Borderline Pathology: An Integration of
Cognitive Therapy and Psychodynamic Therapy

Francois Louw

University of the Witwatersrand

Gillian Straker

University of Sydney

Because of the belated interest in borderline pathology outside of the psy-
choanalytic arena, there is a relative paucity of integrative models for the
treatment and understanding of borderline pathology. This article compares
and contrasts the practice of A. T. Beck and A. Freeman’s (1990) cognitive
treatment of borderline pathology with O. F. Kernberg’s (Kernberg, Selzer,
Koenigsberg, Carr, & Appelbaum, 1989) psychodynamic approach. An in-
tegrative model of borderline pathology is proposed, building on Beck and
Freeman’s and Kernberg’s shared use of the notion of schema–
representation. The practice implications of the proposed integrative model
are consequently entertained. A brief case study is provided to illustrate the
integration.

The psychotherapeutic treatment of patients with borderline pathology

continues to be one of the biggest challenges facing mental health profes-
sionals (Goldstein, 1993; Oldham, 1991). There are also indications that
borderline pathology is on the increase (Oldham, 1991; Sable, 1997). One
could in fact argue that borderline pathology and the identity diffusion that
is one of its cardinal features constitute the neurosis of the postmodern era;
Millon (1993) went so far as to refer to borderline personality disorder as

Francois Louw, Department of Psychology, University of the Witwatersrand, Johannes-

burg, South Africa; Gillian Straker, Department of Psychology, University of Sydney, Sydney,
New South Wales, Australia.

Francois Louw is now with the Westminster Community Support Team for People With

Learning Disabilities, St. Charles Hospital, London, United Kingdom.

Correspondence concerning this article should be sent to Francois Louw, 88 Cromwell

Avenue, Highgate, London N6 5HQ, United Kingdom. E-mail: Francois.Louw@westminster-
pct.nhs.uk

190

Journal of Psychotherapy Integration

Copyright 2002 by the Educational Publishing Foundation

2002, Vol. 12, No. 2, 190–217

1053-0479/02/$5.00 DOI: 10.1037//1053-0479.12.2.190

a psychosocial epidemic. In light of this, it is therefore not surprising that
borderline personality disorder is one of the most widely researched psy-
chiatric disorders and by far the most extensively researched personality
disorder (Linehan, 1993; Tutek & Linehan, 1993). This contrasts with the
fact that empirically based treatment research on borderline pathology is
sparse (Shea, 1991; Tutek & Linehan, 1993). Shea (1991) and Linehan
(1993) have concluded that little is known about the efficacy of psycho-
logical intervention for borderline pathology.

There seems to be general consensus that the term borderline first

appeared in the psychoanalytic literature (Goldstein, 1995; Pollack, 1990).
Stern formally introduced the term in 1938 (Oldham, 1991; Pollack, 1990)
to refer to a group of disorders intermediate between the neuroses and the
psychoses. Current use of the borderline diagnostic label appears to favor
either a narrow or a broad definition (Oldham, 1991). The broad definition,
associated with the psychoanalytic perspective, refers to diagnosis based on
personality organization or ego functioning. This diagnostic approach is
exemplified by Kernberg’s (1975, 1984b) concept of borderline personality
organization. The narrow definition, associated with descriptive psychiatry
and the Diagnostic and Statistical Manual of Mental Disorders (4th ed.
[DSM–IV]; American Psychiatric Association, 1994) uses a descriptive ap-
proach with operationalized criteria. The format of the descriptive ap-
proach to borderline personality disorder is polythetic; all criteria are of
equal diagnostic significance (Skodol & Oldham, 1991). The debate and
controversy surrounding the validity and reliability of these diagnostic
practices are outside the scope of this article. It is, however, important to
note that the respective diagnoses are similar but not identical: All patients
diagnosed with borderline personality disorder will have a concomitant
diagnosis of borderline personality organization, but not all patients with a
diagnosis of borderline personality organization will meet the diagnostic
criteria for borderline personality disorder (Clarkin & Kernberg, 1993). In
this article we use the term borderline pathology to refer to either border-
line personality disorder or borderline personality organization, depending
on the theorist under discussion.

Since Stern’s description of borderline pathology, a plethora of psy-

choanalytic clinicians have contributed to the now-burgeoning literature in
this area. More recently, Masterson (1978), Rinsley (1982), and Gio-
vachinni (1993) have made significant contributions to the treatment and
understanding of borderline pathology. The work of Kernberg (Kernberg,
1975, 1984b; Kernberg, Selzer, Koenigsberg, Carr, & Appelbaum, 1989) is,
however, regarded by many as the most definitive and comprehensive (e.g.,
Shea, 1991; Swenson, 1989; Westen, 1991). In contrast to their psychoana-
lytic counterparts, cognitive theorists and clinicians have only recently
made a contribution to the treatment and understanding of borderline

Borderline Pathology: An Integration

191

pathology. Cognitive therapy has traditionally focused on acute psychiatric
conditions such as depression (Beck, Rush, Shaw, & Emery, 1979) and
anxiety (Beck & Emery, 1985). The publication of Cognitive Therapy of
Personality Disorders (Beck & Freeman, 1990) thus ushered in a new era
for the applicability of cognitive therapy.

The addition of cognitive therapy to the treatment and understanding

of borderline pathology has also opened an avenue for possible integration
of treatment approaches. Recent years have witnessed the consolidation of
the psychotherapy integration movement (Norcross & Arkowitz, 1992),
but because of the belated interest in borderline pathology outside of the
psychoanalytic arena there is a paucity of integrative models for the treat-
ment and understanding of borderline pathology. There have, however,
been fruitful attempts at psychotherapy integration with personality disor-
ders. Hellkamp (1993) provided an integrative framework for thinking
about and treating what he referred to as “severe mental disorders.”
Hellkamp’s description of the population of patients with severe mental
disorders would include some patients with a borderline diagnosis. Gold
and Stricker (1993) proposed an integrative theoretical and treatment
model for character disorders in general. Their integrative approach draws
on notions from psychodynamic and cognitive–behavior therapy and on
some systemic ideas. Although Hellkamp (1993) and Gold and Stricker
(1993) have made valuable contributions with regard to psychotherapy
integration with personality disorders, their respective models are not spe-
cific to borderline pathology.

1

Two authors who are illustrative of a disor-

der-specific integration are Linehan (1993) and Ryle (Ryle & Beard, 1993;
Ryle & Marlowe, 1995).

Linehan (1993) developed a treatment program for borderline person-

ality disorder that she labeled dialectical behavior therapy (DBT). She used
a dialectic framework (which includes some systemic tenets) to integrate
mainly cognitive and behavioral notions and ideas from Zen Buddhism.
Linehan proposed that borderline pathology is primarily a dysfunction of
the emotion regulation system that results from biological irregularities
combined with certain dysfunctional environments, as well as their inter-
action and transaction over time. More specifically, she proposed that bor-
derline patients are biologically predisposed to emotional vulnerability and
difficulties in emotion modulation. According to Linehan, dysfunction of
the emotion regulation system in borderline individuals is exacerbated by
specific environmental experiences (Linehan, 1993). Pathogenic environ-
ments are characterized by invalidation. There is a negation of the indi-
vidual’s description and analysis of his or her own experiences, specifically

1

Aspects of Gold and Stricker’s (1993) work are incorporated into the integrative per-

spective put forward in this article.

Louw and Straker

192

with regard to causal attributions of emotions, beliefs, and actions. The
individual’s experiences are ascribed to socially unacceptable characteris-
tics or to personality traits. Furthermore, invalidating environments are
generally intolerant of displays of negative affect, at least when such dis-
plays are not accompanied by public events supporting such emotion
(Linehan, 1993). The treatment program that Linehan developed is for the
most part an application of a broad array of cognitive and behavioral
strategies. DBT, however, has a number of distinctive defining features. Its
overriding characteristic is an emphasis on dialectics. The most fundamen-
tal dialectic is the necessity of accepting patients unconditionally within the
context of attempting to teach them to change. Other core features are a
focus on problem solving and observation and management of the contin-
gencies operating in the therapeutic relationship (Linehan, 1993).

