898190

research-article2019
APAXXX10.1177/0003065119898190Otto F. KernbergTHERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

ja Pa

Otto F. Kernberg 67/6

THERAPEUTIC IMPLICATIONS OF
TRANSFERENCE STRUCTURES IN
VARIOUS PERSONALITY PATHOLOGIES

Definitions of specific organizations of transference developments are

proposed for neurotic, borderline, narcissistic, schizoid, symbiotic, and
psychotic character structures. These distinct organizations of transfer-
ence developments correspond to the underlying characteristics of
internalized object relations stemming from the conflictual implications
of split-off, idealized, and persecutory self- and object representations.
The transference structures described have implications for the corre-
sponding application of psychoanalytic technique. Clinical cases illustrate
the relationship between personality structure, transference organiza-
tion, and psychoanalytic techniques.

Keywords: object relations, character pathology, psychoanalytic tech-

nique, transference structures

M y main objective here is to illustrate the development of distinct

organizations of transference presentations in various personality
pathologies. These different transference developments are quite stable
and consistent with neurotic, borderline, schizoid, narcissistic, and
overtly psychotic patients and emerge in transference/countertransfer-
ence developments in the context of psychoanalytic therapies. They are
distinct, specific organizations of the internal world of object relations
that determines the overall type or character of the personality. At the
same time, I propose, they are a linkage between the concrete internal
world of unconscious fantasy (the world of internalized instinctually—

Director, Personality Disorders Institute, New York Presbyterian Hospital,

Westchester Division; Professor of Psychiatry, Weill Cornell Medical College;
Training and Supervising Analyst, Columbia University Center for Psychoanalytic
Training and Research.
Presented to the Association for Psychoanalytic Medicine, January 9, 2018.
Submitted for publication January 18, 2019.

DOI: 10.1177/0003065119898190 951

 Otto F. Kernberg

that is, affectively—invested object relations) and the structural aspects

of character that distinguish personality types.
Transference structures also have implications for the application of
psychoanalytic technique. Certain aspects of psychoanalytic technique
become prominent in particular therapeutic situations, for example, the
importance of countertransference analysis, a required temporary relax-
ation of technical neutrality, or the need to focus sharply on specific
aspects of the patient’s external reality.
I will clarify and illustrate these theoretical proposals, but first will
examine some problems and possible misunderstandings raised by this
extremely condensed overview.
First, with the term transference structures I refer to typical relation-
ships between activated self- and object representations and their corre-
sponding unconscious conflicts: it is a matter not of the content of specific
unconscious fantasies evinced in individual sessions, which obviously are
infinite in their highly individualized and unpredictable nature, but of the
extent to which internal object relations reflect an integrated self, of whether
self and internal objects are clearly differentiated from each other or are
split to the degree that psychic experience is totally fragmented. Thus, I will
be examining the structure of internalized object relations rather than the
specific unconscious conflicts activated in each treatment session.
Second, whatever the predominant personality pathology of a patient,
it is irrelevant for the utilization of a psychoanalytic understanding in each
hour. Each session should be approached “without memory or desire.” The
degree and type of personality pathology is relevant for the overall possi-
bility and indications for psychoanalysis, and related to overall prognosis
(Kernberg 1999, 2018). In that new applications of modified psychoana-
lytic technique have expanded psychoanalysis with psychoanalytic psy-
chotherapies for extremely disturbed patients, and that psychoanalysis and
its derivative treatments are expanding the realm of psychoanalytic thera-
peutics, today personality theory/assessment and the expansion of psycho-
analytic techniques are intimately related subject matters.
Third, in previous work I have attempted to define the body of what
may be called “standard” or “classical” psychoanalytic technique, as well
as its modification for psychoanalytic psychotherapies (Kernberg 1999).
More recently, in Treatment of Severe Personality Disorders: Resolution
of Aggression and Recovery of Eroticism (Kernberg 2018), I have out-
lined the corresponding differences in the application of psychoanalytic
techniques.

952
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

I refer to “standard” or “classical” psychoanalytic technique as the

approach to psychoanalytic treatment shared by the various schools or
theoretical conceptions that today dominate our field: the ego psychologi-
cal, Kleinian, British Independent, relational, neo-Bionian, and (perhaps
even) Lacanian approaches. This commonality is centered on the discov-
ery and resolution of unconscious intrapsychic conflict by means of inter-
pretation of defensive operations directed against unconscious impulses
derived from libidinal and aggressive drives, conflicts activated domi-
nantly in the transference. Transference analysis and the related explora-
tion of corresponding countertransferences, carried out by the analyst from
a predominantly neutral but not “indifferent” position, are the main instru-
ments of this “standard” psychoanalytic technique. While modifications
and expansions of this basic technique tend to be employed more freely in
usual psychoanalytic practice, the Menninger study (Kernberg et al. 1972)
found that the combination of specific modifications of these instruments
may provide effective treatment for patients with severe psychopathology
who do not respond well to psychoanalysis proper. This study showed that
for patients with significant ego weakness, “expressive” psychotherapy
with a combination of transference analysis and external support was more
effective than psychoanalysis proper or supportive psychotherapy. In the
Menninger project, Wallerstein (1986) found that some supportive tech-
niques were quite frequently used in psychoanalysis proper, which led to
their exploration in later work (Rockland 1989).
The Menninger study led to the development of a specific psycho-
analytic psychotherapy, transference-focused psychotherapy (Kernberg
1975), the effectiveness of which could be empirically validated (Clarkin
et al. 2007; Kernberg et al. 2008; Doering et al. 2010; Yeomans, Clarkin,
and Kernberg 2015). Even a more differentiated, specific form of support-
ive psychotherapy has proved helpful, though less so than a truly psycho-
analytic one. These findings signified an expansion of psychoanalytically
based treatments that, independently, was researched in other centers
(Rudolph 2013). In attempting to define the basic technical commonali-
ties of psychoanalysis and psychoanalytic psychotherapies, and to clarify
the differentiation of these approaches, I found interpretation, transfer-
ence analysis, technical neutrality, and countertransference analysis to be
the core psychoanalytic techniques that, combined in various degrees of
modification, permit the definition of specific psychoanalytic psycho-
therapies (Kernberg 2018).

953
 Otto F. Kernberg

Interpretation involves the consistent analysis of defenses, and of resis-

tances as the clinical manifestations of these defensive operations. Transference
analysis has emerged as a dominant therapeutic factor throughout the entire
spectrum of psychoanalytic psychotherapies. Technical neutrality has been
differently applied in alternative psychoanalytic approaches and is a major
differentiating aspect of them. Countertransference analysis has raised the
question whether it should be communicated to the patient, and if so to what
extent. Countertransference “utilization” refers to the internal process of the
analyst, not to its communication to the patient. Analytic understanding of
the transference/countertransference dynamic includes attention to the
related “intersubjective field.”
Careful exploration of other leading technical psychoanalytic aspects
(e.g., character analysis, dream analysis, enactment and acting out, work-
ing through, containment and reverie, repetition compulsion, negative
therapeutic reaction, somatization, termination) reveals that they are
applications of the four basic techniques of interpretation, transference
analysis, technical neutrality, and countertransference (Kernberg 2018).
Fourth, it may rightly be questioned whether one can speak at this
time of a “classical” or “standard” psychoanalytic technique, given the
many psychoanalytic schools and orientations flourishing today. I have
spelled out my own approach (Kernberg 2004b) and characterize it as a a
combined ego psychological and object relations approach that is very
close to or integrated with a Kleinian perspective. In my efforts to arrive
at a general definition of psychoanalytic technique, however, I have
reached for broad technical definitions that incorporate relational per-
spectives as well. In the course of what follows, I will comment on interven-
tions that may raise questions from a Kleinian or relational perspective, and
clarify my reasons for making them. Regarding Kleinian perspectives,
my work and that of my colleagues is quite close to Betty Joseph’s
approach to transference interpretation, but includes additional observa-
tions regarding the object relation involved in projective identification,
and an expansion of the concept of total transference (Joseph 1985).
Regarding the relational psychoanalytic approach, a major issue is the
interpretation and management of countertransference developments,
particularly with severely disturbed patients. I will illustrate what such
dialogue with a relational approach would imply.
Fifth, to illustrate the activation, and transformation during the course
of treatment, of characteristic transference structures in different

