3
Developmental Levels
of Personality Organization
This chapter focuses on what many analysts have seen as the maturational issues embedded in a person’s character—the unfinished or impeded business of early psychological development: what Freud called fixation and what later analysts called developmental arrest. In much analytic writing about personality, it has been assumed that the earlier the developmental obstacle, the more disturbed the person. This belief is a great oversimplification and in some ways is simply wrong (see Fischer & Bidell, 1998; Westen, 1990). But for purposes of introducing a way to think about character that can be clinically helpful, I lay out the traditional overview as well as more recent efforts to account for general differences in psychological health and personality structure.
Historically, analysts have conceived of a continuum of overall mental functioning, from more disturbed to healthier. They have explicitly or implicitly construed individual personality as organized at a particular developmental level and structured by the individual’s characteristic defensive style. The first dimension conceptualizes a person’s degree of healthy psychological growth or pathology (psychotic, borderline, neurotic, “normal”); the second identifies his or her type of character (paranoid, depressive, schizoid, etc.).
A close friend of mine, a man with no experience in psychotherapy, who cannot imagine why anyone would go into a field where one spends hour upon hour listening to other people’s problems, was trying to understand my interest in writing this book. “It’s simple for me,” he commented. “I have just two categories for people: (1) nuts and (2) not nuts.” I responded that in psychoanalytic theory, which assumes that everyone is to some degree irrational, we also have two basic attributions: (1) How nuts? and (2) Nuts in what particular way? As I mentioned in Chapter 2, although contemporary analysts conceive the phases through which young children pass in less drive-defined ways than Freud did, many of their theories continue to reflect his conclusion that current psychological preoccupations reflect infantile precursors, and that interactions in our earliest years set up the template for how we later assimilate experience.
Conceptualizing someone’s unmet developmental challenges can help in understanding that person. Interestingly, the same three phases of early psychological organization keep reappearing in psychoanalytic developmental theories: (1) the first year and a half to 2 years (Freud’s oral phase), (2) the period from 18 to 24 months to about 3 years (Freud’s anal phase), and (3) the time between 3 or 4 and about 6 (Freud’s oedipal period). The approximateness of these ages reflects individual differences; the sequence seems to be the same whether a child is precocious or late blooming. Many theorists have discussed these phases, variously emphasizing drive and defense, ego development, or images of self and other that characterize them. Some have stressed behavioral issues of the stages, others have addressed cognition, still others the child’s affective maturation.
Many scholars (e.g., Lyons-Ruth, 1991; D. N. Stern, 2000) have critiqued stage theories in light of infant research, which has illuminated far more competence in early infancy than most developmental models assume and connects difficulties to parental attachment behaviors rather than presumed developmental phases. Analysts of a postmodern bent (e.g., Corbett, 2001; Fairfield, 2001) point out that models of “normal development” contain implicit cultural prescriptions, inevitably contributing to images of an in-group that is fine and an out-group that is not. Despite these limitations, I think that some notion of expectable psychological stages will survive in our conceptual formulations, as there is something that invites clinical empathy in the idea that we all go through a similar process of growth. In the following, I draw mostly on the ideas of Erikson, Mahler, and Fonagy to explicate the developmental aspect of psychoanalytic diagnosis.
It has never been empirically demonstrated that people with a lot of “oral” qualities have more severe degrees of psychopathology than those with central dynamics that earlier analysts would have regarded as either anal or oedipal, even though Freud’s naming of the first three stages of development by these inferred drive concepts has a lot of intuitive appeal and correlates to some degree with type of personality (depressive people at any level of health or pathology tend to manifest orality; the preoccupations of compulsive people are notoriously anal—see Chapter 13—whether or not their compulsivity causes them major problems).
Yet there is substantial clinical commentary (e.g., Volkan, 1995) and increasing empirical research (e.g., Fonagy, Gergely, Jurist, & Target, 2002; L. Silverman, Lachmann, & Milich, 1982), supporting a correlation between, on the one hand, one’s level of ego development and self–other differentiation, and, on the other, the health or pathology of one’s personality. To a certain extent this correlation is definitional and therefore tautological; that is, assessing primitive levels of ego development and object relations is like saying an interviewee is “sick,” whereas seeing someone as obsessive or schizoid is not necessarily assigning pathology. But this way of conceptualizing psychological wellness versus disturbance according to categories from ego psychology and the later relational theories has profound clinical implications across different character types. A brief history of psychoanalytic attempts to make diagnostic distinctions between people based on the extent or “depth” of their difficulties rather than their type of personality follows.
