Early analysts recognized three levels of functioning: normal, neurotic, and psychotic. Although everyone was expected to function at one of these levels, it eventually became apparent that some subjects did not fit neatly into this threefold scheme. Given a libido withdrawn totally into the self, psychotic subjects, according to Freud, would not respond to psychoanalysis; he considered their degree of withdrawal inaccessible given the usual analytic tools. His new science was a science of the whole mind, but as a therapy, it could be applied only to neurotics. Any subject who showed good reality contact was not psychotic by definition and, therefore, was analyzable, at least in principle.
Eventually, however, early analysts began to notice a more troubled group of subjects who could not be called psychotic, but who failed to benefit from standard psychoanalytic therapy. Georgia and Jenny would probably fall into this group. Because reality testing was preserved, their very existence required explanation. Seeking to define such individuals, Stern (1938) used the phrase borderline group of neuroses, regarded as neurotic but resistant to the couch. He listed and discussed 10 characteristics for these first borderlines, including classics such as a quickness to anger; depression or anxiety in response to interpretive probes about self-esteem; the use of projection to attribute internal anger to hostile sources in the environment; and "difficulties in reality testing," that is, nonpsychotic deficits in judgment and empathic accuracy. With Stern's contribution, the term borderline was on its way to becoming an informal fixture in psychodynamic discussions.
During the 1940s and 1950s, many writers made contributions that would eventually be incorporated into the contemporary view (Stone, 1986). Just after World War II, the term borderline began to appear in the formal analytic literature. Schmideberg (1947, 1959, p. 399) presaged themes still relevant today. The borderline was "not just quantitatively halfway between the neuroses and psychoses," she stated. Instead, "the blending and combination of these modes of reaction produce something qualitatively different." Borderlines are "stable in [their] instability," she argued (or, as Elizabeth quotes her father, Georgia is a troubled soul who "can't get her life together"). All borderlines experience "disturbances affecting almost every area of their personality and life, in particular, personal relations and depth of feeling" (1959, p. 399). Other writers influenced the borderline construct without using the term. For example, Erikson (1956) contributed indirectly to the borderline construct through his discussions of ego identity and early identity formation, an antecedent of the DSM criterion of identity disturbance.
The early 1950s represent an important turning point in the history of the construct (Stone, 1986). Previously, the term borderline was associated with the schizophrenic syndromes. With Knight (1953), however, the construct took a much more analytic cast, allowing it to advance substantially in popularity and take on a more contemporary cast. Knight brought to the foreground the importance of ego weakness and its connection to psychotic episodes, stating that the "normal ego functions of secondary-process thinking, integration, realistic planning, adaptation to the environment, maintenance of object relations, and defenses against primitive unconscious impulses are severely weakened" (p. 165). The ego of the borderline, according to Knight, is "laboring badly" under the stress of traumatic events and pathological relationships. "Integration, concept formation, judgment, realistic planning, and defending against eruption into conscious thinking of id impulses and their fantasy elaborations" are severely impaired, "while other ego functions, such as conventional (but superficial) adaptation to the environment and superficial maintenance of object relationships may exhibit varying degrees of intactness" (Knight, 1953, p. 165). As many writers have noted, borderlines often look much more adaptive or competent than they really are. Georgia, for example, makes good enough of a first impression to get hired but then can't keep her job.
The most important contribution to contemporary psychodynamic conceptions, however, is Kernberg's (1967) idea of levels of organization in personality. Unlike the idea of borderline states or conditions, the idea of a borderline level of organization draws attention to a quality of integration of intrapsychic elements that is stable over time, yet falls midway on the continuum from neurosis to psychosis. All the personality disorders, as well as many psychodynamic character types, can be put on this continuum. Conceived as a level of personality organization, the borderline is thus much broader than the borderline personality described by the DSM. For example, Kernberg puts the schizoid personality at a lower level of borderline functioning. Yet, the socially removed style of schizoids is inconsistent with the intense interpersonal need of DSM borderlines, specifically, their frantic attempts to avoid abandonment, listed as the first diagnostic criterion. What then, does borderline mean as a level of personality organization?