954
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

psychopathologies, I will attempt to present the development of treatment

over time. In illustrating with case material the various types of therapeu-
tic structure developments, I will be limited in presenting the individual
therapeutic hours of each case, thus limiting the process material, always
of highest interest to psychoanalysis. Despite these limitations, I hope to
convey at least some process development that will illustrate my own
analytic work and that of other colleagues.
Sixth, and finally, it is gratifying for a psychoanalyst that the work of
the Cornell Personality Disorders Institute and of other dedicated psycho-
analytic researchers has been effective in alerting the psychiatric com-
munity to the centrality of the nature of the self and its relations with
significant others as the basis for classifying the severity of personality
disorders, as reflected by criterion A of the Alternative Classification of
Personality Disorders in DSM-5. And in stressing the clinical relevance
of the specific personality disorders that emerge in the psychoanalytic
psychotherapies with them, psychoanalysis is pointing to the need to
combine categorical criteria with dimensional (severity) considerations. I
hope that these remarks will clarify the complexity that follows.
Classical psychoanalytic theory proposed that the symptoms of neu-
rotic constellations and character pathology derive from unconscious
infantile conflicts between drives and defensive operations. In the sim-
plest terms, psychoanalytic treatment could be described as the system-
atic interpretation of defensive mechanisms to permit the gradual
emergence into consciousness of the previously repressed drive deriva-
tives in all their symptomatic expressions. This would permit a subse-
quent elaboration by the conscious ego, in the light of the adult capacity
for integration and sublimation, of impulses not previously tolerated, as
well as the related resolution of symptom formation.
Contemporary object relations theory has reformulated these basic
psychoanalytic concepts in terms of the analysis of both drive derivatives
and defensive operations as reflecting the internalization of relationships
between self and others under the dominance of peak affect states repre-
senting these drive derivatives and defensive impulses (Fairbairn 1954;
Klein 1946, 1958; Winnicott 1965; Jacobson 1964; Kernberg 1985, 2004a,b;
Greenberg and Mitchell 1983). In other words, rather than defining, for
example, an obsessive character trait of excessive amiability as a compro-
mise formation between an unconscious aggressive impulse and a defen-
sive reaction formation directed against it, psychoanalytic object relations

955
 Otto F. Kernberg

theory assumes that the aggressive impulse is constituted, in fact, by an

aggressively framed internalized relationship between a dangerous, hos-
tile object representation and an enraged self-representation. The corre-
sponding defensive mechanism of surface friendliness reflects an
internalized object relation involving a submissive self-representation
relating to a powerful but protective and benign object representation.
Both impulse and defense are represented by corresponding internalized
object relations. Psychoanalytic technique, from an object relations the-
ory viewpoint, now consists essentially in the systematic interpretation of
both defensive and impulsive internalized object relations as they are
reflected in the patient’s pathological interactions with significant others,
particularly in the transference.
The transference constitutes the optimal field in which these repressed
or dissociated internalized object relations are activated, with analyst and
patient assuming, in the patient’s experience, the roles of the correspond-
ing self- and object representations both defensive and impulsive. The
predominant affect in their interaction reflects the underlying fantasized
interaction between self and object that is being enacted in the transfer-
ence. Within this framework, psychoanalytic technique can be defined as
the systematic interpretation of the defensive internalized object relations
represented by the defensive aspects of the transference, to be followed
by the gradually predominant activation of the impulsive object relation-
ship in the transference.
Freud observed that all we know about drives are representations and
affects; in the light of contemporary object relations theory, we might say
that drives are represented by the dyadic relations between self- and
object representations framed by a significant libidinal or aggressive,
“positive” or “negative” affect. These units are the building blocks of
intrapsychic life. Eventually, they consolidate into ego, superego, and id
as overall integrated structures, but at the same time their component dis-
crete units of activated self- and object representations constitute transfer-
ence dispositions that are the object of concrete interpretive interventions
by the analyst.
The application of object relations theory to the study of psychic
development has revealed the structural developments of two major stages
of psychic life: first, an early stage of intrapsychic development in which
internalized object relations are sharply dissociated or split according to
their positive, rewarding or negative, aversive characteristics (Kernberg

956
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

1985; Kernberg and Caligor 2005). The temperamentally given, positive

affect systems of eroticism, attachment, and play bonding, which may be
viewed as jointly constituting the libidinal drive, are sharply split from the
negative affect dispositions represented by the fight/flight and separation-
panic systems that jointly may be considered to represent the aggressive
death drive (Panksepp and Biven 2012). The sharp splitting of internalized
object relations, depending on the positive or negative affect systems that
have been activated, determines a lack of integration of a total self, and a
lack of integration of total representations of significant others, so that
idealized and devalued aspects of the self are dissociated, as well as ideal-
ized and persecutory aspects of significant others.
This early stage of development constitutes the condition described
by the Kleinian school as the paranoid-schizoid position, and what has
been defined more recently as the early developmental stage of identity
diffusion (Kernberg 2012). Pathological fixation at this stage is the funda-
mental structural characteristic of borderline personality organization
(BPO). Under normal circumstances, a second stage of development
gradually sets in over the first few years of life, characterized by integra-
tion of the self, which now incorporates both positive and negative self-
representations, and of the internalized representation of objects, with
both the idealized and the persecutory aspects of significant others being
“toned down” into more realistic representations. This combination of an
integrated self, surrounded by an integrated world of significant others,
constitutes normal identity, characteristic of both normality and neurotic
personality organization (NPO).
While contemporary object relations theory views all defense mecha-
nisms, as well as impulse-driven behavior, as corresponding to underly-
ing defensive and impulsive internalized object relations, a major
difference at the borderline level, in contrast to a neurotic level of person-
ality organization, is reflected in the predominance of primitive defensive
operations centering around splitting, with consciously available but
affectively split internalized object relations, idealized and persecutory,
representing, respectively, defense and underlying impulse. The interpre-
tation of the defensive operations at the borderline level deals with defen-
sive and impulsive relations alternatively conscious and sharply split
from each other activated in the transference.
In the case of neurotic personality organization, defense centers around
repression and its related advanced defensive operations. This implies the

957
 Otto F. Kernberg

interpretation of unconscious developments in the transference as uncon-

scious elements of both defensive and impulsive activation, and justifies
the classical view of the interpretation of unconscious impulsive contents
on the basis of a previous elaboration of the defensive preconscious and
unconscious defenses. In all cases, however, psychoanalytic technique
implies, in essence, the interpretation of internalized object relations, par-
ticularly, but not exclusively, in transference activation, and the related use
of clarification, confrontation, and interpretation proper of the correspond-
ing object relationship between self- and object representations under the
impact of a dominant affect state. Unconscious conflicts always emerge in
the treatment situation as conflicts between a “defensive” self and object
relations enacted with role distribution between analyst and patient, and
“impulsive” self and object relations similarly enacted in the transference. In
all cases, analysis of the transference permits us to identify the correspond-
ing unconscious conflicts, link the corresponding distorted interactions in
the transference to parallel problems in the patient’s external reality, and
eventually trace them back into the patient’s unconscious past.
Thus, the essential aspects of psychoanalytic technique across the
entire spectrum of borderline and neurotic personality organization
involve interpretation—with the differential features that interpretation
takes in cases with normal identity or identity diffusion. As the term
implies, transference analysis refers to the analysis of the dominant sce-
nario of activation of past conflictual internalized object relationships as
they appear in the interaction with the analyst. From a position of techni-
cal neutrality, the analyst describes the conflicts activated in the transfer-
ence from a “third, excluded party” viewpoint. Such a position of technical
neutrality, however, does not deny the importance of the activation of
countertransference reactions and, on the contrary, uses the analysis of
countertransference as part of the internal clarification of the transference
by the analyst. Countertransference is used as an important, at times cru-
cial, aspect of the information, including verbal and nonverbal communi-
cations by the patient, that permits transference interpretation. As
mentioned, I have proposed elsewhere (Kernberg 2018) that these four
technical interventions—interpretation, transference analysis, technical
neutrality, and countertransference utilization—may be considered the
basic techniques of psychoanalysis and psychoanalytic psychotherapy,
and, in their systematic use throughout all psychoanalytic modalities of
treatment are the essence of psychoanalytic technique.

958
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

It should be added that to systematically employ psychoanalytic tech-

nique, including the four basic technical instruments and their derivative
technical interventions, a certain therapeutic setting must be established
that permits the development of the transference and the gradual deepen-
ing of the nature of therapeutic information and interventions. This setting
of the therapeutic interaction is constituted by the clinician’s maintaining
him- or herself in an attitude of evenly suspended attention, and by the
patient’s being instructed to carry out free association. Again, this is not
the place to analyze in detail the implications of the analyst’s technical
position and the importance of free association by the patient, but suffice
it to say that establishment of the therapeutic frame will facilitate the acti-
vation of all the technical interventions I have mentioned.
In addition to the four basic analytic techniques, we may add the
basic supportive techniques that, in contrast to analytic exploration, tend
to reinforce the patient’s defensive operations and compromise forma-
tions that may facilitate immediate improvement and adaptation to inter-
nal and external reality. In other words, if we add the techniques of
supportive psychotherapy, which are often combined with analytically
derived therapies, we will then have a list of all the psychoanalytically
derived techniques and use it as a potential profile allowing the differential
description and classification of the entire spectrum of psychotherapeutic
approaches derived from and including standard psychoanalysis. These
supportive techniques include abreaction, cognitive support, affective
support, direct environmental intervention, and reeducational reduction
of transference distortions as learning experiences to be transferred “out-
side” to the patient’s external reality. This profile permits us to differentiate
among the principal applications of psychoanalytic approaches, standard
psychoanalysis, transference-focused psychotherapy, mentalization-
based therapy, general psychodynamic psychotherapy, and supportive
psychotherapy (see Table 1).