HISTORICAL CONTEXT: DIAGNOSING LEVEL
OF CHARACTER PATHOLOGY
Before the advent of descriptive psychiatry in the 19th century, certain forms of mental disturbance that occurred with any frequency in what was considered the “civilized world” were recognized, and most observers presumably made distinctions between the sane and the insane, much as my nonpsychological friend distinguishes between “nuts” and “not nuts.” Sane people agreed more or less about what constitutes reality; insane people deviated from this consensus.
Men and women with hysterical conditions (which included what today would be diagnosed as posttraumatic problems), phobias, obsessions, compulsions, and nonpsychotic manic and depressive symptoms were understood to have psychological difficulties that fall short of complete insanity. People with hallucinations, delusions, and thought disorders were regarded as insane. People we would today call antisocial were diagnosed with “moral insanity” (Prichard, 1835) but were considered mentally in touch with reality. This rather crude taxonomy survives in the categories of our legal system, which puts emphasis on whether the person accused of a crime was able to assess reality at the time of its commission.
Kraepelinian Diagnosis: Neurosis versus Psychosis
Emil Kraepelin (1856–1926) is usually cited as the father of contemporary diagnostic classification. Kraepelin observed mental patients carefully, with the aim of identifying general syndromes that share common characteristics. In addition, he developed theories about the etiologies of those conditions, at least to the extent of regarding their origins as either exogenous and treatable or endogenous and incurable (Kraepelin, 1913). (Interestingly, he put severe bipolar illness [“manic–depressive psychosis”] in the former category and schizophrenia [“dementia praecox”—believed to be an organic deterioration of the brain] in the latter.) The “lunatic” began to be understood as a person afflicted with one of several possible documented illnesses.
Freud went beyond description and simple levels of deduction into more inferential formulations; his developing theory posited complex epigenetic explanations as preferable to Kraepelin’s basic internal–external versions of causality. Still, Freud tended to view psychopathology by the Kraepelinian categories then available. He would describe a man troubled by obsessions (e.g., his patient the “Wolf Man” [Freud, 1918; Gardiner, 1971]), as having an obsessive–compulsive neurosis. By the end of his career, Freud began to discriminate between an obsessional neurosis in an otherwise nonobsessive person and an obsession that was part of an obsessive–compulsive character. But it was later analysts (e.g., Eissler, 1953; Horner, 1990) who made the distinctions that are the subject of this chapter, among (1) the obsessive person who is virtually delusional, who uses ruminative thoughts to ward off psychotic decompensation; (2) the person whose obsessing is part of a borderline personality structure (as in the “Wolf Man”); and (3) the obsessive person with a neurotic-to-normal personality organization.
Before the category of “borderline” emerged in the middle of the 20th century, analytically influenced therapists followed Freud in differentiating only between neurotic and psychotic levels of pathology, the former being distinguished by a general appreciation of reality and the latter by a loss of contact with it. A neurotic woman knew at some level that her problem was in her own head; the psychotic one believed it was the world that was out of kilter. When Freud developed the structural model of the mind, this distinction took on the quality of a comment on a person’s psychological infrastructure: Neurotic people were viewed as suffering because their ego defenses were too automatic and inflexible, cutting them off from id energies that could be put to creative use; psychotic ones suffered because their ego defenses were too weak, leaving them helplessly overwhelmed by primitive material from the id.
The neurotic-versus-psychotic distinction had important clinical implications. The gist of these, considered in light of Freud’s structural model, was that therapy with a neurotic person should involve weakening the defenses and getting access to the id so that its energies may be released for more constructive activity. In contrast, therapy with a psychotic person should aim at strengthening defenses, covering over primitive preoccupations, influencing realistically stressful circumstances so that they are less upsetting, encouraging reality testing, and pushing the bubbling id back into unconsciousness. It was as if the neurotic person were like a pot on the stove with the lid on too tight, making the therapist’s job to let some steam escape, while the psychotic pot was boiling over, necessitating that the therapist get the lid back on and turn down the heat.