The borderline level is probably best understood in contrast to normality, which helps us better determine where the borderline is deficient. As Kernberg (1994) writes, normals exhibit an integrated concept of self and others, called ego identity, which not only gives coherence to the self but also provides a foundation for a healthy self-esteem and a sense of self-identity that endures across time and situation and gives direction to life goals. Most of us know who we are, our likes and dislikes, our core values, ways we are similar to and yet different from others, and where we're going in life. Moreover, a well-integrated ego identity provides ego strength, the ability to resist pressure or stress, just as a ballast allows a ship to weather a storm. Additionally, Kernberg notes that normals possess a mature, internalized value system, the superego, permitting adaptive adult capacities such as personal responsibility and appropriate self-criticism.
The neurotic level is somewhat similar to normality. A well-developed ego identity supports a deep capacity for interpersonal intimacy and sexual love, with sufficient ego strength to tolerate anxiety, control impulses, and function effectively and creatively in work (Kernberg, 1994). Although neurosis is a form of psychopathology, it is by no means disabling. The distinction between normal and neurotic, according to Kernberg, lies primarily in the presence of unconscious feelings of sexual guilt. Pathologies of aggression are reserved for lower levels of personality organization. In Kernberg's schema, the obsessive-compulsive, depressive-masochistic, and hysterical personalities all function at the neurotic level. Given their intense anger and hostility, neither Georgia nor Jenny can be said to function at this level.
In contrast to the ego identity and cohesiveness of the normal and the neurotic, the psychotic level is characterized by a nearly complete fragmentation or diffusion of identity. Nearly everything we usually think of as personality is lost at this level. The capacity to test reality, to distinguish between self and others, for example, ebbs and flows. The constructs that represent the self and particular persons in the individual's life are not understood as integrated wholes. Instead, aspects of the self may temporarily fuse with aspects of others, resulting in a kind of object-relational kaleidoscope that has little relation to external reality. By definition, the psychotic level is characterized by an absence of ego strength and, therefore, an inability to inhibit even minor impulses from actively intruding into ongoing affairs. Moreover, without an integrated ego, the individual cannot develop a coherent plan of action that allows goals to be profitably pursued, much less balance the inhibitions of the superego with the random eruption of id impulses, which might be triggered by internal stimulation from his or her own stream of consciousness or through actual environmental events. The psychotic level seems too severe for Georgia and Jenny, who exhibit reality contact and a nonfragmented, somewhat fluid, sense of identity.
The borderline level of organization exists between the neurotic and psychotic and, as such, has characteristics of both. Kernberg (1994) refers to a triad of identity diffusion, primitive emotional displays of great intensity, and problems with impulse control. Like the neurotic, the borderline retains the capacity to test reality. In fact, many borderlines function at a high level indistinguishable from neurosis much of the time. However, the neurotic is capable of an array of mature defenses, whereas those at the borderline level are comparatively primitive variants of "splitting." Good and bad images of objects are actively separated. Thoughts such as, "Mommy has some good things about her and some bad things about her," are simply not possible.
In Kernberg's view, these good and bad images form two separate identification systems, either of which may be projected onto the self or outside world. Thus, subjects may switch rapidly between idealizing others (a projection of the good image) and completely devaluing them (a projection of the bad image). Thus, Georgia can't seem to make up her mind whether to blame Elizabeth or smother her with love, and Jenny sometimes adores Vera and sometimes hates her.
Understanding the relationship of splitting to the wider constellation of borderline symptoms requires an understanding of its role in normal development. According to Kernberg, borderlines are fixated at Mahler's separation-individuation phase (Mahler et al., 1975), specifically, in the rapprochement subphase, which runs from about 16 to 30 months of age. But the names and timings are technically unimportant. What is important is that separation-individuation precedes object constancy; the future borderline cannot distinguish between self and other before an image of the nurturing figure as a permanent presence is internalized. As the saying goes: Out of sight, out of mind. The fear is that when Mommy leaves, she will be gone forever, never to return.