T echnical I mplications of the S tructural

A spects of the T ransference

The original development of psychoanalytic technique occurred in the

context of the predominant treatment of patients with neurotic personality
organization. In other words, the various symptomatic conditions and
character pathology focused on by standard psychoanalysis and derived

959
960
Table 1. Dominant techniques in psychoanalytic psychotherapies
Psychoanal TFP DPHP TPOPSY MBT EX-SupP SPY  

Interpretation +++ +++ ++ ++ + ++ - Psychoanal Psychanalysis

Transference Analysis +++ +++ ++ + + + - TFP Transference-focused Psychotherapy
Technical Neutrality +++ ++ +++ + + + - DPHP Dynamic Psychotherapy for Higher
Personality Pathology
Countertransference ++ +++ ++ ++ ++ ++ ++ TPOPSY German Depth Psychology Oriented
Utilization
Abreaction - - - + - + ++ Psychotherapy
Cognitive Support - - - + + + +++ MBT Mentalization-based Psychotherapy
Affective Support - - - + + + +++ Ex-SupP Expressive-Supportive Psychotherapy
Environmental Intervention - - - + - + +++ SPY Supportive Psychotherapy
Transference Reduction & - - - - + - ++  
Export
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

psychoanalytic psychotherapies were presented by a relatively healthier

segment of the population, where standard psychoanalysis was the treat-
ment of choice. In more recent times, the extension of psychoanalytic
approaches to borderline personality organization, the severe personality
disorders, and their complications (including alcoholism and addictions,
perversions or “paraphilias,” and severely regressed patients with border-
line, narcissistic, schizoid, paranoid, and hypochondriacal pathology) has
extended our knowledge to archaic psychic functions and structures. It
has expanded our experience with regressive transference developments
that require particular modifications of psychoanalytic treatment, modifi-
cations derived from the particular constellation of their respective trans-
ferences, and the technical implications evolving on that basis. In what
follows, I will describe some structures of these regressive transference
organizations and their technical implications, contrasting them with the
typical transferences of neurotic personality organization.

Neurotic Personality Organization: The Classic Psychoanalytic Treatment Situation

Classical psychoanalytic technique dealt primarily with patients pre-

senting normal integration of identity, that is, an integrated concept of self,
surrounded, we might say, by integrated concepts of significant others.
This structure, within contemporary object relations theory, corresponds
with the achievement of the depressive position, the integration of the per-
secutory and idealized segments of early experience into integrated con-
cepts of self and others. It also involves full development of the tripartite
structure of ego, superego, and id, a tolerance of ambivalence, and a capacity
for deep and mature object relations (Kernberg and Caligor 2005). Most
patients with hysterical, obsessive-compulsive, or depressive-masochistic
personality structures fit this condition and, in the treatment, present typi-
cal development of regressive transferences involving the infantile self
relating to the various dominant infantile transference objects (Figure 1)
(Caligor, Kernberg, and Clarkin 2007). Within these regressive transfer-
ences the patient usually incorporates his infantile self in a defensive or
impulsive relationship with a significant infantile object projected onto the
analyst. The analysis of this transference proceeds by gradually exploring
and resolving the corresponding unconscious conflicts, preoedipal and
oedipal, involved in their defensive and impulsive relationships. The
patient’s communication is mostly through verbal communication,
although nonverbal communication, somatization, and acting out may at

961
 Otto F. Kernberg

Figure 1.  Neurotic Transference

times prevail. Countertransference rarely acquires the role of a primary,

overwhelming source of information regarding transference develop-
ments, though it is always present as important information.
To illustrate clinically: a woman in her mid-forties with an hysterical
personality structure and significant masochistic features initially devel-
oped a regressive transference as part of which the analyst was experi-
enced as a rigid, demanding, controlling figure representing the patient’s
mother. The patient experienced a sense of deep resentment for being
oppressed, and rebellious competitive impulses toward the analyst-
mother. After a period of gradual working through, that dominant trans-
ference was replaced by the patient’s activation of the relationship with a
loving but weak and unreliable father, who was unable to protect the
patient and take her side in what she had experienced as unfair treatment
by her mother. Still later, the transference evolved into a resentful and
derogatory attitude toward the analyst, now perceived as hypocritical in
his friendly “fatherly” demeanor while being emotionally unavailable. In
a later stage of her analysis, and after working through enormous fears of
experiencing forbidden sexual impulses toward him, the early image of a
powerful and sexually provocative father emerged, with both erotic
impulses in the transference and the fear of being rejected and depreci-
ated, deeply connected with a fantasied sense of inferiority as a woman

962
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

incapable of competing with powerful, dominant women (“mother”).

This rather simplified and condensed history of the development of the
dominant transference patterns of this analysis reveals, however, the con-
tinuity of the patient’s self-concept throughout time, the projection of her
temporarily corresponding object representations onto the analyst, and
the relative stability of the self-concept in terms of the patient’s maintain-
ing a certain capacity for self-reflection throughout the treatment. This
integrated self permitted her to explore, within the psychoanalytic setting,
the particular unconscious conflictual relationships activated in the treat-
ment situation.
The intensity of countertransference reaction was relatively moderate
throughout the treatment. A consistently present “observing ego” on the
part of the patient, and the corresponding capacity in the analyst to main-
tain a consistent “split” between specific countertransference responses
and the ongoing availability of his self-reflecting function signaled the
availability of a relatively stable therapeutic alliance.

Structural Aspects of the Transference in Borderline Personality Organization

In the psychoanalytic treatment of borderline personality organiza-

tion, the typical structure of the transference implies a sharp division
between the idealized and the persecutory segments of early psychic
experience and corresponding internalized object relations, with activa-
tion in the transference of split idealized and persecutory relationships
(Kernberg 2004a). Here the lack of an integrated concept of self and the
intensity of primitive affects facilitate the rapid activation of these split
object relationships, reflected in the alternation of intensely positive and
negative transference reactions, not only the corresponding unconscious
conflicts but, at the same time, rapid role reversals in their expression.
Self- and object representations are interchanged in the transference.
Alternately, at times the patient experiences himself to be a victim of the
therapist’s aggression, only then to make the therapist the victim of an
aggressive transference. At other times, the therapist appears as an ideal,
protective object and the patient experiences himself in blissful depen-
dency on such an object. This relationship, too, tends to alternate, with
times in which the patient appears in the role of a giving mother and the
therapist is placed in that of a happy, satisfied child, only to revert rapidly
to the earlier persecutory relationship (Figure 2).

963
 Otto F. Kernberg

Figure 2.  Borderline Transference

In this emotionally intense, rapidly shifting activation of contradictory

affective states and corresponding object relations, what may initially
present as an apparent chaos that may only gradually be clarified, patients
do not evince a basic, integrated self that would permit them to reflect on
these activated states. These split object relations may be integrated only
gradually, as the patient’s understanding of his double identification with
self and object and of their relationship grows, with the corresponding
understanding that the projection of these split-off, chaotic internal rela-
tionships onto relationships with others in present reality is the source of
the chaos, regression, and failure the patient has experienced.
Transference-focused psychotherapy (TFP; Yeomans, Clarkin, and
Kernberg 2015) has developed a specific technical approach that focuses
on the dominant object relation in the transference at any point of the
treatment, using the affectively dominant experience of the patient to
diagnose the corresponding self- and object representations. The therapist
pursues the activation of these relationships in the transference through-
out time, diagnoses and interprets the sharp split between the idealized
and the persecutory relation to the same infantile object, and thus attempts
to integrate the self and the concepts of significant others, fostering the
development of normal identity. Kleinian literature has stressed the fun-
damental function of projective identification in the attribution of internal

