is actually more direct; that is, a
direct, heartfelt, precise expression from the analysand may loosen up the
1 For the reasons I described in Chapter 1, I prefer not to define this continuous unconscious influ-
ence as enactment, but prefer instead to preserve the word “enactment” for episodes in which
dissociation is interpersonalized.
analyst in such a way that the necessary states of mind become available.1
In these instances, the analysand’s emotional expression is the foreign ele-
ment that reveals to the analyst a path to freedom.
One of my favorite examples of this sort of intervention comes from
Bollas’s (1983) report of his treatment of a maddeningly controlling patient:
“It is impossible to convey, I fear, just how maddening this woman can be
as the reality of her self presentations is truly so unbelievable as almost to
defy communication altogether” (p. 28). After a lengthy and thoroughly
unsuccessful period of trying to interpret in traditional fashion what he
faced, Bollas began to fantasize, when he was with his patient, that she
would move away, become disillusioned with him and go to someone else,
or have a breakdown and be taken oἀ his hands by the hospital: “More
than a few times I thought I would have to tell her that unfortunately I
would not be able to continue with her and I would tell her about how
private practice has its limits and so forth” (p. 27). However, one day she
made a complaint that made Bollas sit up and listen: She told him that he
had become cold and removed from her. He knew immediately that she
was right, although he had not formulated the observation before: “I had
withdrawn from her and was always on the alert for her next use of herself
as a kind of afflicting event” (p. 28). Bollas then said to his patient:
I am very glad that you have said this, because in a way, I think you are abso-
lutely correct. I have become somewhat cold as you put it and am aware of being
distanced in these sessions, something that I think you are well aware of. But
let’s wonder, shall we, about how this happened. You see, it’s my view that if you
could convince yourself to stop being so God damned traumatic then I could
be quite a bit more at ease with you and we could actually get down to the task
of understanding you. (p. 28)
Later Bollas added that he thought she had brought up his distance from her
because she knew this was something in her life that needed work. The patient
responded to all this with a level of maturity Bollas had not seen before.
1 I certainly do not mean to imply, though, that empathy for each other’s emotional positions is
always the most therapeutic kind of situation that can occur between analyst and analysand, or
even that such a state of relatedness is always a goal. To take such a position would be to deny
the significance of the analyst’s complementary identifications (Racker, 1968) and the authentic
responsiveness they inspire in the countertransference. It would also ignore the crucial impor-
tance of holding, in which empathy cannot and must not be reciprocal; the analyst must simply
accept the analysand’s feelings, positive or negative, sometimes a rather heavy dose of them over
a rather prolonged period. All of us believe that the analyst’s freedom to experience in the coun-
tertransference and the patient’s freedom to experience in the transference (in the patient’s case, a
freedom almost always unencumbered either by the realistic danger of frank retaliation or by any
demand to understand the analyst’s experience) are important aspects of therapeutic action.
At first glance it might appear that it was Bollas’s intervention that

allowed the patient to treat him diἀerently, with the consequence that he
could adopt a state of mind that allowed him to formulate her perspective.
But why did Bollas come up with this very influential intervention? It was
the outcome of his response to the patient’s direct expression of her obser-
vation about him, however much the message may have been clouded by
her status as “afflicting event.” Note that Bollas had not been able to see
his coldness and emotional distance from the patient until the patient
observed them. That means, in the terms I am using, that he was trapped
in an isolated (dissociated), unproductive, and unresponsive state of self,
one in which he had not even thought to formulate either his coldness or
the pain his patient must have felt about it. Bollas’s angry state of mind,
that is, was dissociated from the state of self in which a more productive
perception of the patient would have been possible. It seems that his recov-
ery of the use of his mind to help this woman with her hysteria was at least
partially the result of her intervention, very much as Hannah destabilized
my dissociation with her interpretation of her dream.1
Perhaps we should even wonder if, in cases in which deadlocks are
resolved, patients come to the aid of their analysts very frequently. Take
those cases in which, like Bollas, the analyst breaks the grip of the field
by oἀering a direct emotional response. How does a beleaguered analyst,
caught in and struggling with his own desire to control the patient, sud-
denly find it possible to oἀer a direct, nonblaming, recognizing but non-
controlling observation? Doesn’t the very capacity to make such a response
suggest that the important shift has already taken place? Sometimes the
analyst may find his own way to such a resolution; but very often it must
be the patient who, however maddeningly, sets this kind of mutative event
into motion. Unexpected aid from the patient, it seems, is one of the most
significant of those foreign elements, often announced by that persistent
aἀective chafing, that the analyst must try to allow himself to perceive if
he is to resolve his own dissociation and find his way to understanding.
But no matter whether it is the analyst or the analysand who resolves
the dissociation first, the story does not end here. Understanding is more
mysterious than the mere absence of dissociation; it does not necessar-
ily fall into place as soon as our unconscious reasons to avoid it vanish.
The view that comes to us from Heidegger, Gadamer, Merleau-Ponty, and
1 We shall see in Chapter 4, thaat enactments end when one participant becomes capable of a “new
perception” of the other. The new perception is what Hannah contributed to me, and what Bollas’s
patient contributed to him. Once such a new perception comes into being, neither the patient nor
the analyst can treat the other as they did during the enactment.
others is that there is no way to codify the process by which understanding

is reached. No one can say exactly why understanding comes about when
it does, why horizons fuse now and not five minutes ago or yesterday, why
language becomes able in one moment to contain experience that the
moment before it could not. Even after dissociation has been resolved, the
most we can do is allow history, or tradition, or the speech and conduct
of the other to act freely within us. We cannot decide to understand, even
under the best of circumstances; we can only strive to put ourselves in the
best position for understanding to occur. And so, while the resolution of
the analyst’s dissociation is crucial, it guarantees nothing: It merely means
that the circular movement that may result in the expansion and flowering
of meaning can occur with less obstruction by unconsciously held motiva-
tion. New understanding may follow immediately or it may not. Resolving
dissociations gives language its head, but what language will do then is
beyond our capacity to know.
4
The Eye Sees Itself
Dissociation, Enactment, and the
Achievement of Conflict
Part I: Must the Eye See Itself?
The analyst’s unconscious participation in the therapeutic relationship

interests us today for a very diἀerent reason than it used to. The signifi-
cance of countertransference no longer lies in its status as primary obsta-
cle to the analyst’s perception of the truth. As the goals of psychoanalysis
have shifted from the acquisition of insight to authenticity, the freedom
to experience, and the expansion of relatedness (e.g., Mitchell, 1993), we
have recognized that countertransference is as much what the analyst
does as what she feels, thinks, and fantasizes, and these enactments, as we
have come to call them, have become opportunities as much as obstacles.
Insight remains crucial because it increases our range of choice. But no
longer is the appearance of new understanding always viewed as the heart
of the matter, as it was almost uniformly just a few short decades ago.
Now, at least in some analytic circles, insight is just as likely to be viewed
as a sign that the important change—the shift in analytic relatedness that
allowed the new understanding to arise in the first place—has already
occurred. Ghent (1995), for example, writes:
In the early years of psychoanalysis, the prevailing view was that therapeu-
sis was essentially informational—insight and awareness would bring about
changes in the ways one would experience events and respond to them. Over
time, there has been a subtle shift from the informational perspective to the
transformational, where insight is often retrospective rather than the active
agent. (p. 479)
71
Among many others, the members of the Boston Change Process Study
Group (BCPSG, 2002, 2005, 2007, 2008; D. N. Stern et al. 1998) are working
this vein. Their transformational view of therapeutic action grows from
the application of complexity theory to psychoanalysis, an area bursting
with theoretical and clinical potential.1
And yet, while loosening the constrictions in analytic relatedness has
become the main event in therapeutic action, and despite the fact that
many enactments relax for reasons that seem to have little to do with the
eἀects of our consciously intended interventions, we know no other means
of focusing our eἀort on dealing with enactments than to try to feel our
way to a speakable understanding of where in our experience and interac-
tions with patients enactments exist and what they are about. Even in the
era of transformation, the analyst must know the countertransference but
does not; and what she is deprived of is precisely the practical grasp she is
able to construct in the case of many other aspects of her experience with
the patient. What we need is what it seems we cannot have. In this ironic
sense, insight as an agent of change is alive and well.
And right there we come across the rub that challenged me to write this
chapter: How is it that we ever come to know what we are doing in the
unconscious parts of our relatedness with the patient? How can the eye pos-
sibly see itself? How can we conceptualize countertransference awareness?
Although virtually every contemporary psychoanalyst would agree
that it is crucial for the analyst to find a way to reflect upon as much of
her unconscious relatedness to the patient as possible, we have nothing
even approaching a similar agreement about how that capacity for self-
reflection comes to light. As a matter of fact (and oddly enough, if you ask
me), the question of how it is that analysts put themselves in a position to
know the countertransference is seldom even broached.
By using the word “how,” though (as in “how we know the counter-
transference”), I do not mean to pose a question about technique. I do not
mean to ask what the analyst should do in order to become aware of the
countertransference. Though I shall oἀer clinical remarks and an illustra-
tion, the quandary I want to address is not technical at all, but theoretical:
1 The term “complexity theory” often refers collectively to complexity theory, nonlinear dynamic
systems theory, and deterministic chaos theory. Recent psychoanalytic work in this area (in addi-
tion to what I have cited in the text) includes Galatzer-Levy (1978, 1996, 2004); Ghent (2002);
Grossmark (2007); Harris (2005); Kieἀer (2007); Levenson (1994); Mandel and Selz (1996); Miller
(1999); Moran (1991); Quinodoz (1997); Palombo (1999); Piers (2000, 2005); S. Pizer (1998); Shane
and Coburn (2002); and Spruiell (1993).
The Eye Sees Itself 73
how is it conceivable that we observe our own unconscious participation in

ongoing relatedness? If our unconscious involvement with our patients is
inevitable and continuous, how in the world do we ever develop a convic-
tion we can trust about which interpretations or relational interventions
might be most useful in any particular moment? Why isn’t every clinical
intention simply swallowed up in an enactment? This seemingly endless
circularity is what Edgar Levenson was describing in the title of his first
book, ἀ e Fallacy of Understanding (1972): The analyst’s insights are not
only what the analyst thinks they are; they are also, and more importantly,
participations in what most needs to be understood.
When the issue is put this way, does it not seem that the analyst ought
never to be able to know the countertransference? Where is the “perch”
(Modell, 1991) from which the unconsciously involved analyst somehow
gains a clear-eyed view of the very involvement that is standing in the way
of the view in the first place? From a purely logical perspective, the task of
observing one’s own unconscious involvement with one’s patients seems
to be a contradiction, a “bootstrapping” operation (Mitchell, 1993) of
impossible proportions. “I have met the enemy,” said the immortal Pogo,
now many years ago, “and they is us.” For years after the appearance of
Levenson’s book, many of us did not know whether we were more excited
by its ideas or despairing about what it meant about our control over
our clinical work. But we got used to it, we began to live with it. Perhaps
we got too used to it. It is time to return to the unanswerable question
that Levenson’s work, and much else that came after it in Relational and
Interpersonal psychoanalysis, seems to pose: how can the eye see itself?
The Nature of the Problem
Each participant’s conduct plays a very important role in “locking” the

other participant into the unseen or rigidly perceived patterns of the
transference and the countertransference. I have referred elsewhere to
this phenomenon as “the grip of the field” (D. B. Stern, 1989, 1997), and
Wolstein (1959) refers to it as “the transference/countertransference inter-
lock.” It is partly because of the way the analyst’s experience is aἀected
by the transference that the countertransference is unconscious; and it is
also partly because of the way the analysand’s experience is aἀected by the
countertransference that the transference is unconscious. Because they are
mutually embedded, transference and countertransference are reciprocal

and inseparable parts of a whole.
This interlocking often enough leads to a situation in which alterna-
tive patterns of relatedness are invisible to both participants, and nego-
tiation is therefore difficult, if not impossible. The relatedness has a
certain mutual intransigence that no good intention is enough to relax.
Each person’s attitude toward the other is frozen. When this unyielding
quality is conscious and obvious to both participants—that is, when it is
reflected in both participants’ explicit views of one another, a therapeu-
tic impasse neither can deny—the demand for interpretation (or rather,
reinterpretation) is clear enough. Under such circumstances, both ana-
lyst and analysand are painfully aware of the need for a diἀerent kind
of understanding and relatedness between them. Every clinician knows,
though, that even then—even in the presence of a mutual, ardent, con-
scious wish for it—a diἀerent and useful way of seeing the situation may
still be maddeningly elusive.
The more common and even more difficult situation, however, is for
analyst and analysand to be locked into an unconscious enactment that
neither of them even knows is under way. The rigidity in the interaction,
that is, is not apparent to either the patient or the analyst. Quite often,
when we look backward in time from a wider understanding, the patterns
we conclude were crucial for us to identify—because they were the sources
of difficulty or puzzlement in the analytic work—were not visible at all
during the sessions within which they were most influential. Because the
analyst had no reason to attend to and formulate such patterns during
those sessions, the relevant relatedness was no more conceptually distinct
than the air the analyst breathed. But our ignorance of the intransigence
does not mean that it was not there; it merely means that neither analyst
nor patient saw the appropriateness of characterizing the relatedness as
intransigent until later on.
And so we face what seems to be a contradiction: On the one hand,
the field must change if it is to become possible for analyst and analysand
to learn something new; but on the other hand, it is quite often pre-
cisely because the field needs to change that neither the analyst nor
the analysand can identify what to address in order to provoke these
changes. It seems that the needed alterations in the analytic relatedness
are themselves the conditions for their own accomplishment. We there-
fore appear to be left with no way to account for the fact that psycho-
analysis is not endlessly circular.
Wolstein and the “Private Regions” Solution
This conceptual problem does not arise in such a pointed way for those
analysts who believe that a significant part of the self exists in a nonsocial
realm, in a part of the psychic geography that exists completely apart from
interaction with others, untouched and unmediated by social interchange.
Obviously, these analysts, most of whom think in the terms of modern
Freudian or Kleinian theory, have the same clinical difficulty we all do
in understanding the enactments in which they are involved.1 But at least
in theoretical terms, from their vantage point the problem is less thorny
than it seems from interpersonal and relational perspectives. Theorists
who conceive the self to be partially disconnected from social interchange
can at least hypothetically conceive how the analyst is able to construct an
understanding of the enactments in which she is involved, even if finding
a way to that understanding is painful, difficult, and slow. If there is some
part of self-experience that exists in a nonsocial realm, that is, there exists a
metaphorical high ground, or as Mitchell (1993) puts it, a “platform,” onto
which the analyst can climb to observe the fray between herself and the
analysand; there is a “place” in the self where one can stand, a private place
uninfluenced by any aspect of the other. The analyst can retreat to these
private regions, and from there, even if she meets internal resistance to
grasping her own unconscious involvement, there is at least the theoretical
possibility of a view of the involvement from an uninvolved perspective.
A “private region” that would be useful in this way would need to have
two characteristics. First, it would be asocial, existing outside the influ-
ence of the countertransference, because only a “platform” unaἀected by
the countertransference could allow the analyst to step away far enough
from her unconscious involvement with the patient to be able look back at
that involvement and make it the object of understanding. Second, such
a private psychic “place” would have to be a portion of subjectivity expe-
rienced as the analyst’s own, not as alien or other. In other words, if the
analyst is going to be informed about her unconscious involvement by a
part of her own mind, that part has to feel like it belongs to her. If it does
not feel that way—if it is not “self”—she has no reason to “listen” to it, and
perhaps (as in the case of the unconscious) she cannot.
1 For an examination of these issues in the work of contemporary Freudians, including those of a
liberal bent, see Hirsch (1996); for a view of the same problems in modern Kleinian thinking, see
D. B. Stern (2001) and Mitchell (1997, Chapter 4), who also describes a similarly problematic posi-
tion held by some contemporary interpersonal analysts (1997, Chapter 3).
If we put these ingredients together, we conclude that a clinically usable

private region of self would be, as Winnicott (1960) describes “true self,”
an “incommunicado” aspect of experience that nevertheless feels like one’s
own. But even Winnicott (1949), despite his early interest in the use of
countertransference, did not use the “private region” argument to grapple
with the problem of how the eye could see itself. That line of thinking
could only be developed by someone who appreciated both the embed-
dedness of the analyst in the field and the private region argument for
dealing with it. The single writer to have done this, as far as I know, is
Benjamin Wolstein (1954, 1959), one of the first to take the position that
transference and countertransference are inevitably reciprocal, though he
is seldom credited as he should be in this respect. Beginning in the 1970s,
Wolstein (1971a, 1971b, 1972, 1974a, 1974b, 1975), at about the same time
that Levenson was formulating the idea of the analyst’s inevitable uncon-
scious involvement with the patient, took a drastically diἀerent course,
180 degrees in the other direction. He began to add to his earlier views
a conception of what he called “the psychic center of the self,” a core of
personhood that, whether it founders or is actualized in the course of liv-
ing, cannot be altered in any essential respect. The psychic center of the
self is not necessarily unreachable from the outside (i.e., by contact with
others), but when it does enter social interaction, it can only be denied or
recognized, never changed or influenced. Wolstein came to believe that
the most meaningful aspect of therapeutic relatedness only occurred once
the transference/countertransference interlock had been dissolved. Only
then did it sometimes become possible to communicate directly from
the center of one self to another. Wolstein felt that the most meaning-
ful of these communications concerned matters that the other might not
even be aware of. He looked forward, in particular, to the capacity of his
patients to observe unconscious aspects of the analyst—himself. The suc-
cess of these observations, however, was not necessarily to be judged, as
Sullivan would have had it, by whether they were clearly understood by
the other. What mattered more to Wolstein than consensual validation
was whether what you observed and communicated about the other was
true to your own experience.
For the present purpose, the significance of the psychic center of the
self is that it allowed Wolstein to pose and solve the riddle of the eye see-
ing itself. Wolstein explicitly noted that, as far as he could see, unless we
conceive a kind of experience that remains uninfluenced by the other
under any and all circumstances, there was no way for the analyst or the
analysand to understand their unconscious involvement with each other.
There must be a place in psychic life from which, as we become familiar

with it, we can observe our unconscious involvements. Resolving transfer-
ence/countertransference interlocks, therefore, is not so much the point
of psychoanalytic treatment as it is a prerequisite to love, which is for
Wolstein the capacity and willingness to know and accept one’s deepest
view or sense of the other. In an interview with Hirsch (2000), Wolstein
said that “a uniqueness of self is the most direct way to get over and out
of that interlocking; it makes it possible, it opens up, the love, as distinct
from the intimacy, to let it go” (p. 198). And then a moment later, sound-
ing the same note Bollas (1989) strikes when he describes the notion of
personal idiom, Wolstein adds, “We all have a unique sense of self that is
inborn: Clinical psychoanalytic inquiry doesn’t create it; we find it there”
(p. 199).
If one can accept such an argument—that there is some inviolate, non-
social core in the self—one can work out the problem of countertrans-
ference awareness fairly simply: One simply observes from within the
nonsocial core.
But however appealing Wolstein’s position is, and however wistfully I
regard it, I cannot accept it. It certainly makes sense to posit temperamen-
tal diἀerences in the self, as well as important inborn constraints on, and
potentials for, what the self can become.1 But it is only within a later social
world that these inborn constraints and potentials take on their meanings
and exert their eἀects. The self is a social construction. That is not to say
that social construction has no limits; it does have limits, significant ones
(D. B. Stern, 2000). Nor does this position require us to deny the unique-
ness of each self.
Even if I cannot agree with Wolstein about the nonsocial core of self,
though, I admire his early recognition that we must think out the problem
of countertransference awareness. That recognition is still very much the
exception. Of those clinicians who continue to work and think as if they
have access to an objectivity that allows them an impersonal view (what-
ever that is) of their own unconscious involvement with their patients,
most have not really reflected on how they conceive that objectivity.2
1 Recent influential conceptions of inborn dispositions of this kind include, besides Winnicott’s
(1960) true self and Bollas’s (1989) personal idiom, Kohut’s (1984) idea of the self’s destiny or
nuclear program and the description by Fonagy et al. (2002) of implicit procedural primary emo-
tion states.
2 See Friedman (2000), though, for a thoughtful recent defense of objectivity in psychoanalysis,
though in my view Friedman, too, underemphasizes unconscious social embeddedness and over-
emphasizes the degree to which people are capable of observing their own motives.
“Where” in subjectivity is it? Wolstein sidesteps that problem altogether,

as we all should, by dealing with it without recourse to a mythical objec-
tivity. All one can do, he believes, is know and remain true to one’s deepest
feelings and perceptions, the psychic center of the self.
The rest of us, though, to the extent that we can accept neither Wolstein’s
solution nor a belief in objectivity and impersonal understanding, must
look elsewhere. We know we manage to succeed in what we cannot
explain how we do. In the search for an understanding that Interpersonal
and Relational analysts, too, can accept (the irony here, of course, is that
Wolstein was one of the most prominent of interpersonalists), I turn to
a clinical illustration. I oἀer it with certain misgivings, though, because
if I am not careful to say exactly what I think I am up to, this kind of
presentation (the discussion of a particular enactment) might convey the
impression that the analyst’s unconscious participation in the process is a
sometime thing. For that reason I state my perspective at the outset. Along
with the many other analysts who share my orientation, I take the position
that the analyst’s and the analysand’s unconscious involvement with each
other is inevitable and continuous.
Now, as soon as I say that, of course, I have stepped yet again right
into the middle of the dilemma. How can we maintain that unconscious
involvement is continuous without destroying the significance of the ana-
lyst’s (and the patient’s!) thoughtfulness and capacity for observation? That
is the paradox with which I begin. Unconscious involvement is ceaseless,
and yet we do not doubt that the analyst makes valid and useful observa-
tions and interventions.
But enactment is only one variety of the analyst’s unconscious involve-
ment. There is a great deal of mutual interactive regulation going on between
any two people, and most of it occurs outside awareness (see, for example,
Beebe & Lachmann, 1998, 2002). In the way I want to define the word,
though, mutual regulation is not enactment. Mutual regulation, especially
of aἀect states, is frequently carried out without conscious design, it is
true. Mutual regulation is descriptively unconscious. But it is not dynami-
cally unconscious. Neither is the kind of responsive participation that is a
key part of most treatments, the kind of reparative and facilitative uncon-
scious involvement—accepting, loving, humorous, or playful—that self
psychologists might refer to as self-object functioning. One has to “mean
it” in order for this kind of relatedness to be useful to the other, and so it
has to be more deeply felt than mere conscious decision could make it.
It has to be part of the analyst’s nonconscious involvement. The same is
true of holding and containment and the host of other noninterpretive
participations that have become so important in our understanding of our

work. But none of these participations is a dynamic necessity. Although
the analyst may not consciously think about doing such things (though of
course sometimes he may), neither does he feel compelled to carry them
out. The analyst’s agency is not compromised; he could stop participating
this way if he wished, but generally he judges it best to continue. In the
most important sense, then, all these kinds of relatedness are freely cho-
sen. Enactment, by contrast, is rigid and unyielding for both analyst and
patient; it either feels irresistible to both or is recognized, after the fact, to
have been a (unfelt) necessity.
It is usually not painful for the analyst to feel and think about his
participation in mutual regulation, affirmation, empathic understand-
ing, reparative and facilitative involvements, holding, containment, and
so on. It may not always be possible to say what unconscious sources are
being tapped at these times, and there may be an emotional cost to be
paid for participating in one of these ways in the face of a strong counter-
transferential pull to behave otherwise; but there is little or no internal
resistance to formulating whatever there is to know about such clinical
attitudes and interventions. One is not deprived of the freedom to think.
And so the problem of the eye seeing itself does not come up in these
circumstances. In enactments, on the other hand, the analyst is more or
less blind to what he is doing and feeling, and he is likely to suἀer. He
cannot find his way for some time to a kind of being with the patient
that would relieve both of them. And therefore, while the analyst’s con-
tinuous unconscious relatedness is the rule, the problem I am setting
out to explore arises only in that portion of the analyst’s unconscious
functioning that, because it is dynamically unconscious, deserves to be
called enactment.
An Enactment: Guilt and Narcissism
In deceptively smooth treatments, we find in retrospect that, all along,

each participant was influencing the other to maintain the status quo. In
one case of mine, the patient, a talented but immature man in his 30s who
had managed to deep-six virtually every one of the many academic and
professional opportunities he had had, worked very hard in treatment and
expressed deep appreciation to me, though he also expressed his fear that
I would somehow indoctrinate him and make it impossible for him to
continue the somewhat self-destructive “fringe” life he had been leading.
During one of his high school years, he had been in a once-per-week psy-
chotherapy for his academic problems, but that treatment had seemed
to him to be nothing more than tutoring, and useless tutoring at that.
According to him, his therapist, like his teachers (and himself, for that
matter) had been completely baffled about why this very bright and appar-
ently well-meaning young man, who seemed quite sincere about wanting
to do well, just could not seem to get his work done.
After a year and a half or so, during which time the treatment seemed
immensely productive to both of us, I began to feel a vague sense of unease.
Something bothered me. Over a period of a couple of weeks, I began to
formulate what was the matter. The treatment had begun to feel to me
very subtly less alive, less continuously intriguing and animated. Certain
moments that I knew would have been interesting in the recent past were
now perhaps a little flat or stale, maybe just a bit forced. Simultaneously,
the analysand, while insisting he was trying as hard as he could, was fail-
ing quite spectacularly in certain academic activities that he had felt, with
my tacit agreement, would represent progress for him.
We found out somewhat later that the analysand was doing with me
precisely what he had done with his parents, though as is so often the case,
he was doing it with such utter naturalness that it shaped our interaction
unobtrusively. On the face of it, he was the dutiful, cheerful, and loving
son; but he was simultaneously failing in an unconsciously purposeful way
that he was able to believe consciously had nothing at all to do with his
own intentions. We eventually learned that he had always been angry and
depressed at the expectations he faced from his parents, which had never
seemed to him to have much to do with what he wanted from life or how
he felt; but because of his parents’ narcissistic vulnerability and his loving
feelings toward them, he had felt too guilty to protest directly. Instead, he
acted out in a way that was invisible to him but that hit them where it hurt:
He failed dramatically to accomplish anything that would have pleased
them, a mission that had ruined every one of the opportunities they had
found a way to oἀer him. With me, the patient had begun the relationship
in a genuinely collaborative mode in which we had accomplished a great
deal. I have not felt the need to revise that initial impression. But we also
discovered that, as I had begun to enjoy his collaboration, he had begun
to resent my pleasure, and had revised our history together, beginning to
suspect (without quite realizing that he felt that way) that we had instituted
our collaboration more for my reasons than for his own. He then began to
treat me as if I needed this narcissistic pleasure, as he had felt (again, in an
unformulated kind of way) his parents had. For a while, I had not noticed
this change, or had registered it only as a subtle aἀective shift, most of the
time continuing to enjoy what was fast becoming a pseudo-alliance and
ingratiation, just as his parents, by what I could gather from his reports,
had always taken pleasure in what seemed to be his adoration and good
cheer. It came to my attention in these weeks that I had also been enjoying
his appreciative responses to my interpretations, and I now realized that I
had been making more of them recently than I usually do. The analysand,
in other words, was playing to what he unconsciously fell into assuming
was my narcissism, and I was enjoying it enough that he had reason to
believe that he needed to keep me well supplied if we were to continue to
get along together.
For my present purpose, the point of the illustration is that I responded
to the analysand’s way of relating to me by developing a countertransfer-
ence that, in turn, reinforced his transference; and as a result, as is usually
the case, he and I locked each other so securely into an unconscious set of
interpersonal patterns that it soon became irrelevant, actually, which of us
was responsible for having provoked the interaction in the first place. The
very idea of determining such a thing, as a matter of fact, would have been
(and often or even usually is) nothing more than a blaming operation.
Once the nature of our relatedness came to light, there was the possibil-
ity for change; and so my example illustrates not only the interlocking of
transference and countertransference, but also the breaking of this mutual
grip. How and why that happens is my primary question. Why did I not sim-
ply continue to play out this scenario of narcissism? Where did the vague
discomfort that helped me to see things diἀerently come from? Eventually,
once I had made some sense of my feelings, I drew the patient’s attention
to what I had noticed about the change in the atmosphere between us and
invited him to investigate it with me. The results I have already described.
I turn now to what made this turn of events possible.
Snags and Chafing
If we accept that the analyst and the patient comprise an immensely com-
plex, changing, self-organizing system, it seems virtually inevitable that
some impasses will relax for reasons we understand only in retrospect,
if we ever understand them at all. But that is only one of the possibili-
ties. In another scenario, the analyst is alerted to examine her experience
for clues to her unconscious involvement with the patient—for hints at
the nature of the system, if you will. More often than not, the alerting
signal is something small and subtle. It often has a mildly bothersome

quality. One feels an emotional “chafing” or tension, an unbidden “hint”
or “sense” that something more than one has suspected is going on in
the clinical interaction. Something feels inconsistent, countering an aἀec-
tive expectation we did not even know we had until that moment; it feels
subtly “wrong” or contradictory or just uncomfortable. For the curious
analyst, therapeutic work is often the psychic equivalent of walking along
a forest path in a wool sweater that snags now and then on a branch or
twig. When it does, we stop, investigate, and disentangle ourselves. We
note some kind of change in our feelings, ask ourselves about it, and find
that we are responding to something about the analysand that we have
not yet explicitly noticed. A new perception comes about. This quiet and
self-reflective process is just as important during impasses, deadlocks, or
stalemates, of course, in which the noise and drama of the interaction is
endlessly uninformative, as it is during the more silent kind of enactments
that only become visible in retrospect.
All well and good, says my personal version of what Sullivan (1950)
described as “an illusion, an illusory person, in the sense of a critic, more
or less like what we think the hearer is” (p. 214). This illusory critic is an
essential but less than entirely inspiring amanuensis who insists that I
think through problems in a publicly verifiable fashion. And in this case,
he has an objection: Even if we can all agree that we often find our way
to an awareness of enactments via aἀective signposts, such as the subtle
diminution of aliveness in the treatment I described, why does this con-
tribute to solving the conceptual problem I have set out to address? Do we
not simply face the same dilemma in another guise—the problem of the
eye seeing itself? How can we account for our capacity to notice the signs
of our unconscious involvement? Shouldn’t our capacity to notice what
points to the enactment be swallowed up in the enactment itself? Why
should enactments be black holes in some respects but not others? If they
suck into themselves our capacity to observe them and to know what to do
about them, why in the world would the same fate not befall subtle aἀec-
tive hints at their existence?
Well, we know better than to accept the black hole argument; but we
still have to admit that the objection makes sense in a purely logical way.
Although we know perfectly well that we experience emotional snags and
chafing and that they eventually help us grope toward some kind of grasp
of the most significant of the enactments that emerge in a treatment, so far
we do not have a way to say why any of this is possible.
Part II: A Theory of Enactment
Dissociation and the Multiple Self
The very existence of my internal critic oἀers a hint, though. Why does it
take him (her) to notice that what I have written does not really answer
the question I have asked? Why does he notice that it seems to be contra-
dictory to say that we can experience hints, but that we cannot just leap
directly to new perceptions of the enactment? He notices because he has a
diἀerent agenda than the part of myself that tries to solve the problem. I
am trying to contribute a creative solution; he, on the other hand, is trying
to make sure that what I say meets the standards of public discourse. “His”
is a diἀerent task than “mine”; I (now in the larger sense of the whole per-
son) am divided. These two purposes of mine are identified with what we
call “parts” of ourselves or, in the current parlance of Relational thinking,
“self-states” or even “selves.”1
I am able to see the inadequacies in my own argument because the
“me” who tries to think creatively acknowledges the “me” who audits my
thoughts—or more properly, because the “me” who tries to think creatively
experiences conflict with the “me” who audits my thoughts.
Is it conceivable that I might not experience the conflict with the other
“me”? Of course. In that case, we could say that these two aspects of me,
these two self-states or personifications, were dissociated from each other.
If I identified with an imagination dissociated from its critic, I would be
grandiose: My ideas would encompass too much significance, apply too
broadly; and by implication I would assume an unjustified self-impor-
tance. If I felt more closely identified with an internal critic who was, in
his turn, dissociated from (did not experience conflict with) my imagina-
tive side, my writing would be flat, uninteresting, overly cautious. In fact,
I might very well never write at all because, in my illusory critic’s estima-
tion, there would always be something that someone could find wrong
with what I would say (and of course he would be right). Just this sort of
dissociation might take place, in fact, in someone who grew up with a par-
ent for whom mistakes were catastrophic. We all know the adult children
of such parents: They are so concerned—consciously, yes, but especially
1 See Bromberg (1998, 2006); Davies (1996, 1997, 1998, 1999, 2004); Davies and Frawley (1991, 1994);
Flax (1996); Harris (1996); Mitchell (1991); S. Pizer (1998); Slavin (1996); Slavin and Kriegman
(1992, 1998); D. B. Stern (1997).
unconsciously—with avoiding the disapproval of critics, inner and outer,

that they are virtually incapable of spontaneous experiencing.
What is the ideal situation? The ideal, and one that many of us actually
do approximate, in at least some significant portion of our experiencing,
is the ongoing and continuous awareness of conflict between self-states,
what Bromberg (1998) calls “standing in the spaces.” It is only when we
can tolerate conflicts between multiple states that we can negotiate the
disagreement between them (S. Pizer, 1998). I hasten to add, though, that
resolution is not necessarily the raison d’etre of negotiation between states
of mind. Negotiation is an ongoing, never-finished weighing of the alter-
natives. If we are willing and able to experience conflicting purposes at the
same time, negotiation is the natural stance for us to take toward them.
On the other hand, we cannot negotiate until conflict comes about.
If I wish to write something that will be of use to others, for instance,
I need to be able to experience simultaneously my illusory critic and my
creative side. I need to be able to think creatively at the same time that I
think critically about thinking creatively. And when diἀerences between
these interests crop up, as they do more often than I would like, I need to
be able to evaluate each purpose in the light of the other. I must either be
able to revise my manuscript or to answer my illusory critic in a way that
satisfies him. I need, in other words, to be able to allow myself to create
the experience of conflict. If I hole up within one or the other of these self-
states and cannot formulate the perspective that would be oἀered by the
other, there is no conflict; I am dissociating.
Dissociated experience, we have learned, does not simply disappear qui-
etly into some hidden corner of the mind. It is enacted.1 I will “play out”
the state of self I cannot tolerate experiencing directly, and I will thereby
unconsciously influence those with whom I relate to adopt a variation on
the same dangerous response that led me to dissociate the self-state in the
first place. In one variety of enactment I embody the traumatized self, in
a continuous and futile attempt to make everything happen diἀerently,
thereby healing myself; but instead I provoke the other person to experi-
ence and behave in ways that, tragically, simply keep retraumatizing me.
In the reciprocal version of this enactment, in a similarly unconscious
attempt to wrest control of the situation, I traumatize the other just as I
have myself been traumatized, but I have little or no appreciation of my
1 See, for example, Aron (2003a, 2003b); Bass (2003); Benjamin (1998); Black (2003); Bromberg
(1998, 2006); Davies (1996, 1997, 1998, 1999, 2003); Davies and Frawley (1991, 1994); B. Pizer
(2003); S. Pizer (1998).
role in doing so. (And if the trauma has been bad enough, I may not even
care [Stein, 2004, 2006].)
We expect the roots of dissociated self-states to be inaccessible; but in
an enactment, the reciprocal state called out in the other is as inacces-
sible to him as my dissociated self-state is to me. My interactive partner
is unaware that his role is anything other than his own choice; he is as
unaware as I am that he is being nudged into it by a reciprocal pattern to
which we both contribute. Each person therefore experiences the other
person as instigator, at fault, imposing his will.
For example, if I dissociate my internal auditor, and my work is gran-
diose, I will be unaware of my overreaching and therefore be unable to
appreciate my role in provoking criticism from others. When criticism
comes it will surprise me, and I will be either devastated (and will perhaps
behave in ways unconsciously designed to inspire guilt) or outraged (how
dare they!). On the other hand, if I dissociate my risk-taking, creative side,
I will be out of touch with the way that the extremity of my self-criticism
suggests to others the magnitude of the contribution I secretly (even from
myself) expect to make.
We do not insist that Sullivan replace his “illusory critic” metaphor with
the description of impersonally defined cognitive processes because it is
intuitively and immediately obvious to us what he means by the metaphor.
In our imaginations we do not limit the characterization of our various
conflicting purposes to the purposes themselves; we characterize our pur-
poses as “parts” of us, as “states” of our selves. We do not simply describe
our internal worlds; we people them. We do not understand ourselves as
concatenations of aἀects, cognitions, and conations, however reasonable
it may be under some circumstances to describe our minds in these terms.
Instead, we sense our fragments as characters. Sullivan’s (1954) theory
describes the various aspects of personality as “personifications,” reflect-
ing his view (he mirrors object relations writers in this one respect) that
the elements of our inner landscapes have human characteristics.
And for good reason. Each of these personifications grew up around
the relatedness to a particular significant person or around the relatedness
to parts of various significant people. When experience is traumatic, the
child cannot bear to experience simultaneously states that were created in
the presence of safety and others created during the appearance of a dan-
gerous, traumatic person, or part-person (“angry-father,” for example).
Dissociation is born. The child, and then the adult, enacts the traumatic
states and lives his “known” life inside the bearable ones.
Dissociation, Enactment, and the Achievement of Conflict
Everything discussed in the remainder of this chapter is an exploration

and expansion of the following proposals about dissociation and enact-
ment. Once I have explored these ideas fully enough, I will return to the
question of how the eye sees itself. I will not be better equipped than I am
now to answer that unanswerable question, but I will be in a position to
make the case that it should be recast, and in that new form it will become
a sensible question.
1. Enacted experience, and thus dissociated states as well, cannot be sym-

bolized and therefore do not exist in any other explicit form than enact-
ment itself. Enacted experience is unformulated experience.
2. Dissociated states, because they are unsymbolized, do not and cannot
bear a conflictual relationship to the states of mind safe enough for us to
identify as “me” and inhabit in a consciously appreciable way.
3. Enactment is the interpersonalization of dissociation: the conflict that
cannot be experienced within one mind is experienced between or across
two minds. The state dissociated by the patient is explicitly experienced
by the analyst, and the state explicitly experienced by the patient is dis-
sociated in the analyst’s mind. Each participant therefore has only a par-
tial appreciation of what is transpiring.
4. Enactment, then, is not the expression of internal conflict. Enactment is
the absence of internal conflict, though the external conflict, the conflict
between the two people in the enactment, may be intense.
5. Enactment ends in the achievement of internal conflict, which occurs
when the two dissociated states, one belonging to each participant in
the enactment, can be formulated inside the consciousness of one or the
other of the two psychoanalytic participants.
I am indebted to Philip Bromberg’s (1998, 2006) thinking in the devel-

opment of these ideas. I might go so far as to say they bear his imprint,
because the relationships drawn in this chapter between dissociation, con-
flict, and enactment also appear in his work. Specifically, Bromberg (1998;
see especially 2000, pp. 564–567) believes that enactment is the result of
dissociation, that conflict does not exist in enactment, and that enactment
dissolves when conflict becomes possible. I have reached these same con-
clusions, starting from the idea of unformulated experience (D. B. Stern,
1983, 1997), a dissociation-based perspective on the unconscious, and a
theory of enactment anchored in that conception (D. B. Stern, 1997; see
also this book, Chapters 3 and 5–7).
I have also drawn on the orientation that began in the work of Jody
Davies and Mary-Gail Frawley-O’Dea (1991, 1994) and has been devel-
oped further in Davies’ more recent writings (1996, 1997, 1998, 1999, 2003,
2004): The patient’s dissociated experience is routinely enacted, and the
analyst’s sole route of access to that experience is the analysis of the trans-
ference–countertransference. Davies’ clinical presentations, especially
recently (2004), are eloquent and nuanced illustrations of the problems all
psychoanalysts face in finding a way into enactments and the mutual dis-
sociations that underlie them.
In the wide-ranging position articulated by Fonagy et al. (2002), dis-
sociation, splitting, and enactment play prominent roles, just as they do in
Bromberg’s and Davies’ thinking and my own (see especially Fonagy et al.,
Chapter 10, pp. 373–433). There are interesting, sometimes striking, simi-
larities between the work of this group and the model I present here, and
they deserve to be detailed. In Chapter 8 of this book, both the similarities
and the diἀerences between these two views are discussed.
Two Background Considerations
Before setting out to explicate the points I have just made, I highlight two
background considerations, each of which supplies a portion of the theo-
retical context or background within which I intend the points I have just
listed to be understood.
Sources of Conflict
When I refer to conflict that remains to be created, I am referring to only a
small part of the conflict that goes on within subjectivity. I am not claim-
ing, for instance, that all conflict emerges from the analysis of the relations
of me and not-me, as if conflict had no other source. It goes without say-
ing that a great deal of conflictual experience is well known to the patient
prior to any sort of treatment. And a great deal of the conflict the patient
eventually comes to appreciate as the treatment unfolds—as a matter of
fact, the largest portion of the conflict created during the treatment occurs
between what Sullivan refers to as good-me and bad-me. These phrases are
umbrella terms for the self-states that originated in the approval and dis-
approval of the significant people in our early lives. Together, they make up
what each of us feels as me, or self. Conflict between good-me and bad-me,
even if it is not initially recognized by the patient, is not terribly difficult to
achieve. Because both aspects of the conflict already exist within the self,
it is often sufficient simply to point out an inconsistency or a contradiction

and ask the patient what it might be about; or perhaps the analyst makes
an interpretation. Although conflicts between good-me and bad-me may
be unformulated, in other words, the states themselves either are already
formulated or are within the range of the patient’s capacity to articulate.
Even when the separation of good-me and bad-me is represented in an
enactment between the patient and the analyst, as happens often enough
(the roles of good and bad may be split between analyst and patient), the
quandary is generally negotiable without much difficulty.
The enactments that derive from the separation of me and not-me,
though, are much more difficult to negotiate, and often enough intracta-
ble.1 Simple interpretation is just not sufficient, at least if by that we mean
the more or less relaxed, professional application of the analyst’s mind—his
wits and intuition. To work with an enactment, the analyst must actually
give himself over to the nonrational, aἀect-laden parts of the experience,
and sometimes for fairly lengthy periods. Not-me dominates the treatment
only episodically, though such episodes are not necessarily rare, depend-
ing on the particular treatment; but addressing them successfully is per-
haps the most important part of the therapeutic action of any treatment,
because it is through the incorporation of not-me that the self (i.e., me)
expands. And so, while including unformulated conflicts between good-
me and bad-me in what I have to say, the enactments I address here are
primarily the enactments most significant for psychoanalytic work, those
rooted in the relations of me and not-me.
The Analyst’s Dissociations