Ryle and Cowmeadow (1992) devised a short-term psychotherapy that

integrates ideas from cognitive psychology and therapy, social psychology,
and object-relations theory, known as cognitive–analytic therapy (CAT).
Their approach to the treatment of borderline pathology stems from this
more general integrative endeavor. The central concept of CAT is the
reciprocal role procedure (RRP). Procedure refers to the sequence of: (a)
mental processes (perception, appraisal, action planning); (b) the act or
role; and (c) the consequences of the action or role, which will be appraised
(under [a]) and confirmed or revised. Role refers to the enacted interper-
sonal behavior and the accompanying communications and emotions. The
essential consequence of interpersonal enactment is the reciprocal role
played by the other (Ryle & Marlowe, 1995). RRPs develop from early
caregiver–infant interactions. Each interaction yields two complementary
roles (self and other), and the developing child elaborates an individual
RRP repertoire that defines options for relating and self-management.
Over time, the child’s array of interpersonal procedures is elaborated and
becomes integrated into complex procedures that are mobilized in a con-
text-appropriate fashion. According to Ryle and Marlowe (1995), the pro-
cess of elaboration and integration goes awry when children are confronted
with inconsistency or overwhelming traumatic experiences, especially if
such experiences are not mediated by caretakers. As a consequence, the
child’s capacity for self-reflection is impaired. This impairment is regarded
as a major factor in the etiology of borderline personality disorder. Ryle
and Marlowe (1995), furthermore, proposed that the maintenance of bor-
derline pathology can be ascribed to three main factors. First, self-states
are precarious and have a narrow range of RRPs. In a given state, the
borderline individual exerts considerable pressure on others to enact the
reciprocal role that will confirm the RRP. Second, when attempts to elicit
role reciprocation fail, the borderline individual switches to another state
rather than revising the procedure. Last, discontinuity of experience and

Borderline Pathology: An Integration

193

impaired recall of one state when in another results from, and reinforces,
the absence of a central self-observing and self-managing capacity in which
the individual is capable of identifying, taking responsibility for, and learn-
ing from the consequences of his or her actions (Ryle & Marlowe, 1995).
Sequential diagrammatic reformulation is at the heart of CAT for border-
line pathology (Beard, Marlowe, & Ryle, 1990). The therapist and patient
jointly create a self states sequential diagram (SSSD) over a period of four
to seven sessions. Beard et al. (1990) described the aim of the SSSD as

to identify and characterise the patients’ principle mental states, and to work out
how transitions occur between them. The end result, characteristically, is a linked
set of state descriptions of which one, the “core state,” represents the long-term,
unresolved psychic pain of the patient. (p. 542)

As the sequence of transitions between self states is jointly worked out, the
patient’s problems become explicable. The jointly constructed SSSD pro-
vides the patient with visible evidence of being understood and mobilizes
new capacities for self-awareness. Continued use of the SSSD over the
course of therapy facilitates further self-observation and provides the basis
for self-continuity and control. For the therapist, the SSSD aids accurate
identification of the patient’s state shifts as well as identification of the
therapist’s own shifting responses, which facilitates avoidance of collusion.
The diagram also guides intervention by indicating where destabilization is
most likely to be fruitful (Beard et al., 1990; Ryle & Marlowe, 1995).
Within the framework of reformulation, the therapist is able to draw on a
wide range of specific methods, for example, cognitive, behavioral, and
gestalt techniques (Beard et al., 1990; Ryle & Cowmeadow, 1992).

In this article we focus on an integration of Beck and Freeman’s (1990)

cognitive treatment of borderline pathology and Kernberg’s (1975, 1984b;
Kernberg et al., 1989) psychodynamic approach. The proposed integrative
model is specific to borderline pathology and its treatment. A synopsis of
both treatment methods is followed by a comparative analysis of their core
assumptions and practice. The comparative analysis will set the stage for an
exploration of conceptual integration. Consequently, the practice implica-
tions of the proposed integrative model will be entertained. Therapeutic
praxis will be illustrated by means of a case study.

BECK’S COGNITIVE THERAPY

Beck and Freeman (1990) provide a cognitive conceptualization of

personality functioning and pathology within an ethological–evolutionary
framework. The main thrust of their argument concerns how personality
processes are formed and operate in the service of adaptation. They spec-

Louw and Straker

194

ulated that the prototypes of one’s personality patterns derive from one’s
phylogenetic heritage. Cognition, affect, and behavior are thus examined in
light of their possible relation to ethological strategies. According to Beck
and Freeman, a person’s genetic predisposition interacts with his or her
interpersonal environment. One’s innate characteristics are thus best
viewed as “tendencies” that can be accentuated or diminished by experi-
ence (Beck & Freeman, 1990).

An individual’s information processing is influenced by his or her be-

liefs and other components of his or her cognitive organization. Beliefs are
organized hierarchically; progressively broader and more complex mean-
ings are assigned at successive levels (Beck & Freeman, 1990). Schemas are
components of a higher order of structuring that attaches meaning to
events; they are viewed as the building blocks of personality. Individuals
with personality disorders tend to process information selectively and in a
dysfunctional manner. This bias in interpretation and subsequent behavior
is shaped by dysfunctional beliefs. According to Beck and Freeman (1990),
three basic beliefs are generally held by patients with borderline person-
ality disorder: (a) “I am powerless and vulnerable,” (b) “The world is
dangerous and malevolent,” and (c) “I am inherently unacceptable.” Bor-
derline individuals display the full range of cognitive distortions posited by
Beck et al. (1979): arbitrary inference, selective abstraction, overgeneral-
ization, magnification, minimization, personalization, and dichotomous
thinking. Dichotomous thinking—the tendency to evaluate experience in
terms of mutually exclusive categories—is particularly pervasive in border-
line patients. There is thus no grading of experiences; consequently, emo-
tional responses and actions are extremes, and there are rapid shifts be-
tween opposing views (Beck & Freeman, 1990). Finally, Beck and Freeman
(1990) view a weak or unstable identity as an important factor in borderline
pathology. This diffusion results in confusion regarding goals, which inter-
feres with sustained effort.

Cognitive therapy of personality disorders is an extension and modi-

fication of cognitive therapy as used for acute psychiatric conditions. Beck
and Freeman (1990) placed a strong emphasis on the necessity for specific
conceptualization of each case. The focus of information gathering is on
identifying the patient’s schemas, specifically, the patient’s self-concept and
the rules and regulations by which he or she lives, as well as his or her views
of others. In general, therapeutic goals are derived from the patient’s core
schemas. There is also a strong emphasis on the therapeutic relationship,
which is characterized by collaboration and guided discovery. The use of
transference reactions within the therapeutic relationship is also advocated
(Beck & Freeman, 1990). The patient’s dysfunctional beliefs vis-a`-vis the
therapist are dealt with by using standard cognitive techniques. The change
technology used in the treatment of personality disorders includes the

Borderline Pathology: An Integration

195

whole range of cognitive and behavioral techniques espoused by Beck for
the treatment of depression and anxiety, for example, the labeling of cog-
nitive distortions, “thought catching,” and distraction techniques (Beck &
Emery, 1985; Beck et al., 1979). Furthermore, Beck and Freeman (1990)
advocated the use of experiential techniques, such as imagery and the
reliving of childhood experiences (by means of role playing and reverse
role playing).

The treatment of borderline pathology focuses on helping the patient

establish a clearer sense of identity, improve control of emotions, and
change maladaptive beliefs and assumptions. Dichotomous thinking is ini-
tially the main focus of intervention as it is central to the borderline pa-
tient’s labile mood and because it amplifies the impact of dysfunctional
beliefs. A reduction in dichotomous thinking could result in a significant
amelioration of the patient’s symptoms and facilitate the task of confront-
ing schemas (Beck & Freeman, 1990).

KERNBERG’S PSYCHODYNAMIC THERAPY

Kernberg’s psychoanalytic approach has its conceptual underpinnings

in ego psychology and object-relations theory (Kernberg, 1984a, 1987).
Within this ego psychology–object-relations framework unconscious con-
flict is conceptualized as occurring between contradictory self- and object
representations (Kernberg, 1992). Opposing poles of self- and object rep-
resentations reflect the drive (libidinal/aggressive) and defensive structure,
respectively. Kernberg (1975, 1984b) has delineated a structural model of
personality organization. Within a Kernbergian framework, intrapsychic
structure and diagnosis are inextricably linked; structure guides diagnosis
and thus treatment modality. Kernberg (1984b) distinguished three broad
classes of personality organization: neurotic, borderline, and psychotic. As
elucidated by Kernberg, the diagnosis of personality is based on three
structural criteria: identity diffusion, level of defensive operations, and
capacity for reality testing. Individuals with a borderline personality orga-
nization lack a single coherent concept of the self, significant others, or
both. Although there is sufficient differentiation of self and other to permit
the maintenance of ego boundaries, the integration of “good” and “bad”
self- and object representations is not achieved. Borderline personality
organization is characterized by a predominance of primitive defensive
operations (primitive idealization, projective identification, denial, om-
nipotent control, and devaluation) centered on the mechanism of splitting.
Reality testing is intact in individuals with borderline personality organi-
zation; however, alterations might occur in relationship with reality and in
feelings of reality.

Louw and Straker

196

According to Kernberg (1975, 1984a), borderline personality organi-

zation develops because of a failure to integrate libidinally and aggressively
determined self- and object representations. This lack of integration de-
rives from the pathological predominance of aggressively determined self-
and object representations and the concomitant failure to establish a suf-
ficiently strong ego core around the good self- and object representations
(originally undifferentiated). Kernberg (1984a) posited that this excess of
aggression is caused by constitutional factors or problems in attunement
between mother and infant or, in most cases, by a combination of both.