964
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

object representations or internal self-experiences to significant others,

particularly in the transference. But the Kleinian approach usually does
not stress an activation of the total object relationship, with enactment of
one aspect of this dyad by the patient, while the complementary object (or
self) is projected onto the analyst (Spillius and O’Shaughnessy 2012).
The following case illustrates this transference structure in the psy-
choanalytic psychotherapy of a woman in her early twenties with border-
line personality disorder and chronic, severe suicidal tendencies, sexual
promiscuity, drug abuse, and failure in her studies. A central issue in her
early childhood was the severe physical abuse to which she was chroni-
cally subjected by a maternal aunt, which her absent father passively tol-
erated. She initially displayed intense hostile reactions when her
expectations and immediate wishes were not satisfied. Verbal attacks on
the therapist escalated into attacks on objects in his office that required
establishing rules limiting the patient’s destructive behavior in the ses-
sions. At other times, the patient bitterly complained about the indiffer-
ence and coldness of the therapist, his sadistic pleasure, as she saw it, in
frustrating her needs, complaining to third parties about mistreatment by
the analyst. Sometimes she presented rapid shifts between aggressive
teasing, making fun of the therapist, provocatively sitting on his desk, and
desperately crying when asked to leave at the end of the session; she
would say she didn’t have enough time to express something really
important to her. At other times, through desperate pleas for phone con-
tact and requests to increase the number of sessions, she showed intense
wishes to depend on the therapist. She expressed fantasies that if the ther-
apist were a kangaroo she would be his kangaroo baby, sitting in his
pouch and watching, in a reassured mood, the world pass by. At one point,
learning about a sudden catastrophe in the life of the therapist, the patient
became extremely concerned, arriving to her session with a huge bunch
of flowers, showing a role of maternal empathy and consolation to what
she thought was the suffering therapist. It was only in the advanced stages
of this treatment that the patient was able to tolerate the emotional aware-
ness of both idealized and persecutory reactions to the therapist, experi-
enced guilt over her aggressive behavior toward him, and wished to repair
the damage that, in her fantasy, she might have caused him.
This may be a typical case where the intensity of acting out induces a
corresponding intensity of countertransference reactions, creates the
threat of countertransference acting out, and poses the general question of

965
 Otto F. Kernberg

how to deal with this development. This patient’s intense rage and frustra-
tion when I did not satisfy her demands for time, attention, or special
privileges coincided with her perception of me as a sadistic, withholding,
torturing object: I became the aunt, and the patient was totally convinced
that I behaved like that aunt. The patient was my helpless, suffering,
enraged victim. And while I attempted to maintain the therapeutic bound-
aries, I could experience the pleasurable refusal to give in to her demands.
To the contrary, when the patient attacked me viciously, insulting me on
one occasion in a public space, on another destroying objects in my office,
I felt the helpless victim of undeserved savagery: now she became the
aunt, and I became the patient as the young, abused girl.
My technical approach consisted in clarifying, at the same time, who
I represented in her massive projective identification, and who she was
identifying with in her response to that projected object. The enacted
transference/countertransference relationship was that between a sadistic
aunt and an enraged, helpless, desperate child. At times she experienced
herself as a sadistic aunt in the transference, enacted this identification,
and I became a helpless, enraged child in my countertransference. At
other times, the relationship inverted to her experiencing herself as a
helpless, mistreated child, while I became the vengeful, sadistic aunt.
Through my interpreting this repeated reversal in our roles, the patient
became able to understand her unconscious identification with both vic-
tim and perpetrator. The activation of this same object relation with role
reversals between self and object in the transference made it possible,
over time, to permit the patient to tolerate her unconscious identification
with both self and object, and recognize in herself what previously could
only be projected. Thus, countertransference analysis and utilization in
the context of the analysis of rapidly shifting transferences permits the
patient eventually to understand also what is going on in the analyst dur-
ing such intense interactions, without having to communicate to the
patient the countertransference reaction as it occurs. This technical
approach, I believe, is different from the relational approach of communi-
cating intense countertransference developments to the patient as they
evolve, and it may add to the Kleinian interpretation of projective identi-
fication in focusing sharply on the patient’s experienced reaction to the
analyst perceived under the effect of intense projective identification.
“While you see me as sadistically commanding you around, you experi-
ence yourself as my helpless, impotent, enslaved victim.” The central

966
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

focus is on helping the patient understand the activation of the self and
object relationships.
There are cases in which the severity of acting out will require limit
setting to protect the structure of the treatment, thus threatening technical
neutrality. Technical neutrality may temporarily be relaxed or abandoned
altogether, but must then be interpretively reinstated. These developments
require the therapist to carry out very intense and ongoing internal work
with powerful countertransference reactions typical in response to the
activation of primitive transference developments in the treatment situation.
While borderline patients may not tolerate a standard psychoanalytic set-
ting, the flexibility of transference-focused psychotherapy permits sys-
tematic use of the four basic analytic techniques, so that these patients are
provided a psychoanalytic treatment within a setting different from that
of standard psychoanalysis.

Structural Organization of the Transference in Narcissistic Personality Disorders

Another type of typically structured transference developments is

encountered with narcissistic personality disorder. Narcissistic personali-
ties function at a broad level of pathology, with indication for standard
psychoanalysis for those who evince a relatively normal capacity for
work, and/or a stable, if superficial, love relationship in the context of an
ordinary social life (Kernberg 2004a, 2014). At the pathological extreme,
severely regressed narcissistic personalities present with a total break-
down of their capacity to work or to maintain any intimate love relation-
ship. They show a typical combination of absence of tenderness or
emotional involvement in the context of intense sexual urges and promis-
cuity, with a concomitant breakdown in their social life. The most severely
regressed narcissistic personalities are often diagnostically confused, at
least initially, with regressed patients presenting a borderline personality
disorder. Careful mental status examination usually reveals the presence
of a pathological grandiose self, the essential clinical characteristic of this
personality disorder.
The transference structure of narcissistic personality disorder pres-
ents a very typical development that persists stubbornly throughout
months of psychoanalysis or psychoanalytic psychotherapy. It reflects the
typical relation between the enactment of a pathological grandiose self
and a projected condensation of the devalued, worthless aspects of the
self and the devalued, worthless aspects of significant others (Figure 3).

967
 Otto F. Kernberg

Figure 3.  Narcissistic Transference

Here the transference development appears as a relation between an

omnipotent, omniscient grandiose self and the devalued self-representation,
usually projected onto the therapist, but with an ever threatening role
reversal. The feared transference reversal then projects the grandiose self
onto the therapist, while the patient enacts the devalued self-concept. The
grandiose self does not relate to internalized valued object representa-
tions, but exists in a strange isolation. Its only requirement, which is
essential, is the admiration of significant others, needed to reconfirm its
grandiosity and ensure its survival. Admiring objects, including the thera-
pist in the transference, may be briefly idealized in an effort to incorpo-
rate what is admirable in them and potentially envied by others, but are
devalued and depreciated once they are not needed to implicitly confirm
the pathological grandiose self. These developments dominate the trans-
ference for extended periods. Occasionally the patient’s objective failure
in reality or in fantasy he cannot deny brings about a sudden reversal of
the relationship, so that the patient projects his grandiose self onto the
therapist while identifying himself with the devalued self-representation
usually projected onto others. By now it is well known how these grandi-
ose, self-centered individuals suddenly become extremely insecure and
dependent on reassurance from others in their social environment, only to
rapidly revert to their original position of grandiosity.
In the transference this pattern is enacted in a controlling and devalu-
ing attitude toward the therapist, while attempts are made to maintain the
therapist in a condition of sufficient appreciation to avoid considering the
treatment totally useless. An authentic respect, interest, and appreciation
for the therapist, to the contrary, would be dangerous, putting the patient
immediately in a position of intolerable inferiority. At the origin of the
structure of the grandiose self lies an internalization of what these patients

968
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

experienced in early childhood as powerful and admired aspects of sig-

nificant others, and their identification with aspects of themselves fos-
tered in a parental environment in which admiration for a strikingly
positive feature of the child replaced authentic love and concern from the
parents.
These component features that have activated the pathological gran-
diose self will gradually emerge, and in positive therapeutic develop-
ments this will allow that self to gradually decompose into its component
idealized self- and object representations. This development will in turn
activate the corresponding primitive object relations in the transference,
idealized and persecutory split off from each other, and transform the
narcissistic structure into a generally borderline one, a fundamental step
toward improvement. When, as the pathological grandiose self is dismantled,
component object representations are activated that reflect identification
with unethical aspects of parental images, these patients may evince anti-
social features and dishonesty in the transference, which may complicate
the development of normal superego functions in advanced stages of
treatment. These are cases that present psychopathic transferences, which
must be gradually transformed into predominantly paranoid transferences
by analyzing the paranoid fears lying behind patients’ dishonesty.
Paranoid transferences may then be interpretively transformed into pre-
dominantly depressive transferences, in the context of identity
integration.
But even when such complicating antisocial conditions are not pres-
ent, the systematic analysis and decomposition of the pathological gran-
diose self usually takes months of “microanalysis” of the subtle ways in
which the corresponding transference developments evolve. Subtle yet
intense enactments of regressive part object relations occur as the patho-
logical grandiose self is dismantled and carry a risk that the analyst will
attribute the enactments of his countertransference reactions to “here-
and-now” interactional processes, neglecting the profound early object
relations being replayed in the transference.
To illustrate I offer the case of a successful biological researcher, a
man in his early fifties, effective, dominant, and controlling in complex
business affairs but with no close friendships, a rather isolated social life,
and a loveless marriage characterized by total sexual indifference. Chronic
promiscuity evincing little tenderness was his dominant source of sexual
pleasure. This patient treated his wife like a slave who looked after his