The second background consideration concerns the reciprocal dissocia-
tions the patient calls out in the analyst (and that the analyst calls out in
the patient, because that, too, can happen). Even though in one significant
sense the analyst’s reciprocal dissociations are reactive to the patient, they
are not implanted in the analyst’s mind as if they were alien objects, the
way they are frequently conceived in the terms of projective identification.2
Instead, reciprocal dissociations are inevitably dynamically meaningful
1 I have presented elsewhere comparative clinical examples of the relatively negotiable enactments
characteristic of unformulated conflict between good-me and bad-me, and the more severe and
troubling enactments—impasses or “deadlocks”—that come about when the patient shifts into
not-me. Unformulated conflict between good-me and bad-me is dissociated in what I have else-
where called the “weak” sense, while unformulated conflict between me and not-me is dissociated
in the “strong” sense (D. B. Stern, 1997; see also this book, Chapter 3).
2 See Chapter 1 for a discussion of these issues.
events in the life of the dissociator. When the analyst participates in an

enactment, it is because she dissociates; and when she dissociates, it is
because she finds herself in circumstances that make her vulnerable in a
way she can manage, for the time being, only by dissociating. Enactment
takes place between two separate subjectivities, each acting on some kind
of combination of her own interests and what she understands to be the
interests of the other. The patient cannot provoke such a dissociation if
the analyst is not vulnerable to it. The analyst’s dissociation is therefore
as much a product of her own life as is the patient’s; and so the creation of
conflict and the negotiation of an enactment requires growth from the ana-
lyst in just the way it requires growth from the patient. The analyst’s role
is not defined by invulnerability, in other words, but by a special (though
inconsistent) willingness, and a practiced (though imperfect) capacity, to
accept and deal forthrightly with her vulnerability.
These were themes Heinrich Racker (1968) pursued throughout his
work. Racker (1957), of course, is the author of the famous aphorism
that “‘the myth of the analytic situation’ is that analysis is an interaction
between a sick person and a healthy one” (p. 132). He believed that the
analyst routinely develops a “countertransference neurosis,” contracted via
identifications with the patient’s internal objects, and that the fate of the
treatment hangs on the analyst’s capacity to resolve it. If the analyst char-
acteristically denies his own aggressiveness, for instance, he is unlikely to
feel empathic when the patient is feeling aggressive. Instead, the analyst is
likely to identify with those of the patient’s internal objects that scold or
reject the patient for having angry feelings or behaving aggressively. The
patient, that is, influences the analyst (at least the analyst who is vulner-
able, which is all of us some of the time) to re-create the original interper-
sonal circumstances that lie behind the creation of the patient’s internal
object world. It is easy to see that these “complementary” identifications
are one well-traveled route to what, in contemporary terms, we call dif-
ficult or sticky enactments, or impasses.
Racker did not see impasse as an expectable part of treatment, of course.
It was a diἀerent time in psychoanalysis. For him, it was the analyst’s
responsibility to observe and analyze his complementary identifications,
thereby restoring his capacity to identify with the patient’s ego (concor-
dant identification). Though Racker does not put it this way, we could say
that the cure of the countertransference neurosis, and the transference
neurosis as well, depends on the analyst’s capacity to stretch his identifica-
tion with the patient’s objects and to encompass the patient’s self as well.
The analyst has to be able to tolerate both perspectives at once. Once we
have said that, we have not only restated the thesis of the multiple self but
also the idea that the self is healed by the creation of conflict, by bringing
together the part that resides in the patient with the part that has been
called into existence inside the analyst.
And so, while I owe my greatest debt to writers of the current genera-
tion for the ideas I am presenting in this chapter, I am also indebted to the
earlier writers from whose work the thinking of the current generation
has arisen. Sullivan has been a particularly influential intellectual ances-
tor. But that is probably more because I have my beginnings in interper-
sonal psychoanalysis than because Sullivan is the only relevant writer of
that era. The object relations writers, and Klein, and Racker, and Bion,
and all the writers since Bion who have used Bion’s way of thinking about
projective identification, are important sources as well. All these writers,
in one sense, write about psychoanalysis as a treatment for painfully frag-
mented subjectivity, as the creation of a new tolerance within the patient
for the diἀerent parts of his own self. For Racker and Bion, just as for
Bromberg and Davies, the estranged parts of the patient’s self are called
out in the experience of the analyst, and it is from there that they are
brought home.
Transcending “Single-Mindedness”: Dissociation

and Enactment as the Absence of Conflict
The idea that enactments can be conceived not to be the outcome of con-
flict may surprise analysts unfamiliar with interpersonal theory. Since
the work of Sullivan, it has been possible to conceive the foundations of
psychopathology (or, in the terminology Sullivan preferred, problems in
living) as the absence of internal conflict. Unlike Kohut, who, for very
diἀerent reasons than Sullivan, explicitly demoted the significance of
conflict in his theory, Sullivan did not directly address the question of
conflict. However, he did introduce dissociation as the primary defense,
and he conceived “real” events as the reason that defenses arise in the first
place. “In fact,” writes Bromberg (1996), “Sullivan’s (1954) theory of inter-
personal analysis, reduced to its essentials is, in my view, a theory of the
dissociative organization of personality in response to trauma” (p. 215).
Actually, Bromberg’s characterization applies more readily to the early
Sullivan than to Sullivan’s later work. Early in his work, Sullivan (1940)
characterizes dissociation very broadly, in a way that suggests for it a for-
mative role in the development of the self. He wrote, for example:
For the expression of all things in the personality other than those which were
approved and disapproved by the parent and other significant persons, the self
refuses awareness, so to speak. It does not accord awareness, it does not notice;
and these impulses, desires, and needs come to exist dissasociated from the self,
or “dissociated.” (pp. 21–22)
In his later writings, though, Sullivan tended to restrict the range of

dissociation, using the concept primarily to define the drastic defensive
measures typical of what he called “schizophrenia” (in his hands a much
broader diagnostic category than we use today, apparently including many
people we would not consider psychotic). Referring back to the book con-
taining the passage I just quoted, Sullivan (1954) wrote, “Dissociation is
unfortunately made rather too important in Conceptions, in which I did
not take enough time to emphasize all the other things that go on besides
dissociation” (p. 317, footnote).
When contemporary writers on dissociation link their work to Sullivan,
as both Bromberg (1998, 2006) and I (D. B. Stern, 1997) have done quite
explicitly, it might be argued that we are really referring primarily to
Sullivan’s early work. That, at least, would seem to be the position Sullivan
himself would take. However, Sullivan did continue in his later work to
argue that personality (that is, the self-system) is formed by our eἀorts
not to reexperience what has hurt us before; and for that reason it is quite
supportable to argue that the work Sullivan did near the end of his life,
along with the early work, remains a model of psychopathology organized
by trauma and dissociation, not by conflict. In neither part of his career
did Sullivan consider that which is unbearable and dissociated, or not-
me, to exist in conflict with the bearable experience that makes up the
self (good-me and bad-me). Not-me is global, primitive, unsymbolized
experience banished from the self; and in that exiled state, while it threat-
ens to break through when circumstances are right (or rather, wrong), it
does not compete with self-experience for conscious representation. (If
circumstances become so dire that not-me becomes present and know-
able in explicit awareness, the consequences are not good. For the most
troubled people they are what Sullivan often called “grave.”) There is no
drive in this theory, and so there is no urgency for discharge, no press for
registration in consciousness, no derivatives, no return of the repressed.
Dissociated experience is simply absent, gone, unformulated, unknowable
in the ordinary course of things (D. B. Stern, 1989, 1991, 1997, 2002a).
Because no one in the 1930s and 1940s had thought yet about the link
between dissociation and enactment (the concept of enactment would not
be formulated for several decades), the eἀect of dissociated experience on

the psyche, as Sullivan understood it, was silent and invisible under most
circumstances. Dissociation determined where experience dared not go
without unbearable anxiety. But this psychic prohibition was not codified
or symbolized anywhere in the mind. Rather, the personality was struc-
tured around it, the way a painting can be structured around unpainted
spaces on the canvas. By a kind of reverse tropism requiring no particular
expenditure of eἀort, the self-system simply turned away from the kinds
of experience that had been dissociated. As an illustrative metaphor, think
of a road through the countryside: If you don’t stray from it, you have no
reason to suspect the existence of anything you can’t see as you walk along
it—and no reason to suspect yourself of excluding anything either. That
is the eἀect of the self-system: to keep experience on safe, well-worn, pre-
dictable paths. And so, for all intents and purposes, dissociated experience
in Sullivan’s theory is simply absent.
In Freudian terms, defense grows from unconscious conflict. The pur-
pose of the Freudian defenses is to keep conflict from becoming conscious
by allowing only one pole of the conflict to be represented in awareness.
In the view that I have developed from Sullivan’s work, it remains the case
that defense protects us from the experience of conscious conflict, but it
does so in a very diἀerent way: Conflict is avoided by simply refusing to
construct one part of the experience, the part we refer to as dissociated.
As Sullivan’s theory suggests, that experience just isn’t there. It is not that
it is “moved” to a hidden location in the mind or changed in such a way
that it is unrecognizable—it is simply not allowed to come into being. It
remains unsymbolized, unformulated (D. B. Stern, 1983, 1997, 2002a).1
In that unformulated state, while it does not necessarily threaten to break
into consciousness, it does remain a source of trouble because, as we have
become all too aware, the price for defensive control over consciousness is
that the dissociated experience is enacted. (It is perhaps worth repeating
that this point was not available to Sullivan.)
There is another reason than the theoretical, though, to think that dis-
sociated self-states do not exist in conflict with tolerable ones, a reason
even better than theory: experience itself. Consider the fact that there is
no conscious experience of conflict in enactment. One might even say that
the absence of the experience of conflict is enactment’s defining aspect.
1 For a fuller theoretical description of the process described in this paragraph, see D. B. Stern
(1997) Chapters. 6–7; for clinical accounts of enactments in these terms, see D. B. Stern, (1997)
Chapters 10 and 12 and this book, Chapter 3.
Enactment is the limitation of both participants’ experience to one pole of

what would otherwise be a conflict; in enactment, it is precisely the point
that we and the patient are each trapped within a single perception of the
other. We cannot perceive anew; each of us is “single-minded.” We can-
not experience a way of understanding the interaction that would conflict
with the perception that traps us. Taken together, the patient’s and the
analyst’s explicit experiences comprise a conflict, but this is a fragmented
conflict, located across the divide of two minds, not contained within one.
Internal conflict can be absent, and that absence of conflict, when it is
clinically salient, appears as enactment.
The absence of conflict in each participant’s experience of enactment
continues until one of the two participants, usually (but by no means
always) the analyst, is able to simultaneously occupy in consciousness
both the tolerable state and the dissociated one. Only when that happens
can the conflicting purposes that organize the two states be negotiated.
Only then, actually, can the conflict even be constructed within one mind;
and therefore, only then can it be directly experienced. The experience of
internal conflict by either the analyst or the patient, in other words, is the
necessary and sufficient condition for the negotiation of an enactment.
Intrapsychic and Constructivist Conceptions of Dissociation
Consider for heuristic purposes how my participation in my vignette

might look from a more traditional intrapsychic perspective on disso-
ciation and the multiple self, what I call the “splitting” hypothesis. In
splitting, tolerable and dissociated self-states both exist “in” the mind.
The relationship between these two states is conflictual in the traditional
psychoanalytic sense, with one state available to consciousness and the
other existing out of awareness. Consider the possibility that during the
time I was enjoying my patient’s superficial show of “progress,” I could
not bear the conflict between, on the one hand, my desire for narcissis-
tic pleasure and, on the other, my understanding (dissociated but fully
formed and present somewhere in my unconscious mind) that I would
have to sacrifice it in order to do good work. In this view, I could not
tolerate experiencing consciously the conflict that the interaction poten-
tiated in me, and so my solution was to dissociate the intolerable state,
which then continued to “be there” in my mind, now conflicting uncon-
sciously with the state I consciously inhabited.
In these intrapsychic terms, the analyst’s part of the enactment takes

place because of internal dynamics set in motion by interaction with
the patient. The patient’s corresponding part comes about in the same
way. In splitting, enactment is the disavowal, by each participant, of the
internal conflicts between their respective dissociated self-states. Both
parts of the conflict continue to exist in each person’s mind, but they are
kept apart; they may sometimes be experienced alternatively, but never
simultaneously.
The place held in Freud’s theory of repression by the banishment of
experience to the unconscious is taken, in the splitting scheme, by the
process of creating internal divisions. The shards of subjectivity that result
from this splitting then appear in awareness as successive, mutually exclu-
sive states. This is a theory of defense organized along the “horizontal”
axis of the psyche (successively experienced states), not in the “vertical”
dimension of repression (“down” into the repository of the unconscious).
In this one way, the intrapsychic splitting model resembles the model of
dissociation and enactment I am proposing here. In all other significant
respects, though, the splitting model remains similar to the conception of
defense as repression: Dynamics are based on unconscious internal con-
flict between diἀerent parts of the mind, and this conflict is already in
place when external events make it immediately relevant. In the splitting
model, the interpersonal field is assigned only the role of shifting the bal-
ance of forces, of sparking or “setting oἀ ” internal events that are already
poised to take place, having been given their essential shape in the dis-
tant past. Unconsciously motivated events between people (enactments),
in other words, are understood to be grounded in conflicts within them.
Unconscious internal conflict therefore precedes enactment.
In the conception of enactment I prefer, on the other hand, the experi-
ence of conflict within people is likely to be the outcome of clarifying the
nature of the enactments between them. As shorthand, and to emphasize
the point that the symbolization of an enactment always remains to be
accomplished, I refer to this latter, Interpersonal/Relational model as the
constructivist conception of dissociation and enactment.
A Comparison of Splitting and the Constructivist Account
Perhaps the most radical tenet of interpersonal theory is that the inter-
personal field is a primary influence on the contents of consciousness
(Bromberg, 1998, 2006; Levenson, 1972, 1983, 1991; D. B. Stern, 1997;
Sullivan, 1940, 1954). The field contributes both facilitations and limita-
tions of experience, influencing which states of mind or self can be created
and occupied by patient and analyst in any particular moment. That is
why dissociation, the most significant of these limitations on experienc-
ing, is not conceived in interpersonal theory as disavowed intrapsychic
conflict. It is, rather, the subjectivity we never create, the experience we
never have.
In splitting, as in repression, we are said to know in one part of our-
selves what we don’t know in another. We unconsciously refuse to expe-
rience what we actually do know in some hidden part of ourselves. In an
enactment conceived in constructivist terms, on the other hand, mean-
ings are split, but not between diἀerent parts of one mind. They are split,
rather, between the psyches of two people: The analyst experiences one
part of the meaning and enacts the other; and the patient experiences
the part the analyst enacts and enacts the part the analyst experiences.
The two minds are mirror images of each other; they fit together like the
two halves of a broken plate. What we hope will eventually become one
person’s consciously experienced conflict is played out between two peo-
ple. In the meantime, analyst and patient are each tempted to conclude
that only she sees the truth of the situation; only she is badly treated by
the other.
Think back once more to my clinical illustration. Think of the separa-
tion between the part of me that wanted to enjoy my patient’s “progress”
and the part that felt guilty about sacrificing my capacity to observe and
thereby letting my patient down. From the constructivist perspective, the
latter self-state (the guilty self-state) did not exist in symbolic form any-
where in my mind until that strange aἀective deadness crept into the ses-
sions and, in making me realize that “something” was there, aroused my
dormant curiosity about what was happening. Under the circumstances
provoked by aἀective chafing, the guilty self-state came into being (was
formulated), and conscious internal conflict was finally possible. My pre-
vious “single-mindedness” was not due to the denial of an existing conflict
inside me; it was due to an (unconsciously) insistent absence of curiosity
on my part, created and fostered by my participation in the enactment.
From a constructivist position the primary defense is the unconsciously
motivated refusal to create or articulate experience, a turning away from
the possibilities (D. B. Stern, 1983, 1989, 1997, 2002a, 2009). When one
does not deploy curiosity, experience goes “unmade” and is therefore lit-
erally absent. It is not “parked” or secreted in some corner of the mind;
rather, it is never articulated or constructed in the first place. Dissociated
self-states, therefore, are potential experience, experience that could exist

if one were able to allow it; but one cannot, and unconsciously will not (D.
B. Stern, 1997, 2009; see also this book, Chapters 3 and 5–8).
The interaction of present circumstances with our deepest aἀects and
intentions creates every moment of experience anew. We seldom directly
experience what we do to participate in constructing our own experience,
though. No matter how intellectually convinced we become of our cre-
ative role, our experience—what we actually undergo—has an unbidden
quality. The future comes to us; it is “found”; it “arrives.”
Because the next moment is unformulated, it may be shaped in many
diἀerent ways—but not in just any ways at all. There are significant con-
straints, ranging from tight to loose, on the experience we can construct
without lying or succumbing to madness (and even in madness the con-
straints do not disappear—instead, their expression becomes bizarre).
The concept of unformulated experience does not constitute a denial that
reality exists; it is, rather, a claim that reality is not a given, but a set of
limits on what experience can become without being false. Even tight con-
straints, though, such as those on our freedom to formulate the meaning
of an enactment, still leave plenty of room for multiple interpretations.
Dissociation, then, is the unconscious refusal to consider a certain range
of the possibilities that might be articulated or formulated in explicit expe-
rience (i.e., the possibilities within the constraints), a shutting down of the
curiosity that might have revealed them. The possibilities we are free to
construct in any given moment depend on the meanings the interpersonal
field of that moment has for us.
The Analyst’s Restoration of Self-Tolerance
In my clinical illustration we could say that I directly experienced narcis-

sistic pleasure and enacted a way of letting down the patient (I accepted too
easily that things were going well), while my patient enacted his attempt
to please me (he did not realize he was encouraging me to feel happy with
his “progress”) and directly experienced what it was like to be let down by
a parent/analyst who was all too ready to be fooled into believing things
were hunky-dory (that is, at every step of the way, my patient knew better
than to believe his own presentation). It was not until I found my way to an
awareness of my unconscious participation, the way I actually was letting
the patient down, and to the guilt that I could then formulate, that I was
in a position to experience the conflict and to negotiate it. (But remember
that my descriptions of what I was enacting could not have been formu-
lated until the enactment resolved into a conflict within my own mind.)
I knew on the basis of experience, though, that it was far too simple to
condemn myself for my unconscious part in the enactment. In fact, if I
had succumbed to self-hatred over it, I would simply have been renewing
the enactment in a diἀerent form. My self-hatred would then have been
as rigid and single-minded as my narcissistic pleasure had been before.
In order to create the kind of experience within which I could set about
trying to negotiate the enactment, I needed to be able to create the joint
experience of two conflicting states: my guilt about failing the patient, and
my sense that I had done my best.
Over time, I worked my way to a position in which I regained the capac-
ity to feel, in my work with this particular patient, narcissistic pleasure
in the exercise of my analytic capacity. In the terms of the problem the
patient’s family did not solve, we might say that it took time for me to find
a way to respect my patient’s freedom without giving up the possibility of
acting like and feeling like a good parent. Neither his parents nor I had
been able to maintain the feeling that, however the patient worried us, we
were doing our level best, and that that was all we could do, even if it was
not good enough.
It took me time, in other words, to return to a tolerance of myself, to
an appreciation of the full measure of my experience. Perhaps we lose
and regain that tolerance and appreciation every time our analytic capac-
ity is compromised. Perhaps the analyst’s rediscovery of her tolerance of
herself is one way to describe the negotiation between the analyst’s self-
states. In this case, the beginning of a change was my development of a
sense that the atmosphere of the sessions was not what it had been. The
snags and chafing were hints of an unknown emotional presence, a stir-
ring. It required work, eἀort, to formulate a feel-able awareness of that
presence, one that was describable; and that awareness of what had been
absent, in concert with what I already felt and knew, finally came to con-
stitute a conflict. For a while, at the beginning of these events, the snags
and aἀective chafings were the only sign of more going on than met the
eye, the sole registration in my awareness of something that could even-
tually become a conflict. It was my clinical interest in those signs, my
everyday devotion to the clinical task, that eventually brought my own
dissociation to light and allowed me to experience a conflict where none
had been before.
I can now go back and answer a question I posed some time ago. If
enactments suck into themselves our capacity to observe them and to
know what to do about them, why does the same fate not befall the snags
and chafing that hint at their existence? If the eye cannot see itself, why
should it be able to see hints of itself?
Actually, most hints of this kind probably do escape our detection. But
we catch a few of them, and it is our unremarkable devotion to the analytic
task that allows those exceptions.1 Snags and chafing are the nascent signs
of a conflict we are in the midst of creating, a conflict that, even if we do
not theorize it that way, we know must be directly and consciously experi-
enced if the enactment is to be negotiated.
In my example, becoming able to formulate the state of mind I had
dissociated aἀorded me, for the first time, the opportunity to make a deci-
sion about whether to continue the enactment. And of course, once I had
the choice, it was obvious. An enactment loses its teeth as soon as one of
its participants can imagine an alternative to it. On the other hand, had
I not been able to breach my own dissociation and eventually experience
conflict, the enactment would simply have continued unabated until it
reached some kind of terminus of its own. Perhaps eventually, feeling let
down yet again by one to whom he turned for help, the patient would have
left treatment.
Pain and the Creation of Freedom
With clinical experience, analysts learn to value uncomfortable aἀective

cues. We value these nuances as alerts to moments of interchange that
deserve more thought and study, to experience that we need to feel and
know as precisely as we can, and that may eventually allow us to get a foot
in an experiential door that we cannot even clearly see yet. Although early
in our professional lives these strange promptings may often be merely
painful or disturbing, the same interactive hints later become for us inti-
mations of freedom. It is our capacity to be interested in these intuitions,
the devotion to our clinical work rooted in our desire for freedom—free-
dom for the patient, yes, but no less, freedom for ourselves—that moti-
vates us to tolerate the painful experience. Over the course of our careers,
previously uncomfortable reminders of the existence of what we do not
1 “Unremarkable devotion to the analytic task,” while it is accurate as far as it goes, is also a sani-
tized expression of what I have in mind. The work of an analyst can be demanding, after all, and
no one chooses to do something difficult every day without getting some kind of a charge out of it.
Psychoanalysts must be people for whom doing treatment stimulates the kinky and often slightly
naughty or guilty satisfaction that Lacan calls jouissance.
understand become precious. We come to value freedom more and secu-

rity less. Or perhaps, with experience, it just takes less to make us secure,
so that we can tolerate more easily our desire for freedom. In time, we
work toward what Symington (1983) calls “the act of freedom”—the ana-
lyst’s liberation from the previously unconscious grip of the field—less and
less by trying to satisfy the patient’s needs or by knowing the truth, more
and more by feeling our way into the experience we are having with the
patient and the patient with us. I am quite sure, for instance, that in the
first years of my practice, though I would have sensed that something was
wrong in the treatment I have used as an illustration, I would not have
had a very well-developed idea of what to do with this impression. I would
have felt it as a warning, not as an opportunity.
In the midst of a relational storm that might otherwise obscure it, the
analyst’s desire for the patient’s freedom and her own sometimes allows
her to see, understand, and accept the patient’s help.1 If Searles (1976) is
right, this help is not only moving but also mutative. We could say that the
analyst’s desire to cure the patient allows the analyst to accept the patient’s
desire to cure her.
On some level, in some unformulated, inarticulate way, did my patient
want me to notice the subtle change in the atmosphere between us? I sus-
pect that he did. He wanted me to love him enough to feel his subtle with-
drawal from me and to care about it without becoming anxious over it.
But was he also giving me my own chance for “cure” in the transference,
my chance to be a good “parent,” by sacrificing my narcissism and coming
through for him? Perhaps he was.
Reformulation
To this point, I may have made it seem that it is only one’s experience
of the dissociated self-state—the formulation of the experience that had
been unformulated—that changes as conflict is achieved. But that is not
the impression I want to leave. In the resolution of an enactment, one’s
experience of the safer state, the state that one had more or less comfort-
ably occupied in awareness, changes as well. The conscious and explicit
experience one had of oneself and the other during the enactment, that is,
also becomes something other than it was.
1 See Aron (1991, 1996); Blechner (1992); Gill (1982); Hoἀman (1998); Mitchell (1993, 1997); Searles
(1975, 1976); Singer (1971); Tauber (1954, 1979); Wolstein (1983); see also this book, Chapter 3.
As conflict appears and a dissociated self-state comes into aware-

ness, the newly formulated experience recontextualizes what had been
consciously and explicitly experienced before. That recontextualization
makes unavoidable the reformulation of that experience. Once the older,
consciously accessible experience shares awareness with an alternative
perception, it simply has to feel and seem diἀerent than it did before. Its
meaning has to change, at least a little bit, because its context has been
rearranged, it is situated diἀerently in the mind. As enactment ends, then,
neither the dissociated state nor the (previously) safer, explicitly experi-
enced state remains as it was. Each recontextualizes and participates in
redefining the other.
In my own case, my easy pleasure in doing the work with my patient,
first recontextualized by my dawning awareness that it was narcissis-
tic, suddenly seemed unacceptable. I felt it as a symptom of sorts. As
time passed and I regained my tolerance of myself, finding it possible
to experience guilt and pleasure simultaneously, even if it was not com-
fortable, my pleasure was recontextualized yet again, this time by my
growing sense that both my states—my pleasure and my guilt—were
understandable responses to the patient’s part of the enactment. And
so, though my experience of pleasure was explicit all along, it did not
remain the same pleasure: Its meaning and the way it felt to me changed
as my previously dissociated state become possible to formulate. What
began as naive pleasure first morphed into a symptom of my narcissism
and eventually took its place, in my mind, as part of the enactment
between the patient and me. The same kind of process went on regard-
ing my explicit experience of the patient: I began by accepting his mar-
velous progress at face value, feeling encouraged by it, then moved to
a position in which his progress appeared false and made me feel like
a vain failure and annoyed with him that he had taken me for a ride.
In the end, his show of accomplishment, like my pleasure in it, was
recontextualized in a way that made it more clinically productive and
tolerable for both of us.
Part III: Redefining the Riddle: Single-Mindedness and Internal Conflict
To conceive how the eye sees itself, then, we are drawn back to the origins
of psychoanalysis, to its roots in internal conflict. We must renew the hon-
ored place of conflict in our theories, but the internal conflict we need to
conceive is not conflict as Freud understood it: It is not the conflict between
drive and defense, or among the id, the ego, and the superego; and it is not
the conflict between consciousness and the unconscious. What we need
is a conception of consciously accessible conflict as personal and social
meaning, a conflict that goes on simultaneously within us and between us,
a conflict of purposes, interests, and desires.1
Furthermore, and even more crucially, we must reconsider the assump-
tion that every aspect of subjectivity is subtended by conflict. We need to
recast our thinking to reflect the view that, even in circumstances of great
emotional pain, internal conflict can be absent, and that its absence can be
the source of the pain, a problem that needs to be addressed by creating it
anew. The repetition compulsion, in other words, is not necessarily main-
tained by a rigidly enacted conflict between conscious and unconscious
aims, but by the absence of the conflict we need to be able to experience
if we are to sense the availability of choice. We need to take the point of
view, more than a little strange in the terms of traditional psychoanalysis,
that, in the case of dissociated self-states, conflict is not a given but a goal.
Without denying that every moment of life is conflicted, we must accept
that there are times when do not experience enough internal conflict, that
a significant part of the pain in human relatedness occurs because con-
flicts that might be actualized within us are not.
Conscious internal conflict is necessary because, if we are to back away
far enough from what is happening with the other to create the oppor-
tunity for reflection, for “seeing” the events in question, we need more
than one perspective. We need an alternative interpretation (an alternative
experience, I might rather say, though I mean the same thing by the two
words), and an alternative interpretation inevitably conflicts with the one
we already have. In the terms of dissociation, we can say that the sensing of
one’s own state of mind requires a second state of mind to serve as a back-
ground against which the first can become a figure. Of course, the first
also serves as background to the second, so whenever one becomes capa-
ble of sensing and reflecting on one state of mind, one is capable of sensing
and reflecting on two. Without an alternative perspective to set against
one’s previous single-mindedness, a new perception is simply impossible
to accomplish. And so the achievement of conflict and the articulation of
1 For example, Bromberg (1998, 2006); Davies (1996, 1997, 1998, 1999); Ehrenberg (1992); Hoἀman
(1998); Levenkron (2006); Mitchell (1993, 1997); Renik (1993a, 1993b); Slavin and Kriegman
(1998).
the meaning of an enactment are, in fact, the same event. The point can be
made from either direction: We can say that the alternative perspective of
a newly achieved conflict allows the formulation of the unformulated; or
we can say that the formulation of the unformulated is, in itself, the cre-
ation of the alternative perspective that subtends new conflict.
The creation of internal conflict is also the creation of a sense of initia-
tive. Desire in the absence of a conflicting alternative is nothing more than
compulsion, and compulsion negates the feeling that one is choosing one’s
own life. In deconstructing enactment one therefore escapes a certain kind
of psychic slavery. The fact that the motive organizing one’s enslavement is
frequently the attempt to dominate the other (Benjamin, 1990, 1999, 2000)
makes the enslavement no less constricting. In the stark, two-dimensional
world of enactment, the dominant take and keep power, but they lose their
freedom as surely as the oppressed do. In this sense, enactment as I am
portraying it—enactment based in dissociation—takes place along the
same lines as what Benjamin refers to as the reversible (“doer–done to”)
complementarity. Neither the patient nor the analyst is any more capable
of experiencing himself creatively than he is of recognizing the other. Both
participants fail to appreciate the fullness of both of their minds.
The most important outcome of a successful analysis is the firm and
unthinking conviction that one’s life is one’s own, that oneself and no one
else is living it. Frequently this feeling that one’s life is the creation of one’s
own mind—which in dryer terminology we can describe as the sense of
agency—arises from our access to the experience of conflict, because when
we are able to face the necessity for choosing the perspective we will take
on the problems that face us, we are able to feel our own hand on the til-
ler.1 In enactment, by contrast, experience either feels as if one is helpless
to shape or influence it (and one may desperately wish to do just that) or
as if it is being imposed by the other. Sometimes it just goes by unnoticed.
All these kinds of experience, but especially the sense of being forced, are
among the factors most responsible for feelings we all often have in enact-
ments, feelings of powerlessness and lack of ownership of our own minds.
We feel enslaved; we feel made to live this way, and we feel as if we cannot
1 It bears repeating, though, that the perspectives from which we choose are not constructed on
a merely conscious basis. The availability of perspectives is a matter of intentionality far deeper
than conscious decision making. The range of interpretations (experiences) we allow ourselves is a
function of a curiosity that goes beyond what we can decide to be interested in. I want to avoid any
implication that agency is only a matter of the growth in our capacity to make conscious choices:
Our sense of agency also arises from our perception of our freedom to experience, a perception
that is often created, ironically enough, by our surprise at what comes to us unbidden (see D. B.
Stern, 1991, 1997, 2002a).
help it, as much as we know that that is not so. Dissociation prevents that
precious feeling of fully inhabiting life that Winnicott (1960) describes as
true self, “the sense of being real” (p. 149).
I began this chapter by setting myself the insoluble riddle of how the
eye sees itself. I then promised to reformulate the question in a way that
could be answered, and I have done that. It turns out that knowing the
countertransference seems impossible only as long we are thinking from
the position of single-mindedness. When we are able to create only one
state of mind, it seems and feels as if, to observe itself, that mind must
somehow twist around and think itself from an impossible elsewhere—
the bootstrapping problem. In the achievement of conflict, we create an
alternative to the rigid perceptions that have locked us into the status quo;
we create multiple consciousness. As single-mindedness is replaced by two
or more internal positions, one part of us becomes capable of observing
another part without having to perform metaphysical contortions; the
mind can feel and reflect on what it could only live out blindly before.
And so we do not have to solve the problem of the eye seeing itself after
all. What seemed the impossible dilemma of countertransference aware-
ness becomes, instead, the thinkable problem of how we transcend single-
mindedness by creating the experience of conflict.
When Does Freedom Happen?
But transcending single-mindedness is more easily said than done. We

always hope that thinking new thoughts makes us more available for new
experience. I certainly have that hope for the ideas I have introduced here.
But as much as new ideas may help to prepare the ground for new experi-
ence, no idea, by itself, creates for us new perceptions of the patient as we
are sitting there needing those new perceptions (whether we know it or not)
so very badly. The problem of clinical freedom has no generic solution.
That is perhaps not so much a bane as a boon. If we believed we knew
just what ought to be done, as analysts probably once did feel more often
than they do now, if we still had a noncontextual theory of technique,
psychoanalysis would eventually become nothing more than practicing
the right way—one more instrumental treatment, boilerplate dictated by
DSM. Despite the annoyance, frustration, rage, envy, sorrow, guilt, shame,
and humiliation that we and our patients feel during enactments, it is a
reminder of the mystery of living, and thus of human dignity, that free-
dom from the grip of the field is beyond our capacity to create by an act
of will.
Experience is sufficiently manifold and complex that we find ourselves
simply living out most of it as it comes to us. In the overall scheme of
things, actually, our capacity and inclination to reflect on experience are
the exceptions. And it is not just the unconscious and the dynamics of
relatedness that stand in our way of understanding our involvements with
others. If we wanted to tot up the difficulties, we would also have to factor
in (though of course that is exactly what we cannot do) the (nonrelational)
contingency of existence, Lacan’s register of the Real, so inconvenient for
theories, wreaking havoc with our intentions to create order and regular-
ity. What will happen next? Will the treatment be interrupted by accident,
illness, financial reversal? One or both of the participants might even die
before they finish their work together. It is simply impossible to predict
that, when the knob on the door of my office comes oἀ in my hand as I am
trying to let my patient out of the room, she finally and suddenly grasps
that I am human. Nor do I have a snowball’s chance in hell of understand-
ing why this was the event to do the trick. Why could it only be now?
Why the doorknob, for heaven’s sake? Why didn’t the same thing happen
when I simply could not wait and had to interrupt the session to go to
the bathroom? ἀ at was pretty undeniably human, was it not? There did
not seem to be anything essentially diἀerent going on between us then,
and the symbolic meanings my patient and I can conjure for the episodes,
hard as we have tried, do not enlighten us in this respect. In the end, who
knows? John Lennon’s most famous aphorism comes to mind: Life is what
happens while you are making plans. Breaking the grip of the field can be
what happens while you think you are up to something else altogether.
Why do we suddenly manage to feel aἀective snags and chafing now?
Why these snags and chafing and not all those others we must be missing
along the way? Why are we able to accept the patient’s aid today when we
could not yesterday? Sometimes it is possible, either at the time or after
the fact, to understand why it was this one moment, or these 2 days, or 3
weeks, that allowed the achievement of conflict. But often it is not possible
to know why the grip of the field relaxes when it does. All we can do to
create the outcomes we desire is to prepare for them; all we can do is to
try to open ourselves to an awareness of whatever aἀective clues drift our
way. But why it is that our preparation, our work, and our best intentions
are only sometimes sufficient remains a mystery. We immerse ourselves
in the writings of psychoanalysis and in the study of our own experi-
ence and that of our analysands, and we practice over the years with the
balance of discipline and innovation that seems right to us. In some hours
we transcend single-mindedness: The result is good. In other hours we
are as mired in enactment as the session ends as we were at the begin-
ning. It is our accomplishment of freedom that makes an hour good, but
often enough, as long as we are working to capacity with a deeply involved
analysand, we do not really know why freedom comes to us when it does.
5
Partners in Thought
A Clinical Process Theory of Narrative
The Diary of a Castaway
I had been thinking about the problem of narrative in psychoanalysis

for some time when I ran across a television screening of ἀ e Incredible
Shrinking Man, a B movie scripted by the science fiction writer Richard
Matheson and released in 1957. I had not seen the film for almost 40 years,
but I remembered it fondly enough to see if it held up. I imagined that
immersing myself again in the atmosphere of one of those awful, inno-
cent 1950s science fiction movies would be nostalgic. Unexpectedly, it was
a good deal more than that, and not only because the movie was better
than I understood when I first saw it. I had long felt that new narrative in
psychoanalysis is not simply the outcome of the analyst’s objective inter-
pretation, as Schafer (1983, 1992) and Spence (1982, 1987) portrayed it,
but is instead the unbidden outcome of unconscious aspects of clinical
process. Oddly enough, by helping to direct and cohere my thoughts on
this point, ἀ e Incredible Shrinking Man jump-started the Interpersonal/
Relational psychoanalytic understanding of narrative construction I oἀer
in this chapter. The understanding I present, though, is based not only in
a certain kind of theory; it is also rooted in a personal sense of clinical
process. And so the tone I have adopted is personal as well. This chapter
should probably be read as a statement of convictions; but I maintain the
hope that my convictions will resonate with the reader’s own.
The plot of ἀ e Incredible Shrinking Man rests on an absurd 10-second
encounter between a man on his boat and a small radioactive cloud that just
happens to be drifting aimlessly around the ocean. During the moments
it takes the cloud to approach the boat, envelop it, and pass beyond, the
107
man’s wife is inside the cabin, fetching bottles of beer. She returns to find
that her husband’s chest is speckled with some kind of glitter. In the cra-
zily concrete way of these films, the wife, having been inside, is spared the
glitter, along with the later eἀects the glitter will have on her husband.
The man brushes oἀ the sparkly stuἀ and mumbles something to his wife
about some strange fog. There is no further discussion of the matter, but
the typically weird music accompanying this inexplicable moment certi-
fies that something mysterious and sinister has come to pass.
That is the set-up for the rest of the movie, in which our hero learns
that he is shrinking. Wrenching losses ensue, one after another, until
finally, when he has become so small that he lives in a dollhouse, his wife,
still dutiful, and a gargantuan in his newly proportioned world, bids him
good-bye one day, goes out of the house, and accidentally lets in the cat.
In a terrifying scene, the cat wrecks the dollhouse trying to get at the tiny
man who, in escaping, manages to shoulder open the door to the base-
ment, but then falls oἀ the side of the steps a full floor down into a basket
of laundry. No one knows he is there. In fact, his wife and the rest of the
world, to all of whom he has become famous as the Incredible Shrinking
Man, believe that the cat got him.
It is only then that the movie hits its stride. It turns out that its improbable
beginnings have been nothing more than a means of entry for Matheson,
the scriptwriter, who really wants to tell a Robinson Crusoe story. And
a great story it is. It is the story of a tiny, abandoned man, thought to be
dead, marooned in his own basement with no chance of rescue, horri-
bly alone, living in a matchbox, climbing ordinary stairs, each one now
turned into a towering cliἀ, with equipment fashioned from the materials
he finds, feeding on cheese left to catch mice, having to invent ways to
cross chasms that are nothing more than the mouths of empty cardboard
boxes, prey to a monstrous spider he fights with a needle he has found
in a discarded pincushion, and threatened by a flash flood from a leaky
boiler. In the end, after these compelling and strangely moving adven-
tures, intricately imagined, and filmed with a notable attention to detail,
and special eἀects surprisingly good for the era, the man becomes so small
that he can finally escape into his own backyard through the screen mesh
covering the basement window. He is now too small for us to see, but we
know he is there. We imagine him standing in a forest of towering grass
blades, shrinking to nothing, as he oἀers us the final lines of his tale. At
the very end, in the moments before he winks out altogether, the camera
pans upward and the tiny hero, gazing at the star-filled heavens, thinks
that the infinitesimal and the infinite are much closer to one another than
Partners in Thought 109
he has imagined before. Surprisingly enough, it is a moment of serenity,

acceptance, and dignity. After the trauma, humiliations, and cynicism he
has suἀered in the first months of his disease, he has not only returned to
himself, he has transcended what has befallen him. It is hardly routine for
survivors of trauma to find their suἀering a provocation to grow, even if
they manage to accept and live with their experiences; and of course the
story of the shrinking man is a fiction, and a fantastical one at that. But
this fictional little man has grown.
For the most part, the hero tells his own story. Yet during the first part
of the film we have no explanation for why we are privy to the tale. We
eventually find out that the tale is actually a diary of the events it depicts,
written by the hero himself. In the course of his adventures, just prior to
the episode with the cat and the dollhouse, the hero, cynical and miserable
to the point of desperation, begins to write. This is the line in the movie
that made me sit up and take notice: “I was telling the world about my
life,” the shrinking man reads to us from his diary, “and with the telling
it became easier.”
It does not require specialized training or experience to recognize the
truth in this simple statement. If there is mystery here, it is mystery we are
so used to living with that it does not surprise us. The fact that narrative
plays a natural role in creating a meaningful life in even a B science fiction
movie puts us on firm ground in agreeing with those many writers and
scholars who tell us that we shape personal meaning by organizing our
experience into meaningful, sequential episodes.1
But the intuitively obvious is not enough. What does the diary actually
do for the shrinking man? Why does it help him to tell his story? How does
it help him?
The narrative of the strange events of the shrinking man’s life sup-
plies him with a coherent and felt experiential order that he has lost in the
rush of bizarre happenings. Prior to constructing his tale in the explicit
terms of his diary, he has become an object in his own life, a figure suf-
fering chaotic, incomprehensible events for no apparent reason and with
little feeling. The emergence of meaning from what has felt to him like
senselessness, helplessness, and despair confers agency and therefore dig-
nity. He is once again a subject. After his fall into the laundry basket, the
tiny man creates his experiential world, his story of the obstacles he faces
and either accepts or overcomes, in such a way that his end has authentic
1 See Bruner (1986, 1990, 2002); Ferro (1999, 2002, 2005, 2006); Polkinghorne (1988); Ricoeur (1977,
1981); Sarbin (1986); Schafer (1983, 1992); Spence (1982).
pathos. After months of a growing sense of chaos and nihilism, he ends his
life a deeply thoughtful and aἀectively alive human being.
In creating his diary, the shrinking man also creates a relationship with
imaginary others who then serve as witnesses of what he “tells” them.
The movie grips us, despite its flaws, partly because we recognize at some
level the help that this witnessing oἀers him: we ourselves become his wit-
nesses. I turn now to a perspective on what it means to have and to be a
witness. I will return to the case of the shrinking man once these ideas are
in place.
Witnessing
We first learned about the significance of witnessing from studies of