Kernberg (1984b) views expressive psychodynamic therapy as the

treatment of choice for borderline personality organization. Expressive
psychotherapy rests on the same pillars as psychoanalysis: technical neu-
trality, interpretation, and the development and resolution of transference.
In the case of expressive psychodynamic therapy these specifications are,
however, modified. In expressive therapy technical neutrality is an ideal
that is strived for, not adhered to, as in psychoanalysis. Interpretation is
modified in the sense that genetic interpretation is postponed until more
advanced stages of treatment. Furthermore, the use of clarification and
confrontation as interventions that aid eventual interpretation predomi-
nate in the early stages of treatment. Finally, analysis of transference is not
as systematic as in orthodox psychoanalysis. The overall goal of psycho-
therapy with borderline patients is to help them to develop multidimen-
sional, cohesive, and integrated images of themselves and of others. The
task of therapy is thus to coalesce part–object and part–self representations
(Kernberg et al., 1989). Kernberg places a strong emphasis on the estab-
lishment of a therapeutic contract at the outset of therapy (Kernberg et al.,
1989). Contracting pertains to therapist and patient responsibilities, respec-
tively, as well as to potential acting out, based on the patient’s history. The
intricacies of the contracting process, including subsequent utilization when
acting out presents itself, was elaborated by colleagues of Kernberg (Yeo-
mans, Selzer, & Clarkin, 1992).

Expressive therapy focuses on the analysis of transference by means of

the interpretive method. As such, analysis involves exploring the here-and-
now reactivation of past internalized object relations by means of clarifi-
cation, confrontation, and interpretation. Kernberg (1987) noted the im-
portance of exploring the patient’s experience both outside the therapeutic
hour (so called extratransference) and in the therapeutic relationship itself.
The transference manifestations of borderline patients initially appear cha-
otic because of the rapid oscillation between contradictory part–self and
part–object representations. The therapist will gradually be able to discern
repetitive patterns of specific object-relations units. These units, or self–
object dyads, are typically played out in alternating ways by means of
projective and introjective mechanisms (Kernberg et al., 1989). Once the

Borderline Pathology: An Integration

197

dominant object relation involved in the here-and-now therapeutic inter-
action is identified, it is communicated to the patient. Such transference
interpretations include the labeling of the part self- and object represen-
tation as well as the linking affect. Kernberg (1993) summarized the psy-
chotherapeutic strategy in terms of a three-step procedure:

1. Diagnosis of an emerging primitive part–object relationship in the

transference and the interpretative analysis of the dominant uncon-
scious fantasy structure that corresponds to the particular transfer-
ence activation.

2. Identification of the self- and object representation of the particular

transference and the typically oscillating attribution of self-
representations and object representations by the patient to him- or
herself, and to the therapist.

3. Linking of the object relationship activated in the transference with

an entirely opposite split-off one, activated at another point in the
therapeutic process.

COMPARATIVE ANALYSIS

Our comparison of Beck and Freeman’s cognitive approach with Kern-

berg’s psychodynamic approach will highlight points of similarity and dif-
ference in relation to ontological and metapsychological assumptions, etio-
logical models, and therapeutic praxis.

Beck and Freeman’s metapsychological position is grounded in cogni-

tive–information-processing theory, tempered with an ethological–
evolutionary perspective. Kernberg’s metapsychological stance is rooted in
an integrative synthesis of Freudian drive theory and British object rela-
tions. Kernberg’s view of drives, however, radically departs from that of
Freud and contemporary ego psychology. His conceptualization of drives
incorporates more modern notions from ethology and research in the field
of affect and infant observation (Kernberg, 1992). Although Beck and
Freeman’s and Kernberg’s metapsychological frameworks are radically dif-
ferent, both draw on ethological notions.

The ontological assumptions of the two approaches stand in stark con-

trast. The cognitive approach assumes agency and free will. Humans are
essentially rational beings, and there is primacy of cognition over affect.
The mind is assumed to be an “open book”; humankind has access to the
processes and content of the psyche. An explicit concept of motivation
seems to be absent from Beck and Freeman’s assumptions regarding the
nature of person. The Darwinian notions of survival and fit can, however,
be inferred as motivational constructs, in light of Beck and Freeman’s

Louw and Straker

198

metapsychological allegiance to ethology and evolutionary theory. Kern-
berg’s view of person is more deterministic than that of Beck and Freeman.
Personhood is born of a dialectical interplay between conscious and un-
conscious forces. Humans are driven by unconscious processes that defy
rationality. A conscious “unknown” thus assumes center stage in humans’
thinking, feeling, and acting. In contrast to Beck and Freeman, Kernberg
conceptualization of motivation is more complex and elaborate; it inte-
grates drive psychology and object relations. Kernberg’s allegiance to clas-
sical drive theory is, however, controversial and contentious. Although
Kernberg uses drive terminology, his conceptualization is a radical depar-
ture from Freudian drive theory (Greenberg & Mitchell, 1983; Summers,
1994). According to Kernberg (1992), object-directed affects (also referred
to as drive derivatives) precede and give rise to the sexual and aggressive
drives. Drive “derivatives” provide affective valence for the internalization
of interactional units (self- and object representations). During the course
of development, self- and object representations with similar affective va-
lences cluster together and are eventually organized into the dual drives as
superordinate motivational systems. Drives thus play less of a motivational
role in Kernberg’s conceptualization, in contrast to classical drive theory.
Drive derivatives (affects), however, serve a critical organizing function.

Kernberg’s and Beck and Freeman’s respective views with regard to

the etiology of borderline pathology reflect their aforementioned theoret-
ical proclivities. According to Beck and Freeman, borderline pathology is
caused by biased information processing. Although Beck and Freeman
(1990) posited that “tendencies” toward biased information processing can
be inherited, they did not specify the nature of such predisposing factors or
operationalize this notion with regard to borderline pathology. “Distorted”
cognizing—and, ultimately, dysfunctional behavior and affect—“are
formed and operate in the service of adaptation” (Beck & Freeman, 1990,
p. 22). The maladaptive schemas and cognitive distortions that characterize
borderline personality disorder are thus learned in an environment or con-
text where they serve adaptive purposes. According to Kernberg (1975),
borderline pathology results from a failure to integrate all-good self- and
object representations with all-bad self- and object representations. Kern-
berg postulated that this lack of integration is caused by a disturbance of
the mother’s availability, which is due either to a constitutional excess of
aggression or to maternal problems with parenting, or a combination of
both. The influence of factors internal to the individual, however, predomi-
nates in Kernberg’s conceptualization. The affective state under which
internalization–introjection takes place is central. Furthermore, once inter-
nalization has taken place, all subsequent perception is distorted by un-
conscious fantasy based on internal objects.

Finally, in this section we compare and contrast the practice of cogni-

Borderline Pathology: An Integration

199

tive therapy with the practice of expressive psychodynamic therapy. The
style of the therapist and the therapeutic relationship in cognitive therapy
is characterized by collaborative empiricism. The treatment endeavor is a
joint one. Within this collaborative relationship the therapist plays an edu-
cative and directive role. The patient’s education and learning take place
both inside and outside of the therapy room. Self-directed change by means
of between-session tasks and assignments is integral. Beck and Freeman
advocate a relationship focus in the treatment of borderline personality
disorder if the patient’s distorted perceptions of the therapist and conse-
quent behavior interfere with treatment. The aim of cognitive therapy is to
change biased information processing, and a cognitive mechanism of
change is assumed. All interventions—behavioral, experiential, and so
forth—aim at cognitive modification. Therapists practicing expressive psy-
chodynamic therapy strive to be technically neutral. In contrast with cog-
nitive therapists, psychodynamic therapists will thus avoid actively inter-
vening in the patient’s life, for example by giving advice or making
suggestions, or by directing the patient in what to do. When the therapist
has veered from neutrality, the reason that necessitated the deviance is
explored and interpreted. The therapeutic relationship, by means of analy-
sis of the transference, is central to the therapeutic endeavor. The major
vehicle for change is the interpretive method, rather than more active–
directive methods. Interpretation involves highlighting the unconscious
motive or function of the patient’s behavior and perceptions. The therapist
has a privileged perspective vis-a`-vis the patient’s unconscious and is thus
a more authoritative figure than the cognitive therapist. The notion of a
working alliance between therapist and patient, which in some aspects
parallels the notion of collaborative empiricism, is assumed to be tenuous
with borderline patients. Expressive psychodynamic therapy aims to inte-
grate split-off parts of the patient’s personality. The very nature of self- and
object representations makes this a dual affective and cognitive endeavor.

Psychodynamic exploration and interpretation thus focus on both the

cognitive and affective aspects of specific self- and object representations.
The focus on cognitive–affective change stands in contrast with cognitive
therapy’s exclusive emphasis on the cognitive.

THEORETICAL SYNTHESIS

From the preceding comparative analysis it is clear that there are

fundamental theoretical and practice differences between cognitive
therapy and expressive psychodynamic therapy. Before we proceed, we
outline the conceptual strategy used to bridge these differences. First,
where there are convergences between Beck and Freeman’s cognitive ideas

Louw and Straker

200

and Kernberg’s psychodynamic notions, we highlight and link them. Sec-
ond, where necessary, we use notions from other theoreticians and ap-
proaches to bridge theoretical gaps. Finally, we use theoretical constructs
from other theorists when both Kernberg’s and Beck and Freeman’s no-
tions are deficient within the context of generating a conceptually sound
integration.