It became common for supervisors to recommend that with healthier patients, one should attack the defenses, whereas with people suffering from schizophrenia and other psychoses, one should support them. With the advent of antipsychotic drugs, this formulation lent itself to a widespread tendency not only to medicate—often the compassionate response to psychotic levels of anxiety—but also to assume that medication would do the covering over and would be needed on a lifetime basis. Therapists were advised not to do any “uncovering” with a potentially psychotic person: That might disturb the fragile defenses and send the client over the edge again. This way of conceptualizing degree of pathology is not without usefulness; it has opened the door to the development of different therapeutic approaches for different kinds of difficulties. But it falls short of a comprehensive and clinically nuanced ideal. Any theory oversimplifies, but this neurotic-versus-psychotic division, even with Freud’s elegant structural underpinnings and their therapeutic implications, offered only a start at a useful inferential diagnosis.
Ego Psychology Diagnosis: Symptom Neurosis, Neurotic
Character, Psychosis
In the psychoanalytic community, in addition to a distinction between neurosis and psychosis, differentiations of extent of maladaptation, not simply type of psychopathology, gradually began to appear within the neurotic category. The first clinically important one was Wilhelm Reich’s (1933) discrimination between “symptom neuroses” and “character neuroses.” Therapists were learning that it was useful to distinguish between a person with a discrete neurosis and one with a character permeated by neurotic patterns. This distinction lives on in the DSM, in which conditions labeled “disorder” tend to be those that analysts have called neuroses, and conditions labeled “personality disorder” resemble the old analytic concept of neurotic character.
To assess whether they were dealing with a symptom neurosis or a character problem, therapists were trained to pursue the following kinds of information when interviewing a person with neurotic complaints:
The former alternative in each of the above possibilities was presumptive evidence of a symptom problem, the latter of a character problem (Nunberg, 1955). The significance of this distinction lay in its implications for treatment and prognosis. If it was a symptom neurosis that the client suffered (equivalent to “Axis I disorder without comorbid personality disorder”), then one suspected that something in the person’s current life had activated an unconscious conflict and that the patient was now using maladaptive mechanisms to cope with it—methods that may have been the best available solution in childhood but that were now creating more problems than they were solving. The therapist’s task would be to determine the conflict, help the patient understand and process the emotions connected to it, and develop new resolutions of it. The prognosis was favorable, and treatment might be relatively short (cf. Menninger, 1963). One could expect a climate of mutuality during therapy, in which strong transference (and countertransference) reactions might appear, but usually in the context of an even stronger degree of cooperation.
If the patient’s difficulties amounted to a character neurosis or personality problem, then the therapeutic task would be more complicated, demanding, and time consuming, and the prognosis more guarded. This is only common sense, of course, in that trying to foster personality change obviously poses more challenges than helping someone get rid of a maladaptive response to a specific stress. But analytic theory went beyond common sense in specifying ways in which work on a person’s basic character would differ from work with a symptom not embedded in personality.
First, one could not take for granted that what the patient wanted (immediate relief from suffering) and what the therapist saw as necessary for the patient’s eventual recovery and resistance to future difficulties (modification of personality) could be seen by the patient as compatible. In instances when the patient’s aims and the analyst’s conception of what was ultimately needed were at variance, the analyst’s educative role became critical. One had to start by trying to convey to the patient how the therapist saw the problem; that is, “making ego alien what has been ego syntonic.” For example, a 30-year-old accountant once came to me looking to “achieve more balance” in his life. Raised to be the hope of his family, with a mission to compensate for his father’s failed ambitions, he was hardworking to the point of drivenness. He feared that he was missing precious years with his young children, whom he might enjoy if only he could stop pushing himself relentlessly to produce at work. He wanted me to develop a “program” with him in which he agreed to spend a certain amount of time per day exercising, a certain amount playing with his kids, a certain amount working on a hobby, and so forth. The proposed program included designated space for volunteer work, watching television, cooking, doing housework, and making love to his wife.
In the meeting that followed our initial interview, he brought in a sample schedule detailing such changes. He felt that if I could get him to put this program into effect, his problems would be solved. My first task was to try to suggest that this solution was part of the problem: He approached therapy with the same drivenness he was complaining about and pursued the serenity he knew he needed as if it were another job to do. I told him he was very good at doing, but he evidently had had little experience with just being. While he grasped this notion intellectually, he had no emotionally salient memory of a less compulsive approach to life, and he regarded me with a mixture of hope and skepticism. Although simply telling his story had provided some short-term relief of his depression, I saw him as having to get used to the fact that to avoid this kind of misery in the future, he would need to bring into conscious awareness and to rethink some of the major assumptions that had governed his life.