trauma, in which witnessing of some sort is usually considered an essen-
tial prerequisite to the capacity to narrate one’s own experience. I believe
that the need for witnessing became visible first in this context because
it was in the impact of trauma that some of the most damaging eἀects of
the absence of the witness were first observed: Without a witness, trauma
must be dissociated; and once the isolated trauma suἀerer gains a witness,
the experience of the trauma becomes more possible to know, feel, and
think about (e.g., Boulanger, 2007; Brison, 2002; Laub, 1992a, 1992b, 2005;
Laub & Auerhahn, 1989; Richman, 2006). I will be discussing witnessing
here as a routine part of everyday, nontraumatic experience that I believe
begins in the earliest stages of development.
In fact, although Fonagy et al. (2002) do not use the language I am
using here, what they tell us about the beginnings of the self can be read
as the proposition that the witness precedes us. As they put it, “we fathom
ourselves through others” (p. 2). Caretakers identify certain feelings and
desires in the infant and treat the infant accordingly. This treatment begins
to organize the infant’s relatively inchoate world in the terms of narrative,
and self-states begin to cohere in and around these earliest stories. In one
sense, then, we are called into being by acts of recognition by the other. We
learn we are hungry because the other feeds us at a moment when we are
having a certain uncomfortable feeling; and so we then have a story that
goes with that feeling: “I am hungry.” We learn we are sad because the other
comforts us at a moment when we are having a diἀerent distressing feel-
ing; and so we then begin to have a story that goes with that feeling: “I am
sad.” This is one way we begin to tell and live stories; there are other ways.
All the various tributaries to narrative sum to the creation of experience:
Hungry is what you are when you need to be fed; sad is what you are when
you need to be comforted. As Harry Stack Sullivan (1940) writes over and
over again, we know ourselves via reflected appraisals. Fonagy and his co-
writers describe the same thing: “At the core of our selves is the represen-
tation of how we were seen” (p. 348); and “At the core of the mature child’s
self is the other at the moment of reflection” (p. 380).1
As development proceeds, we eventually gain the ability to formulate
our experience for ourselves, internalizing the capacity that first belonged
primarily to our caretakers. But we do not outgrow the need, paraphras-
ing Winnicott, to see our reflections in our mothers’ eyes; the need only
becomes more sophisticated. We may no longer need the other actually to
show us the meaning of our experience, as we did when we were infants;
but if we are to know our own experience in reflective terms, if we are to
be able not only to construct narratives, but to be aware of the narratives
we construct, we do need to believe that we are known by the other. We
need to feel that we exist in the other’s mind, and that our existence has
a kind of continuity in that mind; and we need to feel that the other in
whose mind we exist is emotionally responsive to us, that he or she cares
about what we experience and how we feel about it (Bach, 2006; Benjamin,
1988, 1990, 1995). This is what it is, I believe, to have a witness. Without
a witness, even an imaginary witness, events either fail to fall into the
meaningful pattern of episode that is narrative, or we merely enact our
stories blindly, unable to think about them or know what they feel like.
Our witness is our partner in thought.2
1 Fonagy and his collaborators and Sullivan are among the contributors to what has become an
extensive literature describing the structuring of the infant’s and young child’s world by the
relationship with the mother. Some of this literature falls under the rubric of mentalization
(Chasseguet-Smirgel, 1990; Fain & David, 1963; Fain, David & Marty, 1964; Green, 1975; Lecours
& Bouchard, 1997; Luquet, 1987; Marty, 1990, 1991; MacDougall, 1985). Other work grows from
an interest in the recognition of otherness (Benjamin, 1988, 1990; Bion, 1962, 1963; Eigen, 1981;
Lacan, 1977; Modell, 1984; Segal, 1957; Winnicott, 1971). A third relevant line of thought is rooted
in the study of mother–infant interaction and the growth of the interpersonal field (Beebe &
Lachmann, 1988, 1994; Sander, 1962, 1988; D. N. Stern, 1977, 1985; Sullivan, 1940/1953). All of
these branches of the literature are part of the context from which grows my interest in witness-
ing and its place at the roots of personality. Last in this list of citations, but certainly not least, is
Poland’s (2000) lovely and innovative paper on witnessing in psychoanalysis, in which witnessing
is contrasted with interpretation and is characterized as the activity by which otherness is recog-
nized. The influence of Poland’s paper is ubiquitous in this chapter.
2 It has been conclusively demonstrated (if the point actually needed to be demonstrated) that
thought and rationality should not be equated. Thought is creative and eἀective only when thor-
oughly imbued with feeling (e.g., Damasio, 1994). Although thought and feeling are inseparable in
this way, we do not have a single word that allows reference to both. Whenever I refer to “partners
in thought,” I mean to refer to both thought and feeling. The partnering I am describing is at least
as much an aἀective phenomenon as it is a cognitive one.
The witness, while it may feel like a single presence, may nevertheless be
composed of part(s) of one’s own mind or of the other’s, or of both simul-
taneously. The witness is the state(s) of self or other who one imagines is
best suited to fulfill the partnering purpose at the particular moment in
which the need arises. It is not a simple internalization of the historical
mothering one. An internalization of a loving parent who has grasped and
known one’s continuity is probably a necessary condition for the develop-
ment of the capacity to witness oneself, but it is not sufficient. The witness
begins as that kind of internalization but becomes a changing amalgam of
history, fantasy, and current reality. It is not a structure of the mind, but
a function, or better, a way of being. Its composition is limited by one’s
experience, of course, but within those limits the witness changes as con-
tinuously as the events witnessed; the particular selection of parts of one-
self or the other recruited to witness on any one occasion depends on that
occasion’s context. It is not only the witness who is in flux, however; the
one who is witnessed is, as well, since the state of self in need of witness-
ing also changes with context. However complex it may be to describe the
phenomenon in the third person, though, in phenomenological terms the
matter is simpler: The witness is the one imagined, consciously or sub rosa,
to be listening.
To have the ongoing sense that our story exists in someone else’s mind
(even if that someone else exists within our own mind) we must first (and
very often in imagination) continuously “tell” that other person what we
are experiencing. We construct what we know of ourselves by identify-
ing with the other and “listening” through his ears to the story we are
telling. We know our stories by telling them to ourselves, in other words;
but we can do that only by listening to ourselves through the other’s ears.
Psychoanalysts work in just this way: They listen to patients in the way
that allows patients to listen to themselves.1 To convince yourself of this
1 Some form of this point is widely recognized, probably by dozens of writers. Eshel (2004) describes
“I-dentification” as “the analyst’s thoroughgoing identificatory experiencing of the patient’s most
painful and terrifying experiences,” which “renders them tolerable, liveable, enables them ‘the
possibility of being’” (p. 331). Farber (1956) writes, “in listening we speak the other’s words. Or, to
put it another way, the analyst is able to hear only what he, potentially at least, is able to say” (p.
145). Laub (1992a) says, “The listener has to feel the victim’s victories, defeats and silences, know
them from within, so that they can assume the form of testimony” (p. 58). And finally, or rather
first, there are many passages of this sort from the work of Winnicott. This one is representa-
tive: “An example of unintegration phenomena is provided by the very common experience of the
patient who proceeds to give every detail of the week-end and feels contented at the end if every-
thing has been said, though the analyst feels that no analytic work has been done. Sometimes we
must interpret this as the patient’s need to be known in all his bits and pieces by one person, the
analyst. To be known means to feel integrated at least in the person of the analyst” (1945, p. 150).
point, just think about how often, during and after your own analysis,
you found yourself at odd times during your day imagining that you were
telling your analyst something. I remember when I first noticed it happen.
Sometime after that I realized how frequent these tellings were and how
often they went unattended.
This kind of telling and listening, though, arises much earlier in life than
the age at which people typically go into psychotherapy and psychoanaly-
sis. If you have children you remember overhearing them talk to themselves
in their cribs, often quite animatedly, after you put them to bed. They are
organizing their experience of the day, giving it sense. But to whom are they
talking? Not to “themselves,” at least not exactly in the sense in which “self”
will be a meaningful idea later in life. At this early age self and other are not
yet conscious and coherent parts of experience; neither self and other, for
instance, can be explicitly reflected on. Besides, why speak out loud if the
only audience is oneself? It is plausible to imagine that babies in their cribs
are talking to their first witnesses: their parents. But these are their inter-
nalized parents, or some of their first internalized objects. These children
are imaginatively listening to themselves through their parents’ ears and
thereby lending their experience a credence, coherence, and depth of feel-
ing it otherwise could not have (Nelson, 1989). As a matter of fact, what we
are hearing when we listen to babies creating coherence in those minutes
before sleep may very well be part of the process of self-formation.
The diary of the shrinking man, like what patients say to their analysts,
is an explicit kind of telling, with the diἀerence, of course, that the shrink-
ing man’s audience, like the audience listening to the little child in his crib,
is imaginary. Like the child, the shrinking man is writing to some figure
in his inner world. Imaginary audiences are very common. But explicit
telling is not. Most telling of the sort I am describing here, the kind of
telling that allows one to listen to one’s own thoughts, is implicit. It goes
on hazily, not very specifically, seldom noticed except, in a leftover from
our crib days, in the states that take place just before sleep in adult life or
at other times when we are alone, when we sometimes notice that we are
formulating our thoughts by addressing some ill-defined other. Most of
the time, though, it is as if we were telling, and as if we were being listened
to and then listening to ourselves. But the activity is no less crucial for
being hazy and imagined. In order for this process to come about in the
first place, probably we must be fortunate enough to have had parents who
left us able to believe, in at least some states, that there exist others, espe-
cially certain imaginary others, who are continuous presences interested
in knowing our experience (Bach, 2006; Benjamin, 1988, 1990, 1995).
When life feels arbitrary, senselessly cruel, or meaningless, as it did for

the shrinking man before he began his diary, one is liable to be aware of
no story at all. Events seem arbitrary and do not fall into narrative order.
Aἀect is flattened or diminished; one may consciously feel only a kind of
numbness or deadness. The living, hurt places in one’s mind—actually, the
injured parts of the self, the parts we most need to protect—despite their
influence on day-to-day life, go undiscovered until something happens in
ongoing relatedness that allows us to see that someone else recognizes the
pain we ourselves have been unable to know and feel. Our grasp of our own
previously dissociated experience through what we imagine to be the eyes
and ears of the other is synonymous with the creation of new meaning. As a
coherent narrative of the experience falls into place, there is an awakening,
including an awakening of pain. In fortunate cases, there is also relief. Both
pain and relief illuminate the absence of feeling in what came before.
That was the fate of the shrinking man. Until he began to tell his story,
he was losing courage by the day and becoming increasingly angry and
cynical. But once he began writing his diary, his imagined readers, who
“listened” to him “tell” his tale, seemed to help him contact his dissoci-
ated vitality and make it once again part of the mind he felt as “me.” That
change was enough to bring back his determination to face whatever was
in store for him. For now I merely note the following point: imagined wit-
nesses can be as eἀective as real ones.1
All right, I thought, the diary allowed the shrinking man to know his own
story. But so what? Why did the character even want to go on living? Why
didn’t the shrinking man kill himself, or at least think about it? Wouldn’t I
have thought about it if I were he? Was that omission a failing of the script?
The man may as well have been the last human being. He was permanently,
completely, hopelessly alone. Wouldn’t absolute, inescapable aloneness
inevitably lead to despair? Or did the screenwriter know something? Should
we consider the hero’s perseverance to be a consequence of the value that
telling someone his story of isolation brought back to his life?
For another take on the question, I turned to my copy of Robinson
Crusoe (Defoe, 1957), a story that gripped me as a boy, gripped me earlier
and even more deeply than the story of the shrinking man. The first part
of the book is a journal of Crusoe’s years living alone on a deserted island,
the sole survivor of a shipwreck. (Crusoe writes until he runs out of pre-
cious ink.) In the usual manner of diaries, the document is written as if
1 I must defer to the future an exploration of the significant diἀerences between imagined witnesses
and real ones and between the process of witnessing under these two sets of circumstances.
Crusoe were addressing someone, and you soon fall under the spell: It is
as if it is you to whom Crusoe is telling what happened to him. I remember
feeling an intimacy with Crusoe when I read the book the first time; I felt
I was there on that island, just as I felt I was there in that basement with
the shrinking man. It was one of the most thrilling reading adventures of
my childhood. I remember marveling that Crusoe could live so fully by
himself, and now, with the reminder supplied by my recent experience
with the shrinking man, I also remember feeling, even as a boy, that the
diary must have made Crusoe feel less alone.
By writing their diaries and being able to believe in the interest in their
experience held by those imaginary others to whom they wrote, Crusoe
and the shrinking man created partners in thought, imaginary others
with whom to share life. We all create partners in thought, all the time.
In most of life, though, real flesh-and-blood others are so ubiquitous, and
the stories of our lives fall together in such an unnoticed way that it is
much harder to appreciate both the significance of narrative and the role
of witnesses in its creation. The ongoing reciprocal process by which we
quite implicitly oἀer one another the reassurance that we understand well
enough to continue to serve as witnesses generally goes unnoticed, just
keeps on keeping on, like the Boston Change Process Study Group’s (2002,
2005, 2007, 2008; D. N. Stern et al., 1998) “implicit relational knowing,”
unless or until misattunement interrupts the flow and forces us to attend
to the break in our confidence in the other’s responsive emotional pres-
ence. The very isolation of Crusoe and the shrinking man oἀers us the
opportunity to grasp the role of their narrative creations in giving their
lives meaning, and the conception of the witness allows us to understand
why writing their diaries helped them as it did.
Although witnessing is mentioned often in the trauma literature, Sophia
Richman’s (2006) work on “transforming trauma into autobiographical
narrative” contributes observations with more pinpoint relevance to what
I am trying to say than others I have read. Remember what the shrinking
man said about his diary (“I was telling the world about my life, and with
the telling it became easier.”) and compare it to Richman, who tells us this
about autobiography and trauma: “By sharing the creation with the world,
there is an opportunity to come out of hiding, to find witnesses to what
had been suἀered alone, and to begin to overcome the sense of alienation
and isolation that are the legacy of trauma survivors” (p. 644). Richman
also quotes Joan Didion’s observation that writing can make experience
coherent and real. Didion made the remark during a television interview
with Charlie Rose in which she was talking about the memoir she wrote
about the death of her husband: “What helped me to survive was writ-
ing this book, because otherwise I wouldn’t have been able to understand
what I was going through” (p. 648). And finally, Richman agrees that the
witness may be imaginary. Here is what she writes about her father’s mem-
oir of life in a concentration camp: “I believe that in order to write what
he did, he had to conjure up a reader who had an interest in his story and
could function as his witness” (p. 646).
Narrative Freedom and Continuous Productive Unfolding
It is as true in the clinical situation as it is anywhere else that, by the time

our best stories are spoken, they just seem right, convincing generations of
psychoanalysts that it was the content of what they said to their patients—
that is, clinical interpretation—that was mutative. I share the view of those
who see the matter otherwise. The real work has already been done by
the time a new story falls into place.1 Because they and I are tackling the
same problem, I appreciate the work of the many writers who understand
the therapeutic action of interpretations as relational. Mitchell (1997), for
example, writes, “[I]nterpretations work, when they do, … [because] the
patient experiences them as something new and diἀerent, something not
encountered before” (p. 52). But that is not the position I am taking here.
I am arguing that the appearance in the treatment of mutually accepted
new content or newly organized content, which is generally narrative in
form, is not usually the instrument of change at all; it is rather the sign
that change has taken place. It is true that a new understanding is the
fulfillment of possibility; but it is to the creation of that possibility, not the
shape of its fulfillment, that we must look for the source of change. The
important thing about a new understanding—and this applies no mat-
ter whether it is the analyst or the patient who oἀers it—is less its novel
content than the new freedom revealed by its appearance in the analytic
space, a freedom to feel, relate, see, and say diἀerently than before. This is
the likely explanation for the widely recognized observation that former
analysands, even those who credit their treatments with saving or renew-
ing their lives, remember few of the interpretations their analysts made. It
was not the interpretations, per se, that helped, but the freedom that made
1 For example, Boston Change Process Study Group (2002, 2005, 2007, 2008); Bromberg (1998,
2006); Ghent (1995); Pizer (1998); Russell (1991); 2007, 2008; D. N. Stern (2004); D. N. Stern et al.
(1998). See also this volume, Chapters 3, 4, 6 and 8.
the interpretations possible in the first place. What is remembered from a

successful treatment, as a matter of fact, is much less the analyst’s words
or ideas than something about the appearance of that freedom, something
about what particular important moments felt like, something sensory,
perceptual, and aἀective. The new story, then, is not the engine of change
but the mark change leaves behind. Or perhaps this is better: The new story
does not create change, but shapes the way we represent it to ourselves.
But as much as I agree with that statement, it is also a bit of an overstate-
ment. In the attempt to acknowledge that claims for the mutative eἀects
of narrative interpretations have been overstated, we could find ourselves
throwing out the baby with the bathwater. We must admit that each new
story along the way is not only the mark of change, but also helps to pro-
voke the next round of curiosity, and thus to open new narrative freedom
and the stories that follow. Each new story is simultaneously what change
leaves behind and part of what brings about the next generation of clinical
events. In fact, we can say this in a stronger form: Each new story belongs
to the next generation of clinical events.1
And so when we observe that patients may not remember the events of
their treatments primarily in narrative terms, we must also acknowledge
that memory for narrative is not necessarily the best index of narrative’s
influence. The aἀective changes that take place in treatment, and that are
memorialized in the new narratives that fall into place there, are reflected
in our ways of remembering the past, creating the present, and imagining
the future. It is in these eἀects that we see the most profound influence of
new stories. Narratives are the architecture of experience, the ever chang-
ing structure that gives it form. Without narrative, aἀect would be chaotic
and rudderless, as shapeless as a collapsed tent; and without aἀect, narra-
tive would be dry and meaningless.
We see in new narrative freedom a deepened capacity of the patient and
the analyst to dwell in each other’s minds, to collaborate in the analytic
task, to serve as each other’s partners in thought. Any new understanding
in the clinical situation is testimony that these two people have become
better able to “tell” each other their stories and to “listen” to their own
1 This emphasis on the creation of new narrative freedom is not meant to suggest that either charac-
ter or any of the other kinds of continuity in the personality are unimportant. But looking at char-
acter in Relational terms does require us to conceive it as multiple, not singular. That is, character
must be defined in context: Under thus-and-such circumstances, a particular person’s conduct
and experience are liable to be defined in a particular way that is at least partly predictable. But we
cannot guess what anyone will do or experience if we do not know something about the nature of
the interpersonal field in which that person is participating.
tellings through the ears of the other. I mean “tell” and “listen” in the
special way that goes on in imagination and that depends both on being
able to believe that you have an unshakeable existence for the other and on
recognizing yourself in your imagination of the other’s picture of you.
The freedom to create a new narrative in the clinical situation, or to
find value in a new narrative that has been created by the other, is a spe-
cific instance of the general case of narrative freedom. Most of this new
grasp of things emerges without conscious eἀort, unbidden, like implicit
relational knowing, from the ongoing relatedness between patient and
analyst. As long as there is no obstruction of the capacity of each person
to serve as witness to the other, narrative freedom is the expectable state
of aἀairs, and the capacity of analyst and patient to reveal new experience
through an ever-renewed curiosity deepens over time as their intimacy
grows. There is a sense of continuous productive unfolding. Under these
conditions, there is a more or less uninterrupted flow of new aἀective expe-
rience and understanding for both patient and analyst. Old stories hove
into view, are destabilized, and dissolve; new stories fall into place. The
process is often smooth and pleasurable. This kind of clinical work goes
on much of the time with many patients, more often with some patients
than others. Although the process may be punctuated with minor diffi-
culties—hesitations, bumps, and snags—the overall nature of the work is
an ever richer and more thorough exploration and experience of the tol-
erable part of both the patient’s experience and the analyst’s. The analyst
generally feels (and is) valued, skilled, and useful, and the patient feels
helped. The analyst’s unconscious involvement with the patient is present,
but seldom problematic. It serves as a contribution, not an obstacle, allow-
ing the analyst to oἀer a diἀerent take on the patient’s experience from
the one the patient started with, a novel view that is generally experienced
as helpful by the patient. There is the satisfying sense of a job well done.
Continuous productive unfolding is, in the analyst’s mind, what Hoἀman
(1998) would refer to as the unconstricted interplay of ritual and spon-
taneity, what Knoblauch (2000) and Ringstrom (2001, 2007) would call
improvisation in therapeutic relatedness, and what Winnicott (1971), the
font of such thinking, would call play.
Not-Me
This relatively smooth and productive clinical process lasts as long as expe-
rience feels tolerable. But a very diἀerent, more troubling, and sometimes
even destructive kind of relatedness takes place when the experience

evoked in the mind of either patient or analyst, or of both, is not tolera-
ble—that is, when the state that threatens to emerge into the foreground
and shape consciousness is not recognizable as oneself. Such a state of
being is not-me (Bromberg, 1998, 2006; D. B. Stern, 2003, 2004; Sullivan,
1953/1940), and in ordinary life it exists only in dissociation, apart from
what feels like me.1 Not-me has never had access to consciousness, and
in its dissociated state it has never been symbolized: It is unformulated
(D. B. Stern, 1997), a vaguely defined organization of experience, a primi-
tive, global, nonideational aἀective state. It does not exist within the self
because it has never been allowed to congeal into one of self’s states. We
can say it this way: Not-me would be a self-state if it were to move into the
foreground of experience. But if that were to happen, not-me would not feel
like me. The experience would be intolerable; and so not-me remains dis-
sociated. I must not, cannot be not-me. The threatened eruption of not-me
into awareness jeopardizes my sense of being the person I am. In both my
own work on dissociation and the work of Philip Bromberg (1998, 2006),
not-me has never been formulated; dissociated experience has that quality
in common with conceptions such as Bion’s (1962, 1963) beta functioning
and beta elements and Green’s (2000) nonrepresentation.2
Not-me originates as a response to unbearable fear or humiliation, the
experience of having been the object of a powerful other’s sadism. It is
the sense that one is once again that stricken person: terrorized and ter-
rified, sometimes to the point of immobility or helpless, destructive rage;
contemptible, sometimes to the point of a self-loathing that yearns for the
destruction of self or other; shamed and horrified, sometimes to the point
of losing the desire to live or creating the desire to kill; weak, sometimes to
the point of a shameful and utterly helpless surrender that feels as if it can
1 “Me” and “not-me” are ideas more substantial than their colloquial names might suggest to those
unfamiliar with their long history in the literature of Interpersonal psychoanalysis. The terms
were devised by Harry Stack Sullivan (1940) as a means of representing the parts of the personal-
ity that exist within the boundaries of what is accepted as self (“me”) and what is dissociated from
self (“not-me”). The contemporary literature of dissociation, primarily the last twenty years of
work by Philip Bromberg (collected in volumes published in 1998 and 2006; see also Chefetz and
Bromberg, 2004), has lent the ideas new life. Recently they have also played a central role in my
own thinking. See also this book, Chapters 3, 4, 6–8.
2 As in the case of beta elements and nonrepresentation, dissociated material cannot be addressed
by traditional defensive operations because the dissociated has not attained symbolic form:
“Unformulated material is experience which has never been articulated clearly enough to allow
application of the traditional defensive operations. One can forget or distort only those experi-
ences which are formed with a certain degree of clarity in the first place. The unformulated has not
yet reached the level of diἀerentiation at which terms like memory and distortion are meaningful”
(D. B. Stern, 1983, p. 73).
be prevented only by suicide or held at bay only by committing mayhem.

One will not, cannot be this person, because when one was, life was not
bearable; and yet, if not-me enters consciousness, one is that person.
Every personality harbors not-me, although of course the degree of
trauma that has been suἀered by diἀerent people varies enormously. The
impact it would have for not-me to emerge into awareness and become “real”
depends on the severity of trauma and the consequent degree to which not-
me is vicious, loathsome, terrifying, terrified, or abject, and the degree to
which the whole personality is unstable and vulnerable. For those who have
suἀered severe trauma and whose vulnerability is therefore unmanageable,
the eruption of not-me can be catastrophic, provoking massive aἀective
dysregulation or psychotic decompensation. For those who are less trou-
bled, the consequence is nevertheless awful enough to be avoided.
Enactment: An Illustration
When not-me is evoked by the events of clinical process, continuous

unfolding is replaced by some variety of enactment. In the following
example, for heuristic purposes I describe more about both my own expe-
rience and the patient’s than I knew as the interaction was taking place.
My patient was late, and I was taking advantage of the extra minutes to
have a snack. When the patient arrived, I was enjoying what I was eating
and wanted to finish it, and it therefore took me a few seconds longer to
get to the waiting room than it would have if I had simply been waiting
in my chair. The patient was standing there, waiting for me, when I got to
the waiting room. He had not sat down, which I took as a sign that he was
eager to come in. Perhaps I should not have allowed myself my little delay. I
was faced with a small incident of my selfishness. In a defensive attempt to
avoid self-criticism (an insight available only in retrospect), I said implic-
itly to myself, without words, “Well, for heaven’s sake, the patient was late.
What’s wrong with using the time as I see fit?” But I was aware of greeting
the patient without my customary warmth.
The patient, because of his relationship with his demanding and easily
disappointed father, has an intense vulnerability to humiliation. The expe-
rience of being snubbed (my lackluster reception) made him worry (sub
rosa) that he was a burden or a disappointment to me, thereby threatening
the eruption into awareness of not-me. In the patient’s mind, my greeting
confirmed what he feared: My contempt was leaking; I had tolerated him
up to now only because he paid me. The secret was out. He had always had
to contend with the danger of being a loathsome, contemptible boy, and he

must not, could not, be that boy.
What happens at such a juncture? My patient had to do whatever was
possible to avoid the eruption of contemptible not-me in awareness. His
usual defensive maneuvers were of no use now; the danger was upon him.
In our prior work, I had been quite careful to respect his vulnerabilities,
but I had momentarily failed in that respect in greeting him as I had. In the
past, the patient had also defended himself by (unconsciously) influencing
the relatedness with me, making sure never to disappoint or provoke me,
and thereby avoiding any possibility of facing this kind of stark “evidence”
of my contempt for him. But his usual ways could not help him now.
The last-ditch defense, when not-me is imminent, is the interperson-
alization of the dissociation, or enactment: “I am not contemptible, you
are contemptible.” The patient now claimed that most of the time, when
I seemed authentically interested, I had been merely pretending. I hadn’t
really cared—that was now clear for the first time. Other therapists didn’t
pretend as I had; they really did get to know and care about their patients.
My patient began to cite moments from the past that he now believed lent
credence to the interpretation that I just was not very good at my job, that
I should have chosen a field in which my limitations would not have hurt
those I served.
I struggled with my aἀective response to being the object of contempt,
feeling unhappy, hurt, and on the verge of anger. I was feeling the very shame
that the patient was so eager to avoid. But I was nowhere near such under-
standing at this moment, and I said something (I don’t remember exactly
what it was) that protested my innocence. I knew that I sounded defensive.
This situation could have moved in either of two directions at this point.
In one scenario, I come to terms with my own aἀective reaction to the
patient and tolerate it. Under those circumstances, following my defensive
reaction, I would grope toward a therapeutically facilitative response to
the patient, although such a response probably would not occur immedi-
ately after the patient’s provocation, because anyone’s initial response to
an accusation is likely to be defensive. This is actually what happened in
this case, and I will tell that part of the story just below. But it is also com-
mon in this kind of situation, when the analyst is seriously threatened, for
the patient’s enactment of a dissociated state to call out a dissociated, or
not-me, state in the analyst. A mutual enactment ensues. With my patient,
such a scenario might have looked like this: In the same way the patient
has begun to feel that it is not he who is contemptible but I, I now succumb
(even if I “know better,” which of course I usually do) to the strongly felt
sense that I am not doing anything problematic—it’s just that the patient is
impossibly sadistic. I will almost undoubtedly feel uncomfortable in this
position, probably guilty about being a bad analyst, but I will see no way
out of it for the time being. Mutual enactments, which are not as uncom-
mon as the traditional psychoanalytic literature might be read to suggest,
may go on over significant periods and often pose a genuine threat to the
treatment (D. B. Stern, 2003, 2004, 2008).
Enactment, Witnessing, and Narrative
Thinking in narrative terms reveals that enactment of either kind—that

is, either with or without the dissociative participation of the analyst—
is even more than the unconsciously motivated inability of the patient
or analyst to see each other clearly and fully. As enactment rigidifies the
clinical relatedness, it also interrupts each person’s capacity to serve as
witness for the other. Even if the analyst does not respond with a recipro-
cal dissociation, in other words, the patient loses, at least temporarily, the
capacity to allow the analyst to be his partner in thought. The patient also
temporarily loses the desire, and probably the capacity, to be the analyst’s
partner. When the analyst does respond with a reciprocal dissociation, of
course, the situation is both more troubled and more difficult to remedy.
In either case, the eἀortless, unbidden creation of narrative that went on
during continuous productive unfolding now grinds to a halt.
One way to define states of self is as narratives: Each state is an ever-
changing story. Or rather, as I have already suggested, because self-states
are not simply experiences or memories, but aspects of identity, each state
is an aspect of self defined by the stories that can be told from within it.
Our freedom to tell many self-stories at once—in other words, our free-
dom to inhabit multiple states of being simultaneously—is what gives to
the stories that express the ways we know ourselves and others the plastic-
ity to change with circumstances. The many states that compose “me” not
only participate in shaping the circumstances of my life, but are, in the
process, themselves reshaped. This continuous interchange and renewal is
the hallmark of the self-states that make up “me.”
But not-me cannot be told. Not-me remains insistently, stubbornly,
defensively unformulated, not yet shaped or storied at all, isolated, exist-
ing in dissociation and thereby rendered mute. This is the situation within
enactments, both solitary and mutual: neither analyst nor patient knows
how to narrate the significance of what is transpiring; neither knows the
meaning of the transaction nor the feelings and perceptions that make it
up. And so those events remain coded only in procedural terms, in action.
If not-me is to come within our capacity to tell, then me, the self of the dis-
sociator, must somehow expand to accommodate or contact it.
I continue now with the events that actually took place with my patient.
I felt defensive and ashamed in reaction to the patient’s accusations. My
defensiveness was apparent to me and, I told myself, probably to my
patient; but I did not respond with a reciprocal dissociation of my own. I
pulled myself together and said something on this order: “I was taken by
surprise by what you said [the patient’s accusations against me]—I didn’t
know where that was coming from. But now I’m asking myself if the way
you felt might have to do with something you sensed during the last ses-
sion, or when you came in today. Did you notice something I said or did?
Because I did. This may not be the important thing, but I did notice that
I didn’t greet you as I usually do.” Despite my reaction to the patient’s
accusations, in other words, I was able to consider the possibility that I
might have played a role in setting the patient’s complaints in motion. In
this context, at least, I was able to conduct an inquiry without succumb-
ing to an answering dissociation and enactment. I did not shut down the
narrative possibilities, in other words, as the patient had no choice but to
do from within his own dissociative process, but instead returned to being
curious, relatively open to whatever emerged in my mind.
Neither the patient’s dissociation nor his enactment was particularly
rigid as these things go, although the situation certainly might have moved
in that direction if I had failed to gain a perspective on my own reaction
and remained defensive. But I was fortunate in this case, because, sensing
that I was no longer threatened, the patient showed some interest in my
foray. But he was still suspicious, and he said, “Well, but then why did you
get defensive?” referring to what I said in response to his accusations. I
answered, again from within my relocated stability, that I did believe that I
had been defensive, and that it is often hard for anyone not to be defensive
in the face of strong criticism.
The patient softened and (to my surprise, to tell the truth) seemed to
begin to search himself for something that might be responsive to what I
had said. He eventually was able to say that my greeting had indeed stung
him. The atmosphere cooled further. The patient had little difficulty now
in seeing that my defensiveness could be understood, from within my
perspective, as a response to his own critical remarks. More important,
the patient had now lived through an episode in which his brief certainty
that he was a burden to me, and that my caring was inauthentic, was
disconfirmed. This was not primarily a cognitive signification for him. The
patient could feel or sense what it was like for me to be with him through
the course of his accusations. That was important; but more important
yet was that the patient felt for one of the first times the confidence that
I had felt hurt or angry with him without losing track of my warm feel-
ings about him (or losing track of them only very temporarily). In a small
but crucial way, the patient was now someone other than he had been.
Over the following months, other new experiences of this kind opened in
front of him because his growing confidence in my openness to his experi-
ence and my own made it possible for him to begin to listen imaginatively
through my ears to his own feelings of being a burden; and in the pro-
cess, those experiences gained substance and reality for him, on the one
hand, and became less shameful and more bearable, on the other. Stories
about these things emerged in his mind with increasing frequency, some
of them articulate and others implicit. Over time, not-me became me. For
my part, through my experience of my reaction to his stinging criticism, I
also became more able to witness the patient; and beyond that, I came to
depend in a new way on the patient’s capacity to witness me—the way, for
example, he eventually accepted my reactions to his criticisms.
Dissociations are not breached by insight, nor are enactments dissolved
through verbal understanding. Interpretation is not the analyst’s key inter-
vention. Enactments end as a result of a change in aἀect and relatedness,
which provokes a change in each participant’s perceptions (and stories)
of the other and himself (See Chapters 3, 4, 6, and 8). Insight into this
changed state of aἀairs, when it plays a role, comes later. Historical recon-
struction often does take place after the appearance of the new story, and
it can be quite helpful. But therapeutic action lies in becoming a diἀerent
person, usually in a small way, in the here and now. The expansion of the
self takes place in the present, in small increments. As enactment recedes,
the treatment moves back into continuous productive unfolding, and new
narratives once again begin to appear unbidden in the analytic space. The
new stories my patient and I have told as the treatment has moved on have
been more and more often about the contemptible little boy.
Returning to the Castaways
But if the analyst is so crucial to the patient, how do we understand

Robinson Crusoe and the shrinking man? They had no analytic relation-
ship, no relationships of any kind. (Crusoe did eventually have Friday, but
that was years into his saga.) Now it may be clearer why I claimed earlier
that enforced isolation makes these characters such good illustrations of
my thesis. Their creators’ suggestion that the characters grew and changed
despite their circumstances is not mistaken, nor is it by any means a refu-
tation of the point that we are profoundly social beings. On the contrary,
such stories could not demonstrate the necessity of witnessing more clearly
than they do. It seems likely, actually, that some kind of imaginary witness
is invoked in all tales of enforced isolation, real and imaginary.
In the movie Cast Away, the character played by Tom Hanks, alone and
shipwrecked on an island, finds a soccer ball floating in the surf, paints a
face on it, and begins to talk to it, using the conversation as a kind of ironic
commentary to himself on the matter of his own loneliness. He calls the
ball “Wilson,” after the name of the sporting goods company that made it.
But as the years pass, irony turns delusionally earnest, and Wilson eventu-
ally becomes the castaway’s dear friend, continuous companion, and con-
fidant. Years after that, the shipwrecked man escapes from the island on
board a raft he has made himself. In the calm that comes after a storm at
sea, and dying of thirst and exposure, he sees that Wilson, whom he had
tethered to the mast for protection, has fallen oἀ and is drifting away from
him across the swells. The movie’s one truly devastating moment comes
when the castaway sees that in his weakened state he cannot rescue his
“friend” without losing the raft and drowning, and he calls out piteously
after the swiftly disappearing Wilson, pleading for forgiveness.
Let me oἀer one last example, just to put it on the record that factuality
reflects castaways’ need for a witness just as well as fiction does. I recently
read a dreadful story in the New York Times (Onishi, 2007) about a man
in Tokyo so poor that he had not eaten in weeks and so alone that no one
either knew or cared. In his last days he kept a diary. Among the last entries
before his death from starvation was his expression of the wish for a rice
ball, a snack sold in convenience stores across Japan for about a dollar: “3
a.m. This human being hasn’t eaten in 10 days but is still alive. I want to
eat rice. I want to eat a rice ball.” The very fact that the diarist wrote at all
testifies to his imagination of an audience. But note also that he speaks of
himself in the third person. Is it credible that he would have done that if
he really imagined that he was addressing only himself? Could there be
a more eloquent expression of the need to listen through the ears of the
other? It was preserved even as this man was dying.
To know what our experience is, to think and feel, we need to tell the
stories of our lives, and we need to tell them to someone to whom they
matter, listening to ourselves as we do the telling. If we have to make up
our audience, so be it. Our need for a witness goes so deep that imaginary
witnesses must sometimes suffice.
Witnessing One’s Self
We are familiar with the idea of internal conversation between parts of

ourselves. If we can hold an internal conversation, can one part of our-
selves serve as a witness for another? We have seen that Richman (2006)
believes so. Laub (1992a, 1992b, 2005; Laub & Auerhahn, 1989) does too.
He suggests that massive psychic trauma, because it damages the pro-
cesses of association, symbolization, and narrative formation, also leads to
an absence of inner dialogue, curiosity, reflection, and self-reflection. And
what does Laub believe is responsible for this inner devastation? He says
it is the annihilation of the internal good object, the “internal empathic
other” (Laub & Auerhahn, 1989), partner in inner dialogue and narrative
construction. Laub (1992b) tells the story of Menachem S., a castaway of
sorts, a little boy placed in a labor camp who somehow managed to sur-
vive the Holocaust and, miraculously, to find his parents afterward. He
had spent the war talking and praying to a photograph of his mother that
he kept with him. “Mother indeed had promised to come and take him
back after the war, and not for a moment did he doubt that promise” (p.
87). But the mother and father he refound, “haggard and emaciated, in
striped uniforms, with teeth hanging loose in their gums” (p. 88), were
not the parents he had maintained in his memory. Mother was “diἀerent,
disfigured, not identical to herself” (p. 91). Having survived the war, the
boy now fell apart. Laub writes, “I read this story to mean that in regain-
ing his real mother, he inevitably loses the internal witness he had found
in her image” (p. 88).
Richman’s (2006) experience is once again germane. Here she describes the
inner presence to whom she wrote during the time she was working on her
own memoir (2002) of her childhood as a hidden child during the Holocaust:
The internalized other (the projected reader) was an amorphous presence with-
out distinguishing characteristics, but seemed to be an interested observer, a
witness, someone who wanted to know more about me and my life. Perhaps the
amorphous presence represented my mother, my first reader-listener, who lived
to hear my school papers and received my writing with unwavering admiration.
(p. 645)
Something on this order is what happened for the castaways I have cited,
for the toddlers in their cribs, for all of us, much of the time, day to day.
And so we see that the experience of the castaways is hardly unique; it is
what we all do routinely. It is the castaways’ enforced isolation, as a matter
of fact, that throws the process of witnessing into high relief.
But just as Laub’s internal empathic other can be destroyed by trauma,
we cease to be able to invoke the imaginary internal witness as soon as
the experience we must witness touches on parts of us that hurt or scare
us too badly to acknowledge, or that are injured in a way so central to our
makeup that awareness of them threatens the remainder of the personal-
ity. The imaginary internal witness becomes unavailable, in other words,
when the one who must be witnessed is not-me. And yet this is precisely
the part of us that, if we are to grow, we must somehow learn to bear
and to know. In such cases it is crucial to have a witness outside our own
minds. In such cases we not only profit from seeing a psychoanalyst, we
need one.
Final Thoughts
The psychoanalytic accounts of narrative with which we are most famil-

iar (Schafer, 1983, 1992; Spence, 1982) are written as if the stories them-
selves are what matter. Problems in living are portrayed as the outcome
of telling defensively motivated stories of our lives that deaden or distort
experience, or of skewing experience by rigidly selecting one particular
account. Therapeutic action revolves around the creation, through objec-
tive interpretation based on the analyst’s preferred theory, of new and bet-
ter stories—more inclusive, more coherent, more suited to their purpose.
In the accounts of narrative by Schafer and Spence, while there is room for
a good deal of flexibility in the way the analyst works, clinical psychoanal-
ysis is defined by its technique, and its technique, in one way or another, is
defined by the way interpretation is employed.
Schafer (1992) believes that psychoanalytic clinical work is very much
like text interpretation. This “text” is both “interpenetrated” and “cohab-
ited” by patient and analyst. But it remains a text. Consider what the ana-
lyst does with the patient who “talks back” (i.e., the patient who tells the
analyst what he thinks of the analyst’s interpretive oἀerings):
[T]he analyst treats the analysand in the same manner that many literary critics
treat authors—with interest in what the analysand says about the aims of his or
her utterances and choices, but with an overall attitude of autonomous critical
command rather than submission or conventional politeness, and with a readi-
ness to view these explanatory comments as just so much more prose to be both
heard as such and interpreted. (p. 176)
It is hardly controversial for a psychoanalyst to claim that what the patient

says often has meanings that the patient does not know. But there now
exists a substantial body of literature that does take issue with the claim
that an analyst can ever adopt “an overall attitude of autonomous criti-
cal command” (e.g., Bromberg, 1998, 2006; Hoἀman, 1998; Mitchell,
1993, 1997; Pizer, 1998; Renik, 1993; D. B. Stern, 1997). This large group of
writers, most of whom identify themselves as Relational or Interpersonal
analysts, takes the position that the relationship of patient and analyst
is one of continuous, mutual unconscious influence. Neither the patient
nor the analyst has privileged access to the meanings of his or her own
experience.
This is the broad perspective within which the view developed in this
chapter belongs. Although it remains undeniable that refashioned narra-
tives change lives, the source of this change is the patient’s newfound free-
dom to experience, an expansion of the self, created through events of the
clinical interaction that are only partially under our conscious control.
It is not so much that we learn the truth, but that we become more than
we were. Our greatest clinical accomplishments are neither interpreta-
tions nor the stories they convey, but the broadening of the range within
which analyst and patient become able to serve as each other’s witnesses.
This new recognition of each by the other is a product of the resolution of
enactments and the dissociations that underlie them, and the resulting
capacity of analyst and patient to inhabit more fully each other’s experi-
ence, to listen more frequently through each other’s ears. As dissociation
and enactment recede, patient and analyst once again become partners in
thought, and now the breadth of their partnering has grown.
Instead of thinking of narrative as a consciously purposeful construc-
tion, we should recast it as something on the order of a self-organizing
system, in which outcomes are unpredictable and nonlinear (e.g., Galatzer-
Levy, 2004; Thelen & Smith, 1994). Clinical process is the medium—or
to use the language of nonlinear systems theory, the event space—within
which narrative stagnates, grows, and changes: The destabilization of old
narratives and the emergence of new ones are the outcomes of unpredict-
able relational events. I hope I have explained my perspective well enough
by now to substantiate the claim I made at the beginning: New narratives
in psychoanalysis are the emergent, co-constructed, and unbidden prod-

ucts of clinical process.
Without denying for an instant the necessity for careful conceptualiza-
tion or clinical discipline, I intend what I have said to serve as an argu-
ment against the claim that clinical psychoanalysis can be defined by any
specification of technique. Psychoanalysis is, rather, a very particular way
that one person can be of use to another, a way that depends on our pos-
session of common practices but also on our awareness that those prac-
tices are often inadequate to the experience that makes up our immersion
in clinical process. For the analyst who believes that the recognition and
resolution of enactments is central to clinical psychoanalysis, the personal
is unavoidably linked with the professional, a point that reinforces some-
thing we have known at least since the work of Racker (1968): If the patient
is to change, the analyst must change as well. In the end we find, as is so
often the case, that when the mind is locked, relationship is the key.
6
Shall the Twain Meet? Metaphor,
Dissociation, and Co-occurrence
The Meaning of Metaphor
ἀ e American Heritage Dictionary (2000) oἀers this definition of

metaphor:
1. A figure of speech in which a word or phrase that ordinarily designates

one thing is used to designate another, thus making an implicit compar-
ison, as in “a sea of troubles” or “All the world’s a stage” (Shakespeare).
2. One thing conceived as representing another; a symbol: “Hollywood
has always been an irresistible, prefabricated metaphor for the crass, the
materialistic, the shallow, and the craven” (Neal Gabler, New York Times
Book Review November 23, 1986).
From this definition, especially from its second part, we might almost
conclude that metaphor is synonymous with symbolic function. That is
certainly not the meaning most of us were taught in school. We tend to
think of metaphor in the terms of the first definition only, the tamer and
less expansive of the two: metaphor as a figure of speech, a purely linguis-
tic phenomenon.
But it turns out that matters are not that simple today, not since the
claims of Lakoἀ and Johnson (1980, 1999; see also Modell, 2003, who has
done much to bring this work to psychoanalysis), who have revolutionized
our understanding of metaphor and its place in cognition. For over 25 years,
these writers have been making a strong case for the view that, far from
being a mere figure of speech, metaphor lies right at the heart of thought.
Lakoἀ and Johnson characterize metaphor, surprisingly enough,
as a phenomenon of the body. The most basic metaphors, the “primary
131
metaphors” from which all the others are derived, are based in univer-
sal bodily experiences, especially those of infancy and childhood. Certain
bodily experiences are inevitably linked to certain other experiences best
described as aἀective and cognitive, resulting in “cross-domain conceptual
mapping,” or the linking of “subjective experience” (thoughts and feel-
ings) with sensorimotor experience.
In Lakoἀ and Johnson’s view, thought begins with the primary meta-
phors and builds from there, with ever more complex metaphors being
constructed on the basis of the metaphor-saturated experience that has
come before. Actually, to say that experience is saturated with metaphor
is only half the story. It is true as far as it goes; but for Lakoἀ and Johnson
metaphor is also the very structure of experience.
When two thoughts, meanings, or perceptions are identified simul-
taneously, the possibility of representing or symbolizing one of them by
reference to the other comes about. Generally we call this phenomenon
association, as in the sense of safety I feel when I think of the scent of
anise cookies baking in my grandmother’s kitchen. Thereafter, without
any thought of my grandmother, the smell of anise provokes the same
feelings. Anise, that is, comes to represent comfort and safety. Lakoἀ and
Johnson (1999) write that the meaning (comfort and safety) is mapped
onto the sensual experience (the smell of anise), forever thereafter lending
meaning to the odor. This mapping is generally unidirectional, though:
Although I inevitably feel safety and comfort when I smell anise, the feel-
ing of safety or comfort makes me think of the scent of anise only if atten-
tion is drawn to the connection (e.g., “When you think of that feeling of
safety, is there any particular scent that comes to mind?”).
Lakoἀ and Johnson’s explanation for the rise of the most basic meta-
phors, which they believe are universal and body-based and arise very
early in the lives of all people, regardless of culture, rests on the same
kind of simultaneity of the metaphor’s terms that, in my own smaller and
more personal context, made the smell of anise a metaphor for comfort.
Consider just two of the primary metaphors (here summarized and para-
phrased from Lakoἀ and Johnson [1999, p. 50]):
1. Aἀection is warmth: The subjective judgment of aἀection is repeatedly