The concept of schema–representation is proposed as a transtheoreti-

cal construct; it is used as an explanatory concept in both psychoanalytic
(e.g., Bowlby, 1988; Horowitz, 1988; Kernberg, 1992) and cognitive think-
ing (e.g., Beck & Freeman, 1990; Young, 1990). From a cognitive perspec-
tive, schemas are hypothetical structures that guide both the processing of
information and the implementation of action. These templates, or generic
cognitive structures, are extracted in the course of exposure to particular
instances of a phenomenon. Working from a psychoanalytic perspective,
Perlow (1995) defined a schema as

an amalgamation of memories regarding an object, which functions as an anticipa-
tory set for future interaction. As such, a mental representation of an object refers
to a “schema” which organizes experience and provides a context both for present
perceptions and phantasies, and for the recall of past memories. (p. 2)

As is evident, the two respective definitions are strikingly similar. Psycho-
analytic and cognitive thinking thus seem to converge at a cognitive level.
This convergence was noted by Turner (1993), who pointed out that the-
orists such as Bowlby, Horowitz, and Luborsky have constructed cognitive
revisions of the psychoanalytic model. Turner further noted that cognitive
psychology’s tenets of tacit information processing and a feed-forward
mechanism have opened the portal for dynamic theorizing in cognitive
therapy. Psychoanalytic and cognitive notions of schema–representation,
however, differ in important ways. A major point of divergence is psycho-
analytic emphasis on the affective aspects of interaction versus the cogni-
tivist emphasis on the purely cognitive (Perlow, 1995). From a psycho-
analytic viewpoint schemas are thus cognitive–affective in nature. Further-
more, the cognitive endeavor is geared toward exploring schemas in terms
of their representation of reality, however distorted such representation
might be. Psychoanalysis is concerned with the formative influences of
internal factors: drives, phantasies, and so on (Perlow, 1995). Representa-
tions are viewed as heavily influenced by drives or drive derivatives. Rep-
resentation is an amalgam of (a) a real object, (b) drive–affect, (c) defen-
sive distortion of drive, and (d) the external object’s reaction to being
related to vis-a`-vis the representation.

Although Kernberg’s self- and object representations and Beck and

Freeman’s cognitive schemas show considerable conceptual overlap, their
differences seem to be crucial. These convergences and divergences are

Borderline Pathology: An Integration

201

illuminated by means of a comparison with Safran’s (1990) notion of an
interpersonal schema. Safran “contextualized” a cognitive viewpoint by
integrating it with an interpersonal perspective. According to Safran, an
interpersonal schema is abstracted on the basis of interactions with attach-
ment figures and allows the individual to predict interactions in a way that
increases the probability of maintaining relatedness with these figures.
Safran rooted the notion of interpersonal schema in Bowlby’s (1988) at-
tachment theory and Sullivan’s (1955) interpersonal view of psychoanaly-
sis; relatedness is assumed to be a biologically wired-in propensity because
of its importance in the survival of the species. According to Safran, inter-
personal schemas are generalized representations of self–other relation-
ships; schemas are intrinsically interactional in nature. Furthermore, Safran
hypothesized that information relevant to attachment has an important
affective component and that it is thus coded, at least in part, in expressive–
motor behavior. Safran thus proposed that interpersonal schemas are best
conceptualized as cognitive–affective structures as opposed to purely cog-
nitive structures.

Safran’s (1990) and Beck and Freeman’s (1990) respective notions of

schema are functionally similar; they are formed and operate in the service
of adaptation. Safran, however, explicitly placed emphasis on interpersonal
adaptation. Furthermore, Safran’s conceptualization presupposes a unitary
structure of self- and other representations, whereas Beck and Freeman
posited separate schemas for self- and other representations, respectively.
However, scrutiny of the three basic beliefs that Beck and Freeman view as
generally held by patients with borderline personality disorder reveals im-
plicit relatedness. For example, the belief “I am inherently unacceptable”
implies unacceptability in relation to an other. Although structurally au-
tonomous, self- and other representations seemingly interface with regu-
larity from a phenomenological perspective. Although Beck and Freeman
prioritize the cognitive, they do not deny affect as a facet of human expe-
rience. They merely view affect as represented by separate, autonomous
schemas. Schemas representing affect interface with schemas representing
beliefs and assumptions, according to Beck and Freeman, whereas for
Safran cognition and affect are structurally fused.

Beyond the thorny issue of origin, Beck and Freeman and Kernberg

agree that schemas, once formed, operate to distort perception. Further-
more, both Beck and Freeman and Kernberg posit internal processes as
central to the maintenance of schemas. Beck and Freeman refer to these
processes as cognitive distortions, whereas Kernberg referred to them as
defense mechanisms.

The centrality of internal mechanisms is reflected in the therapeutic

endeavor; Kernberg’s interpretation of splitting and cognitive therapy’s
focus on dichotomous thinking. Both views tend to de-emphasize the ex-

Louw and Straker

202

ternal–reality in their notions of maintenance. Safran (1990), in contrast,
emphasized maintaining factors in the “real” world by means of the con-
cept of dysfunctional interactional cycles. Drawing on interpersonal theory,
Safran posited that all behavior occurs in interactional contexts and, as
such, invites reciprocity, that is, a pull toward complementary behavior.
Safran, furthermore, suggested that individuals with maladaptive and rigid
interpersonal schemas will have rigid and stereotyped interpersonal reper-
toires. Such rigid interpersonal behavior is likely to elicit responses from
others that are schema consistent. There is thus less diversity in the inter-
personal experience of individuals with personality pathology and
concomitantly less opportunity for experiences that will disconfirm mal-
adaptive interpersonal schemas. Schemas are thus maintained by cogni-
tive–interpersonal cycles that are self-fulfilling in nature (Safran, 1990).

The conceptualization of differences in a dialogical, complementary

relationship, rather than in opposition, thus reveals that Kernberg’s con-
cept of self-and object representations, Safran’s concept of interpersonal
schema, and Beck and Freeman’s notion of separate and autonomous but
interfacing schemas are strikingly similar. Generically speaking, the inter-
personal schemas/self- and object representations of patients with border-
line pathology are dysfunctional cognitive–affective scripts, which manifest
in a lack of an integrated coherent sense of self and others and consequent
interpersonal difficulties. These schemas–representations develop from an
interaction between the individual and significant others. Schemas are also
maintained in a reciprocal fashion in interactional contexts. The rigidity of
schemas is thus a function of both intrapsychic and interpersonal processes.
Patterns of psychic experience and relating to others are thus derived from
past relationship experience but are continuously operating in the present
in a continuous manner. This notion of continuous construction contrasts
with that of the more traditional psychoanalytic notion of fixation–
regression, which postulates that psychic stasis is due to fixation at particu-
lar developmental junctures or regression to previous developmental ep-
ochs (Zeanah, Anders, Seifer, & Stern, 1989).

Kernberg and Beck and Freeman, however, have divergent views with

regard to the nature of the intrapsychic processes that maintain schemas.
Beck and Freeman view schema maintenance as a function of information
processing, whereas Kernberg invokes motivational notions in his expla-
nation of the fixedness of self–object representations. Disagreement with
regard to the necessity and utility of motivational constructs thus needs to
be bridged. Westen (1988) suggested that an understanding of the elicita-
tion and management of affect can provide a bridge between notions of
information processing and defense. Westen (1988) demonstrated how
such a synthesis can be achieved by examining what he referred to as a
cognitive–affective mismatch. This notion suggests that an individual estab-

Borderline Pathology: An Integration

203

lishes a “set goal,” or ideal state, with respect to a stimulus or situation and
that a discrepancy between the ideal state and cognized reality (a “mis-
match”) realizes an affect. The affect performs a feedback function, which
activates control mechanisms designed to minimize the affect. Information
processing can thus function in the service of affect regulation.

A further related issue is the level at which information processing/

defense takes place, as well as the location and hence accessibility of sche-
mas–representations. The polarity of cognitive surface and psychodynamic
depth needs to be transcended. A three-tier model of personality function-
ing (Stricker & Gold, 1988) allows for the conceptual integration of dif-
ferent types or levels of psychological experience that are seen as dynami-
cally interrelated in a circular manner. Causation is thus multidirectional,
in the sense that a psychological phenomenon in any of the tiers can be the
result or cause of a phenomenon at another level of experience. Overt
behavioral problems are located at Tier 1; maladaptive conscious cognition
and emotions are located at Tier 2; and unconscious motives, conflicts, and
schemas–representations are located at Tier 3. Any psychological phenom-
enon is manifest across all three levels of experience. In this model, be-
havior and conscious cognition are thus not viewed as inevitably and ex-
clusively caused by unconscious processes.

Thus far we have outlined the characteristics of maladaptive schemas.