Second, in working with someone whose character was fundamentally neurotic, one could not take for granted an immediate “working alliance” (Greenson, 1967). Instead, one would have to create the conditions under which it could develop. The concept of the working or therapeutic alliance refers to the collaborative dimension of the work between therapist and client, the cooperation that endures in spite of the strong and often negative emotions that may surface during treatment. Empirically, a solid working alliance is associated with good outcome (Safran & Muran, 2000), and its establishment (or restoration after a rupture) takes precedence over other aims.
Patients with symptom neuroses feel on the side of the therapist in opposing a problematic part of the self. They rarely require a long period to develop a shared perspective. In contrast, those whose problems are complexly interwoven with their personality may easily feel alone and under attack. When the therapist raises questions about lifelong, ego-syntonic patterns, their whole identity may feel assaulted. Distrust is inevitable and must be patiently endured by both parties until the therapist has earned the client’s confidence. With some patients, this process of building an alliance can take more than a year. Trying too quickly to take on what the therapist sees as obvious problems may damage the alliance and impede the process of change.
Third, therapy sessions with someone with a character rather than a symptom problem could be expected to be less exciting, less surprising, less dramatic. Whatever the therapist’s and patient’s fantasies about unearthing vivid repressed memories or unconscious conflicts, they would have to content themselves with a more prosaic process, the painstaking unraveling of all the threads that had created the emotional knot that the patient had until now believed was just the way things had to be, and the slow working out of new ways of thinking and handling feelings.
In the development of personality disorders, as opposed to the appearance of neurotic reactions to particular current stresses, there are long patterns of identification, learning, and reinforcement. Where the etiology is traumatic, “strain trauma” (Kris, 1956) is implicated, rather than the “shock trauma” (one unassimilated, unmourned injury) celebrated in Hollywood’s early, enthusiastic portrayals of psychoanalytic treatment (see, e.g., Hitchcock’s Spellbound). As a consequence, one could expect that in the therapy of character neuroses, both parties would have to deal with occasional boredom, impatience, irritability, and demoralization—the patient by expressing them without fear of criticism and the therapist by mining such feelings for empathy with the patient’s struggle with a difficult, protracted task.
This distinction between neurotic symptoms and neurotic personality remains important, even in instances where one cannot do the long-term work (e.g., D. Shapiro, 1989) that character change requires. If one understands one’s patient’s inflexible personality issues, one can often find some way of making a short-term impact that avoids the person’s feeling misunderstood or attacked. For example, knowing that a woman has a central psychopathic streak alerts the therapist that in trying to interfere with some damaging pattern, it is better to appeal to her pride than to her assumed concern for others.
For a long time, the categories of symptom neurosis, character neurosis, and psychosis constituted the main constructs by which we understood personality differences on the dimension of severity of disorder. A neurosis was the least serious condition, a personality disorder more serious, and a psychotic disturbance quite grave. These formulations maintained the old distinction between sane and insane, with the sane category including two possibilities: neurotic reactions and neurotically structured personalities. Over time, however, it became apparent that such an overall scheme of classification was both incomplete and misleading.
One drawback of this taxonomy is its implication that all character problems are more pathological than all neuroses. One can still discern such an assumption in the DSM, in which the criteria for diagnosing most personality disorders include significant impairments in functioning. And yet some stress-related neurotic reactions are more crippling to a person’s capacity to cope than, say, some hysterical and obsessional personality disorders. A man I know suffers from agoraphobia, ego alien but severe. He has warm relations with friends, enjoys his family, and works productively at home, but he never leaves his house. I see his life as more constricted and deadened than that of many people with personality disorders and even psychoses.
To complicate the issue still further, there is also a problem in the other direction: Some character disturbances seem to be much more severe and primitive in quality than anything that could reasonably be called “neurotic.” One can see that there is no way in such a linear, three-part classification to differentiate between distortions of character that are mildly incapacitating and those that involve fairly dire consequences. A problem can be characterological and of any level of severity. The line between benign personality “traits” or “styles” and mild personality “disorders” is quite blurry. On the other end of the continuum, some character disorders have been understood for a long time as involving such substantial deformities of the ego that they are closer to psychosis than neurosis. Psychopathy and malignant forms of narcissistic personality organization, for example, have long been recognized as variants of human individuality, but until fairly recently, they have tended to be considered as somewhat outside the scope of possible therapeutic intervention and not easily placed on a neurotic–character disordered–psychotic continuum.