paired with the sensorimotor judgment of warmth because babies feel
warm while being held aἀectionately. The aἀective experience is mapped
onto the sensorimotor event, so that it makes sense for us to say “They
greeted me warmly,” for instance.
Shall the Twain Meet? Metaphor, Dissociation, and Co-occurrence 133
2. Important is big: The subjective judgment of importance or significance

is repeatedly paired with the sensorimotor judgment of size because
babies and children find “that big things, e.g., parents, are important
and can exert major forces on you and dominate your visual experi-
ence,” so that it makes sense, for instance, to say that, “Tomorrow is a
big day.” That is, when we think of important, we think of big.
Notice that in these examples, as in my example, the aἀect (what Lakoἀ and
Johnson call the subjective judgment) is mapped onto an experience in the
sensorimotor domain, resulting in a unidirectional symbolic process. The
feeling of aἀection evokes the feeling of warmth, and the sense of impor-
tance evokes the perception of size, but the reverse does not occur: Large
size does not automatically evoke the sense of importance, and higher tem-
peratures do not evoke the feeling of aἀection.1
Transfer and Transference
Here is what we read about the etymology of the word “metaphor,” from
the same definition in ἀ e American Heritage Dictionary:
Middle English methaphor, from Old French metaphore, from Latin metaphora,
from Greek, transference, metaphor, from metapherein, to transfer : meta-,
meta- + pherein, to carry.
1 The work of Lakoἀ and Johnson is important for psychoanalysts for a number of reasons, only
one of which has to do with a reevaluation of the significance of metaphor. Another contribution
made by these authors is to oἀer a way of theorizing all experience to be based in the body (the
title of their 1999 book is Philosophy in the Flesh: ἀ e Embodied Mind and Its Challenge to Western
ἀ ought) without resorting to a drive concept. However, as much as I appreciate their thinking
and its potential contribution to psychoanalysis, I do not share their epistemological stance. It
is true, in one respect, that Lakoἀ and Johnson do away with objectivism. They take the position
that truth cannot exist in objective form in the world outside our minds because much of what we
call truth, even most of it, is created in the shapes of the metaphors that we think with. The mind
therefore has at least as much to do with shaping what we take to be truth as the “objective world”
does. In fact, it does not make sense to refer to “the world outside our minds,” because mind and
world are a unity. This far I can go, and even appreciate. I am thoroughly in agreement with Lakoἀ
and Johnson’s critique of Cartesianism. But Lakoἀ and Johnson end up replacing one objectivism
with another. They present their theory of metaphor as the new objective truth, one more theory
meant to supplant those that have gone before. The problem of endlessly overturning one “objec-
tive truth” and substituting another is precisely what inspired the insights of hermeneutics and
postmodernism. Given their presentation of their theory of metaphor as the new truth, it is not
surprising that Lakoἀ and Johnson explicitly challenge postmodern, constructivist views.
It seems that “metaphor” has come full circle. We learn from this definition
that the origins of the word lie in the transfer or the “carrying” of meaning;
if we listen to Lakoἀ and Johnson, it has arrived back at those origins. Isn’t
the mapping of one meaning onto another a kind of transfer, or carrying
over? “Mapping” has a certain precision in Lakoἀ and Johnson’s usage, but
for psychoanalytic purposes, because “transfer” evokes “transference,” I
prefer it to “mapping,” as in “a metaphor is created when the substance of
one meaning is transferred to another.”
It seems to me that there exists a very interesting relationship between
transference and metaphor. Think of what we used to refer to as “psychotic
transference,” which is not a symbolic or metaphorical process at all, but
an insistence that two people are identical and therefore interchangeable:
My analyst is my father; no other interpretation will do. Clinical related-
ness is much more viable, of course, if the transference is (again in the
old-school term) “neurotic”: I feel that my analyst is like my father, but I
continue to recognize that analyst and father are separate. Transference is
clinically useful when its meaning is metaphorical; transference is prob-
lematic, on the other hand, when it is a literal equivalence. This point
seems straightforward enough, but it breaks down into greater complexity
when it is closely examined.
In order to feel as if my father is my analyst, I must feel that the two peo-
ple are alike in some key respect. In other words, the as-if kind of transfer-
ence requires me to have created a category in which father and analyst
both belong. But the creation of a category is based on more than the iden-
tification of similarities between its items; a category is also defined by
its items’ diἀerences from one another. Without diἀerences to separate
the members of a category, the grouping would not be a category at all,
but would instead simply collapse back into equivalences. The diἀerences
between the items in a category serve as the context that makes their simi-
larities meaningful. It seems that metaphor and category bear a significant
relationship to each other.
Modell on Metaphor and Category
Arnold Modell (1990, 2003), who has made a pioneering eἀort in bringing
contemporary thought about metaphor into psychoanalysis, recognized
the relationship between metaphor and category in the course of think-
ing through a new way of understanding the old idea of complexes. “A
‘complex,’” he wrote in 2003, looking back at his earlier work of 1990, “can
be defined as an organized group of ideas and memories of great aἀective

force that are either partly or totally unconscious.” Modell took the posi-
tion that it is metaphor that “organizes emotional memory. Inasmuch as
category formation is an aspect of memory, metaphor provides the link
between emotional memory and current perceptions” (2003, p. 41). There
occurs the recognition, in other words, that something in the past bears
a meaningful relationship to something in the present. This link creates
what is essentially a category. Then comes the linchpin of the argument:
A similarity based on a metaphoric correspondence is the means through

which emotional categories are formed. Unconscious emotional memories
exist as potential categories, which, in the process of retrieval, are associatively
linked to events in the here and now by means of metaphor and metonymy. (pp.
41–42)1
We know from Lakoἀ and Johnson (1999) that metaphor—one mean-

ing standing for another—is created by the simultaneity of two experi-
ences. When a memory and a present experience occur simultaneously in
our minds, a meaningful link may be established between them. It is in
this way that metaphoric correspondence is at the same time the creation
of an emotional category.
Modell also tells us that trauma can make it impossible for the suἀerer
to situate memories of the traumatic past in the context of the present,
resulting in the familiar concreteness or literalness about the past that
clinicians see every day in the experience of those who have been trau-
matized and in the inability of these people to integrate past and present.
Modell writes that, in the experience of many trauma suἀerers, “It can be
said that the metaphoric process was foreclosed or frozen” (p. 41; emphasis
in the original). In trauma, that is, the past exists as a concrete record and
cannot be contextualized in the present. Note that this drains meaning
from the present because the present cannot be enriched by association
1 Metaphor and metonymy are two forms of symbolic representation. In metaphor, one object or
concept stands for another. It is immaterial whether the two items bore any meaningful relation
to each other prior to being incorporated in the metaphor, and so, if one does not know why the
two terms came together, metaphor often appears arbitrary, like anise and comfort. Metonymy
diἀers from metaphor in that the two experiences, objects, or concepts are meaningfully related
prior to becoming part of the metonym. Usually the thing that is symbolized subsumes the thing
that serves as the symbol, as in “Soldiers serve the flag” (a national flag stands for the entire coun-
try); or “The king has the scepter” (scepter stands for sovereignty). Metonymic relations result in
categories in the same way that metaphoric ones do, but because their terms must bear a prior
relationship to one another, metonymy is a less flexible means of symbolization. The mere tempo-
ral coincidence of the terms, what I will call “co-occurrence” later in this chapter, is sufficient to
create the potential for metaphor.
with some portion of the past. But the foreclosing of metaphor also drains
meaning from the past—or rather, from the reconstructions of the past
we undertake on the basis of what we learn and experience in the present,
a process analogous to what Freud (1895, 1918) called deferred action, or
Nachträglichkeit. The “frozen” (Modell’s word) metaphoric process, that
is, dooms the past to be frozen, as well, because a frozen aspect of the past
cannot be reformulated in a way that allows it to serve present purposes.
A frozen memory in that sense is a “thing-in-itself,” a concretization that
cannot contribute to future meaning, rather like what Bion means by a
beta element.
I take Modell’s “unconscious emotional memories,” which he describes
as “potential emotional categories,” to be exactly the kind of experience
we see in psychotic transference. Such experience remains isolated. It can-
not become part of a category. And so one cannot relate to it in a way
that would make it possible to know it; one cannot reflect on it. In order
to know or reflect, one must be able to feel something that would be put
into words as some version of, “Oh, yes, that was one of the times that I
felt (or did or thought, etc.) …” or “Oh, yes, that was like the feeling I had
when …”. To relate to an experience as an experience of a certain kind, a
member of a certain emotionally defined category, in other words, is what
allows us to reflect on that experience. As in the case of transference, the
very belongingness of the experience to the category allows us to appreci-
ate its uniqueness. If an experience cannot belong to a category, as trauma
cannot, if it must exist in isolation, as a concretization, a “singularity,” it
must remain a thing-in-itself and cannot be cognized, known, or felt. It
seems, then, that the process of creating associative links between other-
wise separate experiences is crucial to psychic growth.
Metaphor and Witness
If a tree falls in the forest with no one to hear, does it make a sound? It cre-
ates vibrations in the air, yes, but to make a sound it must be heard; and it
cannot be heard unless there is a witness to hear it.
The same is true in the rest of life: Without a witness, no experience
can be “heard.” We are familiar with this point in the case of trauma:
Before she herself can feel trauma as real experience, the traumatized
person either needs another person to know what she has gone through
and care about it, or needs to be able to imagine such a person (e.g.,
Felman & Laub, 1992; Laub, 1995). To be truly alone with experience,
to be able to imagine no one who cares about it, is to be unable to make
emotional sense of it.
But one can be truly alone in the presence of another, even in the pres-
ence of one willing and able to witness. We cannot allow someone to
witness unless we feel safe enough in their presence (again, the presence
may be real or imagined—see Chapter 5) to be able to allow a relationship
between diἀerent parts of ourselves. We must be able to take two vantage
points at once, one of them the traumatic experience and the other some-
thing else, another kind of experience that relates to the trauma, but that
is nevertheless diἀerent. We must be able to contextualize the traumatic
experience, in other words, either with other memories with which it has
something in common or with aspects of the present that resonate with it.
The trauma must become part of a metaphor or an item in a category. Only
when we know the experience as our own in this way, which requires that
the interpersonal field feel emotionally safe and responsive enough that we
can let ourselves think, and think freely, can the other successfully witness
the experience and thereby help us make it real. And this, of course, is the
problem with some survivors of trauma: How can they find their way to
safety in the presence of the other? The other’s willing responsiveness is
necessary, but not necessarily sufficient, and the traumatized often have
no idea how to make it so. For that matter, the traumatized often have little
reason to want to make it so.
When trauma becomes part of metaphor, we become able to sense
it as a figure against a ground. To begin to feel trauma as our own and
to know it as part of our lives, we must be able to see it against parts of
life similar enough to be recognizable, yet diἀerent enough to remain
separate. If we cannot do this, we remain distant from the pain and the
meaning. We may be able to oἀer a factual account of the events, but
the feelings and the part they play in our story remain unformulated,
unconscious.
Witnessing is prevented by the enactments that take place in any treat-
ment, such as the one in the clinical vignette presented below, because the
emotional atmosphere of enactments, by definition, does not feel safe. The
participants in an enactment are unlikely to be emotionally responsive to
one another in the way witnessing requires. The rigidities of enactment, we
shall see, can be defined as singularities, each participant locked into a par-
ticular set of perceptions of the other and himself. Enactment is not only a
rigid kind of relatedness but also the absence of metaphorical thought.
Metaphor and Dissociation
Metaphor and category do not come ready-made; they are products. What
comprises their raw material then? Where do metaphors come from? Why,
in other words, do associative links happen at all?
I have already mentioned Lakoἀ and Johnson’s (1999) simple answer:
simultaneity. Whenever two experiences take place at the same time, there
is the potential for the meaning of one to be mapped onto, or transferred
to, the meaning of the other, as in “The scent of anise reminds me of safety
and comfort.”
I have been making the point for many years (D. B. Stern, 1983, 1989,
1991, 1997) that the experience that feels most as if it belongs to us arrives
in our minds unbidden. Even if we know better, it feels as if we do not
create the unbidden. It feels more as if we allow it—and even then, this
“allowing” is not something we can become aware of doing. We feel
instead as if we are conduits for the unbidden. We do not know what we
do to make it happen, but we do learn (when we believe that we can bear
whatever will come to us) to stay out of the way of its creation. Lakoἀ and
Johnson’s simultaneity account dovetails nicely with this bit of phenom-
enology. Simultaneity, like unbidden experience, is something that hap-
pens to us, after all. It is like high tide, depositing in our minds collections
of things that we might never otherwise think to put together.1 To the
degree that we are able to allow our minds freedom—that is, to the extent
that we are, in the deepest sense, curious—we mine this flotsam, mostly
nonconsciously, for useful co-occurrences of meaning, and then we allow
those to percolate and develop into experience that it becomes appropriate
to call metaphor.
Lakoἀ and Johnson do not try to address the active, nonconscious
imagination that selects only the most meaningful co-occurrences. Nor
do they address the possibility that this selection process ought to be as
eἀective in rejecting our awareness of co-occurrences as it is in accepting
it.
1 This expression is an approximation. Although we cannot control what our minds do, we can
prepare ourselves by immersing ourselves in our field of study or practice, thereby making it more
likely that the co-occurrences available to our minds are relevant to the problems we face.
In fact we know that the rejection of an awareness of co-occurrences,

along with the potential metaphorical experience that could have devel-
oped from them, is routine. Think of what happens in the minds of trauma
suἀerers. Do we really think that the co-occurrences that would allow the
past to be contextualized in the present do not happen in the lives of these
people? Do we believe that these people never have the opportunity to link
their traumatic memories with the present? Of course not. The co-occur-
rences happen all the time. What does not happen is awareness of these
co-occurrences—metaphor. (Remember Modell’s words: in the experi-
ence of trauma suἀerers, “the metaphoric process [is] foreclosed or frozen.”)
Metaphor, in other words, is the actualization of the potential made avail-
able in co-occurrence.1
We need a word here to represent the prevention of actualization,
the stunting or blocking of the process by which co-occurrence enters
awareness and becomes metaphor. We do not have far to look. What we are
describing is the prevention of the conscious connection, linking, or asso-
ciation of two experiences. The word we are looking for is dissociation.
Dissociation is conceived in more than one way. In the field of trau-
matology, starting with Pierre Janet (Ellenberger, 1970), dissociation
represents the mind’s fall-back operation, a desperation move that
takes place automatically when all else fails, a nonvolitional shutdown
that protects the mind from coming apart. In this usage, which is no
doubt still the most widespread, dissociation is not employed actively,
as an unconscious defensive operation. It happens to people; they suἀer
it passively.
The other contemporary usage of the term was inspired by the intersec-
tion of the first meaning with psychoanalysis. In this model, dissociation
is an unconscious defensive process, an active process and not a passive
one, and used in a much wider range of situations than those that are gen-
erally considered traumatic.2 In fact, from this perspective dissociation
has supplanted repression as the primary defensive operation. The mind
is therefore theorized not as a vertical organization of consciousness and
unconsciousness, but as a horizontally organized collection of self-states,
states of being, or states of mind, each in dynamic relation to the others.
1 Metonymy, therefore, because it does not originate in co-occurrence, is apparently irrelevant to
the line of thought I am developing about metaphor (see footnote 2).
2 This line of work was pioneered by Harry Stack Sullivan (1940, 1954, 1956) and developed into
a psychoanalytic perspective by Philip Bromberg (1998, 2006) and others (Bucci, 2007a, 2007b;
Davies, 1996, 1997, 1998, 1999, 2004; Davies & Frawley, 1991, 1994; Howell, 2005; Mitchell, 1993;
D. B. Stern, 1997, 2003, 2004, 2009).
Some states coexist comfortably—that is, from within some states, I

can bear to feel that certain other states are also “me.” From within a com-
fortable, self-assured state of being, for instance, although I may not like
having to accept the part of me that tends to feel envious of others under
certain circumstances, I can at least tolerate knowing and feeling that both
are me. If I could not tolerate that knowing and feeling, the states would be
dissociated, which would mean that they could not be experienced simul-
taneously and would remain sequestered from each other. From within
the parts of my experience that I accept as me, the rejected part of my
experience is alien to me. I cannot and will not tolerate its presence within
what feels like it is mine. It is not me, or, in Sullivan’s (1953/1940) term,
“not-me.” I am safe as long as that state is isolated and disconnected from
the multitude of experiences that invoke me.1
And so when the actualization of co-occurences is blocked or pre-
vented and metaphor does not emerge, it is not because a central, unitary
self is unable to know the experience that would result if the metaphor
came to life. The situation is more complicated than that. The reason
metaphor is rejected is that the two experiences in question invoke
states of self that are dissociated from each other. To accept the meta-
phor would require simultaneously accepting me and not-me. I would
have to accept that I am what I cannot be, what I refuse to be. What is
rejected, then, when metaphor is blocked, is not memory but a certain
experience of who I am.
Clinical Illustration
Now I return to the topic of transference and countertransference and,

as promised, to enactment. I present a long case vignette to illustrate the
points I have made.
Two years ago, I began working with an unusually attractive, charming,
socially adept, intelligent, and well-educated woman who, despite always
having assumed that she would marry and have children, could not seem
1 I have defined two kinds of dissociation: dissociation in the strong sense; and dissociation in the
weak sense, passive dissociation, or narrative rigidity. For definitions of these terms, see D. B.
Stern (1997, 2009) and this book, Chapter 1. In this chapter, I am using dissociation only in the
first, “strong” sense. That usage means that what is defended against is actively and unconsciously
rejected; it is specifically selected to remain unformulated for defensive reasons. In dissociation in
the weak or passive sense, on the other hand, psychic material goes unformulated because we are
so attached to creating our lives along the lines of a particular narrative that alternative meanings
simply go unconsidered.
to make relationships take that direction. Now approaching the age of 40,
she was worried about her future. To begin with, I was baffled at her lack
of success, and despite myself, I began to wonder if perhaps her problem
was that her positive attributes threatened most of the men she met. (I did
not yet understand that she threatened me.) But this explanation did not
seem to me to be a very good explanation, because it seemed unlikely that
she could have threatened all the men she met. And besides, that inter-
pretation would ignore whatever her own contribution might be. At this
point, though, I could do no more than refer to “her contribution” in the
abstract. Nevertheless, I pointed out to her, somewhat dutifully, that only
if she could discover ways in which she was unconsciously creating and
maintaining her own distress would there be realistic hope that she could
change this part of her life.
Abstract principles hardly motivate people to enter psychoanalysis.
That is doubly true for those patients, such as this woman, whose capac-
ity for self-observation is limited. (I was later to find, happily, that this
impression needed to be revised.) The analyst needs to be able to cite a
concrete example of some way in which the patient unconsciously under-
mines herself. More than that, the analyst needs to believe the example he
cites. With this woman, I could neither cite nor believe. I was struggling
with the thoroughly nonrational perception that she was flawless. I knew
better, of course. I was even able to refer back to the example of her lack of
psychological mindedness as an illustration; but, unsurprisingly, I could
not convince myself: The perception that dogged me was a feeling, not a
reasonable perception. I did not yet see that her perfection was itself the
point: She was like the perfect princess who lived at the top of the glass
mountain. Like the suitors who tried to ride their horses up the mountain
to reach the princess, I could find no point of purchase, no way to talk to
her that would create some kind of sense of relatedness with her. It was
easy for me to feel inadequate in her presence.
I was reduced to hashing and rehashing with her the end of the relation-
ship that had finally brought her into treatment. She was in genuine pain
about this, and she appreciated my suggestion that her pain was less about
the man himself than about her worry that her hopes for the future were
dimming fast. But this idea was hardly sufficient to carry the treatment. I
could sense that, unless I found a way to help her deepen the work—which
is to say, unless I found a way into a discussion of the less than perfect
parts of her experience, but without shaming her about them—the treat-
ment was going to end shortly. She would feel better, at least temporarily;
and if I had nothing more to oἀer, she would leave.
I had ceased anticipating this patient’s visits with pleasure soon after we
had started meeting, and at this juncture, 2 or 3 months in, I was becom-
ing quite familiar with the feeling that I was not a very competent analyst
for her. My rehashing of her recent relationship seemed vapid, superficial,
and intellectualized to me; and while it was not difficult for me to connect
my feelings of inadequacy with her impenetrability, I also imagined, with
moderate discomfort, that she agreed with my assessment of my eἀorts.
Actually, it is not true to say that I could observe nothing beyond this
woman’s perfection. I have mentioned being impressed with her impen-
etrability. I had also noticed the defensive quality of her continuous, brittle
good cheer. She could cry about her pain, and she could be angry, but only
if some objective situation in the outside world merited it. I had the sense
that sadness or anger under any other circumstances would feel unjustifi-
able to her and would probably represent a weakness in her eyes. It would
shame her. She could not be vulnerable to me, in other words, and I felt
sure that I was not the only one with whom she felt this way. As a matter
of fact, I imagined that this might be exactly the problem she was encoun-
tering in maintaining a romantic relationship. I could not just oἀer the
patient that observation, though, not unless I had something to say that
would help her make use of it in a way that did not potentiate the shame I
could sense in the wings.
Time was running out. This was not a person who could discuss her
frustration with the treatment or with me in a productive way. To do so
would seem unacceptably hostile to her, rather like criticizing her marvel-
ous parents (one of many attitudes that had made it difficult to get the
treatment moving). Or rather, even if she were able and willing to talk
about her frustration, it would do her no good unless I could say some-
thing that would give that frustration a diἀerent or broader meaning than
it had now, something that would bring some life into the work for her. If
I could not do that, then talking about her frustration would simply be a
prelude to her departure.
And so one day, having failed to come up with the perfect (!) interpreta-
tion and with the time left in the treatment swiftly draining away, I took
a deep breath and stumbled into an attempt to say something authentic
to this woman about my reaction to her presentation of herself. I did not
know where I was going or exactly what I would say when I got there. I
talked to her for a couple of minutes about feeling that there must be parts
of her that she was not pleased about, that maybe she didn’t even like,
because everyone has parts like that. Yet (I told her), I didn’t seem to be
able to get to know her that way. I told her that I felt she was having a very
hard time being vulnerable with me, letting me really know her. I told her
that, while vulnerability could be uncomfortable for anyone, I thought it
must be particularly uncomfortable for her. I could see, I said, that unless
we were able to move what we were doing in the direction of me getting to
know her in a way that would no doubt make her feel vulnerable, the treat-
ment was going to end, because she was going to cease seeing any value in
it. I told her I knew how frustrated she must be with what we were doing,
and, like me, how little she must be able to figure out how to make things
diἀerent between us.
These thoughts did not come smoothly, nor did I express them that way.
I struggled with them. And of course I was watching her reaction. She
seemed interested in what I had to say about vulnerability, and she agreed
outright with my estimation of her frustration. These things were good;
but still I could not see how I was going to identify something in her expe-
rience, something that she could see at least as well as I could, that would
open what we were doing into a psychoanalytic treatment.
At some moment, as I was talking, she appeared to me to change.
It was quite subtle. She seemed softer and more open. That descrip-
tion, though, “softer and more open,” was not available to me in the
moment, only later on, when I thought back on it. In fact, I was not
even aware of the presence of my new perception of her until, in retro-
spect, I tried to understand what had happened in the moments before
I finally found myself able to say what occurred to me next, which was
something new about her experience, something that I thought she
would recognize and that might just help us into a more analytic kind
of relatedness. The thought formed itself as I was speaking. Actually, I
am quite sure that its possibility was created by the prior subtle change
in my perception of her, which was, in turn, created by some change
in her own affective state. As I spoke, I think that my novel percep-
tion was also helped along by my patient’s facial expressions, through
which she expressed a frank, friendly, and inquisitive interest in what
I was saying.
But I am describing these moments with more precision than I expe-
rienced at the time. The truth is that I surprised myself—I didn’t know
what was coming until I was in the process of uttering it. I said, “I think
you must be lonely. I think you must always have been lonely.” Seeing her
shock and recognition and the tears welling in her eyes, I was encouraged
to continue: “I wonder if you have ever felt really known by anyone.”
She wept, but this was not the hard crying that had accompanied her
angry descriptions of the way her boyfriend had treated her. She hid her
face in her hands. After a minute or two of silence she looked up at me and
said simply and sadly, “I am lonely. I’ve always been lonely.” After another
silence, she confirmed that, indeed, she had never felt that anyone had
known her, not even her parents, who were so very proud of everything
she had accomplished, but equally eager not to know more than that about
her. Her mother had actually physically turned away from her, she told
me, on the few occasions when the patient had tried to talk to her about
less than sunny matters. (Over the next months she revealed, unsurpris-
ingly, that there actually had been quite a few unsunny matters.)
The session ended. It was obvious to both of us that we had started to do
something quite diἀerent. As she walked in the door for her next session,
she said as she sat down and smiled at me, “Now we have something to
talk about.” It was unnecessary to say it.
In later sessions, although she sometimes returned to her previous pre-
sentation of herself, my patient also found her way surprisingly often to
moments of vulnerability and authenticity. She is one of those people who
had no idea that it was even possible to talk in the way that successful
psychoanalytic treatment demands. She had more capacity to think about
herself than I had seen before. In fact, she had more capacity of that kind
than she had known about herself. I began to know her diἀerently, and I
have grown quite fond of her.
It is worth noting that at the same moment that I saw her loneliness, I
suddenly regained my sense of competence. Immediately I had the feel-
ing that things were going to be all right in this treatment, and I regained
my customary clinical confidence. These were the signs, from within my
experience (we already know the signs from within hers), that a mutual
unconscious enactment had dissolved, an enactment in which she was
perfect and I was badly flawed.
Dissociation and Enactment
Over the past few years I have been theorizing links between dissocia-
tion and enactment, reported in other chapters of this book. Enactments
require that the patient, or both the patient and the analyst fall into a
mutual dissociation. It is generally not important (and not a good idea
to try to establish, either) who begins the process. In any case, one par-
ticipant treats the other in a way that threatens to force acknowledgment
of not-me. Whenever that happens, in treatment or any other part of life,
the response is enactment: in eἀect, one’s behavior says “I’m not x; you’re
x,” where “x” is some not-me aspect of one’s personality. In this case, let
us say for heuristic purposes (that is, I am starting with the patient’s con-
tribution only for the sake of simplicity; I could just as well start with the
analyst) that the patient, prone to the concern that treatment would reveal
her unacceptable imperfections, felt threatened with unbearable shame,
which for her was not-me. She could not tolerate that possibility, and so
she began enacting the dissociation by presenting to me her well-honed
perfect persona and thereby made me feel the shame of imperfection.
I do not conceive the analyst’s part of an enactment to be the outcome
of having been colonized by an alien part projected by the patient, as in
projective identification. Each participant’s dissociation emerges instead
from the interaction of her own private motives with the unconscious
influence of the other. When both participants are vulnerable to the same
themes, a state of aἀairs that is obviously not routine but which neverthe-
less takes place with great frequency, the result is a reciprocal enactment
of the same themes.
I was for a time vulnerable enough to the way my patient treated me—
which is to say that to some degree this kind of shame had to be not-me
for me as well—to take on the shame unthinkingly and without being able
to deal with her veiled contempt. My experience was, like hers, a singular-
ity, a concretization that I could not use in my mind, which remained a
thing-in-itself and therefore could not be felt or known. And of course my
singularity was reciprocal to her own. We fit each other like a hand and a
glove, or better, in a metaphor I am fond of and have used before, like the
two halves of a broken plate.
On the other hand, my shame was not intense enough to push me
toward responding with a countervailing unconscious enactment of the
intention to make her feel ashamed. That happens often enough between
analysts and patients (see, for example, the case of Hannah in Chapter 3),
and it would have been easy with this woman. The analyst who must feel
competent at all times, in other words, might have dealt with this woman
by revealing her insecurity or the brittleness of her good cheer.
Dissociation and the enactment are both breached when either par-
ticipant develops a new perception of the other, a new perception that, in
coming into being, makes it possible to experience the other (and there-
fore, herself as well) in more than one way (see Chapter 4). In this case, that
moment came when I suddenly saw my patient as softer and more open.
In the terms I have been exploring in this chapter, the new perception
makes it possible to attend to co-occurrences that had always come and
gone but had never been allowed to percolate and grow. The moment I
saw my patient in this new way, I was able to contextualize her pursuit
of perfection diἀerently, more poignantly, and of course that changed
the nature of our relatedness. I am quite sure that there had been oppor-
tunities (co-occurrences) for me to recontextualize her treatment of me
(i.e., see her treatment of me metaphorically) before this, but I had been
blind to them. The new perception is evidence that the potential in certain
key co-occurrences has become available for actualization, often for the
first time, and the result is metaphor or category where neither had been
before. The past can be contextualized in the present. In response to my
new perception of her and the diἀerence that that new perception quite
automatically made in my treatment of her (from the moment you see
the other diἀerently, you just cannot treat him or her the way you had
the moment before), my patient’s experience of shame about her imper-
fections, which had been not-me, frozen in an unchanging world, could
begin to be metaphorized, to be situated in a present context that could be
compared to the past, a new context in which imperfection no longer had
to mean the same thing.
I want to reemphasize that it was not some particular memory that the
patient had been unconsciously unwilling to encounter. What she was
avoiding was the perception of herself as a certain kind of person, a person
whose imperfections frightened her, a person who needed the love of her
parents but who questioned whether she would be loved if she could not
fulfill their image of her. The transference, while it had not been psychotic,
had certainly qualified as a singularity, existing apart from her capac-
ity to think. In order to dwell exclusively in the self-state of perfection
and banish the state in which her imperfections had to be acknowledged,
the transference had to be a singularity. Only singularity would do. Now
she could question the nature of her relatedness to me. The transference
became metaphorical, and as such, became clinically viable. (The coun-
tertransference, of course, went through the same sequence of changes.)
Multiple images supplanted singularity. And finally, as her dissociation
(and mine) was breached (over and over, but never as dramatically as
the first time) and her sense of self expanded, I could finally become my
patient’s witness.
7
Opening What Has Been Closed,
Relaxing What Has Been Clenched
Dissociation and Enactment Over Time
in Committed Relationships
People who want to be in long-term romantic relationships, but who have

trouble maintaining loving, sexual feelings toward their lovers, have two
kinds of problems. One kind of problem has to do with preserving loving,
sexual feelings long enough to get into such a relationship in the first place;
the other difficulty, suἀered by those who are already in long-term roman-
tic relationships, concerns keeping loving, sexual feelings alive.1
The first kind of problem is a consistent, rigid inhibition. One’s capacity
to stay sexually and intimately connected to a lover does not rise and fall
according to context, as these parts of relatedness do in a long-term rela-
tionship. One is simply interested until one is not, and then the passion is
over and gone; and because of that, because of the rigidity of this kind of
problem, its structural quality, its (relative) lack of responsiveness to con-
text, it is especially easy to fall into thinking about such difficulties in what
can seem one-person terms. These people stay at such a distance from the
unavoidable and continuous oedipal themes in their relationships with
their lovers that they do not really engage in the struggle.
1 Virginia Goldner (2004, 2006) organized her discussion of the first chapter of Stephen Mitchell’s
(2002) book, Can Love Last?, around the observation that Mitchell’s portrayal of romantic love,
surprisingly enough, seemed more one-person than two-person. That is, Mitchell’s description
of the difficulties of romantic love seemed to Goldner to have an inadvertently intrapsychic cast.
Mitchell’s chapter and Goldner’s discussion then became the basis of an online colloquium orga-
nized by the International Association of Relational Psychoanalysis and Psychotherapy, held in
2005 and called “What Happens When Love Lasts? An Exploration of Intimacy and Erotic Life.”
On the basis of their postings, some of those who participated in the colloquium were invited by
the organizers, Margaret Crastnopol, Ph.D. and Randy Sorenson, Ph.D., to contribute articles to a
collection (Psychoanalytic Dialogues, 16(6)) that addressed the same subject. An earlier version of
this chapter was written for that collection of articles.
147
In referring to “oedipal themes” and “struggle,” I am thinking about

a paper by Davies (2003) in which the successful outcome of the oedipal
situation is conceived not just as the acceptance that one can never win,
but as the acceptance that one will both lose and win. The oedipal conflict
is not successfully resolved by accepting defeat, that is, but by becoming
strong and self-respecting enough to accept instead a lifelong and vital-
izing engagement in the themes of love and rejection.1 The child who
knows that she is loved by a parent who often or even usually chooses his
spouse—but who sometimes chooses the child, too—grows up to be a rela-
tionship partner able to thrive within admiration and love and to tolerate
rejection and narcissistic wounds without being badly thrown. These are
the lucky people who “remember” that they are loved even when love is
not immediately present. Love is the background condition for them, the
aἀective atmosphere within which the most important relationships take
place. When their partners are rejecting, they are disappointed and hurt,
but they are not left in suspense about the future. That is, unlike the less
fortunate—those who cannot get into long-term romantic relationships in
the first place—people who are actually in relationships, at least those in
relationships that are not in what Goldner (2004) expressively refers to as
“defensive lockdown,” find their way through this oedipal terrain. Over
and over again they find their way to and fro between intimacy and sexual
desire, on the one hand, and distance, hurt, and relative lack of interest, on
the other. In diἀerent ways at diἀerent times, they struggle with the other
person as old and new object, as parent and lover.
We know from life, of course, and we see once again from this brief
description, that having problems in a committed relationship can be a
quite diἀerent kind of problem than being unable to get into such a rela-
tionship in the first place. Within successful existing relationships, part-
ners find a way to continue their engagement over love and sex, whereas
those who cannot find their way into a relationship at all very often avoid
the struggle altogether, giving up the ghost as soon as the conflict appears.
Is it perhaps a mistake to place both the engagement of struggle and the
avoidance of it under the same rubric? The back and forth of engagement
tends to be more existential and contextual than avoidance is, and habitual
avoidance tends to be more neurotic, rigid, and stereotyped than engage-
ment is.
1 See also Stuart Pizer (2004), who uses Davies’ ideas to invent a concept he calls “the paradoxical
analytic triangle,” a tool to use in thinking about the interrelatedness of the loving feelings and
rejections in a very long treatment.
Opening What Has Been Closed, Relaxing What Has Been Clenched 149
The problem I set for myself in this chapter is this: How can we think, in
two-person terms—in Interpersonal or Relational terms, that is—about both
kinds of problems, about both the rigidities of those people who cannot find
their way into relationships and the changing issues between couples who
are already in them? Is there a single frame of reference that encompasses
desire both as fluid and contextual and as structural and relatively unchang-
ing? Given the theme of this book, it will not be surprising that, in thinking
about this question, I find myself drawn to the subject of dissociation.
Bad-Me, Good-Me, Not-Me
It seems to me that both kinds of difficulties—the struggles of engagement

and the ongoing status quo of individual characterological inhibition—
relate closely to how broad or deep an experience of self one can tolerate.
The language that suits this kind of thinking is the language of self-states,
which I introduced in the earlier chapters. We might say that both the
emotional oscillation of engagement and the defensive lockdown of habit-
ual avoidance can be described in terms of the degree to which one has the
freedom to move about between states of mind.
In what Bromberg (1998) calls “normal” dissociation, and what I call
narrative rigidity or dissociation in the weak sense (D. B. Stern, 1997; this
book, Chapters 1 and 3), other people’s participation interacts with one’s
own tendencies, predilections, and inner life in such a way that access
between self-states that are otherwise simultaneously present in the mind
is temporarily disrupted. This is what might happen, for instance, when
someone unexpectedly shames us for something with which we are usually
quite familiar in ourselves, but which we do not particularly like. Taken by
surprise, we might react with heat, attributing the disliked characteristic to
the other person and feeling wrongly accused. Or, in even more everyday
terms, perhaps we frequently walk around with a fairly good opinion of
ourselves, our good points in the foreground of our minds, our less savory
aspects outside immediate awareness. But when called upon to acknowl-
edge what we do not like about ourselves, it is relatively easy for us to do
so. It may require a spouse’s irritation, or an analyst’s confrontation, and
we may be grudging about it, but we do know these things about ourselves,
or can stand knowing them. We can tolerate them. Normal dissociation is
relative and often situational. The unconscious defensive need for it is not
urgent. Normal dissociation makes life more comfortable, that’s all. These
mild dissociations are therefore remedied with relative ease, sometimes
simply by a shift in the circumstances that provoked them in the first place
(for a clinical example, see Chapter 3, the case of Daniel).
In the terms of Harry Stack Sullivan’s (1954) that I have been using in
this book, we could say that normal dissociation, or dissociation in the
weak sense, takes place when the part of me that I have come to feel as
“bad,” or bad-me, is not experienced simultaneously with the part of me
I feel as “good”—good-me. Bad-me is the part of myself I associate with
the criticism, punishment, and rejection of my caretakers in childhood (at
a certain point, Sullivan says the child characterizes these caretakers col-
lectively as bad-mother); and good-me is the part of me that was loved, val-
ued, accepted, and praised (by all those parts of caretakers characterized
collectively as good-mother). Even if I don’t like being bad-me or am wor-
ried about the consequences (a drop in self-esteem, or security, according
to Sullivan; but we can broaden the range to include other kinds of less
than pleasant outcomes), I do feel that I am being myself when I am either
good-me or bad-me; and so even if I am reluctant to experience good-me
and bad-me simultaneously, I have little difficulty doing so.
It is not as rewarding to be bad-me, of course; it may even be unpleas-
ant. I may resent the circumstances in which I seem to have no choice
but to be particularly aware of being bad-me, and at times I may feel that
the pain of acknowledging this side of me is more than I will accept. But
my unconscious defensive refusal to acknowledge being bad-me (which
amounts to insisting I am only good-me) is usually temporary, and if it
lasts longer, it is easily breached—because bad-me does still feel like me.
Dissociations between bad-me and good-me are frequent, but they are
hardly crucial to the continued, secure existence of the personality, and so
they are easily destabilized.
Now, if you add to these ideas the relatively uncontroversial clinical
observation that dissociated states tend to be enacted (e.g., Bromberg,
1998, 2006; Davies & Frawley, 1994; see also the earlier chapters in this
book), you can see that dissociations between good-me and bad me are
generally translated into treatment of the other that conveys something
like this: “I am not bad-me, I am only good-me. You are bad, not me.”
If we then translate this scenario into the contemporary oedipal terms
used by Davies (2003), Goldner (2004, 2006), and S. Pizer (2004), we find
ourselves, when we temporarily cannot accept rejection from the other,
treating the other in a way that conveys something like the following: “I
am the loving, generous, reasonable, intelligent, attractive boy my mother
adored, not the selfish, angry, unreasonable, stupid, ugly boy (or the inad-
equate boy who mother thought was ‘less’ than father) from whom she
turned away. No, the bad one is you—you are the selfish, angry, etc. one,
the inadequate one, the outrageous one, not me.”
I know that this set of ideas immediately brings to mind projective
identification. I have addressed this point Chapters 1 and elsewhere (D.
B. Stern, 2001, 2005; see also Mitchell, 1997, Chap. 4). But let me digress
briefly here to add a bit more about the matter.
The diἀerences between my views and those of the contemporary
Kleinians are due, from my perspective, to the diἀerent origins of our
views. Psychoanalytic theories of dissociation are rooted in Interpersonal
and Relational psychoanalysis. From those perspectives, dissociation can
only be enacted if the roles that constitute it are relevant in the psychic
lives of both participants. The contemporary Kleinian analyst is often
expected to be able to experience the impact of the patient’s projections
while simultaneously maintaining a purposeful, conscious analytic stance
toward them. The analyst, that is, however understandable it may be for
him to lapse in this respect, is believed to be capable of not becoming
lost or embedded in the countertransference.1 From Interpersonal and
Relational perspectives, on the other hand, it is acknowledged from the
outset that the analyst’s task is to learn the nature of his quite thorough,
and quite personal, unconscious involvement, which has always already
taken place. There is no consistently maintained “analyst part” that some-
how exists above the fray, separate from the subjective aspects of the ana-
lyst’s personality.
Analogously, from an Interpersonal or Relational vantage point, there
can be no consistently maintained parts of the personalities of persons in
romantic relationships that exist apart from their relatedness from one
another. The analysis of analytic relatedness and of problems in romantic
relationships, at least when such understandings are sought by the partici-
pants themselves, are both examples of what Mitchell (1997) calls “boot-
strap operations.”
Now I return to the minor dissociations that go on within ongoing
romantic relationships. Although intimate sex is not well served by these
minor dissociations, it is also not likely to be precluded for too long at a
time either. As we wend our way back and forth from simplified, wish-
ful, and often angry characterizations of ourselves and our lovers to more
complex and tolerant observations, we first avoid the bedroom and then
slip back into it. These are the ups and downs of sex and love in long rela-
tionships. The crucial thing that allows us to return to intimacy each time
1 See Chapter 4, footnote 6.
is our reclaiming of the capacity to tolerate the experience of good-me and

bad-me simultaneously, which amounts to accepting the conflict between
these two portrayals of ourselves (I am neither as good as I want to be nor
as bad as I sometimes fear I am). Of course, it is only when we can tolerate
conflicting aspects of ourselves in this way that we can, in turn, tolerate
conflicting perceptions of the other. And so it follows that the acceptance
of our own conflicting self-states is what allows us to recapture a fuller,
warts-and-all perception of the other.
The terms Sullivan invents to describe the aspects of ourselves that we
like and dislike are very simple. But their simplicity is deceptive. Consider
the degree of complexity introduced as soon as we admit into the pic-
ture the fact that all dissociations are conditional, in the sense that they
are continuously preserved or challenged by the interpersonal context in
which they take place. That is, if the other is able to sense nondefensively
his or her response to being treated as bad and then bring it to mutual
attention, the interaction will not turn into a mutual enactment.
But if one’s partner in the relationship is vulnerable to a reciprocal,
answering dissociation (“You think I’m bad? No way, it’s you who’s bad”),
a mutual enactment is oἀ and running, in just the same way that I have
described in Chapters 3–6 for the analytic relationship. The enactment will
end only when one partner or the other can find a way back to an internal
experience of the conflict between what she feels is good and bad about her-
self. Once she is capable of that conflict again, she quite automatically treats
the other person diἀerently, because he no longer has to control the other’s
perception. At that point, the dissociation, with its accompanying enact-
ment, which was mild in the first place, vanishes, at least for the time being.
But in the case of “pathological dissociation” (Bromberg, 1998) or “dis-
sociation in the strong sense” (D. B. Stern, 1997; this book, Chapters 1 and
3), the consequences for the relationship are much greater. This is the dis-
sociation that comes into play when certain kinds of experience are spe-
cifically and continuously disallowed or rejected; this is dissociation used
as an unconscious defensive operation. In instances demanding “strong”
or “pathological” dissociation, the disallowed experience is so intolerable
that, despite its presence within the very broad bounds of subjectivity, it
is not acknowledged as part of the self (it is not me) and therefore can-
not be articulated in awareness. It cannot be felt as one’s own experience
without fairly severe, disruptive consequences. This is neither good-me
nor bad-me; this is what Sullivan called not-me, and it is just as disrup-
tive in romantic relationships outside the consulting room as it is in the
analytic situation.
Now, this point is key: Unlike relatedness between parts of me, there is
no experience of conflict between me and not-me (see Chapter 4.) Good-me
and bad-me, remember, can be experienced simultaneously because they
are both parts of the self, and so they can be experienced to conflict with
each other. Not-me, though, must simply be absent from one’s sense of
oneself, forever unsymbolized—that is, unformulated. Because it cannot
be experienced in symbolic form at all, not-me certainly cannot be expe-
rienced to conflict with me.
In the experience of especially vulnerable people, whose defenses are
notably rigid, this absence of simultaneous experience of me and not-me
is sometimes absolute; if not-me emerges in consciousness, requiring the
recognition that I am what I cannot and must not be, the self is so seri-
ously threatened and destabilized that experience can become psychotic.
The emergence of not-me in consciousness does not carry the same degree
of danger for more stable and flexible personalities, of course, but even
they are unable to access or symbolize that part of their subjectivity at
will; and for them, too, it therefore can be strange (Sullivan’s word was
“uncanny”), even excruciating, to find themselves faced with not-me.
Children, we know, develop their capacity to digest experience slowly,
over time. It takes many years of continuous, loving care for a child to
develop a sophisticated enough capacity for experiencing that he is no lon-
ger continuously vulnerable to being overwhelmed by events of everyday
life that will eventually become routine. Prior to the development of that
degree of resilience, and especially in infancy, the child is dependent on
his caretakers to contain and symbolize experience for him (Fonagy et al.,
2002). In those early years, experience is traumatic to the precise extent
that the child’s caretakers cannot bear it, cannot let themselves consciously
and fully experience it. The child is vulnerable to trauma, that is, whenever
the parent is forced to confront not-me. When the parent cannot stand to
feel the experience, the child, who only knows what his own experience is
if the caretaker is capable of bearing it (feeling it, knowing it), is deprived
of what he needs if he is to create his own mind; and so, over time, this
kind of experience becomes as dissociated for the child as it has been for
the caregiver. This is not the kind of dissociation that comes and goes,
like the temporary and relatively mild dissociations between good-me and
bad-me; this is a stable, foundational kind of dissociation around which
the personality comes to be shaped. This is the transmission of psychic
pain and damage across generations.
We hope, of course, that parents manage to experience their most
uncomfortable moments, their moments of not-me, without anxiety so
overwhelming that it prevents their symbolization of experience, because