The adult borderline persistently reexperiences this same overwhelming separation anxiety. Thus, borderlines often seem dependent and clingy and are unable to tolerate being alone for long periods of time. Like Mommy, a spouse or lover might never be seen again. In Jenny, this primitive separation anxiety is recaptured in her relationship with her father. Jenny has no internalized, stable image of her father, therefore, no appreciation of a love that might endure across time and circumstance. As such, she cannot comprehend his ability to simultaneously possess two different kinds of love: the love for Vera and the love for his daughter. Accordingly, Vera is viewed as a substitution rather than an addition to the family, thereby supplanting Jenny. Thus, Vera is hated and hated totally, the extreme of devaluation. Lack of stable internalized images of attachment figures creates considerable anxiety and the concomitant possibility of regression to more primitive ego states, especially for borderlines facing developmental milestones that require separation, such as leaving home for college or enduring a period of occupational training away from their spouse.
What causes a failure of object constancy? According to Kernberg, borderlines have a surplus of aggression, either constitutionally or because of frustration of their early needs by insensitive caretakers. As such, the integration of "identification systems of opposite quality" (Kernberg, 1985a, p. 69) becomes extremely threatening. Were integration to occur, the intensity of rage and hatred directed at the bad image would likely destroy the good image. Even if borderlines were to achieve an integrated conception of their significant others, this image would receive so much rage that the composite would be destroyed or alienated and the good object, the good Mommy, along with it. The situation is not unlike pouring a bucket of black paint and a bucket of white paint together; the black dominates.
As such, borderlines linger in the separation-individuation phase long beyond the development of cognitive abilities that might allow for more sophisticated appraisals, using splitting defensively as a means of keeping good and bad objects apart. Splitting thereby explains the identity diffusion of borderlines and their tendency to suddenly switch from profoundly positive to profoundly negative affects, features that lay the foundation for chaotic relationships, lack of commitment to life goals, lack of insight into the core values that might define personal existence, and an inability to inhibit the expression of strong affects and impulses, including those related to promiscuity and substance abuse. Everything that the ego does as the executive agency of the personality is considerably weakened. As Kernberg (1985a, p. 121) states, such individuals may also exhibit sexual pathologies, and all "evince nonspecific manifestations of ego weakness, that is, lack of anxiety tolerance, of impulsive control, and of sublimatory functioning in terms of an incapacity for consistency, persistence, and creativity in work." All of these characteristics are found in Georgia and Jenny, though in different ways.
Because the borderline, by this argument, defines a level of organization among intrapsychic structures, there is naturally a question of its relation to the character styles of classical psychoanalysis as well as to the DSM personality disorders. Rather than dump all of these into a single level called borderline, Kernberg (1994) provides additional differentiation by subdividing the borderline into two sublevels, one of which has more in common with psychosis and the other with neurosis. The more neuroticlike borderline organization includes the sadomasochistic, cyclothymic, dependent, histrionic, and those narcissistic personalities compensated by grandiosity. The lower level includes the paranoid, hypochondriacal, schizotypal, hypomanic, and antisocial personalities, as well as what Kernberg calls "malignant narcissism." All the personalities at the higher and lower levels express the borderline personality organization but in different ways, depending on the particulars of their character or personality style. In contrast, no stylistic variations exist in the DSM borderline, conceptualized simply as another personality disorder existing alongside the others.
Although Kernberg has been the most influential, other object-relations thinkers have emphasized the related themes of attachment and separation-individuation. Although every child begins life absolutely dependent on caretakers, eventually each child must grow into a separate person. According to Masterson and Rinsley (1975), the growing autonomy of the future borderline challenges the caretaker's desire to maintain closeness. In response, the child develops an intense ambivalence toward the caretaker, usually the mother, sometimes giving in to coercive clinging and sometimes reacting with a negativistic withdrawal. The conflict between wishing to retain the nurturance of the caretaker and developing as a unique individual prevents the child from integrating good and bad images of the mother. Nor can the child, caught in the conflict, consolidate his or her own identity.