What remains to be addressed is the etiology of these cognitive–affective
structures. Certain constitutional factors are assumed to predispose an
individual to borderline pathology. Neither Kernberg nor Beck and Free-
man have fleshed out the specifics of such predisposition in any detail.
Furthermore, both Kernberg and Beck and Freeman view biological pre-
disposition as interacting with environmental (interpersonal) factors. Line-
han’s (1993) view with regard to the etiology of borderline pathology pro-
vides a sound conceptualization that can subsume Kernberg’s notion of
excess aggression and lack of attunement and Beck and Freeman’s notion
of environmental fit. Linehan hypothesized that borderline personality dis-
order results from a dysfunction of the emotion regulation system due to
biological irregularities, combined with certain dysfunctional environments
(characterized by invalidation), as well as their interaction and transaction
over time. The impact of such predisposition thus depends on the fit be-
tween mother and child and the quality of the psychosomatic partnership
(Winnicott, 1960/1990) that they develop. More specifically, individuals
with borderline pathology are biologically predisposed to emotional vul-
nerability and difficulties in emotion modulation. Characteristics of emo-
tional vulnerability include high sensitivity to emotional stimuli, emotional
intensity, and slow return to emotional baseline. Because of the inter-
twined nature of cognition and affect vis-a`-vis schemas, emotional vulner-
ability has a disorganizing and distorting effect on the formation of cohe-

Louw and Straker

204

sive, coherent, and realistic interpersonal scripts. Separate and distinct
schemas are abstracted in the course of differential affective experiences,
and these schemas influence consequent perception and experience in a
biased manner.

CLINICAL IMPLICATIONS

According to Arkowitz (1989), the aim of theoretical integration is to

blend two or more psychotherapies at a conceptual or theoretical level. The
hope is that such a synthesis will create a framework that incorporates the
best elements of the individual psychotherapies on which the synthesis is
based. Although theoretical integration goes beyond eclecticism’s technical
blending, it does not exclude the integration of intervention methods. Thus
we now explore the possible clinical implications of the integrative con-
ceptual model with the aim of broadening the interventive scope afforded
by “pure” cognitive therapy or expressive psychodynamic therapy.

The broad aim of psychotherapy with patients with borderline pathol-

ogy, which flows from the integrative model, is the modification of mal-
adaptive self–other schemas. This specific targeting of schemas is similar to
Young’s (1990) approach in schema-focused cognitive therapy. More spe-
cifically, three core manifestations of maladaptive schemas are proposed as
the focus of intervention: (a) difficulties in interpersonal relationships, (b)
self-functioning (identity diffusion), and (c) affect regulation. The treat-
ment is exploratory and insight oriented, as opposed to action oriented.
Borderline patients are thus assisted in expanding their self-knowledge by
means of increased self-awareness of their behavior, thinking and feeling,
and gaining insight into the machinations of their conscious and uncon-
scious mental functioning.

Given that schemas are interpersonal in nature, the relationship be-

tween therapist and patient is central to the stated therapeutic aim of
modifying and restructuring schemas. For the sake of conceptual clarity, we
discuss the therapeutic relationship and its components separately from the
techniques of the proposed approach. We use a tripartite distinction of-
fered by Gold (1993) in elaborating the therapist–patient relationship. The
three facets of the therapeutic relationship that Gold delineated are (a) the
emotional climate (of the interaction), (b) the interactional stance, and (c)
the role of interactional data. Gold defined the emotional climate as the
“the quality and quantity of affective engagement and involvement be-
tween patient and therapist which are thought to be helpful, necessary, or
ameliorative” (p. 526). Although there are numerous concepts from di-
verse orientations that underscore the defined emotional climate, Winn-
icott’s (1960/1990) notion of a holding environment most aptly captures

Borderline Pathology: An Integration

205

this aspect of the therapeutic relationship. Holding provides the “space”
for development and fostering of a working alliance and the development
and exploration of transference. The working alliance with patients with
borderline pathology is usually tenuous and fragile when treatment com-
mences. Because of their pathological schemas, these patients struggle to
ally themselves with the therapist and thus form a working alliance. Pa-
tients with borderline pathology often act out in ways that undermine the
therapeutic process, such as by not attending sessions and so on. Acting out
necessitates measures that protect the therapeutic space. Kernberg advo-
cated making a contract at the outset of therapy (Kernberg et al., 1989;
Yeomans et al., 1992). Limits and responsibilities that are contractually
agreed on when treatment commences are referred back to in order to
protect the holding environment and thus the therapeutic space. The work-
ing alliance, however, remains fragile because of transference distortions
due to pathological self- and object representations.

The interactional stance is defined in terms of such issues as the thera-

pist’s activity level; the role and responsibilities assigned to patient and
therapist, respectively; and the place of therapy on spectra such as egali-
tarian versus authoritarian, directive versus nondirective, and exploratory
versus didactic (Gold, 1993). Cognitive therapy and expressive psychody-
namic therapy stand in stark contrast when compared along the cited di-
mensions. Beck and Freeman’s cognitive approach is directive, didactic,
and more egalitarian, whereas Kernberg’s approach is nondirective, ex-
ploratory, and more authoritarian. The psychodynamic notion of neutrality
is central to Kernberg’s and Beck and Freeman’s opposing views. Cognitive
therapy has no comparable concept, and the therapist is thus consequently
free to act and direct within the framework of treatment. Kernberg (1984b)
views neutrality as an indispensable prerequisite for interpretive work.
Deviation from this stance is seen as bedeviling the development of trans-
ference. Westen (1997), however, cogently argued that neutrality and ab-
stinence are not the prime features of the analytic situation that promote
transference reactions. Westen (1997) reasoned that it is the analytic con-
text’s tendency to elicit reactions to authority, attachment, and intimacy
that afford the therapist a window into patients’ inner worlds and relation-
ships. Integration thus necessitates an interactional stance that allows for a
synthesis of the directive–nondirective and exploratory–didactic polarities.
Sullivan’s (1955) notion of participant observation proves to be well suited
for such synthesis. According to Sullivan, the therapist “has an inescapable,
inextricable involvement in all that goes on” (p. 19) in the therapeutic
interaction. Sullivan, however, believed that the therapist could be mindful
of his or her participation in the therapeutic interaction. The therapist’s
stance is thus a dialectical oscillation between engagement and reflexivity.
Gold (1993) distinguished between two related ways of conceptualizing

Louw and Straker

206

and working with interactional data: the intrapsychic–transferential mode
and the interpersonal–characterological mode. From an intrapsychic–
transferential perspective, therapeutic interaction is determined by the in-
ternalized needs, wishes, conflicts, self- and object representations, and
defenses of the patient. This mode thus assumes primacy of the internal
and archaic in terms of determinacy of therapeutic interaction, with only
peripheral significance attached to the characteristics and behavior of the
therapist. The interpersonal–characterological mode construes therapeutic
interaction as embodying in vivo the patient’s characteristic and habitual
way of engaging with other people. The patient’s creation of so-called
“opportunity structures,” and engagement of “accomplices” (Wachtel,
1993), which maintain psychopathology, inevitably enter the therapeutic
interaction. The proposed integrative model of borderline pathology views
the continuous re-creation of the past in the present and the effects of such
recreation as mutually reinforcing. A synthesis of the dual modes is thus
called for to encompass this circular notion. The patient needs to gain
awareness and insight into his or her relation to the person of the therapist
“incorrectly” as a significant other from the past, as well as into how he or
she creates interpersonal situations that repeat archaic patterns in a self-
fulfilling manner. The conceptualization and clinical use of countertrans-
ference is of cardinal importance given the relational stance of participant
observation and the circular view regarding the maintenance of maladap-
tive schemas. Countertransference in this sense refers to the totality of the
therapist’s responsiveness in relation to the patient. Countertransference
evoked in the therapist is essential to the understanding of the patient’s
intrapsychic and interpersonal dynamics.

Within the relational context sketched in the foregoing discussion the

therapist pursues the goal of modifying and restructuring the borderline
patient’s maladaptive schemas by means of exploration and insight. The
locus and method of intervention are guided by the three-tier model of
Stricker and Gold (1988). The three tiers can be used to classify, assess, and
locate various aspects of personality pathology. Overt behavior problems
are located in Tier 1, maladaptive and abnormal conscious cognitions and
emotions are located in Tier 2, and unconscious motives, conflicts, and
schemas are located in Tier 3. The overt behavior of the patient, either as
observed or as experienced in countertransference, and as constructed by
means of the patient’s in-session narratives about his or her relationships,
is explored so that he or she can become aware of repetitive behavior
patterns as well as the interpersonal impact of these patterns. A further aim
of exploration is insight into the motives for behavior. The psychodynamic
techniques of confrontation and interpretation are used to further these
aims. The conscious cognition or emotions of the patient are focused on by
means of cognitive techniques and confrontation. Specific cognitive inter-

Borderline Pathology: An Integration

207

ventions, such as the labeling of cognitive distortions, thought catching, and
diarizing are used to expand patients’ awareness of their conscious cogni-
tive and emotional functioning. An advantage of these cognitive methods
is their “portability” from therapy to the patient’s life. The in- and out-
session focus on conscious cognition and affect play a major role in the
development of reflective self-functioning. More structured involvement of
patients in the treatment process outside the therapeutic hour also has
other potential benefits, such as strengthening of the working alliance by
means of explicit elicitation of collaboration. Self-monitoring forms may
fulfill a transitional function (Winnicott, 1953/1982) that can be especially
useful with regard to the feelings of emptiness and aloneness that border-
line patients frequently experience (Adler, 1993; Gunderson, 1996). Un-
conscious schemas–representations and conflicts are explored and changed
by means of the psychodynamic interventions of confrontation, clarifica-
tion, and interpretation. The guidelines spelled out by Kernberg (see the
Kernberg’s Psychodynamic Therapy section) are adhered to with regard to
interpretation.