Object Relations Diagnosis: The Delineation
of Borderline Conditions
Even in the late 19th century, some psychiatrists were identifying patients who seemed to inhabit a psychological “borderland” (Rosse, 1890) between sanity and insanity. By the middle of the 20th century, other ideas about personality organization suggesting a middle ground between neurosis and psychosis began to appear. Adolph Stein (1938) noted that people with qualities he called “borderline” got worse rather than better in standard psychoanalytic treatment. Helene Deutsch (1942) proposed the concept of the “as-if personality” for a subgroup of people we would now see as narcissistic or borderline, and Hoch and Polatin (1949) made a case for the category of “pseudoneurotic schizophrenia.”
By the middle 1950s, the mental health community had followed these innovators in noting the limitations of the neurosis-versus-psychosis model. Numerous analysts began complaining about clients who seemed character disordered, but in a peculiarly chaotic way. Because they rarely or never reported hallucinations or delusions, they could not be considered psychotic, but they also lacked the consistency of neurotic-level patients, and they seemed to be miserable on a much grander and less comprehensible scale than neurotics. In treatment, they could become temporarily psychotic—convinced, for example, that their therapist was exactly like their mother, yet outside the consulting room there was an odd stability to their instability. In other words, they were too sane to be considered crazy, and too crazy to be considered sane. Therapists began suggesting new diagnostic labels that captured the quality of these people who lived on the border between neurosis and psychosis. In 1953, Knight published a thoughtful essay about “borderline states.” In the same decade, T. F. Main (1957) was referring to similar pathology in hospitalized patients as “The Ailment.” In 1964, Frosch suggested the diagnostic category of “psychotic character.”
In 1968, Roy Grinker and his colleagues (Grinker, Werble, & Drye, 1968) did a seminal study documenting a “borderline syndrome” inhering in personality, with a range of severity from the border with the neuroses to the border with the psychoses. Gunderson and Singer (e.g., 1975) continued to subject the concept to empirical scrutiny, and eventually, via both research and clinical findings, and thanks to the elucidation of writers such as Kernberg (1975, 1976), Masterson (1976), and M. H. Stone (1980, 1986), the concept of a borderline level of personality organization attained widespread acceptance in the psychoanalytic community.
By 1980, the term had been sufficiently researched to appear in the DSM (DSM-III; American Psychiatric Association, 1980) as a personality disorder. This development has had mixed effects: It has legitimated a valuable psychoanalytic concept but at the price of losing its original meaning as a level of functioning. The concept of borderline psychology represented in the DSM drew heavily on the work of Gunderson (e.g., 1984), who had studied a group that most analysts would have diagnosed as having a hysterical or histrionic psychology at the borderline level. Kernberg (1984), one of the originators of the concept, began having to differentiate between “borderline personality organization” (BPO) and the DSM’s “borderline personality disorder” (BPD).
I am probably fighting a losing battle in trying to preserve the original meaning of the term “borderline” (as I did, for example, in the Personality section of the Psychodynamic Diagnostic Manual [PDM Task Force, 2006]), but I think a lot has been sacrificed in equating the term with a particular character type. The concept of “borderline” as a level of psychological functioning had evolved over decades of clinical experience, coming to be generally viewed as a stable instability on the border between the neurotic and psychotic ranges, characterized by lack of identity integration and reliance on primitive defenses without overall loss of reality testing (Kernberg, 1975). I worry that with the DSM definition having become accepted, we are losing a way of talking about, say, obsessional or schizoid people at the borderline level (e.g., the “quiet borderline” patient of Sherwood & Cohen, 1994). If all our empirical research on borderline phenomena applies narrowly to the more self-dramatizing, histrionic version of borderline-level personality organization, we are left in the dark about the etiology and treatment of other personality disorders at the borderline level.
By the second half of the 20th century, many therapists struggling to help clients that we now see as borderline found themselves drawing inspiration and validation from writings of analysts in the British object relations movement and the American interpersonal group, who looked at patients’ experiences with key figures in childhood. These theorists emphasized the patient’s experience of relationship: Was the person preoccupied with symbiotic issues, separation–individuation themes, or highly individuated competitive and identificatory motifs? Erikson’s (1950) reworking of Freud’s three infantile stages in terms of the child’s interpersonal task made a significant clinical impact, in that patients could be conceptualized as fixated at either primary dependency issues (trust vs. mistrust), secondary separation–individuation issues (autonomy vs. shame and doubt), or more advanced levels of identification (initiative vs. guilt).