then the child herself will eventually be able to take on the same capacity.
In those fortunate instances, the experience ends up feeling to the child
like “mine,” and what could have become not-me becomes me instead.
Childhood is an apprenticeship in the creation of mind, and one’s caretak-
ers are the masters of the guild.1
The (pathological) dissociation between me and not-me is maintained in
the same way as the (normal) dissociation between good-me and bad-me:
by enactment. But because the dissociation of me and not-me is continu-
ous and absolute, not contextual and relative, and because one’s sense of
basic security in the world depends on maintaining these dissociations, the
enactments that derive from the separations between me and not-me are
more rigid, intense, urgent, and less amenable to change or resolution than
the enactments that support the dissociation of good-me and bad-me.
Enactment, Dependence, and Knowing Your Partner
In an ongoing love relationship, what is the nature of our dependence on

our partner? Don’t we come to rely on that person (among other things)
to be the same as he or she has been, to deal with us in the ways with
which we are accustomed, ways that perhaps we have come to need? Why
might we need this familiarity? Might not our lover’s familiarity allow us
to preserve our own? Might not it be the case, as Benjamin (1988, 1995,
1998) tells us, that we need our lovers to recognize in us what we recognize
in ourselves (and to leave unrecognized in us what we prefer not to see)?
1 This description is more or less consistent with the thinking of Winnicott and Bion, but it prob-
ably has most in common with (besides Sullivan) Fonagy et al (2002), whose work might serve as a
detailed exegesis of Sullivan’s (1953/1940) developmental theory of infancy and childhood. In the
model proposed by Fonagy et al., mind is conceived to be created by the interaction of the infant’s
inborn potentials (the “primary” or “constitutional” self) with caregivers’ provision of a sensitive
mirroring response to what they believe they can understand of the infant’s aἀects and intentions.
It is the caretakers’ mirroring that structures the infant’s inborn somatosensory experience into
the experience of aἀects and eventually into the capacity for a deeper and more thorough reflec-
tive processing (mentalization) of the causal mind states (aἀects and intentions) that lie behind
the infant’s own behavior and the behavior of other people. The distinctive aspect of this way of
thinking is its jettisoning of the Cartesian privileging of internal experience. In most models of
the development of self, mind is assumed and inborn, and so the development of mental represen-
tations of self and others and the ego structures that make representation possible in the first place
are an expansion of this inborn capacity. The radical aspect of the model proposed by Fonagy et
al., and the aspect that makes it especially relevant here, is that mind is not assumed and inborn:
It is created from the outside in. The mind that results is sophisticated to the degree that early
caretakers’ mirroring responses correspond to the structures of the constitutional self.
Might not that recognition from the other give us the reassurance that we
are who we want to believe we are, and that we will continue to be—and
that we are not who we prefer not to be? Might not our lover’s sameness,
in other words, allow us predictable, reliable experiences of good-me and
bad-me? Might not we be involved, that is, in long-term enactments of
good-me and bad-me, enactments that preserve the stability and comfort
of both partners? And might not our familiarity with our lovers, in just
the same way, reinforce their sense of who they are and aren’t, their sense
of self?
Of course, even if this portrayal makes sense, it still leaves each mem-
ber of any couple in the position of sometimes having to put up with cer-
tain rather predictable discomforts. It may make us feel and act angry,
for instance, when, in the face of our insistence that we are innocent and
it is wrong to accuse us, our partner refuses to acknowledge being in
the wrong. We may suspect that our partner has a point, which may be
uncomfortable, making us feel guilty. We may, in fact, blame our partner
for that guilt, too, right along with blaming him for his unwillingness to
acknowledge responsibility for being wrong. But even all this discomfort
is a small price to pay for having a reliable means of reassuring ourselves
of being good-me. (Even if you refuse to acknowledge the truth, I can still
believe that I’m not bad, you are.)
These enactments take time to fall into place and become habitual,
which is one reason for the immense sense of freedom and excitement
we experience at the beginning of a new relationship. We are raw and
open at the beginning, and the themes of oedipal struggle occupy us fully,
intensely, and spontaneously. The relatedness is not yet patterned, as it will
be. (I think she likes me! She’s incredibly exciting, and it’s so wonderful
that she seems excited by me! Does she like my hair? Is she angry with me?
Does she love me? Maybe she really wants to be with him.)
The hopes, fears, exultations, and despairs at the beginning of a rela-
tionship have a special intensity; but as painful as those first weeks and
months can be, they are also one of the most exciting parts of life. Lovers
are less intimate at the beginning than they will be later on, but ironically,
they also can be less defended. Their openness to being aἀected by hurt
or admiration (for instance) may be greater than it will become. They are
freer to be completely delighted with each other. They know each less well,
love each other much less deeply, and yet in certain respects are perhaps
most thoroughly available to each other.
But what happens to that ongoing struggle, to involvement and change
and engagement, as we come to depend on sameness, on the enactments
that develop over time? Struggle dims. There is less and less change, life
stays closer to the baseline. We reassure ourselves that the other’s mind is
completely mapped. We become locked into patterns of enactment.
Of course, this is an extreme characterization. In most relationships, the
struggle does go on, even as it dims, and there may be as many moments of
spontaneity as there are pallid repetitions of the familiar. The point I want
to make is that, to the extent that we become dependent on characteristic
enactments of good-me and bad-me, the vitality of the struggle and the
excitement it generates diminish over time.
Of course, if these points make sense for good-me and bad-me, how
much more must they be true for me and not-me? How much more depen-
dent must we be on our partners and spouses to maintain the enactments
that protect us from an exposure to what we cannot bear to be, must not be?
The least fortunate people, whose personalities are the most rigid, because
they must protect themselves from the experience of the uncanny, from
the eruption of not-me, defend themselves by creating and tolerating rela-
tionships that will contribute to their crucial dissociations. These people,
in order to keep not-me at bay, tolerate relationships that feel continuously
dead and boring or hateful and destructive. The more urgent the need
to isolate one’s own not-me in the other, the more hateful and intensely
sadistic the enactments necessary to accomplish the task. All enactments,
especially those of not-me but even including bad-me, are adversarial; but
it is much more likely, in enactments that protect one from having to bear
the unbearable, for the relatedness to degenerate into violence, physical or
emotional. There is no reason to hold back when the consequences of the
enactment’s failure (i.e., when a particularly virulent not-me threatens to
come home to roost) are worse than any outcome of the interpersonaliza-
tion of the dissociation. The dread of not-me can be compelling enough to
lead to murder, sometimes of a virtual stranger, sometimes of a spouse or
other family member, as Stein (2006) shows in cases culled from Federal
Bureau of Investigation files of violent crime. Fonagy and his collaborators
(2002) describe patients, incapable of mentalization, who are reduced to
protecting themselves from an outer world that, because the patient oper-
ates in the mode of psychic equivalence, seems as literally dangerous as
their inner world of fantasy. In other cases, as Fonagy et al. tell us simply,
“The child who recognizes the hatred or murderousness implied by the
parent’s acts of abuse is forced to see himself as worthless or unlovable”
(p. 353). If not-me is hated, worthless, unlovable, the outcome can be the
enactment of the perception that it is instead you (the other) who deserves
to be treated murderously.
It seems reasonable to say that relationships in defensive lockdown are

more likely than others to be cemented and maintained by enactments
of pathological dissociation, those in which the relationship must stay
exactly the same if the me’s of its participants are to remain safely iso-
lated from their not-me’s. Relationships in which the oedipal struggle is
alive and kicking (probably the majority), on the other hand, are at least
partly maintained by the security oἀered by habitual enactments of nor-
mal dissociation. It is primarily these relationships we would expect to be
most open to change, to giving up the old ways, because breaking up their
enactments should lead to nothing more dire than discomfort.
Locked Out of Relationships
But what about the people who can’t get into relationships in the first
place, people who start relationships with great excitement and passion,
but who can sustain those feelings only temporarily? What about Stephen
Mitchell’s (2002) case of Brett, for example, who was the immediate impe-
tus for the original version of these remarks (D. B. Stern, 2006)? I have
seen a number of people like Brett, who, at some point in the first few
months of a relationship, typically lose their passion for their lovers and
then begin to want to see other prospective lovers, who are often emo-
tionally unavailable. Typically, the first relationship then ends, sometimes
with great guilt on the part of the restless partner, who is tortured by the
unmistakable pattern, but helpless to feel otherwise.
With the background of dissociation theory, I would now say that these
are not actually people who cannot get into relationships. People like Brett
get into them all the time! The problem, rather, is that the Bretts of this
world can allow relationships to take only one course: They are stuck in a
rigid pattern with their partners, and the particular nature of this pattern, in
which, for one reason or another (diἀering with the person concerned), inti-
macy threatens the eruption of not-me, makes relationships too dangerous
to maintain once a certain level of intimacy is reached. (I turn to an example
of such a pattern in a moment.) Brett was so distressed and helpless about
this kind of problem that it was his primary motive to enter treatment.
That degree of distress and that sense of helplessness about the prob-
lem are not uncommon among people who know they cannot tolerate
romantic intimacy; and so we can speculate with a degree of confidence
that what lies behind this relationship pattern is not the avoidance of bad-
me but rather the avoidance of not-me. If individuals with this kind of
problem were able to bear experiencing whatever stops them from taking
relationships any further than they do, they would have done that, even if
it caused discomfort, because they very often badly want the problem to
change. If, in terminating relationships at such an early point, they were
turning away from experience that belongs to the self (bad-me), it seems
likely that their motivation to change the problem would have allowed
them to feel and own both sides of the conflict. They would have been able
to disrupt the enactments that had maintained the dissociations, because
the worst consequence would have been the experience of parts of them-
selves that they did not like. The fact that these people cannot seem to
manage to disrupt the enactment, though, as badly as they often wish to
do so, encourages us to believe that it is something more than bad-me that
feels more dreadful and that they are keeping at bay, something that does
not feel to them like themselves at all.
Perhaps, for the sake of illustration (I will use the example of a het-
erosexual man), a fictional patient’s mother was intensely exciting to
her little boy, but felt so terrorized by her own loving feelings that our
patient’s infantile expressions of love and his need for her love for him
simply fell outside her range of perception, oἀ the radar. The patient’s
mother, that is, dissociates tender feeling. We could say that our patient
has a version of what André Green (1976) calls a “dead mother,” one in
whose eyes and mind the baby is unable to find himself. The mother’s dis-
sociation of loving feelings, and her enactment of the same deadness to
love that she herself (let us say) experienced as a baby and child, provokes
our patient’s own dissociation. The patient feels unlovable—not hated, as
in Fonagy’s case, but not answered with love either. No child can grasp
his love and his need for it in the absence of a response to it. To feel love
for someone else becomes humiliating because one anticipates that one’s
needs will not be responded to; and to want the love of the other is debas-
ing for the same reason. Over time, the child becomes as dead to love as
the mother.
Our patient’s loving feelings and his desire to be loved become not-me.
As an adult, he perpetually looks for his exciting mother, and he finds
her over and over in his lustful adventures; but these new mother-women
face him with the dilemma that they often can love, and they provoke
our patient’s inchoate aἀection for them. These loving women awaken the
patient’s desire for the return of this feeling, for what his mother could nei-
ther experience in her own life nor oἀer to him. But the long-dead desire,
if the patient found himself experiencing it now, would fall into the realm
of the uncanny—dreadful, frightening, or even loathsome.
As soon as this patient begins to feel, probably without awareness,

slight stirrings of aἀection for the women he meets, or the whiἀ of a desire
for their love of him, he loses interest, begins to characterize his women
friends as dull, and feels like moving on. His growing feeling that they are
dull and boring dampens whatever aἀection they might have been in the
process of developing, thereby closing the circle and giving him that much
more reason to be afraid of arousing love the next time around.
If this man is to find his way into a relationship that lasts, one with
mutual love, it seems likely that it will only be through the repeated appear-
ance and transcendence of just this kind of enactment between him and
his analyst. Such an accomplishment would probably require the patient
to experience dread and at least the danger of humiliation (and perhaps
the real thing); and in the course of the enactments that would have to
take place, the analyst would no doubt be the object of feelings ranging
from deadness to rejection to contempt. In a treatment of great success,
the patient would eventually become more able to experience simultane-
ously the states in which he loves and the states in which he fears loving,
thereby tolerating a conscious internal conflict over how he feels and what
he will do about it. But of course there is always the very real danger that
the enactments will either result in the treatment ending or that the treat-
ment will endure, but with the deadness that the patient experiences with
his ex-girlfriends. In such an outcome, in other words, love and the desire
for love would remain not-me. And we can hardly blame the patient for
that. Until the conflict is tolerable to him, no other outcome makes emo-
tional sense.
The only diἀerence between our fictional patient and the unfortunate
souls I referred to earlier who maintain relationships of the most hateful
and destructive kinds is that the enactments our fictional patient gets into
do not lock him into relationships; they end relationships. Remember the
question I set for myself at the beginning of this chapter: How can we
think, in two-person terms, about both kinds of problems people have in
making love last: (1) the rigidities of those people who cannot find their
ways into relationships, and (2) the changing issues between couples who
are already in them? In answer to this question, we can link these two
kinds of problems through dissociation and enactment: The problems in
basically loving and satisfying relationships often derive from the enact-
ment of mutual dissociations of good-me and bad-me; but problems such
as my fictional patient’s problem, and the problems of those damaging
relationships that seem to be maintained on some other basis than love,
are likely to be the outcomes of the enactment of mutual pathological
dissociation, or dissociation in the strong sense—that is, the mutual dis-

sociative sequestering of me and not-me.
One final caveat, however. I do not want to leave the impression that
in long-term relationships the healing of pathology is the only route to a
broadening or deepening of the relationship over time. There is also such
a thing as courage: courage to challenge the status quo in the interest
of “more life” (Corbett, 2001), courage to formulate and think about the
stories we are always implicitly telling ourselves about our lives, cour-
age to imagine all those other stories about our partners and ourselves
that, merely because we are so attached to the stories we do tell, we
are continuously in the process of occluding. The danger accompanying
this kind of open and accepting experiencing is the danger of sacrificing
security; and it is true more often than not that, over time, security does
trump desire. But it is not always that way. And when it is not, when
freedom and desire win, it is because we are willing to put ourselves at
risk by experiencing more readily and deeply than before what we do
not like about ourselves—or, in the most significant instances, what we
have not until that uncomfortable moment even been able to imagine
is ourselves.
8
On Having to Find What You
Don’t Know How to Look For
Two Views of Reflective Function
One of the most influential views of reflective function in the history of

psychoanalysis, and clearly the most influential view of that subject in the
current literature, is that of Fonagy and his collaborators (2002). Their
work has reignited interest in reflection and reinvigorated its study. Since
the beginnings of our field, self-reflection has lost a great deal of ground to
relational eἀects in our theories of therapeutic action. That is as it should
be, and Fonagy et al. would be the first to agree. In fact, perhaps their
most significant theoretical contribution has been their detailed proposal
that the development of mind, including reflective function, is itself thor-
oughly relational. Reflective function is a relational event, with relational
roots. As a result of this work, reflection is no longer a dusty subject on
the back of the shelf. It is now as fascinating as it was long ago, and it is
back in the center of psychoanalytic interest. The work of Fonagy et al.
has deeply aἀected my own work, a fact reflected in the frequent citations
of mentalization theory in this book. Their description of the process of
mentalization and their observations about how it develops are with me
whenever I sit down with patients. I use their observations and their con-
ceptions every single day.
I, too, have devoted a good deal of attention to the place of reflective
function in clinical work.1 My interest in reflection has focused on the for-
mulation of newly explicit experience from its unformulated origins. Like
Fonagy et al., I have understood the process of formulation as a relational
event. But unlike Fonagy et al., and, for that matter, unlike most writers on
1 These writings have appeared over the past 25 years. The first of these, from the early 1980s into the
early 1990s, were revised and collected in my earlier book (D. B. Stern, 1997). Other articles have
appeared since that book and are reprinted as chapters in this book.
161
the subject, I believe we do not and cannot control reflection, at least not
the instances of reflection that matter the most clinically. The important
reflections in a treatment come upon us; we do not have the simple choice
to create them. They arrive of their own accord.1 I shall explore one aspect
of my views in this chapter and compare those views with the perspective
oἀered by Fonagy et al.
Socrates and the Inebriate
I begin, though, far from psychoanalysis, with Plato’s (1956) famous report
of Socrates’ dialogue with Meno, often known as Meno’s Paradox.
Meno: But how will you look for something when you don’t in the least know
what it is? How on earth are you going to set up something you don’t know as the
object of your search? To put it another way, even if you come right up against it,
how will you know that what you have found is the thing you didn’t know?
Socrates: I know what you mean. Do you realize that what you are bring-
ing up is the trick argument that a man cannot try to discover either what he
knows or what he does not know? He would not seek what he knows, for since
he knows it there is not need of the inquiry, nor what he does not know, for in
that case he does not even know what he is to look for.
Meno: Well, do you think it is a good argument? (pp. 128–129)
Socrates says he does not. Neither do psychoanalysts, who search all the
time for what they do not know how to describe. (The fact that we do not
know what to look for has never stopped us!)
Socrates’ reasons for adopting this view are not literally relevant to
modern readers, although the dialogue, interpreted metaphorically, is
often cited in discussions of the contributions of memory to thought in the
present. (Socrates’ argument centers on certain consequences for pedagogy
that follow from the doctrine that souls are immortal and spend eternity
being reborn into one new life after another.) But whatever Socrates’ rea-
sons for it, his conclusion is the same one that would be reached by most
readers of this chapter: We do somehow manage to find what we don’t
know how to look for. Of course, the conclusion is the easy part. The trick
is how to do it and how to think about it.
Here we can take a hint from a much less lofty source: The venerable
joke about the man who has had too much to drink and who, on his way
1 I have given these points particular emphasis in Chapter 12 of Unfomulated Experience (D. B.
Stern, 1997) and in thiis book, Chapter 5.
On Having to Find What You Don’t Know How to Look For 163
home in the middle of the night, accidentally drops his house key in the
street. A passerby comes upon the man, who is on his hands and knees
searching under a streetlight. The passerby begins to help look. After a
few minutes without success, the Good Samaritan asks whether our main
character can be any more specific about where he might have dropped
the key. The man points oἀ into the darkness. Astonished, the passerby
asks why, in that case, the two of them aren’t searching over there; and the
answer, delivered with a roll of the eyes, as if it ought to be totally obvious,
is, “Well, that wouldn’t do any good, would it? It’s dark over there.”
And so I begin with two claims, both of which I think are familiar
to all clinicians: (1) In psychoanalysis and psychotherapy, although often
enough we do eventually find what we most need, for long periods we have
no idea what that is or even how to recognize it; and (2) This means that
we have to learn how to search where we are least able to see—in the dark
places. I would go so far as to say that to be a psychoanalyst is to love the
dark places and the recalcitrant meanings that lurk there—to love even the
way that those meanings evade us. Of course, we, like those with whom
we work, are often tempted to avoid the dark. But with another part of
ourselves, we also develop a certain calm in the dark—a comfort with the
discomfort, if you will, an acceptance—because we have come to have the
confidence that what we need will eventually somehow emerge from this
part of our experience.
By “dark places” I mean to refer to those parts of experience we can-
not access. It is not just that we cannot understand darkened meanings.
Unlike the unfortunate man searching for his keys, who knew what he was
looking for, psychoanalysts and other psychotherapists often have to wait
long periods to learn what it is we seek. Our problem is therefore more dif-
ficult than being faced with an impenetrable meaning. The experience we
need is actually absent. In the most literal sense, we really don’t know what
we’re looking for. And yet we try to keep ourselves continuously ready to
find it.
Let me put this in the terms of reflective function: We have the con-
fidence that our capacity to think about ourselves and other people will
eventually come in very handy; but most of the time, before the fact, we
just cannot imagine when and how the opportunity to think productively
in this way is likely to occur. And even when the opportunity arises, it
does not arise planfully. At such moments, we are not able to decide to
think; instead, we find that we have already thought. This position, I think,
may be diἀerent from the position taken about reflective function by most
contemporary analysts, including Fonagy et al., in whose thinking the
analyst generally seems able to call at will upon her powers of reflection. I
will have to cover a good deal of preparatory ground before I will be in a
position to discuss this diἀerence. I will return to it later, after introducing
mentalization theory.
Mentalization Theory and Relational

Dissociation Theory: Commonalities
In their important 2002 book, Fonagy et al. tell us that it is their inten-
tion to link psychic development and philosophy of mind, “to capture and
specify the process by which infants fathom the minds of others and even-
tually their own minds.” That is, “we fathom ourselves through others” (p.
2). Mind is created, not inborn, and it is created by the interaction of the
baby with the parenting ones. Mind is an interpersonal construction, a
product of a certain very particular kind of reflected appraisal.
This is music to my ears, with my roots in the interpersonal develop-
mental thinking of Harry Stack Sullivan. As a matter of fact, the account
by Fonagy et al. is remarkably similar to Sullivan’s; and where it diἀers,
it frequently could serve as a creative specification and elaboration of
Sullivan’s thinking.
There are many other areas where the work of Fonagy et al. overlaps
with my own views. They argue for the continuing centrality of verbal
meanings in psychoanalysis, for instance, while simultaneously making it
abundantly clear that a large proportion of the most important events in
psychoanalysis takes place outside the reach of verbal language and may
never be brought within it. This is also a position I favor (see especially
Chapter 4). We all also agree that mind or self is inevitably fragmented.
Bromberg (1998, 2006; Chefetz & Bromberg, 2004) and I (D. B. Stern, 1997;
see also Chapters 6–10 of this book) each express that point with an updated
version of Sullivan’s (1954) personifications, me and not-me. Fonagy et al.
prefer a diἀerent set of terms: “Self,” which refers to the part of one’s sense
of who one is that, because it connects to inborn potentials, is authentic and
real; and “alien self,” a sense of oneself as “other” that comes about because
it is not connected to the inborn potentials of the constitutional self. The
basic affinity of our two perspectives is unmistakable.
But I also may diἀer with mentalization theory. This chapter was writ-
ten partly to think through the question of whether those diἀerences are
real, and if they are, how to characterize them. I am going to take the
view that it is at precisely the moments when the analyst is most in need
of her capacity to mentalize that she has least access to it. These moments
are mutual enactments, and Fonagy et al. agree with me that the analyst’s
ability to mentalize is often compromised at such times. What we may
or may not agree about is: (1) the degree to which the analyst’s capacity
to mentalize is absent at such times; and (2) the eventual clinical value of
these mutual unconscious entanglements with our patients.
Fonagy et al. are highly sympathetic to the plight of the analyst who
participates in an enactment, and they accept that some enactments, even
rather distressing ones, are unavoidable, especially when working with
borderline patients. But in my reading of their work, these writers also
believe that mutual enactments are problematic and regrettable. For me,
on the other hand, while the potentially damaging aspects of the analyst’s
unconsciously motivated participation are obviously regrettable, it is also
the case that enactment is not only inevitable, but also often constitutes
the therapist’s most important clinical opportunity. Before I present my
own view, though, I will shape what I think is the relevant underlying
issue here as a problem in therapeutic action, and then present my under-
standing of the view oἀered by the authors of mentalization theory.
Chicken-and-Egg in Therapeutic Action
We have known for many decades that it is too simple to ask whether
therapeutic action is based on insight (or reflective function) or relational
eἀects, because the two ingredients interact with each other. For the pres-
ent purpose, I will consider the issue a chicken-and-egg dilemma. Like all
such dilemmas, this one has no absolute solution. But there are diἀerences
of emphasis in the two approaches to the problem, and they are important.
Here is the dilemma: Does mentalization pave the way for a shift in the
relationship, or must relational effects precede mentalization? I begin the
exploration of this problem by summarizing what Fonagy et al. tell us, in
their 2002 book, about the role of mentalization in therapeutic action.
Therapeutic Action in Mentalization Theory
The authors write that “An important aim of psychotherapy, if not its cen-
tral aim, is the extension of mentalization” (p. 14). This aim is achieved in
a number of ways, but all of them depend on the analyst’s own capacity to
mentalize. It is the analyst’s more or less continuous capacity to mentalize,
say Fonagy et al., that eventually has the eἀect, via internalization, of
extending the patient’s capacity to mentalize. The analyst’s mentalization
is available to the patient whenever the analyst reflects on the patient’s
experience, or even when the analyst merely labels the patient’s experience
or describes it. For some patients, such moments are among the first times
they have thought about what they think or feel. These are people for whom
it is not habitual, at least under certain circumstances, to reflect on their
own experience or the experience of others. Assuming that the patient
experiences the analyst as trying to be useful or helpful, it sometimes does
not even matter if the analyst’s reflection is wrong. These instances, like
those other moments when the analyst is right, show the patient what it is
like to treat experience, especially the experience of aἀects and intentions,
as symbolic representations, not as concrete or literal copies of the world.
A symbolic representation is understood to be a subjective rendering, and
therefore open to modification; an experience understood as a literal copy
of the world, on the other hand, is a direct avenue to the truth that can be
modified only by lies or distortion.
The patient also takes in the analyst’s reflective function when the ana-
lyst deals with the patient on the basis of the analyst’s mentalization of her
own experience, as in, “Oh, now I see. I thought you meant ‘X,’ but now I
see you meant ‘Y’.” In saying things like this, the therapist is once again
showing the patient that she (the therapist) is treating experience as a rep-
resentation. Experience can change on its own without a corresponding
change in the world; it can be negotiated. Each time the analyst mentalizes
with the patient, the patient is just a bit more able to treat his own experi-
ence that way, as well as the experience of others in his world. He is more
often able to transcend the mode of psychic equivalence, in which experi-
ence is perceived and treated as a slavish copy of the outside, and the out-
side as a copy of the inside. In psychic equivalence, the inside of oneself and
the outside, experience and world, are identical, so that frightening fanta-
sies are just as threatening as frightening realities, and frightening external
situations cannot be diἀerentiated from the state of one’s own mind.
With the capacity for mentalization, experience is no longer a photo-
copy, but a picture of the world, a picture that we know could have been
constructed diἀerently, even if we may not have immediate access to alter-
natives. Patients who can mentalize, for example, know that their percep-
tions of their analysts, however ironclad those perceptions may seem, may
very well not be the only ways their analysts could be seen. Patients in the
mode of psychic equivalence, on the other hand, have what we used to
call “psychotic transferences”: As far as these people are concerned, their
analysts simply are what they are perceived to be. There is no ability to
create the “as-if” quality.
Once we understand that we can make many diἀerent pictures of the
same reality, and that all of them have their own validity, the inner and
the outer worlds become more and more independent of each other. As the
process goes on, experience less frequently feels to the patient as if it were
merely imposed on him. He feels less helpless and victimized, and also less
omnipotent. His sense of his own limits, the limits of others, and most of
all, his sense of agency, of living his own life, are facilitated, and he real-
izes that life can be chosen, rejected, negotiated, and discussed much more
often than he had understood.
Mentalization, or reflective function, is an undeniably significant part
of therapeutic action. I don’t imagine anyone believes otherwise. The ques-
tion I want to raise is not whether reflective function is important; rather,
I want to focus on when analysts are capable of responding this way, and
why. Can analysts always mentalize?
We cannot; everyone also agrees on this point. When analysts are
blindly involved in enactments, they can mentalize neither their own
experience nor their patients’. The term mutual enactment might just as
well be defined, as a matter of fact, as the mutual inability of patient and
therapist to mentalize.
During an enactment, from the perspective of mentalization theory,
because the analyst reacts directly to the patient’s projective identification,
the patient’s expectation that the inside and the outside worlds must be the
same—that is, psychic equivalence—is reinforced. If, on the other hand,
the analyst is able to accept the patient’s projective identification without
reacting directly to it, and especially if the analyst can convert it into a men-
talization that can then be given back to the patient, a mutual enactment
does not take place. (In my own way of thinking, an enactment takes place,
but it is not mutual.) The patient is encouraged to move just a bit further
from psychic equivalence toward the beginning of reflective function.
Now, what I am about to say is crucial to the point I am trying to reach:
In this theoretical scheme, no matter how understandable mutual enact-
ments may be, and no matter how sympathetic we are to the analyst’s role
in them (and Fonagy et al. are very sympathetic), they are regrettable to
the extent that they interrupt the analyst’s capacity to mentalize.
Let me oἀer illustrations of these views, in two brief quotations from
the 2002 book.
[I]t is necessary to accept that in order for the patient to stay in mental proxim-
ity, the therapist must occasionally allow herself to become the vehicle for the
alien part within his self. If she is to be of any use to him, she has to become what
he needs her to be. Yet if she becomes that person fully, she can be of no help
to him. She aims for a state of equipoise between these two positions—allowing
herself to do as required yet trying to retain in her mind as clear and coherent an
image of the state of his mind as she is able to achieve. (p. 370; emphasis added)
The therapist is portrayed to be in control of the enactment: she “must

occasionally allow herself to become the vehicle for the alien part within
his self.” This is not something that is portrayed to be happening to the
analyst; she “allows” it. It seems to be assumed in this passage that the
therapist has the ability to select the clinical stance toward the patient that
she thinks is best. In this “state of equipoise,” she allows “herself to do as
required yet [tries] to retain in her mind” a clear and coherent image of
the mind of the patient.
And now consider this passage:
Some enactments on the part of both therapist and patient are an inevitable
part of this work, since the patient cannot remain psychologically close to the
therapist without externalizing the alien parts of the self. It is at these moments,
when the therapist is enacting the split-oἀ parts of the patient’s experience, that
the patient’s true self may be most accurately observed. Sadly, almost invariably
the therapist will have extreme difficulty in communicating insight and under-
standing at such times. It is far more likely, in the middle of such turmoil, that
the therapist’s rage or terror or both will obscure her vision of the patient. (p. 14;
emphasis added)
In this description, the analyst’s participation in an enactment of the

patient’s alien self is clearly regrettable. Participating in an enactment,
however inevitable or unavoidable it may sometimes be, deprives the ana-
lyst of the tools she needs to be useful to the patient.
There exist in our field two broad ways of understanding the relation-
ship of mind and clinical interaction. In the first, older perspective, held
by many contemporary Freudian and Kleinian analysts, but also by some
Interpersonalists, there are parts of the mind, especially the analyst’s
mind, that usually (though not always) exist apart from the unconscious
influence of the mind of the other. In the analyst’s case, there exists an
internal “perch” (Modell, 1991) from which the analyst can expect to be
able to observe the interaction more or less objectively. This certainly does
not mean that the analyst is not aἀected by the countertransference. But
even when she is aἀectively moved by the transference, she should usually
be able to observe her own disturbed aἀective state from this other part
of her own mind, and then use this disturbance as vital information in
shaping her understanding and interventions. The analyst, that is, should
usually be able to observe and reflect not only on the transference, but also
the countertransference.
It should go without saying that many analysts who work from some
variety of this position are nevertheless quite tolerant of those times when
the analyst’s capacity to mentalize her own countertransference fails. These
analysts know that, at times, they will be blindly involved in enactments—
especially, of course, with more disturbed patients. Fonagy et al. are at the
head of this humane group. Their writing is replete with examples of enact-
ments that are presented in a way deeply sympathetic to therapists who find
themselves participating in interactions shaped by various disavowed aspects
of their patients’ minds. Along with many others, Fonagy et al. understand
these enactments to be the eἀects of the patient’s projective identifications, a
familiar defense employed periodically by almost everyone, but the habitual
defense of those with borderline personality organization.
Mentalization theory adds a detailed and original understanding of
how the predilection for projective identification begins and what sustains
it. I find especially compelling the descriptions of how parental aἀect mir-
roring can go awry and the dire consequences that ensue: the absence of
a firm boundary between self and other; the intolerable and continuous
threat of being dominated by the alien self; and the inability to construct
the secondary representations required to process experience, especially
aἀect, to the depth at which it would be manageable (that is, the inter-
ruption of capacity to mentalize). When people can’t mentalize, it means
they can’t convert threatening experience—again, especially aἀect—into
a form in which it can be tolerated and thought about. If you can’t think
about it, the only solution is to get rid of it, and projective identification
is the way to do that. In the face of this onslaught, this attempt by the
patient to lodge unwanted aspects of his mind in the therapist’s mind, the
therapist is understandably drawn into reacting, often quite dramatically.
Today we know to encourage therapists who are otherwise capable and
committed to their work not to feel too self-punitive about their participa-
tion in these episodes.
But note that, no matter how accepting the attitude about being drawn
into enactments may be, from this point of view such enactments must
be understood as failures—entirely understandable failures, yes, but fail-
ures nonetheless. They must be seen as failures because the analyst has
not been able to use the part of his mind that, in a better or less challeng-
ing world, could have been used to maintain his capacity for more or less
objective observation.
Now hearken back to the questions I asked a few minutes ago: “Does
reflective function pave the way for relational change? Or must relational
eἀects precede reflective function?” The position Fonagy et al. emphasize
seems to be that mentalization comes first. The analyst must be able to men-
talize if the patient is to relate diἀerently, because a change in relatedness
requires the new depth of processing that only mentalization provides. We
might also state this position by saying that, in concert with the way psy-
choanalysis is usually written, internalization (in this case, internalization
of the analyst’s ability to mentalize) precedes behavioral change.
In making these points, I certainly do not mean to deny that there is
plenty of room in mentalization work for the productive use of enact-
ments. If a patient stays in treatment following a mutual emotional storm,
the mentalization of the experience behind the enactment, especially the
aἀect, would be the most important and immediate clinical goal for any
analytic therapist. The point I want to emphasize is that the therapist’s
unconscious (relatively) uncontrolled aἀective involvement with the
patient, very often reflected in less than fully controlled clinical conduct, is
not itself understood to be productive and desirable. It may be necessary,
but it is a necessary misfortune, not an opportunity.
Therapeutic Action in Relational Theory
I mentioned above that there exist in our field two ways of understand-
ing the relationship of mind and clinical interaction. You already may
have guessed that, in my view, Relational theory is the alternative under-
standing. From most Relational perspectives, the analyst is continuously
involved in an unconscious way with the patient. There is no part of the
mind that exists apart from the influence of the other. Every one of the
analyst’s thoughts and feelings, and all of her conduct, are constructed
by her and are therefore personal and open to the patient’s unconscious
influence. Even those parts of the analyst’s conduct that she selects or
constructs on the basis of a theory of technique are personal, because of
course every choice could have been made otherwise, and no theory of
technique prescribes what we actually do from one moment to the next
in any more than a very loose way. Whatever content is being discussed is
also being played out in some way in the therapeutic relatedness; and this
means that, if analysts assume that their interpretations are nothing more
than what they appear to be at face value, they are liable to be participat-
ing in ways that do not illuminate the transference but reinforce it instead.
And so, without realizing it, it happens that the analyst of a masochis-
tic patient makes sadistic interpretations of masochism; the analyst of a
seductive patient who is nevertheless fearful of sexuality makes seductive
interpretations of the patient’s fear; the analyst of a narcissistically vulner-
able patient interprets the narcissism in a way that wounds the patient’s
self-regard. (This seminal observation was first made by Edgar Levenson
[1972, 1983].)
But this is only half the point. Just as inevitably as the patient uncon-
sciously aἀects the analyst, the analyst unconsciously influences the
patient. From this point of view, the relationship is understood to take
place between two unconsciously intertwined subjectivities. There is no
refuge in the analyst’s mind from the patient’s influence, no socially iso-
lated perch from which to observe.
One point that becomes immediately apparent for analysts working
from this perspective is that the motivation for mentalizing with patients
is liable to be just as complicated as the motivation for any other interven-
tion and just as likely to be part of mutual unconscious process, visible or
not. Telling a patient he seems angry, sad, or pleased about something, for
instance, can serve any number of purposes, many of them the analyst’s
own purposes, developed at least partially, and unconsciously, in reaction
to the patient. And it is not only the content of the analyst’s reflection that
may be an unconscious participation in an ongoing enactment, of course,
it’s even more likely to be the manner in which the content is delivered,
which may occur entirely outside the analyst’s awareness. Or, for whatever
reason, the analyst’s perception of his manner may be quite diἀerent from
the patient’s. And when it comes to mentalizing one’s own reactions and
sharing the process with the patient, even if one’s reactions seem innocu-
ous at face value, the situation is even more fraught with complication.
Think of my earlier example: The analyst says to the patient, “Oh, I thought
you meant Y, but now I see you meant X.” Innocuous at face value, isn’t
it? But can’t we imagine these words taking on many diἀerent emotional
resonances, depending on the context within which they are spoken and
the expressive qualities with which they are given voice? There may be a
corrective to the patient embedded in what is meant to pass as acceptance;
or the analyst’s tone of voice may be subtly patronizing, signifying that
“X” is a less interesting meaning than “Y” would have been. And so on.
The exercise of the analyst’s reflective function can never be assumed to be
a simple or straightforward matter.
I should add, too, that, although it is certainly the case that these ongo-
ing mutual enactments are liable to be most dramatic and disturbing in
the treatment of borderline patients, I do not mean my remarks to be spe-
cific to work with those patients. What I am saying is meant to apply to all
treatments. It is likely that the clinical difficulty for the therapist, and her
level of discomfort, declines with milder psychopathology, but the prin-
ciples are the same. The therapist can certainly hope, in other words, that
her interventions have a particular impact on the patient; but she can never
rely on her conscious intentions alone. She must always be asking herself
what else than what she knows might she be doing or saying? How else
might she be involved with the patient than she has considered? What role
might her seemingly neutral clinical conduct be playing in that involve-
ment? From a Relational perspective, it is as crucial for the therapist to be
continuously curious about the unconscious meanings in her own experi-
ence as it is for her to be curious about those in the patient’s.
The Interpersonal Field
The mutual influence I have described tends to lock the two participants
into certain reciprocal patterns of experience and conduct. It does not
matter to me what term we use for the field of influence comprised by
the interrelation of these two subjectivities. Sullivan’s (1940) term was
“the interpersonal field,” although he meant something slightly diἀer-
ent from what we mean today; Mitchell (1988) called it “the relational
matrix”; Wolstein (1959) referred to “the transference/countertransfer-
ence interlock.” There are others. Whichever term we choose, the point
is that the clinical situation is mutually constructed and conducted,
though asymmetrical (Aron, 1996). That is, while maintaining roles
that diἀer in expertise, clinical experience, power, and so on, patients
and therapists also relate in a mutual way, as two human beings. Even
if they wanted to, they could not do otherwise. For this reason, the
patient’s transference tends to draw the analyst into a certain kind of
countertransference; and the analyst’s transference has the same eἀect
on the patient. Sometimes, over time, these meaningfully related inter-
active patterns, when they are motivated by both parties’ dissociations,
become mutual enactments. Both parties play roles in maintaining rigid
patterns of this kind.
It follows from this point of view that the therapist’s expertise cannot be
rooted in invulnerability to the patient’s influence or in the ability to grasp
the nature of the patient’s influence and resist being transformed by it.
Sometimes therapists indeed are invulnerable, but that is not necessarily a
state to be desired. We need our vulnerabilities. And sometimes therapists
do sense the patient’s influence before responding to it. That, too, though,
is not inevitably a good thing, because it can sometimes deprive patients of
their habitual ways of making relationships, what Feiner (1982) has called
their “ticket of admission.”
The analytic expertise that allows us to practice over many decades
without burning out grows from a special kind of acceptance of our vul-
nerability and a consequent ability to observe how we have already been
influenced. To reach something new, we must usually find a way to tran-
scend or deconstruct the old; and to do that requires, often or even usually,
that, without our awareness, we create the old in the therapeutic related-
ness and then make it visible, available for reflection by both patient and
analyst. To articulate the old is to create the new. Like water rushing in
to fill an empty space, the new comes about by itself, in a way we may not
even be able to imagine, when room is made for it by our reflection on the
old (see the theory of narrative proposed in Chapter 5).
If this is so, then the most important question is what happens to make
it possible to reflect on the old, on what has been invisible. What makes it
possible, that is, for the new to leap into being? Whatever this is, it is seldom
reflection. Reflection is usually an outcome. By the time you can reflect on
the old, it has already lost its bite. New reflection opens up other areas for
consideration and change, and in that way the process of change proceeds
outward, in ripples. It is often, but not always, the case that taking advan-
tage of the therapeutic potential of new perceptions requires their men-
talization. But the process usually begins in something that comes before
reflection. The process begins in something that makes reflection possible.
My candidate for this “something” is the development of a new percep-
tion of the other (see Chapters 3–6). I am not alone in taking this position
(see Bromberg, 1998, 2006). The new perception in question may be the
analyst’s new perception of the patient, or it may the patient’s new percep-
tion of the analyst. That happens, too. Most often, as in the example I will
oἀer you in a moment, it is difficult to say who had the new perception
first. But whatever the origin of a new perception of the other, once you
have it, especially when the aἀective tone of that perception is diἀerent
from the tone of the previous relatedness, you can’t help relating to the
other in a new way. You couldn’t relate in the old way if you tried—at least,
not authentically. Among other freedoms that become available at such a
moment is the freedom to mentalize what is newly visible and therefore
first accessible to mentalization. The new perception itself remains to be