Because of the predominantly conceptual focus of this article, discus-

sion of the technical blending has been brief and schematic. Readers are
referred to the literature (e.g., Lockwood, 1992; Westen, 1991) for more
detailed accounts of possible integration of cognitive and psychodynamic
interventions.

CASE ILLUSTRATION

We briefly illustrate the proposed integration with a brief case study.

Before we proceed with a discussion of the chosen case, it is important to
point out that the seamlessness of integration on paper belies clinical re-
ality. As Wachtel (1991) so aptly stated, “Eclecticism in practice and inte-
gration in aspiration is an accurate description of what most of us in the
integrative movement do most of the time” (p. 44). The reality of flesh-
and-blood encounters with borderline patients necessitates clinical acumen
and flexibility that draw on experience and ideas that are not always inte-
grated with our stated models or frameworks. The aim of this case study is
to illustrate the clinical understanding of material afforded by the integra-
tion. The centrality of interpersonal schemas and of splitting or dichoto-
mous thinking in understanding the patient and the therapeutic process are
highlighted. The patient’s schemas are inferred by the therapist when they
are activated in the here-and-now therapeutic interaction as well as from
the patient’s narratives about his or her interactions in the world. Recog-
nition and understanding of splitting and other cognitive and defensive
operations is gained in a similar fashion. Intervention per se is guided by

Louw and Straker

208

three overarching treatment strategies: (a) labeling activated schemas, (b)
labeling cognitive and defensive operations, and (c) describing the motive
or motives for the activation of particular schemas and cognitive and de-
fense mechanisms.

James,

2

a 30-year old extremely obese man, sought therapy because of

his obsessive infatuation with a male colleague at his place of work. More
specifically, James feared that his sexual fantasies would spill over into real
life. His distress with regard to his “rupturing” fantasy had escalated to a
point where suicidal ideation was present. Relatedly, James voiced im-
mense anguish with regard to the duplicitous nature of his sex life. He
worried about the asexual nature of his marital relationship and feared that
his wife would abandon him if the marriage were not consummated. On the
other hand, he felt compelled to seek anonymous sexual encounters with
men. James worded his expectation with regard to treatment as: “getting
out of the gay lifestyle and all the problems that go with it.” In this regard
it also emerged that James viewed his marriage (his first and only commit-
ted heterosexual relationship) as a vehicle to heterosexuality.

A DSM–IV (American Psychiatric Association, 1994) diagnosis of bor-

derline personality disorder was arrived at after three sessions of clinical
interviewing and history taking. James met the following diagnostic crite-
ria.

1. Unstable and intense relationships. James had a history of stormy

relationships characterized by vacillation between idealization and
devaluation.

2. Identity disturbance. James experienced major conflict and confu-

sion with regard to his sexual identity. He was attracted to men, but
he found his homosexual inclinations abhorrent. He perpetually
struggled with “becoming straight and normal or giving in to the gay
lifestyle.” Issues with regard to identity, however, run much deeper
than sexual orientation. James did have an integrated and coherent
image of himself more generally.

3. Impulsivity that is potentially self-damaging. James intermittently

binged on food, but no purging was involved. James also frequently
engaged in unprotected anonymous sex.

4. Suicidal and self-mutilating behavior. James had a history of para-

suicide attempts, involving wrist slashing and overdosing.

5. Affective instability. James was emotionally labile, and experienced

difficulty with modulating his moods. His “rollercoaster” moods
often fluctuated among anxiety, dysthymia, and short-lived elation.

2

This is a pseudonym for a patient treated by Francois Louw. The patient granted

permission for use of the material.

Borderline Pathology: An Integration

209

6. Inappropriate, intense, and uncontrolled anger. James was prone to

intense and disproportionate anger. His history was checkered with
uncontrolled outbursts of anger, exemplified by him striking a co-
worker, which caused him to lose his job.

7. Chronic feelings of emptiness. James often experienced his life as

barren, boring, and without excitement.

The therapist initially contracted with James for once-weekly psycho-

therapy sessions. By the third session, it was apparent that one session a
week was inadequate to foster a holding environment. Although James
voiced a sense of relief when an extra session per week was proposed, the
therapist also sensed trepidation on James’s part with regard to this sug-
gestion. Exploration revealed that James feared emotional intimacy, which
for him engendered dependency and vulnerability and, ultimately, the de-
spair of abandonment. James’s fear of emotional involvement reflected the
core beliefs he held about himself and the world: (a) “I am powerless and
vulnerable,” (b) “I am inherently unacceptable,” and (c) “The world is
dangerous and malevolent” (Beck & Freeman, 1990). The interpersonal
schemas that shaped James’s relational experiences were not yet clear to
the therapist.

As the therapeutic process unfolded, James’s difficulties with intimacy

came to be reflected predominantly in fears pertaining to his sexuality.
James questioned whether the therapist could tolerate his homosexual
inclinations. He voiced a fear that the therapist would be repelled by his
attraction to men. Through exploration of the interactional dimensions of
James’s fear of rejection, the therapist inferred the operation of a schema
that can be described as follows: a judgmental, rejecting other relating to an
approval-seeking self, with fear (of rejection) as the linking affect. Informed
by the psychodynamic notion of technical neutrality, the therapist adopted
a nonjudgmental stance with regard to the conflicts pertaining to James’s
sexuality. The therapist did this to allow James’s attributions and fantasies
concerning the therapist to unfold. The notion of internalized homophobia
due to societal discourse with regard to homosexuality was also important
in understanding this schema (especially in light of James’s embeddedness
in a conservative religious community). As therapy progressed and James
sensed that the therapist did not find his homosexuality repugnant, he
began to discuss the vicissitudes of his homosexual experiences. James had
little trouble recalling the thoughts and feelings pertaining to these expe-
riences. When access to material was difficult, the use of visualization and
imagery to aid recall proved useful. This entailed James imagining himself,
with his eyes closed, in a particular situation. Exploration over time re-
vealed the following pattern. Typically, James’s anonymous sexual encoun-
ters were preceded by a period during which he was in denial about his

Louw and Straker

210

homosexual desires. During these periods James prided himself as exem-
plifying a saintly and heroic combatant who had overcome the perverse
and sinful (James’s religious convictions condemn homosexuality as a sin).
Operation of the following schema was inferred from James’s recounting of
the preceding—a victorious self relating to an accepting other, with joyous
pride as the linking affect. Eventually such denial broke down to faltering
suppression. When attempts at suppression wore thin, James becomes en-
gaged in an anguishing battle between what he referred to as his “evil and
perverse needs” and his “need to be normal.” This conflict also tormented
James during his search for a sexual partner and abated only during an
actual sexual encounter and for a brief period thereafter. The short-lived
peace rapidly gave way to angry self-recrimination and intense shame.
James believed that God demanded of him to be heterosexual and that he
was condemned because he defied the Lord’s wishes (schema: a punitive,
rejecting other relating to an unacceptable self, with shame as the linking
affect). James’s causal attributions also were colored by his religious be-
liefs. He vacillated between viewing himself as the helpless victim of evil in
the guise of a powerful sexual force (schema: hapless victim-self relating to
an all-powerful evil other, with helplessness–despair as the linking affect), or
as a sex-crazed fiend who chose vice over virtue in a hostile act of rebellion
against God (schema: perverse, rebellious self relating to a demanding,
judgmental other, with alternating rage and shame as the linking affect).
James’s attributions bred hopeless despair, as, in his words, “Nothing will
ever change; evil is too powerful.” The good-versus-evil ascriptions inher-
ent in the noted schemas and attendant beliefs reflected the pervasive
operation of dichotomous thinking or splitting.