These developmental-stage concepts made sense of the differences therapists were noticing among psychotic-, borderline-, and neurotic-level patients: People in a psychotic state seemed fixated at an unindividuated level in which they could not differentiate between what was inside and what was outside themselves; people in a borderline condition were construed as fixated in dyadic struggles between total enmeshment, which they feared would obliterate their identity, and total isolation, which they equated with traumatic abandonment; and people with neurotic difficulties were understood as having accomplished separation and individuation but as having run into conflicts between, for example, things they wished for and things they feared, the prototype for which was the oedipal drama. This way of thinking made sense of numerous puzzling and demoralizing clinical challenges. It accounted for why one woman with phobias seemed to be clinging to sanity by a thread, while another was oddly stable in her phobic instability, and yet a third woman was, despite having a phobia, otherwise a paragon of mental health.
By the late 20th century there was, both within the psychoanalytic tradition and outside it, a vast literature on borderline psychopathology, showing a bewildering divergence of conclusions about its etiology. Some investigators (e.g., M. H. Stone, 1977) emphasized constitutional and neurological predispositions; some (e.g., G. Adler, 1985; Masterson, 1972, 1976) focused on developmental failures, especially in the separation–individuation phase described by Mahler (1971); some (e.g., Kernberg, 1975) conjectured about aberrant parent–child interaction at an earlier phase of infantile development; some (e.g., Mandelbaum, 1977; Rinsley, 1982) pointed to poor boundaries between members in dysfunctional family systems; and some (e.g., McWilliams, 1979; Westen, 1993) made sociological speculations. Others (e.g., Meissner, 1984, 1988) were integrative of many of these perspectives. With advances in attachment research (e.g., Ainsworth, Blehar, Waters, & Wall, 1978), some writers began to conjecture about the infantile attachment styles that correlated later with borderline psychology. By the 1990s, more and more people were writing about how trauma, especially incest, plays a bigger role in the development of borderline dynamics than had previously been suspected (e.g., Wolf & Alpert, 1991).
Recent empirical studies of borderline personality, most of them using the DSM definition, have looked at all these aspects. There is some evidence for constitutional predispositions (Gunderson & Lyons-Ruth, 2008; Siever & Weinstein, 2009); some for misattuned parenting around attachment and separation issues (Fonagy, Target, Gergeley, Allen, & Bateman, 2003; Nickell, Waudby, & Trull, 2002); and some for the role of trauma, especially relational trauma in early attachment (Schore, 2002) but also later experiences of sexual abuse (Herman, 1992). It is probable that all these factors play a role, that borderline psychology is not a single entity and is multidetermined, like most other complex psychological phenomena. Current psychoanalytic writing, especially about borderline dynamics, has drawn heavily on empirical findings in the areas of infant development, attachment, and trauma. One consequence has been a significant paradigm shift, as unquestioned notions of fixation at a normative developmental phase have been challenged by evidence for different experiences of attachment and for the destructive effects of recurrent trauma even long after the preschool years.
Whatever the etiology of borderline personality organization, and it probably differs from person to person, clinicians of diverse perspectives have attained a surprisingly reliable consensus on the clinical manifestations of problems in the borderline range. Especially when an interviewer is trained in what information, subjective as well as objective, should be observed and pursued, the diagnosis of borderline level of character structure may be readily confirmed or disconfirmed (e.g., through Kernberg’s [1984] structural interview or the later, more carefully empirically validated instrument of his colleagues, the Structured Interview for Personality Organization [STIPO; Stern, Caligor, Roose, & Clarkin, 2004]).
Despite the complexity of the etiologies of borderline conditions, I think it can still be useful to view people with a vulnerability to psychosis as unconsciously preoccupied with the issues of the early symbiotic phase (especially trust), people with borderline personality organization as focused on separation–individuation themes, and those with neurotic structure as more “oedipal” or capable of experiencing conflicts that feel more internal to them. The most prevalent kind of anxiety for people in the psychotic range is fear of annihilation (Hurvich, 2003), evidently an activation of the brain’s FEAR system (Panksepp, 1998) that evolved to protect against predation; the central anxiety for people in the borderline range is separation anxiety or the activation of Panksepp’s PANIC system that deals with early attachment needs; anxiety in neurotic people tends to involve more unconscious conflict, especially fear of enacting guilty wishes.