969
 Otto F. Kernberg

daily needs. In recent years she had gradually rebelled against this situation,
expressed growing unhappiness with the empty nature of their marriage,
and finally had told her husband she was considering divorce. At that
point an anxious collapse of the patient’s grandiosity brought him to treat-
ment, initially to deal with the marital conflict, which rapidly turned out
to reflect deep problems in his sexual life and social interactions. The
patient was diagnosed with narcissistic personality disorder, with psycho-
analysis the treatment of choice.
He quickly developed a transference with the characteristics described
above. He considered the analyst a mediocre, small-thinking “technician”
who was trying to apply the book, knowledge the patient, on the basis of
his readings, felt he had himself possessed all along. An ongoing com-
plaint was that he had been “conned into” a useless treatment. It took
many months of treatment to open up an awareness of his defenses against
intense envy of his wife’s emotional richness and the gratification he felt
the analyst must take in his work. The patient, by contrast, felt himself
involved in constant professional and financial competition in his work,
which gave him no rest or relaxation. Gradually, a very frustrating early
childhood emerged, both parents being experienced by him as insensitive
and unavailable; the gradual development of a sense of successful compe-
tition and triumphant superiority over schoolmates was the only source of
gratification in his childhood. Eventually, components of his grandiose
self could be isolated, and explored in the transference.
The following example illustrates this development. This patient
lived in the same professional environment as his analyst, a mid-size
Midwestern city. He would attentively listen to any gossip that he could
catch regarding his analyst, eventually constructing a story about suppos-
edly inappropriate and ridiculous behavior by his analyst that the patient
then spread among acquaintances. This story made the rounds and finally
came back to him: somebody told him the same story he had spread about
me, the analyst. The patient, frightened by this, decided to “confess” to me
that he was the source of this gossip. Despite intense negative counter-
transference, the analyst was able to maintain the analytic relationship
and gradually, over a period of several weeks, analyzed what had moti-
vated this intensely invested behavior by the patient.
It turned out that it replicated the behavior of his mother, who, com-
ing from a socially disadvantaged environment, chronically felt insecure
in the socially privileged environment of her husband, the patient’s father.

970
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

The patient had a clear sense of his mother as frequently gossiping about
her social acquaintances in order to diminish the importance of people she
envied and felt insecure with. As an aspect of his pathological grandiose
self, he had incorporated this image and source of power of his mother,
now expressed in the transference relation with the analyst. By the same
token, here the relationship between the patient’s pathological grandiose
self and the projected devalued self became transformed into the specific
relation between the mother-identified patient and the projection of his
neglected and rejected self-representation in the transference. In other
words, this specific transference relationship heralded the dismantling of
his pathological grandiose self. The patient now experienced authentic
feelings of shame and guilt over his behavior, and it was a first recogni-
tion of the origin of aggression within himself, in contrast to its usual
projection onto others.
I was quite shocked when I first became aware of the patient’s spread-
ing gossip about me. My first reaction was the wish to terminate his treat-
ment. I felt disappointed and betrayed, consulted with a senior colleague,
and was able to maintain the treatment, but only with an inhibition in my
interpretive interventions. I clearly became aware that the patient was
identifying himself with his gossipy, envious, and depreciative mother,
but it took me some time to realize that I was not simply reacting as the
betrayed and abandoned son. I also developed a devaluating, vengefully
superior reaction in my countertransference. Exploring with the patient
his experience of me as superior and devaluating, and his shameful sense
of having to depend on such an object, clarified for me why having put
me down outside the sessions was a relieving reversal of the situation.
Again I was able to help him understand his identification with both self
and object in this conflictual, highly traumatic experience, as well as the
related defensive function of his grandiosity.

Structural Organization of the Transference in Schizoid Personality Disorders

The concept of schizoid transferences lends itself to confusion because

of two different uses of this concept. The classical psychiatric definition of
schizoid personality disorder clearly describes the characteristic symp-
toms that lend themselves to a diagnostic assessment, and that indeed are
fundamental for assessing typical schizoid transference dispositions that
have been relatively underappreciated in the psychiatric literature. At the
same time, the Kleinian concept of the paranoid-schizoid position (Klein

971
 Otto F. Kernberg

1946), based in part on the description of the psychodynamics of the schiz-

oid position elaborated by Fairbairn (1954), has dominated the psychoana-
lytic literature and proven itself essential in the analysis of the defensive
organization of the entire field of borderline personality organization, that
is, the defensive dissociation or split between idealized and persecutory
internalized object relations. Paranoid-schizoid dynamics are found in the
entire field of patients with severe personality disorders, but are to be dif-
ferentiated from the very specific transferences that characterize the schiz-
oid personality disorder, which require a specific technical approach
originally suggested by Fairbairn (1954), Guntrip (1969), and Rey (1979).
These specific transference dispositions described by Fairbairn correspond
more closely to the classical psychiatric concept of the schizoid and
schizotypal personality disorders.
Typical descriptive characteristics of the schizoid personality disor-
der include social withdrawal, social isolation, a lack of intimate relation-
ships, hypersensitivity to criticism, feeling very easily hurt by others, and
a particular, heightened sensitivity to the feelings and behavior of others,
in contrast to these patient’s social isolation (Akhtar 1992). At the same time,
these patients seem to withdraw into a private, secretive self-affirmation,
and an internal world of fantasy that they control. This self-affirmation
lacks the sense of superiority and depreciation of significant others that is
characteristic of the pathological grandiose self of narcissistic personali-
ties. The schizoid patients described by Fairbairn present, as a dominant
dynamic, a desperate desire for close, dependent relationships but, at the
same time, an extraordinary fear of overwhelming control by the other, of
being swallowed by any close relationship. In contrast to their lack of
capacity for empathic and tender relationships with others, they fre-
quently present a dissociated “explosion,” the activation of sexualized
and aggressive relationships in a dissociated mode.
They miss the modulated activation of affect dispositions and present
a highly specific fragmentation or dispersal of affects. Positive and nega-
tive affects seem equally unavailable, except in sudden dissociated out-
bursts, in contrast to a chronic, apparent unavailability of explicitly
affective experiences. These patients evince, as a predominant structure, a
fragmented sense of self—that is, a disturbing, confusing unawareness of
their present affective self experience, which contrasts with the alternating
activation of idealized and persecutory affective experiences of ordinary
borderline patients. Schizoid patients’ experiences of significant others are
equally fragmented. They find it difficult to divide the world into idealized

972
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

Figure 4.  Schizoid Transference

and persecutory objects: because of a great sensitivity regarding individual

behaviors and interactions that defy a clear differentiation between “posi-
tive” and “negative” objects, they experience others in confusing ways. In
other words, they experience fragmentation of affects rather than splitting
mechanisms. From the viewpoint of Fairbairn’s analysis, that includes
both the descriptive, classical symptoms of the schizoid personality and
their dominant dynamics, the assumption of a purely descriptive psychiat-
ric approach that considers these patients as having no desire for intimate
relationships clearly ignores their deeper psychological reality. Figure 4
presents the schizoid structural organization that manifests in the transfer-
ence as an activation of these fragmented relationships, including the frag-
mentary self and the fragmentary experiences with others.
Practically, these patients appear very distant, with no specific affect
activation in the sessions, thus confusing the therapist as to the dominant
object relationship being enacted in the transference at any particular
point. The therapist may feel confused about his affective reaction to the
patient, which seems to center in the feeling of nonunderstanding or con-
fusion about the situation, matched by the patient’s indications of being
similarly confused, having no clear sense of what it is all about and no
clear sense of his own affective experiences. It is as if, in spite of, or due
to, verbal communications with a trivial, impersonal, meaningless, or dis-
tracting quality, the affective relationship in the session remains strangely
distant. The therapist’s effort to clarify what is in the patient’s mind may
lead the patient to a sense of confusion, a feeling of being invaded, or, if

973
 Otto F. Kernberg

the therapist tentatively suggests that a dominant present relationship

seems to be activated, the patient may evince a frightened sense of being
invaded, controlled, or brainwashed. The therapist may in turn easily feel
that he has engaged in a theory-driven statement rather than doing justice
to the dominant affective relationship. The solution to this situation of
therapeutic uncertainty for the therapist lies in recognizing the relative
failure of verbal communication regarding transference clarification, and
accepting the central function, under these conditions, of the activation of
countertransference reactions.
Countertransference may provide answers if the therapist can let
himself be influenced by the total situation in which he is now engaged
with the patient. It is an intersubjective situation that cannot be traced
back to any particular experience of patient or therapist, but clearly
reflects the nature of the atmosphere created by their actual interaction.
This requires an openness by the therapist to the activation of whatever
dominant affect state develops in him, an openness to the fantasies that
may accompany such an affect state or that may seem the activators of a
certain affect state, and the use of that dominant affect and the related
fantasy material to reexamine the interaction with the patient, in the light
of the patient’s dominant pathology and his external reality at this point.
That gradual, difficult, but feasible analysis may lead to an understanding
of what affective relation is presently dominant, while being dispersed
and fragmented to an extent that initially made it impossible to gather.
Technically, here the same use of the four basic techniques applies,
but with particular caution regarding interpretation and, even more so,
transference interpretation. The therapist may venture interpretations that
may be easily rejected by the schizoid patient, and the therapist must be
prepared to accept such a rejection, with further exploration about what the
patient thinks might be more applicable at the moment than what the thera-
pist is saying. A willingness to retrace one’s own observation, to share with
the patient that one understands the patient’s difficulty in clarifying what
is going on in his mind, as a difficulty parallel to that of the therapist to
clarify what is going on in the interaction, may be a helpful, reassuring
assertion of uncertainty. It should not be difficult to differentiate this devel-
opment from a narcissistic patient’s contemptuous rejection of interpreta-
tions by the analyst. The therapist must therefore be cautious in interpreting
the thinking of the schizoid patient, and must stress his search for clarity in
the patient’s thinking, as well as in the therapist’s own.