As noted, the therapist adopted a stance that would allow the emer-

gence of James’s attributions and fantasies concerning him. At some point
during the first few weeks of therapy, James started to greet the therapist
with a handshake at the end of the session. His manner in doing so was coy
and somewhat sheepish. The therapist was left vaguely puzzled by James’s
manner with regard to these goodbyes. Soon the therapist’s countertrans-
ference response to the end-of-session handshake included an association
of lovers who bid each other farewell. This led the therapist to hypothesize
that an erotic transference might be developing. In the 17th session of
treatment James shared with the therapist what he referred to as “any gay
man’s ultimate fantasy.” This fantasy entailed seducing a heterosexual man
and initiating him into the pleasures of gay sex. In the context of discussing
this fantasy, the cognitive distortion of overgeneralization was inferred
from James’s comment that “All straight men have a weakness.” Given
that James had, on numerous prior occasions, directly or indirectly referred
to the therapist as a heterosexual male, the possible transference commu-
nication in this comment was also noted. The therapist intervened by clari-

Borderline Pathology: An Integration

211

fying whether he was included in the generalization. James became flus-
tered and said that he preferred not to discuss the matter. After a brief
period of uncomfortable silence James stated that he felt that he was being
put on the spot. Following an empathic comment focused on James’s ex-
treme discomfort, James said that he viewed the therapist as unattainable.
On clarification, James revealed that he had had sexual fantasies involving
the therapist. He rapidly added that he was not really attracted to the
therapist, because the therapist was not his type; he preferred older men.
After a brief silence, he added that he had never experienced both emo-
tional and sexual intimacy in the same relationship. The therapist explored
this comment in the context of its associative chain and hence for its pos-
sible transferential commentary. James noted that all his former counselors
got “switched off” when he developed an attraction to them and that this
was “the downfall to many of [his] attempts at salvation.” For the remain-
der of the session James spoke about his obsession with Chris (his col-
league at work) and how it had intensified. He conveyed with glee how
uncomfortable and squeamish Chris became when lustful glances were cast
in his direction. James related that he was absolutely certain that he would
eventually succeed in seducing Chris—if James continued to pursue him,
Chris would eventually succumb. James’s fantasy of omnipotence makes
for a reversal of his impotent-victim status. He becomes the all-powerful
seducer and the other the helpless victim (schema: omnipotent seductive
self relating to a feeble, helpless other, with sadistic pleasure as the linking
affect). James’s comments about Chris were understood in transferential
terms but were not directly explored as such by the time the session ended.
In the next session James opened by stating that he had had an emotionally
trying time since the previous session. Clarification led to James talking
about his attraction toward the therapist and the conflict that it evoked in
him. It also came to light that James’s attraction to the therapist was purely
physical in nature. He saw the therapist as a body that could gratify his
sexual wishes and desires. This contrasted with previous comments in
which James had lauded the therapist for being understanding and accept-
ing. The split between sexual and emotional intimacy that was alluded to in
the previous session was now clearer. Because of dichotomous thinking, or
splitting, the realm of the intimate was cleaved into two mutually exclusive
and separate realms: the emotional and the sexual.

Given that the activation of particular schemas and cognitive and de-

fense mechanisms are understood in the context of affect regulation, thera-
peutic work included exploration of motives. In terms of the preceding,
therapeutic exploration was guided by a curiosity with regard to James’s
fear of emotional attachment and the reason (or reasons) why he separated
sexual and emotional intimacy. The following functional understanding
emerged over time. James’s sexual encounters allowed for a closeness,

Louw and Straker

212

which at the same time precluded the dangers inherent in sustained emo-
tional intimacy. Sexualization was thus a vehicle for controlled attachment.
Sex as bonding appeared to be linked historically to James’s relationship
with a neighborhood adult, Greg, which commenced when James was a
14-year-old budding adolescent. Beck and Freeman (1990) noted how
evocative and useful the “reliving” of childhood experiences can be in
therapeutic work with borderline patients. Exploration of past experiences
is also an important feature of psychodynamic therapy. Greg befriended
James and took him hiking and on other outings. James described how
Greg “became like a father figure” to him. Some months into the relation-
ship, Greg one day coldly demanded anal intercourse, to which James
acquiesced. James avoided Greg for some time after this incident but even-
tually resumed the relationship. James described being like a sex slave to
Greg. Although Greg sexually exploited and abused James, James also felt
cared for and affirmed in relation to him. James maintained an intense and
acrimonious relationship with Greg until the age of 20. In describing the
relationship, James alternately painted Greg as the hated, evil person who
exploited him and robbed him of his heterosexuality or as a benevolent
father figure who took an interest in him and his life. Cleavage of the sexual
and the emotional dimensions thus seemed linked to James’s relational
experience with Greg. By sexualizing relationships, James defends against
the fears he harbors with regard to intimacy. He protects himself from
getting too close and thus becoming dependent and vulnerable. In the
therapeutic relationship, for example, James’s sexual fantasies involving
the therapist increased dramatically when he perceived himself as getting
too close or when he felt vulnerable in relation to the therapist. As noted
earlier, one of James’s fantasies entailed him being the omnipotent seducer
who seduces a helpless other. The psychodynamic notion of identification
with the aggressor proved useful in understanding this fantasy and related
seductive behavior. The reversal of perpetrator and victim status enhanced
James’s self-esteem; a helpless, vulnerable self became a powerful, irresist-
ible self. James’s anonymous sexual encounters were also related to the
management of self-experience. Choosing and being chosen in the context
of seeking a sexual partner left James feeling a sense of worth. Such shoring
up of his fragile self was, however, fleeting, and the ritual of seeking and
being sought needed to be repeated frequently. Motivational dynamics,
such as the ones described so far, are not viewed as functionally exclusive.
Motives, more often than not, operate in concert, as illustrated in the
following example. In the work context, James’s dominant experience was
one of being the brunt of ridicule and abuse. His victim status was fueled
by a combination of excessive personalization and selective abstraction, as
well as projection of his own hostility. James’s omnipotent seductive fan-
tasy vis-a`-vis his colleague in this context can be understood as simulta-

Borderline Pathology: An Integration

213

neously fulfilling self-management and aggressive functions. Changed self-
experience (enhanced self-esteem) is coupled with aggression in the form
of sadistic enjoyment of his colleague’s perceived discomfort.

As can be gleaned from the discussion so far, much of the therapeutic

work focused on the vicissitudes of James’s relationships. Given that
James’s binge eating rarely entered the therapeutic discourse, the therapist
suggested that James keep a diary to chronicle his binge eating. Diarizing
revealed that James usually binged during periods of abstinence from sex,
particularly when the battle between seeking sexual gratification and self-
restraint became unbearable. During a bingeing episode, James felt
soothed by his consumption of food. In addition, the thought of eating as
much as he desired imbued bingeing with a sense of satisfaction. In the
wake of a bingeing episode James would feel physically ill and miserable
and would ruminate about his lack of self-control. In-therapy exploration
furthermore revealed how obesity and a fantasy associated with it acted as
a defense. James believed that his obesity was a turn-off and that it frus-
trated him in seeking sexual partners. He thus saw his obesity as a deterrent
to his sexual exploits. It also came to light that James felt safe from AIDS
when he saw himself as overweight. He associated AIDS with emaciation
and saw obesity as a sign that he was not infected with HIV and had not
developed AIDS. This fantasy was also related to James’s failure to prac-
tice safe sex. James never wore a condom when he sought anonymous
sexual encounters. He in fact described his sexual encounters as “playing
Russian roulette.” If he did contract HIV, it would be his just deserts for
leading such a “decadent” lifestyle and, furthermore, death would put an
end to his baneful existence. The use of diary keeping to ascertain James’s
conscious thoughts and emotions, in the context of binge eating, thus paved
the way for exploration of important related conscious ideas, fantasies, and
feelings.

CONCLUSION

The integrative conceptual framework proposed in this article should

be regarded as a clinical map for the understanding and treatment of
borderline pathology. Disorder-specific conceptualizations and treatment
proposals run the risk of resulting in “paint-by-numbers” psychotherapy
(Silverman, 1996), which denies the uniqueness of each individual who is
afflicted with borderline pathology and the complexity of the therapeutic
process. Disregarding this caveat, the proposed integration provides a
broadened understanding of borderline pathology that affords therapists
more scope and flexibility in their dealings with a difficult patient popula-
tion. In this endeavor the proposed integration joins CAT and DBT. Be-

Louw and Straker

214

cause of a common source in object relations, there is much commonality
between CAT and the proposed schema-focused conceptualization. In
practice, however, the two approaches differ considerably. Whereas CAT
has an explanatory focus, the proposed integration has a strong relation-
ship focus. The use of sequential diagrammatic reformulation can, how-
ever, be used as a useful adjunct to the proposed relationship focus. Be-
cause DBT originates from a synthesis of mainly behavioral and cognitive
notions, within a dialectical framework, it differs more sharply from the
integrative approach offered here. To the extent that DBT promotes cog-
nitive change and mindfulness and focuses on the contingencies operating
in the therapeutic relationship, it does, however, share features that can
usefully be incorporated in the proposed integration. In closing, we should
note that the outlined integration is a framework that is open to assimilat-
ing ideas from other approaches and that its utility lies in fostering further
crosspollination.

REFERENCES

Adler, G. (1993). The psychotherapy of core borderline psychopathology. American Journal

of Psychotherapy, 47, 194–205.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disor-

ders (4th ed.). Washington, DC: Author.