OVERVIEW OF THE
NEUROTIC–BORDERLINE–PSYCHOTIC SPECTRUM
In the following sections, I discuss neurotic, borderline, and psychotic levels of character structure in terms of favored defenses, level of identity integration, adequacy of reality testing, capacity to observe one’s pathology, nature of one’s primary conflict, and transference and countertransference. I focus on how these abstractions manifest themselves as discernible behaviors and communications in an initial interview or in an ongoing treatment. In Chapter 4 I explore implications of these discriminations for the conduct and prognosis of therapy. Again, I want to emphasize that these levels of organization are somewhat artificial, that we can all find in ourselves issues from every level, and that viewing one’s client as organized at one or another of the levels should not distract a therapist from the person’s individuality and areas of strength.
Characteristics of Neurotic-Level Personality Structure
It is an irony that the term “neurotic” is now reserved by most analysts for people so emotionally healthy that they are considered rare and unusually gratifying clients. In Freud’s time, the word was applied to most nonorganic, nonschizophrenic, nonpsychopathic, and non-manic–depressive patients—that is, to a large class of individuals with emotional distress short of psychosis. We now see many of the people Freud called neurotic as having borderline or even psychotic features (“hysteria” was understood to include hallucinatory experiences that clearly cross the border into unreality). The more we have learned about the depth of certain problems, and their stubborn enmeshment within the matrix of a person’s character, the more we currently reserve the term “neurotic” to denote a high level of capacity to function despite emotional suffering.
People whose personalities would be described by many contemporary analysts as organized at an essentially neurotic level rely primarily on the more mature or second-order defenses. While they also use primitive defenses, these are not nearly so prominent in their overall functioning and are evident mostly in times of unusual stress. While the presence of primitive defenses does not rule out the diagnosis of neurotic level of character structure, the absence of mature defenses does. Traditionally, the psychoanalytic literature noted that healthier people use repression as their basic defense, in preference to more indiscriminate solutions to conflict such as denial, splitting, projective identification, and other more archaic mechanisms.
Myerson (1991) has described how empathic parenting allows a young child to experience intense affects without having to hang on to infantile ways of dealing with them. As the child grows up, these powerful and often painful states of mind are put away and forgotten rather than continually reexperienced and then denied, split off, or projected. They may reemerge in long-term, intensive analysis, when analyst and client together, under the conditions of safety that evoke a “transference neurosis,” peel back layers of repression; but ordinarily, overwhelming affects and primitive ways of handling them are not characteristic of persons in the neurotic range. And even in deep psychoanalytic treatment, the neurotic-level client maintains some more rational, objective capacities in the middle of whatever emotional storms and associated distortions occur.
People with healthier character structure strike the interviewer as having a somewhat integrated sense of identity (Erikson, 1968). Their behavior shows some consistency, and their inner experience is of continuity of self through time. When asked to describe themselves, they are not at a loss for words, nor do they respond one-dimensionally; they can usually delineate their overall temperament, values, tastes, habits, convictions, virtues, and shortcomings with a sense of their long-range stability. They feel a sense of continuity with the child they used to be and can project themselves into the future as well. When asked to describe important others, such as their parents or lovers, their characterizations tend to be multifaceted and appreciative of the complex yet coherent set of qualities that constitutes anyone’s personality.
Neurotic-level people are ordinarily in solid touch with what most of the world calls “reality.” Not only are they strangers to hallucinatory or delusional misinterpretations of experience (except under conditions of chemical or organic influence, or posttraumatic flashback), they also strike the interviewer or therapist as having comparatively little need to misunderstand things in order to assimilate them. Patient and therapist live subjectively in more or less the same world. Typically, the therapist feels no compelling emotional pressure to be complicit in seeing life through a lens that feels distorting. Some portion of what has brought a neurotic patient for help is seen by him or her as odd; in other words, much of the psychopathology of neurotically organized people is ego alien or capable of being addressed so that it becomes so.