974
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

In this context, the patient’s rejecting behavior and mistrust needs to be

tolerated, as well as the indications of his hypersensitive reaction to a per-
ceived rejection by the therapist. The patient may give indications of want-
ing to be close, being afraid of it, withdrawing in a suspicious attitude, and
even a preventive explicit rejection of the therapist as a protection against
excessively desired and feared closeness. The nature of the affectively
dominant object relationship defended against by the prevalent schizoid
fragmentation mechanism may vary widely: patients may reveal erotic fan-
tasies behind the apparent emptiness of the session, intense fusional long-
ings, or aggressive, dependent, paranoid affective dispositions, with
variable condensations between oedipal and preoedipal relationships. The
following case illustrates a prevalent schizoid transference.
The patient presented a typical schizoid personality disorder. A
woman in her early twenties, she chronically cut herself with razor blades
to observe the bleeding. She had evinced serious social isolation from
early childhood on, and a total social breakdown in college, where she
could not relate to other students, and where her withdrawal into an
intense world of fantasies prevented her from concentrating on her stud-
ies. Clinically, the extent of her social withdrawal, her cutting off relation-
ship with family and friends, the failure at school, her rapid withdrawal
from several early dating experiences that seemed to cause traumatic
reactions in her, and her almost disorganized way of talking raised the
question whether she suffered from a schizophrenic illness. After extended
psychiatric evaluations it became clear that she presented a schizoid per-
sonality disorder, and the treatment recommended was TFP.
In the first few weeks of treatment, after the usual history was taken,
our interaction evolved into a superficial, almost mechanical repetitive
communication of trivial aspects of her daily living. After the first two or
three months of treatment, I started to find it almost impossible to concen-
trate on anything in the sessions with her, or to use whatever cues seemed
available to direct our interaction into some meaningful communication.
My efforts to inquire into what she was feeling, what her fantasies were,
led to more of the same trivial communications, and I could sense her
irritation with me when she felt I was forcefully attempting to find new
meanings in what she was saying. At the same time, she would come
punctually to all sessions and seemed not to object to the empty content
of what evolved in them. She referred vaguely to her tendency to cut her-
self discreetly, and to watch drops of blood emerging. There was an

975
 Otto F. Kernberg

occasional seductive quality about some of her expressive demeanor, so

tentative and transitory that it was gone by the time I felt there was some
significance in it.
I had occasional fantasies that she focused on the discreet cutting of
her skin to observe blood drops as a way to exhibit herself to me or per-
haps enacting a fantasy of me attacking her sexually, or that I was a father
figure who failed in protecting her, but all efforts to explore her world of
fantasy led nowhere. It was as if whatever I said confused and disorga-
nized her thinking, and my own thinking seemed to get confused at such
moments. I pointed out to her that she experienced my efforts to under-
stand her as invasive, and that it was as if any attempt at clarification was
dangerous.
In one session I had great difficulty maintaining my attention on what
she was saying. I was following my own thoughts, and suddenly remem-
bered a film I had seen six months earlier, Investigation of a Citizen beyond
Suspicion, an Italian film about a district attorney in charge of finding a
sexual murderer. The D.A., who was himself the murderer, would kill
women in the process of having sex with them. One image from that film
came to my mind. A woman reaching orgasm was sitting on top of that
district attorney, when he suddenly pulled out a knife and cut her throat,
blood running over her breasts. This scene came into my mind, with a
combination of excitement and disgust, followed by a kind of frightened
surprise on my part that I should develop such a fantasy, and in the middle
of such a session. I attempted to dismiss the memory of that experience the
next few days. But then I realized that her repetitive comments about her
body, and bleeding openings, and the strangely seductive moments in the
sessions with this extremely inhibited patient, and my sense of “shock”
over my sadistic sexual fantasy, reflected an oscillating countertransfer-
ence identification with both a self- and an object representation involved
in a sadomasochistic interaction. And I understood that my frequent sense
of confusion also represented the effect of the patient’s defensive fragmen-
tation of all emotional experiences. I said there were threatening sexual
thoughts “in the air” that could not be talked about.
Several weeks later, when the patient mentioned there were thoughts
she had difficulty talking about, she referred to a powerful fantasy about me
that she had had repeatedly. She wished I would shoot her, and in killing her
become a murderer. I then would remain, for the rest of my life, feeling
regretful. I would never be able to forget her, and she would remain with me
the rest of my life. She would not mind dying, knowing she would be my

976
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

permanent companion the rest of my life! This sexual fantasy, with regres-
sive oedipal and severely aggressive preoedipal sadomasochistic and self-
destructive implications, became a central focus of our exploration in the
following months. What I want to stress here is the close relationship
between what was developing in the transference and the nature of my
countertransference fantasy in the middle of the apparently fragmented,
dispersed activation of a specific object relationship. This example may seem
unusual, but in fact it is a quite frequent type of experience in countertrans-
ference developments when the therapist tolerates the fragmented relation-
ship of a schizoid transference. The opening up of a specific relationship
gradually transforms the schizoid transference into the more usual transfer-
ences of borderline personality organization, and makes the treatment much
easier to carry out within the general technical approach for patients with
identity diffusion. The specific defensive problem presented by this frag-
mentation or dispersal of affect raises an open question. Is this a purely
psychological development of the intrapsychic world of these patients, or
does it reflect a more basic neurobiological disposition to dispersal of par-
ticular, excessively intense negative affects?

Structural Aspects of Symbiotic Transference Developments

The term symbiotic has been used in ambiguous ways in the literature.
In one use it refers to relationships in which an intense enmeshment
between self and other would not tolerate another relationship to coexist
with this particular enmeshed one. The boundary between self and other is
maintained, but the relationship has an exclusive quality, no other relation-
ship with a “third party” is tolerated, and exchange between self and other
by means of projective identification facilitates alternative identification
with self and other. The term’s other use refers to actual merger between
the concept of self and other, a relationship in which there evolves a lack
of differentiation between self and other, so that self experience and expe-
rience of the other are confused, with an implicit loss of ordinary ego
boundaries. In the second use of the term, an authentic psychotic process
is present, and the loss of reality testing underlies the development of
abnormal perceptions, hallucinations, and delusions. In contrast, in symbi-
otic relationships in which the differentiation between self and object
always is maintained—although they are enmeshed, there may be a rapid
interchange of roles between self and other. I am reserving the term sym-
biotic transference for an intense, entangled involvement between self and
other, but with clear maintenance of boundaries between self and other,

977
 Otto F. Kernberg

Figure 5.  Symbiotic Transference

even when rapid exchange of the role relationship develops. I reserve the
term psychotic transference for cases with loss of differentiation between
self and other, a merger of self and other that implies the loss of ego bound-
aries and reality testing. Figure 5 outlines a symbiotic transference, in con-
trast to Figure 6, which depicts a psychotic transference.
In the ordinary activation of transferences typical of borderline per-
sonality organization, patients may tolerate intense differences of views
with the therapist, and express their conflicts in sharply differentiated
roles within a specific affect-centered relationship between self- and
object representations. In symbiotic relations, patients tolerate no differ-
ence in their view of reality and that of the therapist. The therapist must
agree totally with the patient. Any disagreement indicates either a violent
invasion of the patient’s mind by a therapist who disagrees with him, or
total abandonment by a therapist who ignores him and, by the same token,
violently abandons him. Britton (2004) has described this situation as an
intolerance of triangulation. We also see this development in some
severely regressed borderline patients. The patient has no tolerance of the
therapist’s involvement with anyone else, with any other person or entity,
or other ways of thinking, from which the patient would be excluded.
This may reflect an archaic defense against an early oedipal situation, in
which mother must be totally identified with the baby, and the existence
of mother’s relationship with father must be completely denied; or it may
express intense envious resentment of the life, knowledge, and general
existence of the therapist outside the realm of the patient’s mind. In any