Arkowitz, H. (1989). The role of theory in psychotherapy integration. Journal of Integrative

and Eclectic Psychotherapy, 8, 8–16.

Beard, H., Marlowe, M., & Ryle, A. (1990). The management and treatment of personality-

disordered patients: The use of sequential diagrammatic reformulation. British Journal
of Psychiatry, 156, 541–545.

Beck, A. T., & Emery, G. (1985). Anxiety disorders and phobias: A cognitive perspective. New

York: Basic Books.

Beck, A. T., & Freeman, A. (1990). Cognitive therapy of personality disorders. New York:

Guilford Press.

Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression.

New York: Guilford Press.

Bowlby, J. (1988). A secure base. London: Routledge.
Clarkin, J. F., & Kernberg, O. F. (1993). Developmental factors in borderline personality

disorder and borderline personality organization. In J. Paris (Ed.), Borderline personality
disorder: Etiology and treatment (pp. 161–184). Washington, DC: American Psychiatric
Press.

Giovachinni, P. L. (1993). Borderline patients, the psychosomatic focus, and the therapeutic

process. Northvale, NJ: Jason Aronson.

Gold, J. R. (1993). The therapeutic interaction in psychotherapy integration. In G. Stricker &

J. R. Gold (Eds.), Comprehensive handbook of psychotherapy integration (pp. 525–531).
New York: Plenum.

Gold, J. R., & Stricker, G. (1993). Psychotherapy integration with character disorders. In G.

Stricker & J. R. Gold (Eds.), Comprehensive handbook of psychotherapy integration (pp.
323–336). New York: Plenum.

Goldstein, W. N. (1993). Psychotherapy with borderline patients: An introduction. American

Journal of Psychotherapy, 47, 172–183.

Goldstein, W. N. (1995). The borderline patient: Update on the diagnosis, theory, and treat-

Borderline Pathology: An Integration

215

ment from a psychodynamic perspective. American Journal of Psychotherapy, 49,
317–337.

Greenberg, J., & Mitchell, S. (1983). Object relations in psychoanalytic theory, Cambridge,

MA: Harvard University Press.

Gunderson, J. G. (1996). The borderline patient’s intolerance of aloneness: Insecure attach-

ments and therapist availability. American Journal of Psychiatry, 153, 752–758.

Hellkamp, D. T. (1993). Severe mental disorders. In G. Stricker & J. R. Gold (Eds.), Com-

prehensive handbook of psychotherapy integration (pp. 385–398). New York: Plenum.

Horowitz, M. (1988). Introduction to psychodynamics: A new synthesis. London: Routledge.
Kernberg, O. F. (1975). Borderline conditions and pathological narcissism. Northvale, NJ:

Jason Aronson.

Kernberg, O. F. (1984a). Object relations theory and clinical psychoanalysis. Northvale, NJ:

Jason Aronson.

Kernberg, O. F. (1984b). Severe personality disorders: Psychotherapeutic strategies. New Ha-

ven, CT: Yale University Press.

Kernberg, O. F. (1987). An ego psychology–object relations theory approach to the transfer-

ence. Psychoanalytic Quarterly, 56, 197–221.

Kernberg, O. F. (1992). Aggression in personality disorders and perversions. New Haven, CT:

Yale University Press.

Kernberg, O. F. (1993). The psychotherapeutic treatment of borderline patients. In J. Paris

(Ed.), Borderline personality disorder: Etiology and treatment (pp. 261–284). Washing-
ton, DC: American Psychiatric Press.

Kernberg, O. F., Selzer, M. A., Koenigsberg, H. D., Carr, A. C., & Appelbaum, A. H. (1989).

Psychodynamic psychotherapy of borderline patients. New York: Basic Books.

Linehan, M. M. (1993). Cognitive–behavioral treatment of borderline personality disorder.

New York: Guilford Press.

Lockwood, G. (1992). Psychoanalysis and the cognitive therapy of personality disorders.

Journal of Cognitive Psychotherapy, 6, 25–42.

Masterson, J. (1978). New perspectives on psychotherapy of the borderline adult. New York:

Brunner/Mazel.

Millon, T. (1993). The borderline personality disorder: A psychosocial epidemic. In J. Paris

(Ed.), Borderline personality disorder: Etiology and treatment (pp. 197–210). Washing-
ton, DC: American Psychiatric Press.

Norcross, J. C., & Arkowitz, H. (1992). The evolution and current status of psychotherapy

integration. In W. Dryden (Ed.), Integrative and eclectic therapy: A handbook (pp. 1–40).
Buckingham, England: Open University Press.

Oldham, J. M. (1991). Borderline personality disorder: An introduction. Hospital and Com-

munity Psychiatry, 42, 1014.

Perlow, M. (1995). Understanding mental objects. London: Routledge.
Pollack, W. S. (1990). Psychotherapy (of borderline personality disorder). In A. S. Bellack &

M. Hersen (Eds.), Handbook for comparative treatment of adult disorders (pp. 393–419).
New York: Wiley.

Rinsley, D. (1982). Borderline and other self disorders: A developmental and object relations

perspective. Northvale, NJ: Jason Aronson.

Ryle, A., & Beard, H. (1993). The integrative effect of reformulation: Cognitive analytic

therapy with a patient with borderline personality disorder. British Journal of Medical
Psychology, 66, 249–258.

Ryle, A., & Cowmeadow, P. (1992). Cognitive–analytic therapy (CAT). In W. Dryden (Ed.),

Integrative and eclectic therapy: A handbook (pp. 84–108). Buckingham, England: Open
University Press.

Ryle, A., & Marlowe, M. J. (1995). Cognitive analytic therapy of borderline personality

disorder: Theory and practice and the clinical and research uses of the Self States
Sequential Diagram. International Journal of Short-Term Psychotherapy, 10, 21–34.

Sable, P. (1997). Attachment, detachment and borderline personality disorder. Psycho-

therapy: Theory/Research/Practice/Training, 34, 171–181.

Safran, J. D. (1990). Towards a refinement of cognitive therapy in light of interpersonal

theory: I. Theory. Clinical Psychology Review, 10, 87–105.

Louw and Straker

216

Shea, M. T. (1991). Standardized approaches to individual psychotherapy of patients with

borderline personality disorder. Hospital and Community Psychiatry, 42, 1034–1038.

Silverman, W. H. (1996). Cookbooks, manuals, and paint-by-numbers psychotherapy in the

90’s. Psychotherapy, 33, 207–215.

Skodol, A. E., & Oldham, J. M. (1991). Assessment and diagnosis of borderline personality

disorder. Hospital and Community Psychiatry, 42, 1021–1028.

Stricker, G., & Gold, J. R. (1988). A psychodynamic approach to the personality disorders.

Journal of Personality Disorders, 2, 350–359.

Sullivan, H. S. (1955). The psychiatric interview. London: Tavistock.
Summers, F. (1994). Object relations theories and psychopathology. Hillsdale, NJ: Analytic

Press.

Swenson, C. (1989). Kernberg and Linehan: Two approaches to the borderline patient. Jour-

nal of Personality Disorders, 3, 26–35.

Turner, R. M. (1993). Dynamic cognitive behavior therapy. In T. R. Giles (Ed.), Handbook

of effective psychotherapy (pp. 437–454). New York: Plenum.

Tutek, D. A., & Linehan, M. M. (1993). Comparative treatments for borderline personality

disorder. In T. R. Giles (Ed.), Handbook of effective psychotherapy (pp. 355–378). New
York: Plenum.

Wachtel, P. L. (1991). From eclecticism to synthesis: Toward a more seamless psychothera-

peutic integration. Journal of Psychotherapy Integration, 1, 43–54.

Wachtel, P. L. (1993). Therapeutic communication: Principles and effective practice. New

York: Guilford Press.

Westen, D. (1988). Transference and information processing. Clinical Psychology Review, 8,

161–179.

Westen, D. (1991). Cognitive–behavioral interventions in the psychoanalytic psychotherapy

of borderline personality disorder. Clinical Psychology Review, 11, 211–230.

Westen, D. (1997). Towards a clinically and empirically sound theory of motivation. Interna-

tional Journal of Psycho-Analysis, 78, 521–548.

Winnicott, D. W. (1982). Transitional objects and transitional phenomena. In Playing and

reality (pp. 1–25). London: Routledge. (Original work published 1953)

Winnicott, D. W. (1990). The theory of the parent–infant relationship. In The maturational

processes and the facilitating environment (pp. 37–55). London: Karnac. (Original work
published 1960)

Yeomans, F. E., Selzer, M. A., & Clarkin, J. F. (1992). Treating the borderline patient: A

contract-based approach. New York: Basic Books.

Young, J. E. (1990). Cognitive therapy for personality disorders: A schema-focused approach.

Sarsota, FL: Professional Research Exchange.

Zeanah, C. H., Anders, T. F., Seifer, R., & Stern, D. N. (1989). Implications of research on

infant development for psychodynamic theory and practice. Journal of the Academy of
Child and Adolescent Psychiatry, 28, 657–668.

Borderline Pathology: An Integration

217