mentalized, of course; but it is also now possible to reflect on the unseen
bonds that had constricted the therapeutic relatedness into the patterns of
the preceding enactment.
Although I do believe that the “new perception” is key to the dissolu-
tion of enactments, I do not want to leave the impression that I believe
insight to be inconsequential or nothing more than an epiphenomenon.
Insight about a new perception necessarily follows the events it describes;
but once understanding does arise, it is involved in the creation of the next
generation of clinical events, which would have been diἀerent if the new
understanding had not come about in the first place. And so, in the end,
reflective function and new perception have continuous, reciprocal parts
to play in therapeutic action.
Illustration
I had met with a very bright middle-aged professional man three times
a week for a year and a half. Ron was enthusiastic about beginning an
analysis. He hoped to reignite his career, which had been spectacular, but
was now sputtering. Although I had expected to enjoy working with him,
shortly after we started I found myself dreading Ron’s sessions. He was
seldom openly angry or dissatisfied, but he frequently told me that my
skill was inadequate and that I was not warm enough in manner. He made
these observations very politely, but I felt the bite of them; and he treated
me in a way that conveyed his criticisms even when he wasn’t making
them. He told me repeatedly that his primary interest in life was to be as
generous and kind as possible to his wife and children. He also partici-
pated, apart from the family, in numerous personal and political activities.
I often admired the commitment behind many of these activities, familial
and extrafamilial alike; but what struck me most regularly and forcefully
about them was his presentation of each one as another evidence of his
virtue. There was a continuous and subtle suggestion that his virtue was
notable, especially when compared to others—and I felt that “others” in
this case certainly included me.
Any attempt on my part to point out things in his experience that might
complicate this picture (for example, less positive eἀects) was met with a
rueful smile and a demurral. Ron really wished he could identify what I
saw in his experience, he would say, but he just could not. In the mean-
time, his wife, who seemed quite unreasonable to me when Ron began
treatment, flew into daily, seemingly unprovoked rages at him. It was not
long before I thought I understood how she felt. Ron’s virtue, and espe-
cially his advertisement of it, most particularly his smiling “patience” in
the face of others’ frustration with him, was itself a provocation.
Ron’s critical observations of me were perfectly accurate, of course,
because my reaction to the way he treated me was to feel irritated with
him. I felt that way much of the time I was with him. I was often cool and
distant, and I was certainly unsuccessful in the pursuit of what he called
his “feelings,” which, as far as I could see, were limited to two registers:
warm, loving, and appreciative; or mournful and sad, a reaction to the
regrettable absence of warmth, love, and appreciation in others. I did not
know which I disliked more, him with me or me with him. And yet, of
course, I also felt that I should be finding a way of working with this state
of aἀairs. I felt interested in whatever I could learn about what was hap-
pening between us; but I also felt guilty and self-critical.
I knew that my patient’s reactions and my own were interlocked. But
as is usually the case, knowing this in the abstract was not enough to help
me out of the field’s grip. I talked with Ron many times about his dissatis-
faction with me and about the kind of relatedness that had come to exist
between us. I told him that we needed to find a new way to see what was
transpiring between us. Naturally, he asked how we should do that. I did
my best to answer the question, but nothing I said seemed to set us on a
new path. Ron did begin to tell me, though, that this kind of relatedness
had taken place elsewhere in his life, including in his relationship with his
wife. He usually dealt with it by ending the relationship, he said, although,
for various reasons, he had not done that in his marriage. He spoke often
about leaving treatment, but he agreed that it would be better not to do
that if instead we could find a way to come out the other side of what was
going on between us.
At least now we were sometimes agreeing that we had a mutual prob-
lem. Sometimes, Ron could accept that it was possible that his wife and I
were not merely unreasonable and persecutory. He was willing to go this
far, I think, because he saw that I took his distress seriously, even though
he also still felt that I was culpable. But we did not have a convincing way
to characterize this problem, which therefore remained mutual in only a
hypothetical way. It was clear to me that, without getting some traction in
real experience, even the glimmerings of collaboration between us about
the issue were not going to keep the treatment going for long.
The beginning of a break in the deadlock came about one day when Ron
was once again talking about his doubts about continuing in treatment.
This time, though, he did not say it with anger, but with regret. He looked
diἀerent to me, helpless, despairing. He said he did not want the same
thing to happen yet again: another relationship abandoned because of his
dissatisfaction with it. In response, I said something to him that I had
said more than once before: I told him I felt sure that our relationship was
somehow mirroring a significant relationship from his past, though I did
not know which one. But this time, responding (I think) to Ron’s changed
demeanor, I made the point more softly and with my own feelings of regret.
I added that I, too, must be involved in maintaining what was transpiring
between us, that I could not imagine how it could be otherwise.
Before I had spoken I had gauged that Ron would understand and
accept what I said as an expression of my regret, not as a statement of my
willingness to talk about my past. It turned out I was right. But when I say
that I “gauged” Ron’s response, I do not mean that I actually formulated
this meaning in my mind. Nor did I formulate until a few minutes later the
part of this moment that surprised me most: my confidence that I could
trust Ron not to use what I said as one more reason to criticize me. This
was a new perception of Ron, and I would not have had it unless he had
somehow treated me diἀerently, in some small way I could not specify.
And so it seems likely that he had also perceived me diἀerently. Something
between us had shifted.
At this point, Ron began telling me more about his relationship with
his father than I had yet heard. His father had been much more continu-
ously and severely critical, unnecessarily brutal really, than Ron had let
me know to that point. Ron told me that his hatred and fear of his father
had therefore been more intense, and more troubling to him, than he had
related before. Within a few minutes it was clear to me that my patient had
suἀered much more than I had understood.
Later, I asked Ron what he knew about why he had not told me more
about this aspect of his relationship with his father. He said he had not
been aware of hiding it, but that when it began to come out of him, he
realized that he had always been ashamed of having been treated this way,
and that it was his shame that had kept him silent about it. I was glad Ron
mentalized this shame because it was better coming from him. But if he
had not described it, I would have, because the feeling had been palpable
in the room.
It was not long before his shame and rage at having been humiliated by
his father could be linked to the way he felt I treated him and the way he
treated me. Having been so ashamed, he had wanted to shame me—and
he had—by making it clear that he was more decent and loving than I
was. He would not descend to his father’s overt sadism; but it was not long
before he appreciated the irony that his “goodness” was its own form of
sadism. For Ron, of course, I had been his father; and for me, he had been
a member of my own family. As is often the case in emotionally intense
enactments, our dissociations had been reciprocal.
This session and those that came after it indicated to me that the treat-
ment would probably not founder, and it did not. We moved on to Ron’s
longings for his father (and for me), longings that his rage and humiliation
had always obscured.
I am going to leave my account at this point, because my intention is
only to illustrate my contention that shifts in relatedness precede mental-
ization. I will not take up here the crucial question of why the new percep-
tions became available to Ron and me when they did. I have addressed that
matter in Chapters 3–6, though. It is a very knotty problem. I believe that,
while there are indeed some significant things that can be said about the
reasons for new perceptions, there is a mystery at the heart of the matter.
Seldom if ever do we really know exactly why new perceptions and the
interactive freedom they herald arrive when they do.
It is not clear to me whose new perception of the other came first. I sus-
pect, actually, that any attempt to establish priority is doomed to failure,
because you can follow the interaction back through any number of epi-
sodes, all of which led up to the present moment. There is no marker that
tells you that you have reached the source; and that is probably because the
events cannot realistically be said to have sprung from a single moment.
This is the best account I can oἀer: Ron would not have spoken to me with
the regret he showed if he were seeing me, at that moment, as the sadistic
father. I responded immediately to this new feeling on his part, and a new
perception of him also awakened in my mind. It would be true to say that,
prior to my new ways of seeing Ron, I wanted badly to be able to treat him
diἀerently. But that desire was born less of an empathic grasp of Ron’s
need for a diἀerent response than it was of the narcissistic injury I was
suἀering of being a bad analyst.
Once I perceived Ron diἀerently, whether I wanted to treat him diἀer-
ently or not was immaterial. You might say I was drawn to treat him dif-
ferently; or you might even say I had to. I certainly no longer would have
been able to treat him as I had just a few minutes earlier; nor could I have
rid myself of my sudden empathic appreciation of Ron’s situation, even if
I had wanted to—which, of course, I did not. This is the kind of moment
that inspired the title of this chapter: I found what I had not known how
to look for. I did not make it happen; I was taken by it. In my experience,
that is how all enactments end. We must put ourselves in the best position
we can, which we do by not giving up our attempt to do analytic work; but
after that, we can only wait for freedom to arrive.
Of course, only a small proportion of our opportunities for reflection
with our patients grow from the breach of a mutual enactment. It is nev-
ertheless worth noting that there are very often moments when it becomes
possible for us to say something new to our patients, something we could
not have said a few minutes before. Many of the best interpretations arise
in this way. I think the principle governing the appearance of novelty
in the experience of either the patient or the analyst is always the same,
whether we are talking about the new perceptions that break up intense
enactments or the everyday appearance in the analyst’s mind of reflections
she has not had before: To say something new requires that you be some-
thing new. Sometimes you must change considerably, as I had to with Ron;
and sometimes the changes are so subtle that we do not even notice them.
We are always in the position of having to find what we do not know how
to look for, even when we do not know it. And so I believe that I probably
diἀer with the writers of mentalization theory over the relative emphases
to place on the alternative answers to the question about therapeutic action
with which I began. To me, it more often seems that a shift in relatedness
opens the possibility for new mentalization rather than that mentaliza-
tion creates the opportunity for new relatedness. In the other set of terms
I have used, the point can be put this way: Internalization of the analyst’s
ability to mentalize seldom catalyzes relational change; more frequently, it
seems to me, a shift in relatedness allows the processing of the dissociated
aspects of the social environment to a new depth. But of course we must
also admit that clinical process can move in both these directions.
Was my unconscious involvement with Ron regrettable? Was there a
better way to learn about his humiliation and longing? As far as I am con-
cerned, this part of Ron’s experience might very well not have come to
light if it had not been enacted. I am not the least bit flippant about the
gravity of the issue of the analyst’s unconscious involvement. I feel quite
serious about it; I recognize that it can injure treatments—and patients.
But I also believe that continuous unconscious involvement is consistent
with some of the most significant kinds of analytic discipline. Keep in
mind that I was able to build and maintain enough of a collaboration with
Ron that he stayed until we were able to do the work I have described.
I wish pain on neither patients nor myself, and I do what I can, within
the limits of my job, to avoid arousing unnecessary pain and to relieve
it when it has become unavoidable. But emotional pain comes with the
territory. We all know this, but I think the point is especially clear if you
believe, as I do, that dissociated experience is unformulated experience.
Dissociated experience cannot be represented, consciously or uncon-
sciously; it has yet to attain a shape articulated or realized clearly enough
to allow representation. Mutual enactments are our only route of access to
these parts of our patients’ minds; as enactments end, the experience can
finally be formulated. The idea that enacted experience is unformulated
is what led me, at the beginning of this chapter, to claim not only that we
must find what we do not know how to look for, but that what we are look-
ing for does not exist until we find it.
The therapist has no recourse to being personally involved. Patients
cannot directly experience their dissociated parts, but their therapists can.
In fact, therapists have no choice but to experience their patient’s dissoci-
ated pain; that is what it means, from the therapist’s vantage point, to be
involved in a mutual enactment. Therapists must experience the version
of the patient’s pain that is brought to life in their own psyches. In the pro-
cess, therapists suἀer a good deal of discomfort. In fact, with clinical expe-
rience, we come to recognize what we can sense about such discomforts as
opportunities, as what I have referred to in Chapters 3 and 4 as “emotional
snags and chafing” and as “intimations of freedom.”
We must respect the danger posed by mutual enactments. I know that
things with Ron could have gone diἀerently; he could have left treatment
in despair. In situations such as this one, we work on the razor’s edge. This
is one time that the old saw is true: If enactments don’t kill you, they make
you stronger.
I hope I have managed to convey that the most important instances of
reflective function inevitably surprise us. The parts of life we can already
point at, that lack only a name, are not the places of mystery within which
the most significant change occurs. The mentalization of experience is
unquestionably critical, especially with sicker patients. But to name a
part of experience that you already know is an operation that you can
tell someone else how to carry out. It is basically a technical procedure:
Name it, so that the patient begins to be able to represent it symbolically.
Clinically speaking, if you can already point at something, if you know it
is there, then naming it is not a particularly creative act. Naming experi-
ences we already know is not what excites us about our work. We eventu-
ally find what we most need in what we begin by not knowing how to
experience. We have to find a part of our subjectivity that was not even a
presence in our minds until it was created in unconscious reciprocity with

the patient.
Coda
What exactly can we say, then, about the significance of reflective func-
tion? What I believe is this: Socrates was right, although he was right for
the wrong reasons. We psychoanalysts often do find what we do not know
how to look for. And the unfortunate inebriate was wrong: We do not waste
our time looking only in the light. We try to locate the places of mystery,
the parts of our relatedness with the patient that we just cannot seem to
find words for, that (to begin with) we may not even be able to shape into a
coherent set of events; and we do that even when, or even especially when,
those parts of relatedness are uncomfortable. We immerse ourselves in
this discomfort, feeling it so thoroughly that we know it from the inside.
We try to deny ourselves no part of it; we try to allow ourselves to become
saturated with its eἀects. And then, eventually, at least when we are for-
tunate, we find ourselves surprised by a new perception of the other, of
ourselves, of the two of us together. Something has released, and we are
both more free than we were the moment before.
To my mind, this is not what most psychoanalytic writers mean by
insight—that is, I do not believe that this is what such writers usually
understand to be the way reflective function is used in treatment. Does
that group of writers include Fonagy et al.? I started out, as I wrote this
chapter, believing that the answer to that question was probably yes. That
is, I believed that Fonagy et al. held a diἀerent view of insight than I do,
and that is the conclusion that I have oἀered.
But after discussions with both Fonagy and Target at the conference
on mentalization at which this chapter was first presented as a paper, I
am less sure of the answer to the question. Fonagy and Target each told
me, independently, that they saw their views and mine as consonant. They
both said they found nothing to disagree with in my position that it is
a new perception of the patient by the analyst, or of the analyst by the
patient, that eventually makes it possible to mentalize the frozen events of
a mutual unconscious enactment. Fonagy did emphasize, though, that we
must continue to think about enactments as problematic, and therefore
unwelcome.
We cannot deny that enactments can be problematic, even destructive;

and it is certainly true that the repair of the relationship after an enact-
ment, even if that repair is a crucial therapeutic result, implies that we
(the analyst) have participated in damaging it; and damage is obviously
not what we intend to accomplish. And so we must say, from this point of
view, that enactments are unwelcome.
On the other hand, the material represented via enactment very often
simply cannot be accessed or represented in any other way. I have already
made the points I will oἀer in the remainder of this paragraph and the one
to follow; but it is worth giving these matters particular emphasis, because
they are the heart of the matter. Dissociated experience is unformulated. It
is a global, nonideational aἀective state. It is not until this material appears
in the relational world as enactment that it even becomes possible to men-
talize it; and often enough, of course, mentalization remains prohibitively
difficult under even these more favorable circumstances.
Dissociated material in the patient’s mind frequently first “shows
up” or “becomes visible” via the analyst’s participation in enactments,
participation that begins as unconscious. The analyst’s participation
in enactments is a significant part of what Sandler (1976) called the
analyst’s free-floating “role responsiveness.” The analyst must partici-
pate in this deeply affective, nonconscious way, and then struggle, over
time, to grasp the nature of that involvement. And even then—even
when it becomes possible for the analyst to grasp his own involve-
ment—the analyst cannot consciously choose to engage in the think-
ing that makes that grasp possible. Instead, the thinking arrives in the
analyst’s mind. It is unbidden. That is the only route to the mentaliza-
tion of dissociated material.
It is from within this perspective that I welcome enactments. The free-
dom we are seeking requires us to plunge into the maelstrom. And so,
while we suἀer real pain and sometimes racking self-criticism over the
myriad ways that enactment compromises us as analysts, we also welcome
the opportunities that the necessity to suἀer this pain brings us.
If there is a genuine diἀerence between my views and mentalization
theory, it has to do with the degree of centrality we assign to the nature
of the analyst’s experience in enactments and the kind of clinical use we
make of that experience. It has to do with the consistency with which we
understand the content of the session to be an outcome of clinical process,
a process that is inevitably interactive and contextual and in which both
analyst and patient are continuously involved in an unconscious way. If
there is a diἀerence, it is not a diἀerence in kind, but in emphasis, in the

degree to which we espouse the view that unconscious, interactive process
precedes content in clinical work. But I am also impressed with how much
we share. Few psychoanalytic writers hold views about dissociation and
clinical process as close as those of the mentalization theorists, on one
hand, and of Relational dissociation theorists, on the other.
9
“One Never Knows, Do One?”
The Relation of the Work of the Boston
Change Process Study Group and
Relational Dissociation Theory
I have been reading and teaching the papers of theBoston Change Process
Study Group since the first one appeared (D.N. Stern, et al., 1998). My
enthusiasm for this body of work is due to the basic sympathy between
their views and my own. I begin by detailing some of those agreements,
and then go on, later in the chapter, to take issue with some of BCPSG’s
positions. Because we have so much in common, I want to contextualize
our disagreements within the substantial degree of similarity of our views
in many other respects. The points I discuss bear on all the papers in the
series to date (D. N. Stern, et al., 1998; BCPSG, 2002, 2003, 2207, 2008).
Agreements
First comes the focus BCPSG and I share on the clinical situation, espe-
cially on the details of clinical process, and a certain democratic attitude
toward what transpires in the consulting room, well expressed in Lew
Aron’s (1996) observation that, while therapeutic roles are necessarily
asymmetrical, the human mutuality that lies at the heart of treatment is
not. The democratic attitude and focus on detailed clinical process that is
so characteristic of BCPSG’s work also defined the ground of my training
as an Interpersonal psychoanalyst; and the same attitudes came to shape
the Relational perspective, the second half of my psychoanalytic identity.
Another general point of agreement is the attitude shared by BCPSG and
Relational analysts, including me, that psychotherapy and psychoanaly-
sis are simultaneously social and intrapsychic enterprises, meaning that
183
their contents and processes are inevitably and continuously mutually

constructed.
BCPSG and I share many more specific psychoanalytic positions as
well. We have similar conceptions of what it means to describe something
as unconscious. Here is a representative quotation (BCPSG, 2008):
In spontaneous speech, there is something in mind that wants expressing. Let

us call this “something in mind” an image, in the broadest sense of the term.
The image can be an idea, a movement, a gesture, an aἀect, a vitality aἀect, a
background feeling. None of these are presently in verbal form.
A nonverbal “something in mind” is reminiscent of what I mean by “unfor-

mulated experience” (D. B. Stern, 1983, 1997). Furthermore, BCPSG agrees
with me that language does not come about by matching words to the
“something in mind.” Instead they tell us, just after the description I have
just quoted, “Now comes the messy work, especially in spontaneous dia-
logue. There is an intention (with its goal and structure) to link the image
with words. For almost each phrase, the intention enters into a dynamic
process with the existent repertoire of pieces of language to find the best
fits.” About these “fits,” they write later, “it doesn’t matter if there is a ‘right’
fit. None such exists. It only has to be good enough for eἀective commu-
nication.” Compare that with my view (D. B. Stern, 1983, 1997) that the
meaning in a particular unformulated experience may take any one of the
more precise formulated meanings toward which it moves. Meaning, that
is, becomes creation, not discovery. “Insight into an unconscious wish,”
says Herbert Fingarette (1963), “is like noticing a well-formed ‘ship’ in the
cloud instead of a poorly formed ‘rabbit.’ On the other hand, insight is not
like discovering an animal which has been hiding in the bushes” (p 80).
Years later, in discussing Fingarette’s point, I wrote: “We can now add
that the cloud itself changes with the passage of time, and so the ship inev-
itably disappears; and even before it does, if one keeps looking, perhaps
one sees a tree in the cloud that is even better formed than a ship” (D. B.
Stern, 1997, p. 176). It seems that my views and those of BCPSG accord
concerning the nature of the fit between an expression in the verbal-reflec-
tive domain and its implicit referent.
Another similarity: although BCPSG does say that the verbal-reflec-
tive “emerges from” the implicit, they also make it clear that they do not
intend to characterize the verbal-reflective as a “higher” form of thinking
or experiencing. Instead, they understand these two domains to be in a
relation of continuous mutual influence. Referring to the verbal-reflective
“One Never Knows, Do One?” 185
and the implicit, they say this: “it is not a duet of separate instruments.
Instead one voice emerges and is derivative of the other.” BCPSG cites
Merleau-Ponty in support of this position, as I do repeatedly through my
work. For a long time (D. B. Stern, 1997), and especially in the recent past
(D. B. Stern, 2002a), I have been presenting the verbal and the nonverbal as
the most significant contexts for each other. Each defines the possibilities
of the other. This view and the view of BCPSG also seem to coincide.
BCPSG and I also seem to agree that a substantial part of therapeu-
tic action takes place in the realm of the implicit. In discussions, several
members of BCPSG and I have sometimes disagreed over the degree to
which, after unconscious mutual enactments end, they must be described
in language in order to be mutative. I generally felt more strongly than
they did that the meaning embodied in an enactment must be formulated
in language to be clinically useful, at least often; but it was very clear that
BCPSG and I were on the same page in focusing our primary attention on
the nonverbal negotiation of relatedness (e.g., S. Pizer, 1998) in the treat-
ment situation.
I am stretching a point to call the meaning of unconscious mutual
enactment implicit. It is perhaps closer to my view to say that such mean-
ings are dynamically unconscious, by which I mean two things: (1) they
do not exist relative to verbal language, but apart from it; (2) they are spe-
cifically (that is, defensively) excluded from awareness. Implicit meanings,
in my frame of reference, are those that can be directly represented in
language; dynamically unconscious ones, on the other hand, cannot be.
Dynamically unconscious meanings are what Fourcher (1992) calls the
“absolute unconscious.” BCPSG’s “implicit relational knowing” (IRK) is
not dynamically unconscious, either in BCPSG’s view or in mine, because
IRK is not specifically excluded from awareness; but because of its nature,
IRK cannot necessarily be verbally represented either. It therefore lies, for
me, on the border between the implicit and the absolute unconscious.
And finally, BCPSG and I agree on the following closely related matters:
the emergent quality of meaning; meaning as process, not structure; and
meaning-making as continuous. We can summarize this position as the
view that meaning is a continuous emergent process. As significant as all
our other similarities may be, this may be the most important of them. In
fact, I think that this position is one of the most significant links between
BCPSG and most, perhaps even all, Relational analysts.
I will explain later how BCPSG and I diἀer over the source of meaning’s
emergent quality—that is, over the question of epistemology. But we cer-
tainly do agree about the presence of this emergent quality. BCPSG’s work,
rooted as it is in nonlinear dynamic systems theory, grows from the claim

that the events of psychotherapy and psychoanalysis, and IRK in particu-
lar, are unpredictable, nonlinear, and emergent. The analyst’s task is there-
fore participation in, and sometimes the discovery of, the involvement that
is in the process of developing. For BCPSG, as for Relational analysts, the
clinician is not in a position to grasp this involvement and oἀer interpreta-
tions of it as it transpires, because the involvement is always in the process
of becoming. Knowing must lag behind participation, that is, if knowing
develops at all. Therapeutic action does not rest on the analyst’s revelation
of the truth but rather on participation in an unfolding, emergent process,
what BCPSG started out calling “the ‘something more’ than interpreta-
tion” (D. N. Stern, et al., 1998; BCPSG, 2005). All of this is very congenial
to most Relational analysts, including me.
I like very much the BCPSG idea, derived from the work of Louis
Sander, that intentions are the basic units of psychological meaning, and
that sequences of intentions “give motivated human behavior its psycho-
logical existence, coherence and finally its meaning” (BCPSG, 2008, p.
129). Furthermore, I agree that the “intention unfolding process” (p. 129)
occurs implicitly. This “silence” in the process helps us to understand why
so much of what we identify as “me” occurs unwittingly and yet feels so
much as if it is our own. I have pursued for years the themes that we often
find ourselves in possession of intentions that we did not know we had, and
that what feels most like our own is experience that we have no conscious-
ness of having created (e.g., D. B. Stern, 1983, 1990, 1997). The intention
unfolding process is one of the few conceptions I have come across that
begins to help us think through this common and compelling phenom-
enon. The idea allows us to theorize the continuity across the implicit and
verbal-reflective domains that we intuitively know is there, and that results
in our sense of identity. The intention unfolding process “acts as the refer-
ent to identity … creating a common coinage across levels” (BCPSG, 2008,
p. 130). The process takes place, of course, outside the realm of conscious
will, which is why what we most deeply intend nevertheless has the capac-
ity sometimes to surprise us. We are very often in the position, BCPSG
suggests, of discovering our intentions, not consciously creating them.
Of course, this insight corresponds to a cornerstone of psychoanalytic
treatment: We must not assume that we always know what we desire, but
instead must observe (feel our way into) the desires that already exist, in
some ways just as we observe them in others. Only later, when we are better
acquainted with them and with our previously unconscious reluctances to
allow them to be “me,” will we be able to sense them directly.
How the intention unfolding process, in its definition as a “referent to

identity … creating a common coinage across levels,” applies to the con-
ception of the multiple self piques my interest. Can the intention unfold-
ing process be paired with the idea of multiple self-states in a way that
would further the understanding of how, at least in fortunate cases, people
have the capacity to feel like one person while being many, or as Bromberg
(1998) puts it, how we “stand in the spaces” between self-states?
In the BCPSG (2008) discussion of intention, I like the way that the mar-
velously expressive word “chunk” allows a description of continuity across
streams of otherwise disparate-seeming modes of experience, as in, “The
intention unfolding process arises from a fundamental psychological pro-
cess that chunks the flow of motivated human behavior into intentions” (p.
129). I have trouble with the word “fundamental” in this sentence and else-
where, though, and also with such phrases as “innate mental tendency,” used
to describe the intention unfolding process a couple of sentences later—but I
defer discussion of this point until I address the question of epistemology.
One last note here: Most psychoanalysts are well aware of the work of
Fonagy and his colleagues (2002) on mentalization. Mentalization theory
links firmly with the thinking that BCPSG has done and makes both bod-
ies of work that much more compelling. There seems to be an increas-
ingly wide recognition of the significance in psychological life of inferring
intentions, both others’ and our own.
These are at least some of the similarities between my views and those
of BCPSG. But I have also developed questions as I have read the BCPSG
papers, questions that are that much more important to me because of our
agreement on so many other matters. The first two of these questions, one
concerning enactment and the other about the use of nonlinear dynamic
systems theory, may very well be negotiable. The last question, which con-
cerns science and social construction, probably represents an enduring
disagreement. My discussion of all three questions, though, presents the
thoughts of a friend and fellow traveler.
The Question of Enactment
We have already seen that BCPSG goes this far: They accept that the ana-
lyst is inevitably involved with the patient on both conscious and uncon-
scious levels; and that it is not possible for the analyst to observe his or
her own involvement until after it has taken place—that is, observation
must always be at least a step behind participation. I do not have the sense
across the work of BCPSG, though, that they have emphasized the sig-
nificance of the analyst as a particular personality as much as I and most
other Relational analysts do. And (partly) because of that, BCPSG does
not give unconscious mutual enactment the place in their views that it
holds in most of ours.
BCPSG has not taken the step into conceiving the analytic relation-
ship to be fully personal, with all the problems that an intensive personal
relationship inevitably has. Granted, the personal relationship of analyst
and patient is very strange because it is also a very particular kind of pro-
fessional relationship, with firm boundaries of a professional kind. But it
is a personal relationship. BCPSG’s clinical illustrations do not reveal any-
thing like the kind of enactments that are routine in the Relational litera-
ture. There are minor difficulties between analysts and patients, but these
are not intense personal entanglements. I do not remember an example
from these papers in which either the patient or the analyst felt that the
conduct of the other was more than negligibly problematic.
I suspect that the absence of mutual enactment in the work of BCPSG
has to do, ironically enough, with the source of BCPSG’s conviction
about the inevitability of unconscious mutual influence—a conviction
that is one of the most significant points of agreement between BCPSG
and Relational psychoanalysis. BCPSG seems to take their version of this
conviction from the longstanding involvement of a number of its mem-
bers in some of the first and most innovative mother–infant observation
research.1 In that context, inevitable mutual unconscious influence is gen-
erally conceived under the rubric of mutual regulation.2 In the recent past,
those processes of mutual regulation have begun to be described, by some
of the same writers, between patient and therapist (D. N. Stern, et al., 1998;
BCPSG, 2002, 2005, 2007, 2008; Beebe & Lachmann, 2002; Lachmann &
Beebe, 1996; D. N. Stern, 2004). For the most part, mutual regulation,
whether between mother and infant or patient and analyst, is a continuous
sequence of relatively small, mutual, and nonconscious accommodations
and adjustments.
Relational psychoanalysis, on the other hand, emphasizes interpersonal
disjunctions and resolutions that take place on a larger and more con-
1 For example, Lyons-Ruth (1991, 2003); Lyons-Ruth, Connell, Zoll, and Stahl, (1987); Lyons-Ruth
et al. (1998); Nahum (1994, 2000); Sander (1962); D. N. Stern (1971, 1977, 1985, 1995).
2 See Beebe and Lachmann (1988, 1994); Gianino and Tronick (1988); Jaἀe, Beebe, Feldstein, and
Crown (2001); D. N. Stern (1985, 1995); Tronick (1989, 1998); Tronick and Weinberg (1997).
flictual scale.1 There exists no counterpart in the mother–infant observa-

tion literature to this larger-scale kind of enactment; and therefore, it is
perhaps because BCPSG has such deep roots in that part of the field that
the Relational interest in large-scale enactment has not flowed as naturally
into their thinking.2
From an Interpersonal or Relational perspective, interaction between
patient and analyst is a sequence of successes and failures, like any human
relationship. It is a familiar Relational theme that mutual unconscious
enactment is more an opportunity than a detriment. That is because it is
the part of the patient that the patient does not and will not know (in the
sense that this part of the patient’s experience is dissociated, not available
to consciousness) that has brought the patient into treatment, the part of
the person that, after Sullivan (1954), several of us (Chefetz & Bromberg,
2004; D. B. Stern, 2003, 2004, 2006), notably Bromberg (1998, 2006), call
not-me, exists only as unformulated experience. Because not-me cannot
be given symbolic form (it is defensively maintained in an unformulated
state [D. B. Stern, 1997]), it can enter the treatment only via enactment
(Bromberg, 1998, 2006; see also Chapters 1 and 3–6 in this book). Some
of these enactments do not reach the analyst in a way that provokes the
analyst’s own dissociations (and thus also provoke the answering part
of a mutual enactment); but some of the patient’s enactments do reach
the analyst in that way, setting in motion a mutual enactment. Whether
mutual enactment is the outcome or not, the opportunities oἀered by the
arousal of the not-me part of the patient’s personality can only be actual-
ized if the unconscious parts of the interaction, which can sometimes be
deeply distressing and unpleasant for both parties, become possible for
analyst and patient to “know.” Not-me, that is, must become me for both
participants. Non-self must become self. In the case of mutual enact-
ment, quite common in Relational annals, the analyst must grow, not
only the patient (for a detailed presentation of this process, see Chapter
4).
I do not see in the work of BCPSG an acknowledgment of this part
of the work, which to me lies at its heart. BCPSG might answer that
1 See Aron (2003a, 2003b); Bass (2003); Benjamin (1998); Black (2003); Bromberg (1998, 2006);
Davies (1997, 1999, 2003, 2004); Davies and Frawley (1994); B. Pizer (2003); S. Pizer (1998); D. B.
Stern (2003, 2004).
2 Beebe and Lachmann (2005; Lachmann & Beebe, 1996) write about what they call the sequence
of disruption and repair, a much larger-scale kind of mutual regulation that bears an interesting
relationship to the concept of mutual unconscious enactment. Considering the issues raised by
these writers would take me too far from the thrust of this chapter.
the analyst’s personality and his or her unconscious participation is

an unmistakable part of the concept of IRK. I agree with this point.
IRK is a joint, emergent, and unpredictable production of therapist and
patient, and it depends on the contribution of each—and, from the ana-
lyst’s position, the personal contribution as well as the professional one.
The issue for me is the degree and depth to which the influence of the
analyst’s personal contribution is acknowledged. Despite the inclusion
of the analyst’s personhood in IRK, I do not see in the work of BCPSG
a place for either the “aἀective snags and chafings” (see Chapters 3 and
4) that alert the analyst to his unconscious involvement and reactiv-
ity, or for the sometimes intensely difficulty entanglements that rela-
tional analysts, especially those writing about dissociation, have made
the focus of clinical and theoretical attention. To say this is not to claim
that BCPSG cannot accommodate these ideas and phenomena, only to
point out that, up to now, they have not done so. As a matter of fact, I see
no reason why these central relational concerns cannot be brought into
the kind of theory BCPSG is developing.
Nonlinear Dynamic Systems Theory
It has always seemed to me that psychoanalysis needed to look beyond

its own backyard for inspiration. Nonlinear dynamic systems theory
(NDST), which originated in mathematics and the hard sciences, seems
to hold great promise for many fields, and a growing number of psycho-
analysts, BCPSG prominent among them, feel that the list of those fields
should include our own. If you add to BCPSG’s use of NDST their applica-
tion of Lakoἀ and Johnson’s (1999) seminal work on metaphor, you must
conclude that the body of BCPSG’s work exemplifies the interdisciplinary
ideal. I subscribe to this ideal myself and admire it in the work of others.
NDST is a way of understanding the simultaneous interaction of a mul-
titude of events and influences—hundreds, thousands, even more. It is a
way of grasping how events of almost unimaginable complexity come into
being. And of course this multitude of influences does not merely sum
to an outcome; the influences also interact with one another along the
way. The degree of complexity is astonishing. It is also unpredictable and
nonlinear (i.e., changes do not necessarily take place in an orderly way as
the result of an accumulation of influence, and may actually be quite sud-
den). What often appears to be chaos on the surface of things is, if we look
closely enough, the manifestation of very complex kinds of order. That is
why NDST is also known as chaos theory or complexity theory. NDST is a

growing body of ideas about the rules that govern these kinds of phenom-
ena, which are much more common in our world than we used to believe.
This is the kind of description and explanation NDST oἀers, and that is
what stirs me about it.
Yet I am uncertain about whether NDST is useful when applied to
relational events. I mean “uncertain” literally, not as a euphemism for
disagreement. I can present my reservation only by oἀering a little pre-
liminary explication.
Arguably, Esther Thelen (who died only a couple of years ago, tragi-
cally, in her early 60s) and her collaborator, Linda B. Smith, researchers
in developmental psychology, have done more to bring NDST to the study
of psychology, and through that application, to psychoanalysis, than any
other writers (Smith & Thelen, 1993; Thelen & Smith, 1994). Their book
on cognition and action (Thelen & Smith, 1994) proposes a theory that
accounts for enormously complex phenomena. Yet as speculative as this
work is, Thelen and Smith are rooted in very detailed examinations of
such phenomena as infant locomotion (it turns out that, if you examine
the development of locomotion on the level of minute detail, infants vary
enormously in how they progress toward the common endpoint of walk-
ing) and infant reaching toward objects (“from the messy details of real
time … global order can emerge” [p. 247]). In fact, most research using
NDST in developmental psychology has focused on the study of simple
actions—behaviors in space—because what is needed for research from
the NDST perspective is a vast number of the most minute kinds of obser-
vations, enough to reveal underlying, complex patterns invisible at most
levels of understanding, and for that you need very fine-grained measure-
ment, the kind of measurement that is available by mapping points in
space (Fischer & Rose, 1999). And therefore, action is vastly more practi-
cal to investigate using nonlinear dynamics than more complex parts of
human life such as cognition and aἀect—to say nothing of the unimagi-
nable degree of complexity introduced by considering the interaction of
two of these nonlinear systems (two human beings, for instance). It seems
to me, admittedly a layperson in these matters, that the interaction of two
systems of enormous complexity must increase the number of relevant
influences on the outcome one is studying by a factor that is geometric,
not arithmetic.
That is hardly a reason not to use NDST in psychoanalysis. But it
may be a reason to be cautious. The classical systems theorists—von
Bertalanἀy (1968) and Bronfenbrenner (1979), for instance,1 spent many

years criticizing the narrowness and simple-mindedness of most explana-
tory theory and research in the social sciences, including psychology. But
they were hindered in taking their work further by the fact that, unlike
NDST, classical systems theory did not have clear enough implications for
the development of new theory and research methodologies. That older
generation of systems theorists could only lament the absence of a more
sophisticated appreciation of complexity.
And that is what I worry about today. Are we really able to do more
than acknowledge what we all know is the reality of emergent process and
multitudinous influences on experience and interaction in the consulting
room? Are we able to do more than point at it? Exactly how does NDST
illuminate data for BCPSG that would have been invisible or seemed irrel-
evant otherwise? What specific findings or theory would not have devel-
oped if BCPSG had limited themselves to classical systems theory, for
instance, and had simply said to themselves, “Well, the clinical setting
continuously generates conduct and experience in and between its partici-
pants that is obviously enormously complex. What will we come up with
if we take that view as a starting point and then interpret detailed clinical
process as the continuously emerging outcome of this enormously com-
plex set of processes?” Would BCPSG have been able to create their com-
pelling views of the therapeutic situation using just this much theory, or
would their views have been impossible without specific NDST proposi-
tions? If the answer is that NDST propositions have indeed been essential
to the theory creation of BCPSG, which propositions are they?
We know that BCPSG, like other researchers in our field who use
NDST, do not collect the voluminous data that comprise NDST stud-
ies in (for instance) developmental psychology. In other words, NDST in
1 I am writing here about academic psychology, but if I were writing about classical systems theory
in psychoanalysis and psychotherapy, I would have to mention the contributions of Interpersonal
psychoanalysts and family therapists. Salvador Minuchin wrote that Harry Stack Sullivan was
instrumental in bringing information and communication theory to psychoanalysis, and thus to
family therapy. Many of the most important of the family systems theorists also acknowledge this
influence: Don Jackson’s training included supervision by a number of Interpersonalists, includ-
ing Sullivan; Jay Haley reports Sullivan’s influence on him; and Minuchin was actually trained as
an Interpersonal psychoanalyst and writes that he is clear how central that experience was to his
thinking as a family therapist. Ivan Boszorrmenyi-Nagy and Murray Bowen, two other influential
family therapists, were also heavily influenced by Interpersonal psychoanalysis. I am depending
here on a valuable article detailing the connections between these two groups by Richard Gartner
(1995). Eventually, Edgar Levenson (1972, 1983), in Kavanagh’s (1995) words, took the position,
influenced by von Bertalanἀy and other early systems theorists and amazingly prescient in the
context of the later development of NDST, that “the world is a complex, organismically related set
of events in which there is great order but not of the simple cause-and-eἀect kind” (p. 587).
psychoanalysis is not used to generate research. That is not necessarily a

problem, of course; NDST can just as validly be used for theory construc-
tion or model making as for the generation of new quantitative research.
But if this is the way it is being used, a question arises: Is NDST being
used literally among psychoanalysts, as it is in mathematics or develop-
mental psychology? Or are we using NDST metaphorically, leaving aside
the details of the theory (which are heavily mathematical)? It may be quite
helpful to use NDST metaphorically. I do it myself (see Chapter 5). But it
would be desirable to be clear that the ideas are being used that way, rather
than in the way they are used in research fields in which quantitative stud-
ies of NDST are more practical.
Science and Social Construction
NDST is the source of the concept of emergent meaning in the work of

BCPSG, which I have described as one of the most substantial areas of
agreement between BCPSG and Relational psychoanalysis. But NDST is
also a scientific theory, which means that it rests on an objectivist epis-
temology. Here we reach the discussion of the points of epistemological
diἀerence between my thinking and that of BCPSG that I promised to
undertake earlier in this chapter.
I am not setting out to take issue with science. I am not even setting
out to take issue with science in psychotherapy and psychoanalysis. That
would be tilting at windmills, but without the nobility of Don Quixote;
it would just be wrongheaded. I take issue only with the privileging of
systematic, quantitative empirical research in the investigation of psycho-
therapy process and outcome. The word “privileging” is important here: I
have no argument with the use of such research as one source of informa-
tion about psychotherapy process and outcome.
There are many questions in our field for which science probably should
have a privileged status: the biology of trauma, for instance; or the genetics
of schizophrenia and disorders on the pervasive developmental disorders
(PDD) spectrum; or the chemistry of severe depression. I even accept the
desirability of quantitative outcome studies of psychotherapy and psy-
choanalysis, for three reasons: (1) that kind of research is couched in a
language that the wider culture understands; (2) empirical research may
identify certain ideas that clinicians need to reconsider, such as primary
autism or the schizophrenogenic mother; (3) good clinical ideas sometimes
come from such research (a prime example is the behavioral treatment of
phobia, which may not have originated in quantitative research, but which
was widely adopted largely because of the confirmatory outcomes of such
studies). The problem is that quantitative research on psychotherapy pro-
cess and outcome has seldom been able to reflect the subtlety and sophis-
tication of psychoanalytic clinical work (there are significant exceptions).
One reason for this state of aἀairs is that the simplest treatment methods
and measures are the easiest to quantify. Increasingly, it is exactly these
studies, which often do not even study procedures that would be recog-
nized as psychotherapy by psychoanalytically trained clinicians, that are
used to validate methods for insurance reimbursement. I appreciate the
frustration over clinicians’ resistance to quantitative empirical research by
those in our field who feel that such research must be done but who also
believe that scientific study is just one source of information. I do not sym-
pathize, though, with those who accept only quantitative data as evidence
of treatment validity and eἀectiveness. In fact, I do not sympathize with
any position of dependence on quantitative data for these purposes.
To say that, of course, immediately brings up the question of how we
should think about treatment eἀectiveness, which is a can of worms that
I don’t want to open here. Suffice it to say that I disagree in the stron-
gest terms with the movement called “evidence-based treatment.” I do not
believe that quantitative research should determine how we conduct psy-
choanalytic treatment (see Hoἀman [2006] for an excellent discussion of
the issues).
I have said something about these general problems in order to be
able now to make it clear that I do not include BCPSG in the category
of those I have just described critically. NDST is indeed science and
therefore rests on an objectivist epistemology. But NDST is also about
the necessity to accept unpredictable and unexpected outcomes, which
means to me that the work of BCPSG would be very hard, if not impos-
sible, to use to “scientize” psychoanalytic practice. More important, it
is not BCPSG’s intention to do that. For BCPSG, as for Relational psy-
choanalysts, you cannot know exactly what you are going to do as an
analyst until you come upon the circumstances that will contextualize
your intervention. For all the order we give our time and fee schedules,
when it comes to transactional events we are often flying by the (highly
educated and seasoned) seat of our pants. After my own heart, the motto
of BCPSG could be Fats Waller’s famous aphorism: “One never knows,
do one?”
And yet the science of the BCPSG approach to clinical process is uncom-
fortable for me, and I want to try to say why. This is the most difficult
part of my reaction to BCPSG’s work for me to capture, because the use

of NDST, with its encouragement for us not just to study the nonlinear
and the unexpected, but to expect them, makes irrelevant all of the most
hard-edged of the disagreements between a scientific approach to psycho-
therapy and views such as mine. It is possible that what I have to say will
be a surprise to BCPSG, because, in addition to doing away with linearity
and conventional notions of orderly processes of therapeutic change, they
align themselves with the contemporary critique of Cartesianism, thereby
taking a philosophical position that they may believe is inconsistent with
objectivism. Yet it is hard to see how NDST, when it is used literally and
not metaphorically, can be understood as anything other than objectiv-
ism. The general outline of the argument I will make is based on a broadly
conceived philosophy of social construction.1
In describing philosopher Charles Taylor’s view of language, Timothy
Zeddies (2002) speaks for me:
[T]he web of language comprises a broad experiential and preexperiential back-

ground that does not completely dominate [and is not dominated] by individual
human subjects. The language we speak is not an individual possession but
something that is shared by an entire speech community. Touching any piece
reflects only a miniscule portion of an immeasurable web of common meanings,
practices, and understandings that define and inform a particular speech com-
munity. Articulated speech (spoken or written) reflects only a miniscule portion
of an immeasurable web of common meanings, practices, and understandings
that define and inform a particular speech community. What is actually spoken
or written depends on and is possible because of the vast web of unexpressed
and unarticulated meanings, practices, and understandings that remain in the
background of explicit awareness and immediate experience. (p. 17)
Taylor is only one of many writers to take the position that the verbal and
the nonverbal are inextricably entangled with each other. All the writers
who have participated in the “linguistic turn” in the humanities and the
social sciences take the position that what we say is not only shot through,
but made possible, by what we cannot or will not say. To the extent that the
thinking of all these writers emphasizes the continuous interrelationship
of the verbal reflective and the implicit, BCPSG is right there on the same
page with them.
1 I refer to my social constructionism as broadly conceived because I am not limiting myself to a
Foucauldian analysis of knowledge as a social product and the relations of power, but am also
including modern ontological hermeneutics. Hermeneutics is often equated with constructivism,
but social constructionism is a term that usually seems to refer more specifically to Foucault and
analyses based on his approach.
But Taylor and company are saying more than that. They are also say-
ing that language is used in socially defined circumstances. The people
who use any particular language, or any local variant of it, or even just
people who live in a particular place, constitute a community of language
users who define how and for what purposes they speak, write, and read.
And so language is not only used in socially defined circumstances—lan-
guage is itself a social product. But that is not all. The whole of life, in this
philosophical perspective, is a social product, the implicit right along with
language—any human activity you can think of. This vast accumulation
of social products, which sums to what we know as culture, is in con-
tinuous flux; and it is we who are changing it. But because such change is
often (though not always) slow, and because taking an independent stance
toward our own culture is like the eye seeing itself (the instrument is part
of the object of study), we often do not notice the changes as they occur
(as we make them). As the wise man said, “I don’t know who discovered
water, but it sure as hell wasn’t a fish.” Therein lies the potential for a kind
of enslavement, because ideology—the unseen values that shape a sub-
stantial part of our participation in the world—thrives on invisibility.
How we behave with one another, how we feel, how we think, what we
believe—all of this is a social product. All of it is a creation in context,
and it takes its meaning from that context. There is reason to question
whether our understanding of any taken-for-granted phenomenon we
might choose to examine—the verbal, the nonverbal, the whole kit and
caboodle—would be the same if our subjects came from diἀerent cultures
or lived in diἀerent periods of the history of our own culture. The point
gains in complexity, of course, when we acknowledge the significance
of subject positions. “Our culture” is hardly the same for a gay African
American man of limited means, for instance, as it is for a wealthy, straight
Latina. Even if we set out with what we believe is the intention to under-
stand the other, we know how commonly we run roughshod over cultural
diἀerences, ironing them out into facsimiles of what is familiar. Frederic
Bartlett (1932) oἀers what may be the classic examples of this phenom-
enon, which he calls “conventionalisation,” in his study of the memories
over time of Cambridge undergraduates for a Native American story that,
of course, contained much that made little sense in upper crust England.
After startlingly brief periods, the unfamiliar details and plot elements
remembered by the undergraduates were distorted to fit cultural expecta-
tion or simply disappeared.
Reading any anthropological study of what it is to be a person, or how
people think, or the nature of psychopathology, or any of a thousand
other topics, will convince you (at least they convinced me) that people
are simply not the same everywhere. Let me cite just one example, one I
use in teaching all the time: In a few astonishing pages, Cliἀord Geertz
(1974), the eminent anthropologist, describes what it is to be a person in
three diἀerent cultures: Java, Bali, and Morocco. To a Westerner, these
peoples’ understandings of self may as well be from Mars. I won’t go into
the details. I cite the paper to support the familiar social-constructivist
contention that just because something seems to us like part of the natu-
ral world does not mean that it is. In fact, the parts of human life that
are objectified in this way thereby cease being questionable and, for that
reason alone, become precisely the parts of life that most need to be ques-
tioned. Here is Geertz’s conclusion, which has had wide currency since the
publication of his essay:
The Western conception of the person as a bounded, unique, more or less

integrated motivational and cognitive universe, a dynamic center of aware-
ness, emotion, judgment, and action organized into a distinctive whole and set
contrastively both against other such wholes and against its social and natural
background, is, however incorrigible it may seem to us, a rather peculiar idea
within the context of the world’s cultures. (p. 59)
It is possible, of course, to study anything in more than one way. One

can study the self in Western culture as a manifestation of the natural
world, as Daniel Stern (1985), among many others, has done. Or one can
study the self as one particularly significant creation of the culture we live
in, as Erich Fromm (1941, 1947, 1955) and Philip Cushman (1990, 1991,
1995, Cushman & Gilford, 1999), among many others, have done. I believe
that our field has benefited enormously from the work of all of these writ-
ers, although I believe that the benefit we have derived from Daniel Stern’s
work on the self is due less to his natural-science approach and more to
his astute mother–infant observations and his incisive clinical and theo-
retical analyses of them. One does not need to accept Stern’s observations
as objectivistic or the theory as natural science to find both enormously
useful.
In a hard-hitting critique of Stern’s work on the self from a social con-
structionist perspective, Philip Cushman (1991), who calls for traditional,
decontextualized psychology research on human universals to be replaced
by research defining and studying phenomena in their social context, sug-
gests (in one of his less critical conclusions about Stern’s work) something
similar to my own view:
Instead of universal laws, Stern articulated the shape of the cultural horizon
at this historical moment. He did that by using methods that are valued by his
professional colleagues, and in doing so collected information that is meaning-
ful to us, in order to aid us in efficiently performing professional roles that are
indispensable to our current Western way of life. (p. 209)
One can also study psychoanalysis and psychotherapy in diἀerent ways.