Stressing that the caretaker may be borderline as well, Masterson (1972, 1976) sees the mother as encouraging the child to continue his or her symbiotic clinging. At the same time, the mother threatens to withdraw love should the child strive toward autonomy. This sets up a profound fear of abandonment played out across life, in which dependence brings reward and independence is equated with loss of love. The child is thereby caught in a lose-lose ambivalence between assertiveness and abandonment, creating a foundation for classic borderline symptoms such as unstable relationships, a tendency to search for idealized or romanticized fusions, as well as states of emptiness and depression. According to Masterson, then, it is the mother who is ultimately responsible for creating the borderline pathology, not the constitution of the subject.
Still another twist on the object-relations view is represented in Adler (1985), who sees the borderline as suffering from a failure in object constancy caused by insensitive or inadequate mothering. In other words, borderlines fail to internalize a representation of a caretaker that provides reassurance, a "holding-soothing object," to carry them through times when the caretaker cannot be physically present. What Adler calls an insufficiency theory would appear to explain a number of cardinal symptoms. First, without a caretaker with whom to have meaningful, empathic interactions, the future borderline cannot develop a stable sense of self-identity able to withstand stressful times. Regression to a more primitive ego state is, therefore, a perpetual risk. Second, because the developing self cannot organize itself around positive interactions with the caretaker, it is instead left only with feelings of profound emptiness caused by the absence of positive introjects and a chronic dysphoria (reminiscent of the classic analytic view that depression is caused by object loss). Third, the absence of object constancy explains why borderlines frantically avoid abandonment and require the actual physical presence of their significant other. In effect, they are searching for the holding-soothing object that early development would not provide. We know nothing of Jenny's mother, but we do know that Georgia was regarded as the "black sheep" of the family and compared unfavorably to her older sister, now a prominent attorney. According to Adler, Georgia is searching for a holding-soothing object to love and appreciate her.
The conception of the borderline as a discrete diagnostic entity also has its foundations in the psychodynamic tradition. The first systematic empirical study of a borderline sample was undertaken by Grinker, Werble, and Drye (1968). Using cluster analytic methods, these researchers found four groups, unified by several common characteristics, namely, "anger as the main or only affect, defect in affectional relationships, absence of indications of self-identity and depressive loneliness" (p. 176). The so-called core borderline group exhibited "vacillating involvement with others," "overt or acting-out expressions of anger," pervasive depression, and "absence of indications of consistent self identity" (p. 87).
The most highly developed research program, however, has been developed by Gun-derson and colleagues (Gunderson, 1977, 1979; Gunderson, Carpenter, & Strauss, 1975; Gunderson & Singer, 1975). These authors viewed the borderline as a diagnostic entity clearly distinguishable from schizophrenic syndromes and neurotic conditions. Beginning with a thorough review of prior work (Gunderson & Singer, 1975) and opportunities to carry out a series of empirical studies, Gunderson and colleagues developed the
Diagnostic Interview for Borderlines (DIB) "to achieve diagnostic reliability specifically for borderline patients" (Gunderson, Kolb, & Austin, 1981, p. 896) on the basis of Gunderson and Singer's (1975) literature review of borderline conditions.
One of the best ways to study a construct is by examining the content of established instruments. Where the focus is on a single construct, interrelationships among its various content aspects are readily apparent. The DIB has been revised to increase its specificity and to "refine its format, phrasing, and scoring system" (Zanarini et al., 1989). By surveying these topic areas, clinicians quickly gain an appreciation for how traits of the larger personality pattern hang together. The revised DIB uses 97 items, grouped into 22 summary statements, to assess functioning in four broad areas: affect, cognition, impulse action patterns, and interpersonal relationships. As a significant extension, Zanarini (1993) has suggested that the borderline personality might best be considered an "impulse spectrum disorder" rather than a variant of the affective disorder spectrum. The DSM definition of the borderline personality largely represents a synthesis of Kernberg and Gunderson's contributions. The 22 summary statements of the DIB-R are presented in Table 14.1.
Was this article helpful?