978
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

Figure 6.  Psychotic Transference

case, only a primitive coincidence of thinking, or total and exclusive

availability of the therapist to the patient in his mind, is tolerable. This is
an ideal situation, against which any “betrayal” by the therapist’s other-
ness triggers intolerable rage and resentment in the patient, who fears
destructive invasion or total abandonment by this betrayal.
The clinical conditions under which such a development occurs are
difficult to foresee, though usually this complication presents in patients
with an extreme incapacity to adjust to ordinary social interactions, who
evince an intense aggression rationalized by projective identification and
omnipotent control that severely distorts their intimate relationships.
Once such a symbiotic transference becomes dominant, it can be recog-
nized by its threatening character, the extended duration of a patient’s
incapacity to tolerate contrary ideas in the therapist, desperate efforts by
the patient to maintain control of reality under such conditions, and his
apparently total incapacity to recognize ordinary logic. At this point the
treatment must focus almost exclusively on this very development, the
patients’ incapacity to tolerate any difference of views, and the reasons
why such difference would cause panic in the patient.
The following example illustrates a symbiotic transference in a
patient with the diagnosis of severe narcissistic personality disorder with
overt borderline functioning, that is, almost total breakdown in the capac-
ity for work, intimate relations, and an ordinary social life. She was a
woman in her early forties, treated in psychoanalytic psychotherapy by a
skilled psychoanalyst. At one point one of her brothers died, which she
experienced as a terrible blow. At the cemetery, while her brother was
being buried, the patient broke out in intense crying and dramatic mani-
festations of intolerable psychic pain, approaching the grave and looking

979
 Otto F. Kernberg

as if she would throw herself in after the coffin. Naturally, she created
quite a commotion. People were holding her back, as she became enraged
and angrily accused her family of being insensitive to her grief. She had
to be escorted from the cemetery by family members.
In the psychotherapeutic session following this event, the patient was
still enraged, complaining about the insensitivity and brutality of her fam-
ily at the cemetery. The therapist first expressed his interest and under-
standing of the patient’s emotional reaction. He clarified her sense of rage
over her family’s lack of understanding and real feelings. Then, when he
asked how she understood the reactions to her at the cemetery, she
declared their behavior to have been completely inappropriate and incom-
prehensible. The therapist tactfully tried to confront her with the fact that
it seemed, from everything he was aware of, that her behavior had been
rather inappropriate, disturbing the ceremony that was taking place. To
this she reacted with even more rage, accusing the therapist of being in
cahoots with her family, being totally insensitive, and having no under-
standing of her, In short, he was a total disappointment. She wondered
whether she could stay in treatment with him under these conditions. The
therapist, realizing she was unable to reflect on this experience, limited
his intervention to expressing his understanding of her suffering, with no
further effort to clarify the situation at the cemetery. It took a number of
weeks, and many sessions, for the patient to consider the possibility that
the therapist might have a different view of the situation at the cemetery,
even if she disagreed with him. It took even longer for her to recognize or
even consider that the therapist’s view might actually be valid. Just her
acceptance of the fact that the therapist might think differently, without
that signifying a total attack or rejection of the patient, became a major
issue to be explored in the treatment. In more general terms, this case
illustrates not only intolerance of triangulation—the therapist’s holding to
another view—but the patient’s profound underlying incapacity to con-
sider that she might be taken seriously, respected in her own views, and
appreciated by an early maternal object.

The Structure in Psychotic Transferences

Psychotic transference dispositions are characterized by a lack of dif-

ferentiation between self- and object representations or a lack of differen-
tiation between self and other that is reflected in the loss of reality testing.
This situation has been explored in the psychoanalytic literature, and
described in the experience of intensive psychoanalytic psychotherapy with

980
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

psychotic patients in the United States and the United Kingdom before the
development of psychopharmacological medication, particularly in institu-
tions dedicated to intense psychotherapeutic approaches to psychosis.
Harold Searles (1965) described intensive psychotherapy with schizophrenic
patients as undergoing typical stages: first, a stage of lack of contact or
absence of a specific transference relationship; second, a stage of intense
symbiotic development, evincing the lack of differentiation between self
and other as the typical dynamic of intense psychotic transferences; third, a
phase of differentiation, in which the patient gradually learns to differentiate
himself from the therapist, with recovery of reality testing regarding his
behavior and his contribution to the transference relationship. A fourth phase
follows in which the patient becomes able to integrate the mutually dissoci-
ated aspects of his earlier transference experiences, and experiences a paral-
lel integration of his sense of self and of the nature of the relationship with
his therapist. This integration may then be generalized to other relationships
of the patient. Herbert Rosenfeld (1954) applied a Kleinian perspective to
the analysis of the confusional states and primitive (psychotic) mechanisms
dominating clinical encounters with schizophrenic patients.
Due to pharmacological advances, intensive psychotherapy of psy-
chosis has receded as an important treatment modality, but it may still be
indicated for a subgroup of patients identified by Michael Stone (1983,
1986). These are schizophrenic patients who do not respond to psycho-
pharmacological treatment with restoration of reality testing, have a high
intellectual level, maintain a certain integration of the personality and
significant differentiation of affect states, do not present antisocial fea-
tures, and provide the possibility of intensive long-term treatment under
the provision of a sufficiently structured environment to absorb unavoid-
able periods of acting out.
I once had the opportunity of treating an eighteen-year-old girl who
suffered from paranoid schizophrenia, at a time when intensive psycho-
therapeutic treatment seemed an important aspect of the treatment of such
conditions, when psychopharmacological treatment was still in its early
experimental stages and had not yet become the treatment of choice. I saw
the patient under supervision in a highly regarded hospital specializing in
this kind of treatment. In the first few months of the treatment I had to see
her in a padded cell under constant nursing observation. This patient
would tear up all her clothing and could only with great difficulty be
made to wear anything on her body. She was sitting naked in the cell,
masturbating much of the time, and smelling the fingers she had used in

981
 Otto F. Kernberg

masturbating. When I entered the cell to see her, she ignored me com-
pletely, and continued masturbating.
I need to stress that, in spite of her physical attractiveness, and this
open sexual behavior, there was a total lack of erotic quality in her
demeanor and in the atmosphere of that room. It is difficult to describe,
but there was a totally impersonal quality about the contact she estab-
lished with me, as if I were some strange object rather than a human
being. I was myself totally surprised by the lack of any erotic quality of
her behavior, or of the atmosphere she created in that cell. Her delusions
were that the devil was forcing her to have sex, and invading the world to
force everybody to have sex with him. She was attempting to assure her-
self that the devil had not destroyed her genitals, while believing that her
parents were being held prisoners by the devil, who had divided humanity
into those who were in prison, and those who would be sexually abused.
It was a quite chaotic system of delusions, mixed with other fantasies and
delusions involving nurses and teachers. I was trying to find the sense of
all this delusional material, trying to formulate it to myself as clearly as I
could, conveying an attitude of interest and effort to understand what was
going on in her mind, and attempting to help her clarify whatever was
confusing her. It was painful and laborious work, in the course of which I
found her becoming more and more attentive to me, but giving no indica-
tion that what I was saying influenced her in any way.
Only gradually did I realize that she was evaluating whether I was in
any way a secret representative or emissary of the devil, and hiding that
from her. Eventually we could talk about her suspecting me of not being
honest and straightforward, and that this seemed frightening to her. Then
one day I entered her cell, and there was a clearly erotic atmosphere in the
air. I couldn’t decide what brought this about, but the patient clearly showed
a seductive attitude toward me, and looked at me provocatively while still
masturbating. She told me with a sardonic smile, “I am the devil and you
are the devil,” treating both of us as if we were the devil enjoying seducing
her sexually. It was an intense and disturbing experience because she now
really appeared sexually attractive to me, while at the same time I was
afraid she would assault me. I was reassured by the fact that the windows of
the room were constantly monitored by the nursing service.
My effort in the following sessions was to point out to the patient that
the danger of being sexually assaulted by the devil had now taken the
form of misdiagnosing “us” as the devil, rather than her simply being the
victim, and I an emissary of the devil. In other words, my effort went into

982
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

attempting to differentiate two people out of what at that moment was a

clear condensation of her experience of herself and of me: we both were
the devil and the frightened girl. I believe this vignette illustrates the shift
from an early phase of a psychotic transference (“symbiotic” in Searles’s
terminology) and my early efforts to help the patient move into the phase
of differentiation between self- and object representation.