One can define them as procedures composed of processes belonging to the
natural world; or one can take the position that the processes that compose
them, as well as the treatments themselves, are inventions of our culture
that will change over time and may even disappear. In the former case,
it ought to be possible to specify the underlying, universal processes that
make up psychotherapy; this is what BCPSG attempts to do. In the latter
case, it ought to be possible to describe in detail the processes that we see
in psychotherapy, with an eye to identifying many diἀerent ways that the
people of our time create and change meaning in their lives and many dif-
ferent ways that psychotherapists might therefore use in dealing with their
patients (multiple solutions and contextual knowledge are hallmarks of
this way of doing things, whether we call it hermeneutic or postmodern).
In the former (scientific) case, we agree to change the procedures of
psychotherapy only when we believe the new procedures are closer to
some kind of newly discovered truth about the subject matter. In the latter
(hermeneutic) case, we agree to change the procedures because we sense
that the world is a diἀerent place, and that we are therefore diἀerent from
those who came before us; something diἀerent is demanded of us. In a
similar vein, one can think of the people who practice psychotherapy and
psychoanalysis and the people who come to them for the service they pro-
vide as human beings who are pretty much the same as human beings in
any place or at any time; or one can think of patients and their therapists
as human beings of a particular time and place, shaped according to the
possibilities aἀorded them by the cultures into which they are born. It is
a commonplace, for example, that since the 1960s there has been a broad
shift in our culture regarding our relation to authority. We simply are not
as willing to accept authority as people in our culture once were, when
political leaders, teachers, doctors, and policemen could take respect and
admiration for granted (which, of course, is hardly to say that the abuse of
authority has ceased to be a problem). Because people are diἀerent today in
this way, the kind of psychoanalysis that was practiced prior to the 1960s
is simply no longer appropriate. People today are much less likely than
they used to be to put up with analysts who feel that their patients must
believe that the doctor knows the truth about them. And so in recent years
Interpersonal psychoanalysis has emerged from its relative obscurity, and
Relational psychoanalysis, with its emphasis on mutuality and its accep-
tance that the analyst’s authority can be no more than ironic (Hoἀman,
1998, pp. 69–95; Mitchell, 1997, pp. 203–230), has ascended. Actually, it
seems to me very much to the point that until recently BCPSG would
not have been able to publish their articles in the International Journal of
Psychoanalysis (BCPSG, D. N. Stern, et al., 1998, 2002) or the Journal of the
American Psychoanalytic Association (BCPSG, 2005), because in the past,
what BCPSG does would not have been accepted within the boundaries of
classical psychoanalysis. I do not think that these two mainstream jour-
nals accepted the BCPSG articles only because their editors believe that
BCPSG has come closer to the truth than their intellectual predecessors;
I think the editors of those publications were bending to changes in our
culture’s acceptance of a more democratic clinical practice. The readers
of these two journals have also been aἀected in this way, of course, so
that they are now interested in the papers of BCPSG—which, in circular
fashion, is part of what makes the editors more likely to publish the work
of BCPSG. Of course, the whole process works in both theoretical direc-
tions: The success of Psychoanalytic Dialogues can be partially pegged to
the same factors.
BCPSG favors a natural science point of view. They are looking for the
kind of regularities in psychotherapy that belong to the natural world; they
write of “fundamental psychological processes” and “innate mental ten-
dencies.” My view is hermeneutic; I am looking at psychotherapy, includ-
ing the transactions that take place in it and the experiences and cognitive
processes in the minds of its participants, as a particular social practice in
the here and now of our Western cultures.
The experience of presenting this point over and over again has taught
me that I need to pause long enough to make it clear that my perspective
is not relativistic. I do not take the position that one can say whatever one
feels like saying. There exist constraints on our perceptions and thoughts;
beyond them, our understandings are simply wrong, or crazy. There is a
reality, then, and we sense it as a set of constraints that we must respect if
we are to remain truthful and sane. But the possibilities that reality con-
tains for our experience are manifold, which means that no single version
of reality is possible to select as the correct one. There is always “wiggle
room” in the next moment.1 And if there is no single correct answer, and
if we therefore choose our interpretations for some reason other than mere
accuracy, every answer has a political aspect, visible or not.
I have just said that BCPSG favors a natural science approach to psy-
choanalysis, while I favor a hermeneutic one. But I have also made it clear
earlier that BCPSG and I share some of the most important orientations
about how to do psychotherapy and psychoanalysis. Given this similarity,
why does it matter clinically whether we diἀer over something as abstract
as the choice between natural science and social construction?
Philip Cushman (1990, 1991, 1995; Cushman & Fiἀord, 1999, 2005)
would say that the diἀerence matters because what he calls “decontextu-
alized” psychology research, by which he means research that treats psy-
chology as a natural science, takes an unintentional political position by
treating current social conditions as a manifestation of the natural order,
thereby reinforcing the status quo and the power relations that sustain it.
Social constructionist research, on the other hand, is designed to formu-
late those same conditions as the expression of the values uppermost in
our culture today, thereby exposing power and privilege and giving us a
choice about whether to continue the status quo or change it.
Cushman’s agenda is an important critique of the social sciences in
general, but it really applies no more to Daniel Stern’s research on the
self or the work of BCPSG than it does to thousands of other research
programs. However broad the reach of that kind of critique, though
(Cushman is only one of many to mount it), it remains true that it does
also apply to BCPSG. If we understand the transactions that make up
clinical psychoanalysis as events of the natural world, in the same cat-
egory as thunderstorms and the creation of diamonds, I believe that we
give something up. I believe that viewing clinical practice that way makes
us less likely to question whether what we are doing with our patients
reflects unconsciously embraced values that we might prefer not to actu-
alize—in a word, whether what we are doing with our patients is the
manifestation of ideology. Very few, if any, of those who studied psycho-
analysis in the 1950s, all of them employing a natural science perspective,
1 Because of its multiple possibilities, reality is also never available in unmediated form. That is,
since only one of the possibilities can become our experience, we cannot “know” reality without
first giving it some kind of shape. We must somehow select which of the many possibilities facing
us in every moment will become manifest for us, or formulated. Something must mediate real-
ity to us. This “something” is culture and, on a smaller scale, the relational field and individual
psychology. Culture, relationship, and character, that is, are the means by which we select and
formulate that version of reality that comes to be our experience. For a more extended discussion
of this view, which is rooted in Hans-Georg Gadamer (2004), see D. B. Stern (1997, 2002b).
noticed the influences of gender, sexual preference, race, class, and eth-
nicity on clinical practice; and if they did notice such things, there was
nothing in a natural science view that encouraged them to view these
problematic aspects of relatedness as anything other than phenomena of
the natural world. Women just are passive and receptive; homosexuality
just is psychopathological. It is the natural way. Psychoanalytic writers in
those days seldom had reason to observe the authority relations between
patients and analysts either, so much more rigid than they are today; this
part of analytic relatedness was simply accepted as a part of the natural
order, necessary to create therapeutic regression and the transference
neurosis, which were in turn also considered aspects of the natural world
that had been invisible until uncovered by psychoanalysis.1 Today, in the
same way, if we accept BCPSG’s epistemology, we must accept (to cite
just a few of BCPSG’s most significant contributions) “moving along,”
“now moments,” “sloppy process,” and the “intention unfolding process”
not just as useful ways of formulating our experience of doing analytic
work, which they certainly are, but also as objective observations.
As a means of tying this argument down, let me cite and discuss how
my argument applies to just one passage in the paper I am discussing.
Near the end of their 2008 paper, in moving from the level of the indi-
vidual to the level of communication between two people (after all, the
two-person clinical situation, BCPSG agrees, is the point), BCPSG writes
the following:
The basic problem of the relation of the implicit and the reflective-verbal is
paralleled in the two-person situation in terms of what is spoken and what is
reflectively heard. We consider the spoken to constitute an implicit experience
for the listener for the following reasons. The listener hears the spoken mes-
sage, infers the underlying implicit experience that gave rise to the words, and
feels the diἀerence between the two. He receives a “gestalt.” He must then, in
an act of reflection, make a whole meaning of this gestalt. Again in this act, a
disjunction/coherence is introduced between the implicit experience of hear-
ing/seeing/experiencing the speaker’s performance and the listener’s reflected
meaning. When the listener then becomes the speaker, the process continues,
only in the opposite direction.
The meanings (i.e., packages of implicit, reflective-verbal and their disjunc-
tions) build on each other and reorient the direction as the dialogue advances,
resulting in more global or summarizing intuitive grasps. In other words the
meaning evolves in the course of the interaction. (p. 144)
1 One of the few exceptions, maybe the only one written by a classical analyst, was Macalpine’s
(1950) observation that therapeutic regression is not a “natural” phenomenon at all, but a creation
of the infantile relatedness set up and enforced in the analytic situation.
It is ironic that this precisely rendered passage is a fine description of the

hermeneutic circle! This is exactly the process described by hermeneu-
tic philosophers such as Gadamer (2004) as the source of meaning and
the means by which meaning grows: One partner in the conversation
oἀers a communication, and the other receives some kind of experience
of that meaning, referred to as a “partial” meaning (BCPSG implies that
the “gestalt” received by the listener is also an incomplete meaning); the
receiver must then “complete” the meaning by making it into something
recognizable, exactly as BCPSG says that the receiver must, “in an act of
reflection, make a whole meaning of this gestalt.” In the conception of
the hermeneutic circle, the receiver makes the communication into a rec-
ognizable meaning by relating it to what is familiar—to expectations, or
preconceptions. In felicitous circumstances, what is new in the commu-
nication is revealed by its contrast to what is familiar; in other circum-
stances, the new disappears into preconception. It seems to me that this
hermeneutic process is pretty much identical to what BCPSG is describ-
ing when they say that the receiver, in making a meaning of the “gestalt,”
introduces a “disjunction/coherence … between the implicit experience of
hearing/seeing/experiencing the speaker’s performance and the listener’s
reflected meaning.” That is, some of what is conveyed in the communica-
tion is formulated into a meaning by the receiver and some of it is not. In
either case, the one who has received the communication now goes on to
respond to the first speaker, and as BCPSG says, “the process continues.”
In the hermeneutic literature, although the word “process” in this phrase
would be more likely to be rendered as “conversation,” the meaning would
be the same. I see no meaningful diἀerence between what BCPSG oἀers
here and the idea of the hermeneutic circle.
And that raises the question of why we should bring science into the
question of how treatment is conducted. Unless science adds something to
our understanding that is not available otherwise, what is the advantage of
bringing it to bear? Once again, if the science (NDST, in this case) is being
used as metaphor by BCPSG, the question is moot, because then we are
not talking about science in the literal sense. But I have the impression that
BCPSG would say that they do intend their use of NDST to be literal.
My commentary on this passage has not yet addressed the reason
that I presented it in the first place: science versus social construction
in our investigation of the conduct of psychotherapy and psychoanaly-
sis. Let me approach that problem by asking a question about the pro-
cess of understanding being described in the passage I have just quoted
from BCPSG’s paper: How will it be decided what meanings are being
communicated by the first speaker and whether (or to what degree)
the second speaker has received them? To use NDST to answer that
question in anything more specific than a theoretical sense, we need
to be able to plot data points in an event space. And so we must know
what the data are, in an objectively defined sense. Even if we only use
NDST to create theory, a case can be made that we have to be able to
believe that it is at least hypothetically possible to observe objectively
defined data.
The developmental psychologists I mentioned earlier define actions by
reference to points in space; and that seems straightforward enough. It is
one thing to know objectively what a point in space is, though, and quite
another thing to know objectively what a meaning is. Yet if we are to use
NDST as BCPSG recommends, that is what we must know. We must be
able to specify, in objective, consensually verifiable terms, the relevant
meaning being communicated—or some data point, analogous to a point
in space, according to which we can gauge meaning objectively. Only if it
is possible to define a meaning in an objective way is an analysis of mean-
ing that is rooted in an objectivist epistemology coherent.
BCPSG does make it clear that meaning is continuously changing, and
so they might respond to what I have said by pointing out that meaning,
because it is a relational event, cannot be pinned down, and is therefore
beyond the reach of objective measurement. But flux is not inconsistent
with objective existence. As long as they use NDST in a literal way, BCPSG
is taking the implicit position that, in any particular moment of the clini-
cal interchange, meaning exists in objectively verifiable form. That form
may change over time, may even change very quickly—but it is fixed in any
particular instant. I cannot see how the implication that meaning must be
definable in objective terms can be escaped as long as one embraces an
objectivist epistemology.
It is hard enough to believe that it is even theoretically possible to spec-
ify in an objective way either the meaning that the speaker in the quoted
passage intended to convey or the meaning received by the listener. But
that is just the beginning of the problems of objectivism in this context.
We immediately reach another more difficult and significant conun-
drum: Even if we were to agree that an objectivist conception of mean-
ing were possible, how would we decide what that objective meaning is in
any particular case? How would we decide which, among a multitude of
possibilities, is the “objective” meaning? There is no impersonal way to do

that; there can be no “view from nowhere” (Nagel, 1986). A decision must
be made. Who will decide which version of the meaning that the speaker
conveyed to the listener is the objectively existing one, and who will select
the objective version of what the listener received and then formulated?
In other words, who will have the power to decide what meanings are the
“correct” ones or the ones we study or interpret? Very quickly we come
upon the necessity to acknowledge that the process of understanding is
inevitably social and political.
From the hermeneutic perspective, on the other hand, the decision
about what meanings are being communicated by the first speaker, and the
question of whether (or to what degree) the second speaker has received
them, is made by further hermeneutic inquiry. The questions about the
events in the passage I have quoted, that is, are decided in just the same
way that any other set of questions is decided: by the attempt to under-
stand as best as one can, which requires frankly acknowledged intuition
and interpretation, open to scrutiny and debate (at least when the event
is public).1 The uncertainty does not go away. There is never a “bottom”;
there are never objectively defined meanings that can make the process
transparent and a final answer possible at last. Does that lead to endless
questioning? Yes, that is exactly what it leads to.
I am prepared for BCPSG members to diἀer with my understanding of
the implications of their objectivist epistemology. They may even disagree
that their work should be characterized as objectivist. I am ready to listen
to what they have to say and even to change my mind. I look forward to a
dialogue about these issues. But at least at this point, I see their perspective
in the way I have described. The position that clinical process can be given
an objective significance leads to the conclusion that there is a correct way
to view it, because that is the point of an objectivist epistemology: Truth
inheres in the correspondence between what we say and what exists in the
world, a correspondence that qualifies as objective because it exists prior
to our observation of it. When we accept that truth exists apart from us,
we inadvertently make the political and social influences on the construc-
tion of meaning invisible. I do not believe that anyone’s psychoanalytic
observations represent discoveries of preexisting truths about the world
1 BCPSG also uses the word “intuition” in the same context. But the implication, from within the
NDST perspective, is that what one is trying to intuit is a meaning, however complex and aἀec-
tively nuanced, that has an objective existence.
but are instead creations of new ways of thinking that will, in turn, be
replaced by the next generation of thoughts. I believe that to take any other
view is to take the chance of inadvertently defending invisible ideological
aspects of the status quo and to risk making the revelation of these under-
lying influences more difficult and protracted.
BCPSG might respond to my argument by agreeing with me that
psychotherapy and psychoanalysis are socially defined practices, but
that they (BCPSG) are doing their best, within this socially defined set
of procedures, to make objective observations. I can accept that char-
acterization of their work, but it does not change the point I am trying
to make.
I prefer less certainty than that. I trust that I have conveyed how fas-
cinating and valuable BCPSG’s work is to me. I teach it, and its lessons
instruct me when I am with patients. But I see it as one of many formu-
lations of psychoanalytic practice that could be made. I see the analytic
process itself, including its perceptions, thoughts, and aἀects, its mutual
unconscious enactments, mutual regulatory processes, and moments of
intersubjectivity, as a social construction, valuable in so many ways, but
no doubt serving more purposes than we know (or would be comfort-
able with, if we did know), just as the Western sense of self described by
Geertz does. We all need to keep trying to articulate these hidden pur-
poses, so that psychoanalysis maintains its primary value: its question-
ing stance toward everything. BCPSG and I share Fats Waller’s sentiment
about many things in psychoanalysis, but I would take it a step further to
include our epistemology itself: “One never knows, do one?”
References
Alexander, F., & French, T. (1946). Psychoanalytic therapy: Principles and applica-
tion. New York: Ronald Press.
American Heritage Dictionary of the English Language (3rd ed.). Boston and New
York: Houghton Mifflin.
Aron, L. (1991). The patient’s experience of the analyst’s subjectivity. Psychoanalytic
Dialogues, 1, 29–51.
Aron, L. (1996). A meeting of minds: Mutuality in psychoanalysis. Hillsdale, NJ:
Analytic Press.
Aron, L. (2003a). Clinical outbursts and theoretical breakthroughs: A unifying
theme in the work of Stephen A. Mitchell. Psychoanalytic Dialogues, 13,
273–287.
Aron, L. (2003b). The paradoxical place of enactment in psychoanalysis:
Introduction. Psychoanalytic Dialogues, 13, 623–631.
Atwood, G. E. and Stolorow, R. D. (1984). Structures of subjectivity: Explorations in
psychoanalytic phenomenology. Hillsdale, NJ: The Analytic Press.
Bach, S. (2006). Getting from here to there: Analytic love, analytic process. Hillsdale,
NJ: Analytic Press.
Bartlett, F. C. (1932). Remembering. Cambridge: Cambridge University Press.
Bass, A. (2003). “E” enactments in psychoanalysis: Another medium, another
message. Psychoanalytic Dialogues, 13, 657–675.
Beebe, B., & Lachmann, F. (1988). The contribution of mother-infant mutual
influence to the origins of self and object representations. Psychoanalytic
Psychology, 5, 305–337.
Beebe, B., & Lachmann, F. (1994). Representation and internalization in infancy:
Three principles of salience. Psychoanalytic Psychology, 11, 127–165
Beebe, B., & Lachmann, F. (1998). Co-constructing inner and relational pro-
cesses: Self- and mutual regulation in infant research and adult treatment.
Psychoanalytic Psychology, 15, 480–516.
Beebe, B., & Lachmann, F. (2002). Infant research and adult treatment:
Co-constructing interactions. Hillsdale, NJ: Analytic Press.
Benjamin, J. (1988). ἀ e bonds of love. New York: Pantheon.
Benjamin, J. (1990). Recognition and destruction: An outline of intersubjectivity.
In S. Mitchell & L. Aron (Eds.), Relational psychoanalysis: ἀ e emergence of a
tradition (pp. 183–200). Hillsdale, NJ: Analytic Press.
Benjamin, J. (1995). Like subjects, love objects. New Haven: Yale University Press.
207
208 References
Benjamin, J. (1998). Shadow of the other. London: Routledge.

Benjamin, J. (1999). Afterword to “Recognition and destruction: An outline of
intersubjectivity.” In S. Mitchell & L. Aron (Eds.), Relational psychoanalysis:
ἀ e emergence of a tradition (pp. 201–210). Hillsdale, NJ: Analytic Press.
Benjamin, J. (2000). Intersubjective distinctions: Subjects and persons, recogni-
tions, and breakdowns: Commentary on paper by Gerhardt, Sweetnam, and
Borton. Psychoanalytic Dialogues, 10, 43–55.
Benjamin, J. (2002). The rhythm of recognition: Comments on the work of Louis
Sander. Psychoanalytic Dialogues, 12, 43–53.
Benjamin, J. (2004). Beyond doer and done to: An intersubjective view of third-
ness. Psychoanalytic Quarterly, 73, 5–46.
Bernstein, R. J. (1983) Beyond objectivism and relativism. Philadelphia: University
of Pennsylvania Press.
Bion, W. R. (1962). Learning from experience. London: Heinemann.
Bion, W. R. (1963). Elements of psycho-analysis. London: Heinemann.
Bion, W. R. (1970). Attention and interpretation. London: Tavistock.
Bion, W. R. (1980). Bion in New York and São Paulo. Perth: Clunie Press.
Black, M. (2003). Enactment: Analytic musings on energy, language, and personal
growth. Psychoanalytic Dialogues, 13, 633–655.
Blechner, M. (1992). Working in the countertransference. Psychoanalytic Dialogues,
2, 161–179.
Bollas, C. (1983). Expressive uses of the countertransference. Contemporary
Psychoanalysis, 19, 1–34.
Bollas, C. (1987) ἀ e Shadow of the object: Psychoanalysis of the unthought known.
New York: Columbia University Press.
Bollas, C. (1989). Forces of destiny. London: Free Association Books.
Bornstein, M. (Ed.) (2008). Is unconscious fantasy central to the theory and prac-
tice of psychoanalysis? Psychoanalytic Inquiry, 28(2).
Boston Change Process Study Group (BCPSG). (2002). Explicating the implicit:
The local level and the microprocesses of change in the analytic situation.
International Journal of Psychoanalysis, 83, 1051–1062.
Boston Change Process Study Group (BCPSG). (2005). The “something more” than
interpretation revisited: Sloppiness and co-creativity in the psychoanalytic
encounter. Journal of the American Psychoanalytic Association, 53, 693–729.
Boston Change Process Study Group (BCPSG). (2007). The foundational level of
psychodynamic meaning: Implicit process in relation to conflict, defense,
and the dynamic unconscious. International Journal of Psychoanalysis, 88,
843–860.
Boston Change Process Study Group (BCPSG). (2008). Forms of relational mean-
ing: Issues in the relations between the implicit and reflective-verbal domains.
Psychoanalytic Dialogues, 18, 125–148.
Botella, C. & Botella, S. (2005). ἀ e work of psychic figurability. Hove & New York:
Brunner-Routledge.
References 209
Boulanger, G. (2007). Wounded by reality: Understanding and treating adult onset

trauma. Mahwah, NJ: Analytic Press.
Brison, S. (2002). Aftermath: Violence and the remaking of a self. Princeton, NJ:
Princeton University Press.
Bromberg, P. (1998). Standing in the spaces: Essays on clinical process, trauma, and
dissociation. Hillsdale, NJ: Analytic Press.
Bromberg, P. (2000). Response to reviews by Cavell, Sorenson, and Smith.
Bromberg, P. (2006). Awakening the dreamer: Clinical journeys. Hillsdale, NJ:
Analytic Press.
Bronfenbrenner, U. (1979). ἀ e ecology of human development: Experiments by
nature and design. Cambridge: Harvard University Press.
Bruner, J. (1979). The conditions of creativity. In On knowing: Essays for the left
hand, (expanded edition, pp. 17–30). Cambridge: Harvard University Press.
Bruner, J. (1986). Actual minds, possible worlds. Cambridge: Harvard University Press.
Bruner, J. (1990). Acts of meaning. Cambridge: Harvard University Press.
Bruner, J. (2002). Making stories: Law, literature, life. Cambridge: Harvard
University Press.
Bucci, W. (1997). Psychoanalysis and cognitive science: A multiple code theory. New
York: Guilford.
Bucci, W. (2007a). Dissociation from the perspective of multiple code theory,
Part I: Psychological roots and implications for psychoanalytic treatment.
Contemporary Psychoanalysis, 43, 165–184.
Bucci, W. (2007b). Dissociation from the perspective of multiple code theory, Part
II: Contemporary Psychoanalysis, 43.
Chasseguet-Smirgel, J. (1990). On acting out. International Journal of Psycho-
analysis, 71, 77–86.
Chefetz, R. A. (2003). Healing haunted hearts: Toward a model for integrating sub-
jectivity: Commentary on papers by Gerald Stechler and Philip Bromberg.
Chefetz, R. A., & Bromberg, P. M. (2004). Talking with “me” and “not-me.”
Corbett, K. (2001). More life: Centrality and marginality in human development.
Cushman, P. (1990). Why the self is empty: Toward a historically situated psychol-
ogy. American Psychologist, 45, 599–611.
Cushman, P. (1991). Ideology obscured: Political uses of the self in Daniel Stern’s
infant. American Psychologist, 46, 206–219.
Cushman, P. (1995). Constructing the self, constructing America: A cultural history
of psychotherapy. Reading, MA: Addison-Wesley.
Cushman, P. (2005a). Between arrogance and a dead-end: Gadamer and the
Heidegger/Foucault dilemma. Contemporary Psychoanalysis, 41, 399–417.
Cushman, P. (2005b). Clinical applications: A response to Layton. Contemporary
210 References
Cushman, P. (2007). A burning world, an absent god: Midrash, Hermeneutics, and

relational psychoanalysis. Contemporary Psychoanalysis, 43, 47–88.
Cushman, P., & Gilford, P. (1999). From emptiness to multiplicity: The self at the
year 2000. Psychohistory Review, 27, 15–31.
Damasio, A. (1994). Descartes’s error: Emotion, reason and the human brain. New
York: G. P. Putnam’s Sons.
Davies, J. M. (1996). Linking the pre-analytic with the postclassical: Integration,
dissociation, and the multiplicity of unconscious processes. Contemporary
Davies, J. M. (1997). Dissociation, therapeutic enactment, and transference–coun-
tertransference processes: A discussion of papers on childhood sexual abuse
by S. Grand and J. Sarnat. Gender and Psychoanalysis, 2, 241–257.
Davies, J. M. (1998). The multiple aspects of multiplicity: Symposium on clinical
choices in psychoanalysis. Psychoanalytic Dialogues, 8, 195–206.
Davies, J. M. (1999). Getting cold feet defining “safe-enough” borders: Dissociation,
multiplicity, and integration in the analyst’s experience. Psychoanalytic
Quarterly, 78, 184–208.
Davies, J. M. (2003). Falling in love with love: Oedipal and postoedipal manifes-
tations of idealization, mourning, and erotic masochism. Psychoanalytic
Davies, J. M. (2004). Whose bad objects are we anyway? Repetition and our elusive
love aἀair with evil. Psychoanalytic Dialogues, 14, 711–732.
Davies, J. M., & Frawley, M. G. (1991). Dissociative processes and transference-
countertransference paradigms in the psychoanalytically oriented treatment
of adult survivors of childhood sexual abuse. Psychoanalytic Dialogues, 2,
5–36.
Davies, J. M., & Frawley, M. G. (1994). Treating the adult survivor of childhood
sexual abuse. New York: Basic Books.
Defoe, D. (1957). Robinson Crusoe. New York: Charles Scribner’s Sons. (Originally
published 1719).
Dell, P., & O’Neil, P. (2009). Dissociation and the dissociative disorders: DSM-V and
beyond. New York: Routledge.
Dostal, R.J., (ed.) (2002). ἀ e Cambridge companion to Gadamer. Cambridge:
Cambridge University Press.
Ehrenberg, D. (1992). ἀ e intimate edge. New York: Norton.
Eigen, M. (1981). The area of faith in Winnicott, Lacan and Bion. International
Journal of Psychoanalysis, 62, 413–433.
Elkind, S. (1992). Resolving impasses in therapeutic relationships. New York: Guilford.
Ellenberger, H. F. (1970). ἀ e discovery of the unconscious. New York: Basic Books.
Elliott, A., & Spezzano, C. (1996). Psychoanalysis at its limits: Navigating the post-
modern turn. In O. Renik (Ed.), Knowledge and authority in the psychoana-
lytic relationship (pp. 61–92). Northvale, NJ: Aronson.
Eshel, O. (2004). Let it be and become me: Notes on containing, identification, and
the possibility of being. Contemporary Psychoanalysis, 40, 323–351.
References 211
Farber, L. (1956). Martin Buber and psychoanalysis. In ἀ e ways of the will: Essays
toward a psychology and psychopathology of will (pp. 131–154). New York:
Basic Books.
Fain, M., & David, C. (1963). Aspects fonctionnels de la vie onirique. Revue
Française de Psychanalyse, 27, 241–343.
Fain, M., David, C., & Marty, P. (1964). Perspective psychosomatique sur la fonc-
tion des fantasmes. Revue Française de Psychanalyse, 28, 609–622.
Feiner, A. H. (1982). Comments on the difficult patient. Contemporary
Felman, S., & Laub, D. (1992). Testimony: Crises of witnessing in literature, psycho-
analysis, and history. New York: Routledge.
Ferro, A. (1999). ἀ e bi-personal field: Experiences in child analysis. London:
Routledge.
Ferro, A. (2002). In the analyst’s consulting room. London: Brunner-Routledge.
Ferro, A. (2005a). Bion: Theoretical and clinical observations. Int. J. Psycho-Anal.,
86:1535-1542.
Ferro, A. (2005b). Seeds of illness, seeds of recovery: ἀ e genesis of suffering and the
role of psychoanalysis. London: Brunner-Routledge.
Ferro, A. (2006). Psychoanalysis as therapy and storytelling. London: Routledge.
Fingarette, H. (1963). ἀ e self in transformation: Psychoanalysis, philosophy and the
life of the spirit. New York: Basic Books.
Fischer, K. W., & Rose, S. P. (1999). Rulers, clocks, and nonlinear dynamics:
Measurement and method in developmental research. In G. Savelsbergh,
H. van der Maas, & P. van Geert (Eds.), Nonlinear developmental pro-
cesses (pp. 197–212). Amsterdam: Royal Netherlands Academy of Arts
and Sciences.
Flax, J. (1996). Taking multiplicity seriously: Some implications for psychoanalytic
theorizing and practice. Contemporary Psychoanalysis, 32, 577–593.
Fonagy, P., Gergely, G., Jurist, E., & Target, M. (2002). Affect regulation, mentaliza-
tion, and the development of the self. New York: Other Press.
Foucault, M. (1980) Truth and power. In: The Foucault Reader, ed. P. Rabinow.
New York: Pantheon, 1984, pp. 51-75.
Fourcher, L. (1992). Interpreting the relative unconscious. Psychoanalytic
Frie, R. (1997). Subjectivity and intersubjectivity in modern philosophy and psycho-
analysis. Lanham, MD: Rowman & Littlefield.
Freud, S. (1895). Project for a scientific psychology. In J. Strachey (Ed. & Trans.),
ἀ e standard edition of the complete psychological works of Sigmund Freud
(Vol. 1, pp. 281–387). London: Hogarth Press, 1950.
Freud, S. (1918). From the history of an infantile neurosis. In J. Strachey (Ed. &
Trans.), ἀ e standard edition of the complete psychological works of Sigmund
Freud (Vol. 17, pp. 1–122). London: Hogarth Press, 1955.
212 References
Friedman, L. (2000). Are minds objects or dramas? In D. K. Silverman & D. L.

Wolitzky (Eds.), Changing conceptions of psychoanalysis: ἀ e legacy of Merton
Gill (pp. 146–170). Hillsdale, NJ: Analytic Press.
Fromm, E. (1941). Escape from freedom. New York: Rinehart and Company.
Fromm, E. (1947). Man for himself. New York: Rinehart and Company.
Fromm, E. (1955). ἀ e sane society. New York: Holt, Rinehart and Winston.
Gadamer, H.-G. (1976). Philosophical hermeneutics. Trans. & ed. D.E. Linge.
Berkeley, CA: University of California Press.
Gadamer, H.-G. (1975). Truth and method. G. Barden and J. Cumming, Trans.
New York: Seabury Press.
Gadamer, H.-G. (1965/2004). Truth and method. J. Weinsheimer & D. G. Marshall,
Trans. London: Continuum. (Original work published 1965).
Galatzer-Levy, R. M. (1978). Qualitative change from quantitative change:
Mathematical catastrophe theory in relation to psychoanalysis. Journal of the
American Psychoanalytic Association, 26, 921–935.
Galatzer-Levy, R. M. (1996). Psychoanalysis and dynamical systems theory:
Prediction and self similarity. Journal of the American Psychoanalytic
Association, 43, 1085–1113.
Galatzer-Levy, R. M. (2004). Chaotic possibilities: Toward a new model of devel-
opment. International Journal of Psychoanalysis, 85, 419–442.
Gartner, R. B. (1995). The relationship between interpersonal psychoanalysis and
family therapy. In M. Lionells, J. Fiscalini, C. H. Mann, & D. B. Stern (Eds.),
ἀ e handbook of interpersonal psychoanalysis (pp. 793–822). Hillsdale, NJ:
Analytic Press.
Geertz, C. (1974). “From the native’s point of view”: On the nature of anthro-
pological understanding. In Local knowledge (pp. 55–70). New York: Basic
Books, 1983.
Gerber, L. (1990). Integrating political-societal concerns in psychotherapy.
American Journal of Psychotherapy, 44, .
Ghent, E. (1989). Credo—The dialectics of one-person and two-person psycholo-
gies. Contemporary Psychoanalysis, 25, 169–211.
Ghent, E. (1990). Masochism, submission, surrender—Masochism as a perversion
of surrender. Contemporary Psychoanalysis, 26, 108–136.
Ghent, E. (1995). Interaction in the psychoanalytic situation. Psychoanalytic
Ghent, E. (2002). Wish, need, drive: Motive in the light of dynamic systems theory
and Edelman’s selectionist theory. Psychoanalytic Dialogues, 12, 763–808.
Gianino, A., & Tronick, E. (1988). The mutual regulation model: The infant’s self
and interactive regulation. Coping and defense capacities. In T. Field et al.
(Eds.), Stress and coping (pp. 47–68). Hillsdale, NJ: Erlbaum.
Gill, M. M. (1982). ἀ e analysis of transference: Volume 1. New York: International
Universities Press.
Goldner, V. (2004). Review essay: Attachment and Eros: Opposed or synergistic?
References 213
Goldner, V. (2006). “Let’s do it again”: Further reflections on Eros and attachment.