The Dead Mother Syndrome: Dismantling Transferences

A final type of specific transference disposition may be characterized as

the dismantling of all internalized object relations, and the corresponding
implication of a radical emotional unavailability in the treatment: namely, the
“dead mother syndrome” described by André Green (1993; Kohon 1999).
Here the tragedy is the patient’s lack of capacity for any investment in a sig-
nificant relationship because of the unconscious protective dismantling of all
internalized relationships. These patients usually present a very early trau-
matic background of an absent mother, often due to severe chronic depres-
sion in the mother during infancy or early childhood. In these cases the
unconscious wish to become reunited with their mother in death is reflected
in a sense that only the absence of all actual relationships will facilitate such
a condition. These patients develop a dangerous lack of emotional invest-
ments despite otherwise normal intelligence, capacity to differentiate self
from non-self, capacity for superficial yet adequate social and work relations,
and surprisingly normal superego functions, so that the lack of emotional
involvement becomes evident only in intimate contexts. Their extraordinary
capacity to maintain an apparently friendly but insurmountable distance in
the therapeutic relationship poses a major challenge over many months, even
years, of treatment. These are, fortunately, very rare cases, with reserved
prognosis, and are mentioned here only to contrast them with transference
patterns determined by internalized object relations, patterns that require spe-
cific employment of the four basic psychoanalytic techniques. I will not
explore these cases further here. The specific technical requirements and
challenges of the “dead mother” syndrome have been explored elsewhere.

C onclu d ing C omment

Classical psychoanalytic technique evolved in the context of Freud’s

work, primarily with neurotic patients, showing a predominance of an
integrated tripartite intrapsychic structure, corresponding to normal

983
 Otto F. Kernberg

identity. In light of the expansion of the spectrum of severity of patients’

pathology being treated with psychoanalytic psychotherapy, as well as
with psychoanalysis, and experience with modifications of classical tech-
nical psychoanalytic instruments, we now have available a broad spec-
trum of psychoanalytically based techniques that, jointly, should permit
their selective application across a broad spectrum of pathology. Standard
psychoanalysis may now be described as a specialized, classical form of
that body of techniques, offering a unique potential for further explora-
tion of unconscious conflicts, defensive operations, and structural organi-
zation in both normal and pathological functioning. This classical
technique is of central interest for psychoanalytic education, and for
teaching this standard technique for the treatment of neurotic patients. To
some extent, these applications have already been subjected to empirical
research that confirms the effectiveness of psychoanalytic approaches to
severe personality disorders. At the same time, we may now consider the
integrated body of psychoanalytic techniques as a broad spectrum of
related technical interventions based on psychoanalytic theory that can be
combined and modified according to the specific type of transference
structures that reflect various degrees and types of personality organiza-
tion, and thus make psychoanalytic technique a body of instruments that
have a broad spectrum of application in psychoanalytic psychotherapies.

References

Akhtar, S. (1992). Schizoid personality disorder. In Broken Structures:

Severe Personality Disorders and Their Treatment. Northvale, NJ:
Aronson, pp. 121–147.
Britton, R. (2004). Subjectivity, objectivity, and triangular space.
Psychoanalytic Quarterly 73:47–61.
Caligor, E., Kernberg, O.F., & Clarkin, J.F. (2007). Handbook of Dynamic
Psychotherapy for Higher Level Personality Pathology. Washington,
DC: American Psychiatric Publishing.
Clarkin, J.F., Levy, K.N., Lenzenweger, M.F., & Kernberg, O.F. (2007).
Evaluating three treatments for borderline personality disorder: A multi-
wave study. American Journal of Psychiatry 164:992–998.
Doering, S., Horz, S., Rentrop, M., Fischer-Kern, M., Schuster, P., Benecke,
C., Buchheim, A., Martius, P., & Bucheim, P. (2010). Transference-focused
psychotherapy v. treatment by community psychotherapists for border-
line personality disorder: Randomised controlled trial. British Journal of
Psychiatry 196:389–396.

984
 THERAPEUTIC IMPLICATIONS OF TRANSFERENCE STRUCTURES

Fairbairn, W.R.D. (1954). An Object-Relations Theory of the Personality.

New York: Basic Books.
Green, A. (1993). Le Travail du Négative. Paris: Éditions de Minuit.
Greenberg, J.R., & Mitchell, S.A. (1983). Object Relations in Psychoanalytic
Theory. Cambridge: Harvard University Press.
Guntrip, H. (1969). Schizoid Phenomena, Object-Relations and the Self.
New York: International Universities Press.
Jacobson, E. (1964). The Self and the Object World. New York: International
Universities Press.
Joseph, B. (1985), Transference—the total situation. International Journal of
Psycholanalysis 66:447–454.
Kernberg, O.F. (1975). Borderline Conditions and Pathological Narcissism.
New York: Aronson.
Kernberg, O.F. (1985). Internal World and External Reality: Object Relations
Theory Applied. Northvale, NJ: Aronson.
Kernberg, O.F. (1999). Psychoanalysis, psychoanalytic psychotherapy and
supportive psychotherapy: Contemporary controversies. International
Journal of Psychoanalysis 80:1075–1091.
Kernberg, O.F. (2004a). Aggressivity, Narcissism, and Self-Destructiveness in
the Psychotherapeutic Relationship. New Haven: Yale University Press.
Kernberg, O.F. (2004b). Contemporary Controversies in Psychoanalytic Theory,
Techniques, and Their Applications. New Haven: Yale University Press.
Kernberg, O.F. (2012). Identity: Recent findings and clinical implications. In
The Inseparable Nature of Love and Aggression: Clinical and Theoretical
Perspectives. Washington, DC: American Psychiatric Publishing, pp. 3–30.
Kernberg, O.F. (2014). An overview of the treatment of severe narcissistic
pathology. International Journal of Psychoanalysis 95:865–888.
Kernberg, O.F. (2018). The basic components of psychoanalytic technique and
derivative psychoanalytic psychotherapies. In Treatment of Severe
Personality Disorders: Resolution of Aggression and Recovery of
Eroticism. Washington, DC: American Psychiatric Publishing, pp. 49–72.
Kernberg, O.F., Burstein, E.D., Coyne, L., Appelbaum, A., Horwitz, L., & Voth,
H. (1972). Psychotherapy and psychoanalysis: Final report of the
Menninger Foundation’s Psychotherapy Research Project. Bulletin of
the Menninger Clinic 36(1/2).
Kernberg, O.F., & Caligor, E. (2005). A psychoanalytic theory of personality
disorders. In Major Theories of Personality Disorder, ed. M. Lenzenweger
& J.F. Clarkin. 2nd ed. New York: Guilford Press, pp. 114–156.
Kernberg, O.F., Yeomans, F.E., Clarkin, J.F., & Levy, K.N. (2008). Transference
focused psychotherapy: Overview and update. International Journal of
Psychoanalysis 89:601–620.

985
 Otto F. Kernberg

Klein, M. (1946). Notes on some schizoid mechanisms. International Journal

of Psychoanalysis 27:94–110.
Klein, M. (1958). On the development of mental functioning. International
Journal of Psychoanalysis 39:84–90.
Kohon, G. (1999). The Dead Mother: The Work of André Green. London:
Routledge.
Panksepp, J., & Biven, L. (2012). The Archaeology of Mind. New York: Norton.
Rey, J.H. (1979). Schizoid phenomena in the borderline. In Advances in the
Psychotherapy of the Borderline Patient, ed. J. LeBoit & A. Cappari.
New York: Aronson, pp. 449–484.
Rockland, L.H. (1989). Supportive Therapy for Borderline Patients: A
Psychodynamic Approach. New York: Guilford Press.
Rosenfeld, H. (1954). Considerations regarding the psychoanalytic approach
to acute and chronic schizophrenia. International Journal of
Psychoanalysis 35:35–140.
Rudolph, G. (2013). Strukturbezogene Psychotherapie. Stuttgart: Schattauer.
Searles, H.F (1965). Collected Papers on Schizophrenia and Related Subjects.
New York: International Universities Press.
Spillius, E., & O’Shaughnessy, E. (2012). Projective Identification: The Fate
of a Concept. London: Routledge.
Stone, M.H. (1983). Introductory comments on psychoanalytically oriented
treatment of schizophrenia. In Treating Schizophrenic Patients: A
Clinical Analytic Approach, ed. M.H. Stone, H.D. Albert, D.V. Forrest,
& S. Arieti. New York: McGraw-Hill, pp. 21–64.
Stone, M.H. (1986). Exploratory psychotherapy in schizophrenia-spectrum
patients: A reevaluation in the light of long-term follow-up of schizo-
phrenic and borderline patients. Bulletin of the Menninger Clinic
50:287–306.
Wallerstein, R.S. (1986). Forty-Two Lives in Treatment: A Study of
Psychotherapy and Psychoanalysis. New York: Guilford Press.
Winnicott, D.W. (1965). The Maturational Processes and the Facilitating
Environment. New York: International Universities Press.
Yeomans, F.E., Clarkin, J.F., & Kernberg, O.F. (2015). Transference-Focused
Psychotherapy for Borderline Personality Disorder: A Clinical Guide.
Washington, DC: American Psychiatric Publishing.

21 Bloomingdale Road
White Plains, NY 10605
E-mail: [email protected]

986