Green, A. (1975). The analyst, symbolization, and absence in the analytic setting.
International Journal of Psychoanalysis, 56, 1-22.
Green, A. (1976). The dead mother. In On private madness (pp. 142–173). London:
Hogarth Press.
Green, A. (2000). The central phobic position: A new formulation of the free asso-
ciation method. International Journal of Psycho-Analysis, 81, 429–451.
Greenberg, J. (1991). Oedipus and beyond: A clinical theory. Cambridge: Harvard
University Press.
Greenberg, J. (1999). Analytic authority and analytic restraint. Contemporary
Greenberg, J., & Mitchell, S. A. (1983). Object relations in psychoanalytic theory.
Boston: Harvard University Press.
Grondin, J. (1997). Introduction to philosophical hermeneutics. New Haven, CT:
Yale University Press.
Grossmark, R. (2007). The edge of chaos: Enactment, disruption, and emergence
in group psychotherapy. Psychoanalytic Dialogues, 17, 479–499.
Guignon, C. (2004). On being authentic. London: Routledge.
Habermas, J. (1971). Knowledge and human interests. Boston: Beacon Press.
Harris, A. (1996). The conceptual power of multiplicity. Contemporary
Harris, A. (2005). Gender as soft assembly. Hillsdale, NJ: Analytic Press.
Hirsch, I. (1996). Observing-participation, mutual enactment, and the new classi-
cal models. Contemporary Psychoanalysis, 32, 359–383.
Hirsch, I. (2000). Interview with Benjamin Wolstein. Contemporary Psychoanalysis,
36, 187–232.
Hoἀman, I. Z. (1998). Ritual and spontaneity in the psychoanalytic process: A dia-
lectical-constructivist view. Hillsdale, NJ: Analytic Press.
Hoἀman, I. Z. (2006, January). Doublethinking our way to “scientific” legitimacy:
The desiccation of human experience. Plenary address at the American
Psychoanalytic Association meeting. New York.
Howell, E. (2005). ἀ e dissociative mind. Hillsdale, NJ: Analytic Press.
International Association of Relational Psychoanalysis and Psychotherapy. (2005).
Online colloquium: What happens when love lasts? An exploration of intimacy
and erotic life. New York: Psychoanalytic Connection.
Jaἀe, J., Beebe, B., Feldstein, S., & Crown, C. (2001). Rhythms of dialogue in infancy:
Monographs for the society for research and child development. New York: Wiley.
Kavanagh, G. (1995). Processes of therapeutic action and change. In M. Lionells,
J. Fiscalini, C. H. Mann, & D. B. Stern (Eds.), ἀ e handbook of interpersonal
psychoanalysis (pp. 569–602). Hillsdale, NJ: Analytic Press.
Kieἀer, C. C. (2007). Emergence and the analytic third: Working at the edge of
chaos. Psychoanalytic Dialogues, 17, 683–703.
214 References
Knoblauch, S. H. (2000). ἀ e musical edge of therapeutic relatedness. Hillsdale, NJ:

Analytic Press.
Kohut, H. (1984). How does analysis cure? Edited by A. Goldberg & P. Stepansky.
Chicago and London: University of Chicago Press.
Lacan, J. Ecrits: A selection. New York: Norton.
Lachmann, F., & Beebe, B. (1996). Three principles of salience in the patient-ana-
lyst interaction. Psychoanalytic Psychology, 13, 1–22.
Lakoἀ, G., & Johnson, M. (1980). Metaphors we live by. Chicago and London:
University of Chicago Press.
Lakoἀ, G., & Johnson, M. (1999). Philosophy in the flesh: ἀ e embodied mind and
its challenge to western thought. New York: Basic Books.
Lashley, K. S. (1950). In search of the engram. Society of Experimental Biology
Symposium, 4, 454–482.
Laub, D. (1992a). Bearing witness or the vicissitudes of witnessing. In S. Felman
& D. Laub (Eds.), Testimony: Crises of witnessing in literature, psychoanalysis,
and history (pp. 57–74). New York and London: Routledge.
Laub, D. (1992b). An event without a witness: Truth, testimony, and survival. In S.
Felman & D. Laub (Eds.), Testimony: Crises of witnessing in literature, psycho-
analysis, and history (pp. 75–92). New York and London: Routledge.
Laub, D. (1995). Truth and testimony. In C. Caruth (Ed.), Trauma: Explorations in
memory (pp. 61–75). Baltimore: Johns Hopkins University Press.
Laub, D. (2005). Traumatic shutdown of symbolization and narrative: A death
instinct derivative? Contemporary Psychoanalysis, 41, 307–326.
Laub, D., & Auerhahn, N. (1989). Failed empathy: A central theme in the survi-
vor’s Holocaust experience. Psychoanalytic Psychology, 6, 377–400.
Layton, L. (1998), Who’s that girl? Who’s that boy? Clinical practice meets postmod-
ern gender theory. Hillsdale, NJ: Analytic Press.
Layton, L. (2002). Psychoanalysis and the “free” individual. Journal of Psycho-
Social Studies, 1.
Layton, L. (2004a). Dreams of America, American dreams. Psychoanalytic
Dialogues, 14, 233–254.
Layton, L. (2004b). A fork in the royal road: On “defining” the unconscious and its
stakes for social theory. Psychoanalysis, Culture, and Society, 9, 33–51.
Layton, L. (2005). Notes toward a nonconformist clinical practice: Response to
Philip Cushman. Contemporary Psychoanalysis, 41, 419–429.
Lecours, S., & Bouchard, M. (1997). Dimensions of mentalisation: Outlining lev-
els of psychic transformation. International Journal of Psycho-Analysis, 78,
855–875.
Levenkron, H. (2006). Love (and hate) with the proper stranger: Aἀective honesty
and enactment. Psychoanalytic Inquiry, 26, 157–181.
Levenson, E. A. (1972). ἀ e fallacy of understanding. New York: Basic Books.
Levenson, E. A. (1983). ἀ e ambiguity of change. New York: Basic Books.
References 215
Levenson, E. A. (1984). Follow the fox: An inquiry into the vicissitudes of psycho-
analytic supervision. In A. H. Feiner (Ed.), ἀ e purloined self: Interpersonal
perspectives in psychoanalysis (pp. 111–125). New York: Contemporary
Psychoanalysis Books.
Levenson, E. A. (1988). The pursuit of the particular. Contemporary Psychoanalysis,
24, 1–16.
Levenson, E. A. (1991). ἀ e purloined self: Interpersonal perspectives in psychoanal-
ysis. A. H. Feiner (Ed.). New York: Contemporary Psychoanalysis Books.
Levenson, E. A. (1994). The uses of disorder: Chaos theory and psychoanalysis.
Lionells, M., Fiscalini, J., Mann, C. M., & Stern, D. B. (Eds.). (1995). ἀ e handbook
of interpersonal psychoanalysis. Hillsdale, NJ: Analytic Press.
Loewald, H. W. (1960). On the therapeutic action of psycho-analysis. International
Journal of Psychoanalysis, 41, 16–33.
Luquet, P. (1987). Penser-parler: un apport psychanalytique la thorie du langage.
In R. Christie et al. (Eds.), La parole trouble (pp. 161–300). Paris: Presses
Universitaires France.
Lyons-Ruth, K. (1991). Rapprochement or approchement: Mahler’s theory recon-
sidered from the vantage point of recent research on early attachment rela-
tionships. Psychoanalytic Psychology, 8, 1–23.
Lyons-Ruth, K. (2003). Dissociation and the parent-infant dialogue. Journal of the
American Psychoanalytic Association, 51(3), 883–911.
Lyons-Ruth, K., Bruschweiler-Stern, N., Harrison, A. M., Nahum, J. P., Sander, L.,
Stern, D. N., et al. (1998). Implicit relational knowing: Its role in development
and psychoanalytic treatment. Infant Mental Health Journal, 19, 282–289.
Lyons-Ruth, K., Connell. D., Zoll, D., & Stahl, J. (1987). Infants at social risk:
Relationships among infant maltreatment, maternal behavior, and infant
attachment behavior. Developmental Psychology, 23, 223–232.
Lyons-Ruth, K., Connell. D., Zoll, D., & Stahl, J. (1950). The development of the
transference. Psychoanalytic Quarterly, 19, 501–539.
Macalpine, I. (1950). The development of the transference. ἀe Psychoanalytic
Quarterly, 19: 501–539.
Mandel, A. J., & Selz, K. A. (1996). Nonlinear dynamical patterns as personality
theory for neurobiology and psychoanalysis. Psychiatry, 58, 371–390.
Marty, P. (1990). La psychosomatique de l’adulte. Paris: Presses Universitaires France.
Marty, P. (1991). Mentalisation et psychosomatique. Paris: Laboratoire Delagrange.
McDougall, J. (1985). ἀ eaters of the mind. New York: Basic Books.
McLaughlin, J. T. (1996). Power, authority, and influence in the analytic dyad. In
O. Renik (Ed.), Knowledge and authority in the psychoanalytic relationship
(pp. 189–223). Northvale, NJ: Aronson.
Merleau-Ponty, M. (1964a). Introduction. In Signs (pp. 3–35). Evanston, IL:
Northwestern University Press.
Merleau-Ponty, M. (1964b). Indirect language and the voices of silence. In Signs
(pp. 339–383). Evanston, IL: Northwestern University Press.
216 References
Merleau-Ponty, M. (1973). ἀ e prose of the world. Evanston, IL: Northwestern

University Press.
Messer, S.B., Sass, L.A. & Woolfolk, R.L. (eds.) (1988). Hermeneutics and psycho-
logical theory. New Brunswick, NJ: Rutgers University Press.
Miller, M. (1999). Chaos, complexity, and psychoanalysis. Psychoanalytic
Psychology, 16, 335–379.
Mitchell, S. A. (1988). Relational concepts in psychoanalysis. Cambridge: Harvard
University Press.
Mitchell, S. A. (1991). Contemporary perspectives on self: Toward an integration.
Mitchell, S. A. (1993). Hope and dread in psychoanalysis. New York: Basic Books.
Mitchell, S. A. (1997). Influence and autonomy in psychoanalysis. Hillsdale, NJ:
Analytic Press.
Mitchell, S. A. (2000). Relationality. Hillsdale, NJ: Analytic Press.
Mitchell, S. A. (2002). Can love last? ἀ e fate of romance over time. New York: Norton.
Mitrani, J.L. (1995). Toward an understanding of unmentalized experience.
Psychoanal Q., 64:68-112.
Modell, A. (1984). Psychoanalysis in a new context. New York: International
Universities Press.
Modell, A. (1990). Other times, other realities. Cambridge: Harvard University Press.
Modell, A. (1991). The therapeutic relationship as a paradoxical experience.
Modell, A. (2003). Imagination and the meaningful brain. Cambridge: MIT Press.
Moran, M. (1991). Chaos theory and psychoanalysis: The fluidic nature of the
mind. International Journal of Psychoanalysis, 18, 211–221.
Nagel, T. (1986). ἀ e view from nowhere. New York: Oxford University Press.
Nahum, J. (1994). New theoretical vistas in psychoanalysis: Louis Sander’s theory
of early development. Psychoanalytic Psychology, 11, 1–19.
Nahum, J. (2000). An overview of Louis Sander’s contribution to the field of men-
tal health. Infant Mental Health Journal, 21(1–2), 29–41.
Nelson, K. (Ed.). (1989). Narratives from the crib. Cambridge: Harvard
University Press.
Ogden, T. H. (1994). Subjects of analysis. Northvale, NJ: Jason Aronson.
Ogden, T. H. (2004). The analytic third: Implications for psychoanalytic theory
and technique. Psychoanalytic Quarterly, 73, 167–195.
Onishi, N. (2007, October 12). Death reveals harsh side of a “model” in Japan. New
York Times.
Orange, D. M. (1995). Emotional understanding: Studies in psychoanalytic episte-
mology. New York: Guilford.
Palmer, R.E. (1969). Hermeneutics. Evanston, IL: Northwestern University Press.
Palombo, S. R. (1999). ἀ e emergent ego: Complexity and coevolution in the psycho-
analytic process. Madison, CT: International Universities Press.
Piers, C. (2000). Character as self-organizing complexity. Psychoanalysis and
Contemporary ἀ ought, 23, 3–34.
References 217
Piers, C. (2005). The mind’s multiplicity and continuity. Psychoanalytic Dialogues,

15, 229–254.
Pizer, B. (2003). When the crunch is a (k)not: A crimp in relational dialogue.
Pizer, S. (1998). Building bridges: ἀ e negotiation of paradox in psychoanalysis.
Hillsdale, NJ: Analytic Press.
Pizer, S. (2004). Impasse recollected in tranquility: Love, dissociation, and disci-
pline in the analytic process. Psychoanalytic Dialogues, 14, 289–311.
Plato. (1956). Protagoras and Meno. W. K. C. Guthrie, Trans. London: Penguin Books.
Poland, W. S. (2000). The analyst’s witnessing and otherness. Journal of the
American. Psychoanalytic Association, 48, 17–34.
Polkinghorne, D. E. (1988). Narrative knowing in the human sciences. Albany, NY:
State University of New York Press.
Quinodoz, J.-M. (1997). Transition in psychic structures in light of deterministic
chaos theory. International Journal of Psychoanalysis, 78, 699–718.
Racker, H. (1957). The meanings and uses of countertransference. In Transference and
Countertransference (pp. 127–173). New York: International Universities Press,
1968.
Racker, H. (1968). Transference and countertransference. New York: International
Universities Press.
Renik, O. (1993a). Analytic interaction: Conceptualizing technique in light of the
analyst’s irreducible subjectivity. Psychoanalytic Quarterly, 62, 553–571.
Renik, O. (1993b). Countertransference enactments and the psychoanalytic pro-
cess. In M. Horowitz, O. Kernberg, & E. Weinshel (Eds.), Psychic structure and
psychic change (pp. 135–158). New York: International Universities Press.
Richardson, F., & Zeddies, T. (2001). Individualism and modern psychotherapy.
In B. Slife, R. Williams, & S. Barlow (Eds.), Critical issues in psychotherapy:
Translating new ideas into practice (pp. 147–167). Thousand Oaks, CA: Sage.
Richardson, F., & Zeddies, T. (2004). Psychoanalysis and the good life. Contem-
porary Psychoanalysis, 40, 617–657.
Richman, S. (2002). A wolf in the attic: ἀ e legacy of a hidden child of the Holocaust.
New York: Haworth Press.
Richman, S. (2006). Finding one’s voice: Transforming trauma into autobiographi-
cal narrative. Contemporary Psychoanalysis, 42, 639–650.
Ricoeur, P. (1970). Freud and philosophy: An essay on interpretation. New Haven,
CT: Yale University Press.
Ricoeur, P. (1977). The question of proof in Freud’s psychoanalytic writings.
Journal of the American Psychoanalytic Association, 25, 835–871.
Ricoeur, P. (1981). Hermeneutics and the human sciences. Cambridge: Cambridge
University Press.
Ringstrom, P. (1998). Therapeutic impasses in contemporary psychoanalytic treat-
ment: Revisiting the double bind hypothesis. Psychoanalytic Dialogues, 8,
297–315.
218 References
Ringstrom, P. (2001). Cultivating the improvisational in psychoanalytic treatment.

Ringstrom, P. (2007). Scenes that write themselves: Improvisational moments in
relational psychoanalysis. Psychoanalytic Dialogues, 17, 69–100.
Russell, P. (1991). Trauma, repetition, and aἀect. Presented at the First Symposium,
Massachusetts Institute for Psychoanalysis. Contemporary Psychoanalysis,
42, 601–620.
Russell, P. (2006a). The theory of the crunch: The collected papers of Paul Libbey
Russell, MDH. Smith College Studies in Social Work, 76(1–2), 9–21.
Russell, P. (2006b). The collected papers of Paul Libbey Russell, MDH. Smith
College Studies in Social Work, 76(1–2).
Sander, L. W. (1962). Issues in early mother-child interaction. Journal of the
American Academy of Child Psychiatry, 1, 141–166.
Sander, L. W. (1988). The event-structure of regulation in the neonate-caregiver
system as a biological background for early organization of psychic struc-
ture. Progress in Self Psychology, 3, 64–77.
Sandler, J. (1976). Countertransference and role-responsiveness. International
Review of Psycho-Analysis, 3, 43–47.
Sarbin, T. (1986). Narrative psychology: ἀ e storied nature of human conduct. New
York: Praeger.
Sass, L.A. (1988). Humanism, hermeneutics, and humanistic psychoanalysis:
Diἀering conceptions of subjectivity. Psychoanal. Contemp. ἀ ought, 12:
433-504.
Schafer, R. (1983). ἀ e analytic attitude. New York: Basic Books.
Schafer, R. (1992). Retelling a life: Narration and dialogue in psychoanalysis. New
York: Basic Books.
Searles, H. F. (1975). The patient as therapist to his analyst. In Countertransference
and related subjects (pp. 380–459). New York: International Universities Press.
Searles, H. F. (1976). Psychoanalytic therapy with schizophrenic patients in a pri-
vate practice context. Contemporary Psychoanalysis, 12, 387–406.
Segal, H. (1957). Notes on symbol formation. International Journal of Psychoanalysis,
38, 391–397.
Seligman, S. (2005). Dynamic systems theories as a metaframework for psycho-
analysis. Psychoanalytic Dialogues, 15, 285–319.
Sennett, R. (2008). ἀ e craftsman. New Haven: Yale University Press.
Shane, E., & Coburn, W. J. (Eds.). (2002). Contemporary dynamic systems theo-
ries: Innovative contributions to psychoanalysis. Psychoanalytic Inquiry, 22.
Singer, E. (1971). The patient aids the analyst: Some clinical and theoreti-
cal observations. In B. Landis & E. Tauber (Eds.), In the name of life:
Essays in honor of Erich Fromm (pp. 56–68). New York: Holt, Rinehart
and Winston.
Slavin, M. O. (1996). Is one self enough? Multiplicity in self organization and the
capacity to negotiate relational conflict. Contemporary Psychoanalysis, 32,
615–625.
References 219
Slavin, M. O., & Kriegman, D. (1992). ἀ e adaptive design of the human psyche.
New York: Guilford.
Slavin, M. O., & Kriegman, D. (1998). Why the analyst needs to change: Toward a
theory of conflict, negotiation, and mutual influence in the therapeutic pro-
cess. Psychoanalytic Dialogues, 8, 247–284.
Smith, L. B., & Thelen, E. (1993). A dynamic systems approach to development:
Applications. Cambridge, MA: MIT Press.
Spence, D. P. (1982). Narrative truth and historical truth: Meaning and interpreta-
tion in psychoanalysis. New York: Norton.
Spence, D. P. (1987). ἀ e Freudian metaphor: Toward paradigm change in psycho-
analysis. New York: Norton.
Spruiell, V. (1993). Deterministic chaos and the sciences of complexity:
Psychoanalysis in the midst of a general scientific revolution. Journal of the
American Psychoanalytic Association, 41, 3–44.
Stein, A. (2003). Dreaming while awake: The use of trance to bypass threat.
Contemporary Psychoanalysis, 39: 179–197.
Stein, A. (2004). Fantasy, fusion, and sexual homicide. Contemporary Psychoanalysis,
40, 495–517.
Stein, A. (2006). Prologue to violence: Child abuse, dissociation, and crime. Hillsdale,
NJ: Analytic Press.
Stern, D. B. (1983). Unformulated experience. Contemporary Psychoanalysis, 19,
71–99.
Stern, D. B. (1989). The analyst’s unformulated experience of the patient.
Stern, D. B. (1990). Courting surprise: Unbidden experience in clinical practice.
Stern, D. B. (1991). A philosophy for the embedded analyst: Gadamer’s hermeneu-
tics and the social paradigm of psychoanalysis. Contemporary Psychoanalysis,
27, 51–58.
Stern, D. B. (1994). Empathy is interpretation (and who ever said it wasn’t?).
Stern, D. B. (1996). The social construction of therapeutic action. Psychoanalytic
Inquiry, 16, 265–293.
Stern, D. B. (1997). Unformulated experience: From dissociation to imagination in
psychoanalysis. Hillsdale, NJ: Analytic Press.
Stern, D. B. (2000). The limits of social construction: Discussion of Dyess and
Dean. Psychoanalytic Dialogues, 10, 757–769.
Stern, D. B. (2001). Comments on the clinical material presented by Jill Scharἀ.
Psychoanalytic Inquiry, 21, 499–512.
Stern, D. B. (2002a). Words and wordlessness in the psychoanalytic situation.
Journal of the American Psychoanalytic Association, 50, 221–247.
220 References
Stern, D. B. (2002b). What you know first: Construction and deconstruction

in relational psychoanalysis. In S. Fairfield, L. Layont, & C. Stack (Eds.),
Bringing the plague: Toward a postmodern psychoanalysis (pp. 167–194). New
York: Other Press.
Stern, D. B. (2002c). Language and the nonverbal as a unity: Discussion of
Neil Altman’s “Where is the action in the ‘talking cure’?” Contemporary
Stern, D. B. (2003). The fusion of horizons: Dissociation, enactment, and under-
standing. Psychoanalytic Dialogues, 13, 843–873.
Stern, D. B. (2004). The eye sees itself: Dissociation, enactment, and the achieve-
ment of conflict. Contemporary Psychoanalysis, 40, 197–237.
Stern, D. B. (2005). Introduction. In E. A. Levenson, ἀ e fallacy of understanding
and the ambiguity of change (pp. v–xvi). Hillsdale, NJ: Analytic Press.
Stern, D. B. (2006). Opening what has been closed, relaxing what has been
clenched: Dissociation and enactment over time in committed relationships.
Stern, D. B. (2008). On having to find what you don’t know how to look for. In A.
Slade, S. Bergner, & E. L. Jurist (Eds.), Mind to mind: Infant research, neuro-
science, and psychoanalysis (pp. 398–413). New York: Other Press.
Stern, D. B. (2009). Dissociation and unformulated experience: A psychoanalytic
model of mind. In P. F. Dell, J. O’Neil, & E. Somer (Eds.), Dissociation and
the dissociative disorders: DSM-V and beyond (pp. 653–663). New York:
Routledge.
Stern, D. N. (1971). A micro-analysis of mother-infant interaction: Behaviors
regulating social contact between a mother and her three-and-a-half
month-old twins. Journal of the American Academy of Child Psychiatry, 10,
501–517.
Stern, D. N. (1977). ἀ e first relationship. Cambridge: Harvard University Press.
Stern, D. N. (1985). ἀ e interpersonal world of the infant: A view from psychoanaly-
sis and developmental psychology. New York: Basic Books.
Stern, D. N. (1995). ἀ e motherhood constellation. New York: Basic Books.
Stern, D. N. (2004). ἀ e present moment in psychotherapy and everyday life. New
York: Norton.
Stern, D. N., Sander, L. W., Nahum, J. P., Harrison, A. M., Lyons-Ruth, K., Morgan,
A. C., et al. (1998). Non-interpretive mechanisms in psychoanalytic ther-
apy: The “something more” than interpretation. International Journal of
Stolorow, R. D. (1988). Intersubjectivity, psychoanalytic knowing, and reality.
Contemp. Psychoanal., 24: 331-338.
Stolorow, R. D., Brandchaft, B. and Atwood, G. E. (1987). Psychoanalytic treat-
ment: An intersubjective approach. Hillsdale, NJ: The Analytic Press.
Sugarman, A. (2008). Fantasizing as process, not fantasy as content. Psychoanalytic
Inquiry, 28, 169–189.
References 221
Sullivan, H. S. (1953). Conceptions of modern psychiatry. New York: Norton,

1950.
Sullivan, H. S. (1950). The illusion of personal individuality. In ἀ e fusion of psy-
chiatry and social science (pp. 198–226). New York: Norton, 1971.
Sullivan, H. S. (1954). ἀ e interpersonal theory of psychiatry. Edited by H. S. Perry
& M. L. Gawel. New York: Norton.
Sullivan, H. S. (1956). Selective inattention. In H. S. Perry, M. I. Gawel, & M Gibbon
(Eds.), Clinical studies in psychiatry (pp. 38–76). New York: Norton.
Symington, N. (1983). The analyst’s act of freedom as agent of therapeutic change.
International Review of Psycho-Analysis, 10, 283–292.
Tauber, E. S. & Green, M. R. (1959). Prelogical experience. New York: Basic Books.
Republished by Routledge (New York), 2008.
Tauber, E. S. (1954). Exploring the therapeutic use of counter-transference data.
Psychiatry, 13, 332–336.
Tauber, E. S. (1979). Countertransference reexamined. In L. Epstein & A. H. Feiner
(Eds.), Countertransference (pp. 59–69). New York: Aronson.
Thelen, E., & Smith, L. (1994). A dynamic systems approach to the development of
cognition and action. Cambridge, MA: MIT Press.
Tronick, E. Z. (1989). Emotions and emotional communication in infants.
American Psychologist, 44, 112–119.
Tronick, E. Z. (Ed.). (1998). Interactions that eἀect change in psychotherapy: A
model based on infant research. Infant Mental Health Journal, 19(3).
Tronick, E. Z., & Weinberg, K. (1997). Depressed mothers and infants: The
failure to form dyadic states of consciousness. In L. Murray & P. Cooper
(Eds.), Postpartum depression and child development (pp. 54–85). Hillsdale,
NJ: Guilford.
Tulving, E., & Craik, F. (Eds.). (2000). ἀ e Oxford handbook of memory. Oxford:
Oxford University Press.
von Bertalanἀy, L. (1968). General systems theory. New York: Braziller.
Warnke, G. (1987) Gadamer: Hermeutics, tradition and reason. Stanford, CA:
Stanford University Press.
Winnicott, D. W. (1945). Primitive emotional development. In ἀ rough paediatrics
to psycho-analysis (pp. 145–156). New York: Basic Books.
Winnicott, D. W. (1949). Hate in the countertransference. In ἀ rough paediatrics
to psycho-analysis (pp. 194–203). New York: Basic Books.
Winnicott, D. W. (1960). Ego distortion in terms of true and false self. In ἀe
maturational processes and the facilitating environment: Studies in the
theory of emotional development (pp. 140–152). New York: International
Universities Press.
Winnicott, D. W. (1971). Playing and reality. London: Tavistock.
Wolstein, B. (1954). Transference. New York: Grune & Stratton.
Wolstein, B. (1959). Countertransference. New York: Grune & Stratton.
Wolstein, B. (1971a). Human psyche in psychoanalysis. Springfield, IL: Thomas.
222 References
Wolstein, B. (1971b). Interpersonal relations without individuality. Contemporary

Wolstein, B. (1972). Interpersonal relations without individuality again.
Wolstein, B. (1974a). Individuality and identity. Contemporary Psychoanalysis, 10,
1–14.
Wolstein, B. (1974b). “I” processes and “me” patterns. Contemporary Psychoanalysis,
10, 347–357.
Wolstein, B. (1975). Toward a conception of unique individuality. Contemporary
Wolstein, B. (1983). The pluralism of perspectives on countertransference. In
Essential papers on countertransference (pp. 339–353). New York: New York
University Press.
Zeddies, T. J. (2002). More than just words: A hermeneutic view of language in
psychoanalysis. Psychoanalytic Psychology, 19, 3–23.
Zeddies, T. J., & Richardson, F. C. (1999). Analytic authority in historical and
critical perspective: Beyond objectivism and relativism. Contemporary
Index
A Bronfenbrenner, U., 192
Bruner, J. S., 23
Absence of conflict, 90–93 Bucci, W., 12
Achievement of conflict, 86–87
Action, deferred, 136
Alexander, F., 23 C
Analytic relatedness, 71–73 Cast Away, 125
Anger, 58–60, 175, 176 Category and metaphor, 134–136
Appraisals, reflected, 111 Chafing, emotional, 81–82, 104–105
Aron, L., 8, 172, 183 Character, 8–9, 14
Atwood, 20 Chasseguet-Smirgel, J., 20
Auerhahn, N., 110, 126 Chefetz, R. A., 164, 189
Authentic conversation, 33–34, 40 Collaborative conversation, 40–41
enactment interrupting, 40 Complexity theory, 72
as relatedness, 41 Conceptualized enactment, 12–14
Autonomous critical command, 128 Conflict
Avoidance, 148 achievement of, 86–87
dissociation and enactment as absence of,
B 90–93
reformulation and, 99–100
Bach, S., 111, 113 single-mindedness and internal, 100–103
Bad-me, 87–88, 91, 149–154 sources of, 87–88
Bartlett, F. C., 4, 196 unconscious, 92, 94
Beebe, B., 78, 188 Constraint on unformulated experience, 2–3
Benjamin, J., 8, 28, 102, 111, 113, 154 Constructivism, 7
Bion, W. F., 17, 20, 90, 119 splitting and, 94–96
Bollas, C., 20, 67–68, 77 Constructivist conceptions of dissociation,
Bootstrap operation, 151 93–94
Boston Change Process Study Group (BCPSG), Context
12, 72, 115, 183 enactment and, 145–146
on clinical process, 183–184 in flux, multiple self as, 48–53
Daniel B. Stern agreement with, 183–187 and fusion of horizons, 43–48
on enactment, 187–190 Continuous mutual influence, 184–185
natural science viewpoint, 199–201 Continuous productive unfolding, 116–118
nonlinear dynamic systems theory (NDST) Conversation
and, 190–193, 194–195, 203 agreement in, 27–28
Botella and Botella, 20 authentic, 33–34, 40
Bouchard, M., 20 as collaborative, 40–41
Boulanger, G., 110 hermeneutic frame of reference on, 25–26,
Brandcraft, 20 28–29, 31
Brison, S., 110 prejudice and preconceptions in, 29–30
Bromberg, P., 173, 189 process, 27
on autonomous critical command, 128 questioning in, 30–33
on consciousness, 94 true, 33–34
on dissociation, 13, 90, 91, 149, 150 for understanding, 26, 27–28
on multiple self, 50, 84, 86 Co-occurrences, 138–140, 146
on not-me, 119 Corbett, K., 160
on pathological dissociation, 152 Countertransference, 6, 7, 9, 57
on self-states, 187 neurosis, 89–90
on the mind and self, 164 perception and, 11
223
224 Index
as unconscious, 73–74 Emotional snags and chafing, 81–82, 104–105

Craik, F., 4 Empathy, theories of, 31
Creation of meaning, 3, 7 Enactment
Culture, Western, 197 as absence of conflict, 90–93
Cushman, P., 8, 38–39, 197–198, 200 achievement of conflict and, 86–87
BCPSG on, 187–190
D conceptualized, 12–14
dependence and, 154–157
David, C., 20 dissociation and, 86–87, 90–93, 144–146
Davies, J. M., 87, 90, 148, 150 dissociative, 17–18
Deadlock and dissociation, 57–63 guilt and narcissism, 79–81
path to freedom in, 66–69 imagination and, 65–66
Defensive lockdown, 148 interrupting authentic conversation, 40
Defensively motivated dissociation, 50–51, 123 mentalization theory and, 167–170
Deferred action, 136 mutual, 14–16, 121–122, 179–180, 185
Deflation, feelings of, 53–55 mutual dissociation and, 51
Defoe, D., 114 not-me and, 120
Dependence and enactment, 154–157 problems with, 180–181
Dialogue, 25–26. See also Conversation recognition and resolution of, 129
Didion, J., 115 and the multiple self, 83–85
Direct expressions, 66–69 witnessing, and narrative, 122–124
Dissociation, 13, 21 Experience. See also Memory; Unformulated
absence of conflict and, 90–93 experience
achievement of conflict and, 86–87 co-occurrences, 138–140, 146
as constraint on freedom of thought, 64 dark places of, 163
constructivism and, 96 dissociated, 91–92, 181
deadlock and, 57–63 emerging from relatedness, 4–5
defensively motivated, 50–51, 123 as interpretation, 1, 6–7
enactment and, 86–87, 90–93, 144–146 narrative, 107–110, 124–126
failure of, 64–65 parenting, 153–154
imagination and, 63–65 potential, 1, 7
intrapsychic and constructivist conceptions reformulation and, 99–100
of, 93–94 unbidden, 138
mentalization and, 181 unmentalized, 20–21
metaphor and, 138–140 witnessing, 110–116, 136–137
multiple self and, 83–85 Expressions, direct, 66–69
mutual, 51
mutual enactment and, 14–16 F
normal, 149–150
pathological, 152 Fain, M., 20
people locked out of relationships and, Fallacy of Understanding, ἀ e, 73
154–160 Familiarity with others, 154–155
reciprocal, 88–90 Feiner, A. H., 173
within romantic relationships, 149–154 Felman, S., 137
theory, relational, 164–165 Fingarett, H., 184
weak, 57–58 Fischer, K. W., 191
Dissociative enactment, 17–18 Fonagy, P.
Dreams, 59–60, 61 on dissociation, 16, 20
on enactment, 87, 169, 180
E on link between psychic development and
philosophy of mind, 164
Ellenberger, H. F., 139 on mentalization, 20–21, 187
Elliott, A., 8 on narrative, 110
Index 225
on perception, 11 fusion of horizons and, 45

on relationships, 152, 156 meaning and, 204–205
Formulation of meaning phenomenology of understanding and,
as continuous process, 1–2 45–48
ongoing interaction and, 10–12 Hirsch, I., 77
Fourcher, L., 185 Hoἀman, I. Z., 8, 118, 128, 194, 199
Frawley-O’Dea, M.-G., 87, 150 Hostility, 58–60
Freedom
narrative, 116–118 I
pain and creation of, 98–99
path to, 66–69 Identification, projective, 17–18, 151
problem of clinical, 103–105 Identity, 13
of thought, 64 Illusory critic, 85
French, T., 23 Imagination, 63–66
Freud, S., 92, 94, 101, 136 Implicit relational knowing (IRK), 12
Fromm, E., 197 Incredible Shrinking Man, ἀ e, 107–110,
Frozen memory, 135–136, 139 113–114
Function, reflective. See Reflective function Individual subjectivity, 7–8, 13
Fusion of horizons and context, 43–48 Influence, continuous mutual, 184–185
Future of meaning and relatedness, 21–24 Initiative, 102
Intention unfolding process, 186
G Internal conflict and single-mindedness,
100–103
Gabler, N., 131 International Journal of Psychoanalysis, 199
Gadamer, H.-G., 2, 68, 202 Interpersonal field, 172–174
on agreement, 27–28 Interpersonalization of dissociation, 14
on authentic conversation, 33–34, 40 Interpretation
on collaborative conversation, 40–41 experience as, 1, 6–7
on fusion of horizons, 43 text, 127–128
on mutual questioning, 30 Intersubjectivity, 36–37
phenomenology of understanding and, Intrapsychic conceptions of dissociation, 93–94
45–46
on prejudice and preconceptions, 29–30 J
on psychotherapeutic conversation, 28
on questioning, 30–33 Johnson, M., 131–132, 135, 138
on substantive rightness, 26, 27 Journal of the American Psychoanalytic
on understanding conversation, 25, 26 Association, 199
Galatzer-Levy, R. M., 128 Judgment, subjective, 132–133
Geertz, C., 46–47, 197
Ghent, E., 71 K
Gilford, P., 197
Goldner, V., 148, 150 Knoblauch, S. H., 118
Good-me, 87–88, 91, 149–154
Green, A., 20, 21, 119, 158 L
Greenberg, J., 8
Guignon, C., 33–34, 37 Lachmann, F., 78, 188
Guilt, 58, 79–81, 95 Lakoἀ, G., 131–132, 135, 138
Lashley, K. S., 4
Laub, D., 110, 126, 127, 137
H
Lecours, S., 20
Habermas, J., 45 Levenson, E. A., 33, 73, 94, 171
Heidegger, M., 68 Location of unformulated experience, 3–5
Hermeneutics, 25–26, 28–29, 31 Lockdown, defensive, 148
226 Index
Loewald, H. W., 30 N
Love, 148, 158–159. See also Relationships
Nagel, T., 204
Narcissism, 79–81
M Narrative, 107–110
Mapping, 132, 134 enactment, witnessing and, 122–124
Marty, P., 20 freedom, 116–118
Matheson, R., 107, 108 rigidity, 13, 149
McDougall, J., 20 witnessing, 110–116
McLaughlin, J. T., 8 Nelson, K., 113
Meaning Neurosis, countertransference, 89–90
BCPSG on, 201–202 New York Times, 125
creation of, 3, 7 Nonlinear dynamics systems theory (NDST),
emergent quality, 185–186 190–193, 194–195, 203
fusion of horizons and, 43–44 Not-me, 13–15, 19–20, 21, 118–120, 140, 189
of metaphor, 131–133 mutual enactment and, 121–122
objective, 203–204 narrative and, 122–123
ongoing interaction and formulation of, relationships and, 149–154
10–12
relatedness and future of, 21–24 O
unconscious, 5–6
Memory, 4. See also Experience Objective meaning, 203–204
frozen, 135–136, 139 Objectivity, 77
unconscious emotional, 135–136 Oedipal themes, 148, 150
Meno’s Paradox, 162 Ogden, T. H., 28
Mentalization, 20–21, 164–170, 181, 187. See Ongoing relatedness, 114
also Reflective function Onishi, N., 125
Merleau-Ponty, M., 64, 65, 68, 185
Metaphor P
category and, 134–136
definition of, 131, 133 Pain and creation of freedom, 98–99
dissociation and, 138–140 Parenting, 153–154, 158
meaning of, 131–133 Passive dissociation, 13
transfer and transference, 133–134 Path to freedom, 66–69
trauma and, 135, 136–137 Pathological dissociation, 152
witness and, 136–138 Perceptions, 10–12
Metonymy, 135 reflective function and, 173–174, 177–178
Mind worries and, 19
memory and, 4 Personality, 8–9, 85
relatedness and, 5–10 Pizer, S., 58, 84, 128, 150, 185
Mitchell, S. A., 8, 128, 157, 199 Plato, 162
on private region of the self, 75 Political issues, 37–38
on relatedness, 71, 116 Potential experience, 1, 7
on relational matrix, 172 Preconceptions, 29–30
on relationships, 151, 157 Prejudice, 29–30, 46
Mitrani, 20 Primary metaphors, 131–133
Modell, A., 73, 131, 134–136, 168 Private regions, 75–79
Morality, 36–38 Productive unfolding, continuous, 116–118
Multiple self, 45, 48–53, 83–85 Projective identification, 17–18, 151
Mutual enactment, 14–16, 121–122, 179–180, Psychoanalysis/psychotherapy
185 for academic problems, 80
Mutual influence, continuous, 184–185 analysts’ dissociations in, 88–90
Mutual questioning, 30 conversations in, 28, 34, 35
Mutual regulation, 78–79, 188 countertransference neurosis in, 89–90
Index 227
deadlock in, 57–63 relational theory and, 170–172

difficulties in, 142–144, 174–180 significance of, 180–182
early understanding of, 71 study of, 161
emotional snags and chafing in, 81–82 Reformulation, 99–100
enactment in, 86–87, 90–93, 144–146, Regulation, mutual, 78–79, 188
167–170, 187–190 Rejection, 148
fusion of horizons in, 43–45 Relatedness
journals, 199 analytic, 71–73
mentalization theory in, 165–170 authentic conversation as, 41
morality and, 36–38 conceptualizing enactment and, 12–14
mutual enactment in, 14–16, 179–180 experience emerging from, 4–5
mutual exploration of subjectivity and future of meaning and, 21–24
intersubjectivity in, 36 as goal of therapy, 71
mutual regulation in, 188 interlocking of, 73–74
narrative in, 107–110 mind and, 5–10
natural science view of, 199–201 ongoing, 114
path to freedom in, 66–69 states of, 14
private region of the self in, 75–79 Relational dissociation theory, 164–165
projective identification in, 17–18 Relational theory, 170–172
questioning in, 32–33 Relationships
raising patients’ consciousness, 38–39 bad-me, good-me, and not-me in,
reaching understanding in, 53–57 149–154
reasons for people pursuing, 140–141 dissociations within, 149–154
recognition and resolution of enactments enactment, dependence and knowing your
in, 129 partner in, 154–157
relatedness and meaning in, 22–23 love and rejection in, 148
relational theory and, 170–172 oedipal themes in, 148, 150
restoration of self-tolerance in, 96–98 people locked out of, 157–160
science and social construction in, 193–205 sexuality in, 147, 151–152
social constructionism and, 196 Relativism, 1
study of, 198–199 Releasement, 34
as text interpretation, 127–128 Renik, O., 10, 128
unconscious participation of analysts in, Repression, 94, 95
71–73 Richardson, F. C., 8
Psychoanalytic Dialogues, 199 Richman, S., 110, 115–116, 126
Psychosomatic School of Paris, 20 Ricoeur, P., 45
Psychotic transference, 134 Rightness, substantive, 26, 27
Rigidity, narrative, 13, 149
Q Ringstrom, P., 118
Robinson Crusoe, 114–115, 124–125
Questioning in conversation, 30–33 Romantic relationships. See Relationships
Rose, C., 115–116
R Rose, S. P., 191
Racker, H., 49–50, 89, 129 S

Reflected appraisals, 111
Reflective function Sander, L., 186
chicken-and-egg dilemma in, 165 Sandler, J., 181
dark places of experience and, 163 Sass, 1
enactment and, 167–170 Schafer, R., 107, 127
interpersonal field and, 172–174 Schizophrenia, 91
Meno’s Paradox and, 162 Science and psychoanalysis, 193–205
mentalization theory and, 165–170 Searles, H. F., 99
perception and, 173–174, 177–178 Segal, H., 20
228 Index
Self on “the grip of the field,” 73

mentalization theory and, 164–165 on transformative context, 23
multiple, 45, 48–53, 83–85 on unformulated experience, 1, 3, 7, 20, 86,
mutual regulation and, 78–79 184, xiii, xv
nonsocial core of, 77 on weak dissociation, 57–58
personality and, 8–9, 85 Stolorow, 20
private regions of, 75–79 Subjective judgment, 132–133
witnessing one’s, 126–127 Subjectivity, 7–8, 13, 36–37, 44
Self-hatred, 97 conflict and, 101
Self-interest, 14, 15 true self and, 78
Self-loss, 34 Substantive rightness, 26, 27
Self-states, 13, 85, 187 Sugarman, A., 8
constructivism and, 95–96 Sullivan, H. S., 164, 189
deflation feelings and, 53–55 on consciousness, 94
dissociated and tolerable, 92–93 on dissociation, 13, 90–91, 150
multiple self and, 48–53 on interpersonal dissociations, 152
not-me and, 118–120 on interpersonal field, 172
perfection, 146 on multiple self, 48–49, 82, 85
Self-tolerance, 96–98 on not-me, 119, 140
Sennett, R., 23 on reflected appraisals, 111
Sexuality, 147, 151–152 Symington, N., 99
Shakespeare, W., 131
Shame, 145 T
Single-mindedness, 90–93, 95
freedom and, 103–105 Tauber, E. S., 20
internal conflict and, 100–103 Taylor, C., 195, 196
Smith, L., 128, 191 Text interpretation, 127–128
Snags, emotional, 81–82, 104–105 Thelen, E., 128, 191
Social constructionism, 195–197, 202–205 Tolerance, self, 96–98
Socrates, 162, 180 Transference, 6, 7, 9
Sources of conflict, 87–88 metaphor and, 133–134
Spence, D. P., 107, 127 perception and, 11
Spezzano, C., 8 psychotic, 134
Splitting, 94–96 Trauma, 19–20, 110, 126–127
State of being, 13. See also Not-me dissociation and, 139
Stein, A., 85, 156 metaphor and, 135, 136–137
Stern, D. B., 12, 115, 157, 183, 197 True conversation. See Authentic
agreement with BCPSG, 183–187 conversation
on autonomous critical command, 128 Tulving, E., 4
on consciousness, 94
on constructivism, 95, 96 U
on conversation, 25
critique of, 197–198 Unbidden experience, 138
on dissociation, 13, 21, 63, 91, 152 Unconscious
on empathy, 31 analytic relatedness and, 71–73
on enactment, 24 conflict, 92, 94
on experience, 138 countertransference as, 73–74
on mutual enactments, 122 emotional memories, 135–136
on narrative rigidity, 13, 149 meaning, 5–6
on not-me, 119 private region and, 75–79
on perception, 11 Understanding, 45–47
on projective identification, 151 analytic relatedness and, 71–73
on reflective function, 161–162 enactments and reaching, 53–57
on social construction, 77 Unfolding, continuous productive, 116–118
Index 229
Unformulated experience W
degree of constraint on, 2–3
Weak dissociation, 57–58
location of, 3–5 Winnicott, D. W., 76, 103, 111, 118
as potential experience, 1 Witnessing
theory, 1, 20, 86, 184, xiii, xv enactment, narrative and, 122–124
unmentalized experience and, 20–21 experience, 110–116, 136–137
metaphor and, 136–138
Unformulated Experience, 1, 3, 7, 24, xiii, xv
one’s self, 126–127
dissociation discussed in, 13 Wolstein, B., 5, 20, 73, 76–79, 172
Unmentalized experience, 20–21 Worries, 19
V Z
Von Bertalanἀy, L., 191–192 Zeddies, T. J., 8, 34–37, 195

Stern (2010) Partners in Thought - Working With Unformulated Experience, Dissociation, and Enactment

Uploaded by

Copyright:

Available Formats

Stern (2010) Partners in Thought - Working With Unformulated Experience, Dissociation, and Enactment

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Stern (2010) Partners in Thought - Working With Unformulated Experience, Dissociation, and Enactment

Uploaded by

Copyright:

Available Formats

Partners in

Psychoanalysis in a New Key Book Series

New York London

Chapter 4 originally appeared in Contemporary Psychoanalysis, 40, 197–237, 2004.

Chapter 5 originally appeared in Psychoanalytic Quarterly, 78(3), 701–731, 2009.

An earlier version of Chapter 6 appeared in Psychoanalytic Inquiry, 29(1), 79–90, 2009.

An earlier version of Chapter 7 appeared in Psychoanalytic Dialogues, 16, 747–761, 2006.

An earlier version of Chapter 9 appeared in Psychoanalytic Dialogues, 18(2), 168–196, 2008.

Reprinted with permission.

© 2010 by Taylor and Francis Group, LLC

Printed in the United States of America on acid-free paper

International Standard Book Number: 978-0-415-99969-4 (Hardback) 978-0-415-99970-0 (Paperback)

Library of Congress Cataloging‑in‑Publication Data

Visit the Taylor & Francis Web site at

and the Routledge Web site at

1 The Embodiment of Meaning in Relatedness 1

2 Conversation and Its Interruptions 25

3 The Fusion of Horizons: Dissociation, Enactment, and Understanding 43

4 The Eye Sees Itself: Dissociation, Enactment, and the Achievement of

5 Partners in Thought: A Clinical Process Theory of Narrative 107

6 Shall the Twain Meet? Metaphor, Dissociation, and Co-occurrence 131

For their readings and commentary on various drafts of the chapters of

and content have a mutual influence on each other, sometimes facilitative,

Rethinking Unformulated Experience

Recently I have reconceptualized the theory of unformulated experience.

Chapter 1, in addition to what I have already described about it, contains

the implications of thinking about transference and countertransference

same as my own. It seems that therapeutic action in mentalization theory

here, too, in the beginning, because another implication I want to avoid

that the interpretive aspect, while it remains real enough, is constrained

The Location of Unformulated Experience

Where is unformulated experience? Does the question even make sense?

which we remain confident enough to ascribe to it literal existence. Take

of authorship goes on outside awareness, we are unaware, more often than

Mind and Relatedness: Hand and Glove

Contemporary interest in the relationship between dissociation and enact-

a single skull, has to hold a diἀerent view of analytic relatedness than I

to be replaced with a more flexible conception in which diἀerent aspects

or underneath appearances. Instead, we believe that whatever we can have

Ongoing Interaction and the Formulation of Meaning

aἀect, the “feel” of the situation, that gives an interpersonal perception

related to the original experience).1 Especially significant is the subsym-

To this point I have been discussing parts of our experience of related-

reasons. They are, that is, dissociated. In the Interpersonal/Relational

Dissociative Enactment and Projective Identification

No doubt my fictional example provokes some readers to think about pro-

more complex geography. It encompasses psychic life, of course; but the

Not-Me Does Not Require Trauma

Sometimes, perhaps because of my use of dissociation, readers have been

These possibilities for experience are dissociated for unconscious rea-

Unformulated Experience/Unmentalized Experience

Unformulated material is experience which has never been articulated clearly

Relatedness and the Future of Meaning

Those who theorize the roots of transference to lie in unconscious fan-

chapter, the nature of the interactions in which we participate, and that

is inevitably broader than the transference and countertransference that

We say that we “conduct” a conversation, but the more genuine a conversation

In the chapter following this one, I discuss dissociation and enactment

the nature of true conversation in its uninterrupted form should set us on