People in the neurotic range show early in therapy a capacity for what Sterba (1934) called the “therapeutic split” between the observing and the experiencing parts of the self. Even when their difficulties are somewhat ego syntonic, neurotic-level people do not seem to demand the interviewer’s implicit validation of their ways of perceiving. For example, a paranoid man who is organized neurotically will be willing to consider the possibility that his suspicions derive from an internal disposition to emphasize the destructive intent of others. Contrastingly, paranoid patients at the borderline or psychotic level will put intense pressure on the therapist to join their conviction that their difficulties are external in origin; for example, to agree that others may be out to get them. Without such validation, they worry that they are not safe with the therapist.
Similarly, compulsive people in the neurotic range may say that their repetitive rituals are crazy but that they feel anxiety if they neglect them. Compulsive borderline and psychotic people sincerely believe themselves to be protected in some elemental way by acting on their compulsions and have often developed elaborate rationalizations for them. A neurotic-level patient will share a therapist’s assumption that the compulsive behaviors are in some realistic sense unnecessary, but a borderline or psychotic patient may privately worry that the practitioner who questions the rituals is deficient in either common sense or moral decency. A neurotic woman with a cleaning compulsion will be embarrassed to admit how frequently she launders the sheets, while a borderline or psychotic one will feel that anyone who washes the bedding less regularly is unclean.
Sometimes years can go by in treatment before a borderline or psychotic person will even mention a compulsion or phobia or obsession—in the patient’s view there is nothing unusual about it. I worked with one borderline client for more than 10 years before she casually mentioned an elaborate, time-consuming morning ritual to “clear her sinuses” that she considered part of ordinary good hygiene. Another borderline woman, who had never mentioned bulimia in her abundance of even more distressing symptoms, dropped the comment, after 5 years in therapy, “By the way, I notice I’m not puking anymore.” She had not previously thought to regard that part of her behavioral repertoire as consequential.
Their histories and their behavior in the interview situation give evidence that neurotic-level people have more or less successfully traversed Erikson’s first two stages, basic trust and basic autonomy, and that they have made at least some progress toward identity integration and a sense of initiative. They tend to seek therapy not because of problems in essential security or agency, but because they keep running into conflicts between what they want and obstacles to attaining it that they suspect are of their own making. Freud’s contention that the proper goal of therapy is the removal of inhibitions against love and work applies to this group; some neurotic-level people are also looking to expand their capacity for solitude and play.
Being in the presence of someone at the healthier end of the continuum of character pathology feels generally benign. The counterpart of the patient’s possession of a sound observing ego is the therapist’s experience of a sound working alliance. Often from the very first session, the therapist of a neurotic client feels that he or she and the patient are on the same side and that their mutual antagonist is a problematic part of the patient. The sociologist Edgar Z. Friedenberg (1959) compared this alliance to the experience of two young men tinkering with a car: one the expert, the other an interested learner. In addition, whatever the valence of the therapist’s countertransference, positive or negative, it tends not to feel overwhelming. The neurotic-level client engenders in the listener neither the wish to kill nor the compulsion to save.
Characteristics of Psychotic-Level Personality Structure
At the psychotic end of the spectrum, people are much more internally desperate and disorganized. Interviewing a deeply disturbed patient can range from being a participant in a pleasant, low-key discussion to being the recipient of a homicidal attack. Especially before the advent of anti-psychotic drugs in the 1950s, few therapists had the natural intuitive talent and emotional stamina to be significantly therapeutic to those in psychotic states. One of the finest achievements of the psychoanalytic tradition has been its inference of some order in the apparent chaos of people who are easy to dismiss as hopelessly and incomprehensibly crazy, and its consequent offer of ways to understand and mitigate severe mental suffering (Arieti, 1974; Buckley, 1988; De Waelhens & Ver Eecke, 2000; Eigen, 1986; Ogden, 1989; Robbins, 1993; Searles, 1965; Silver, 1989; Silver & Cantor, 1990; Spotnitz, 1985; Volkan, 1995).
It is not difficult to diagnose patients who are in an overt state of psychosis: they express hallucinations, delusions, and ideas of reference, and their thinking strikes the listener as illogical. There are many people walking around, however, whose basic psychotic-level internal confusion does not surface conspicuously unless they are under considerable stress. The knowledge that one is dealing with a “compensated” schizophrenic, or a currently nonsuicidal depressive who may be subject to periodic delusional yearnings to die, can make the difference between preventing and precipitating disaster. Having carried out or supervised the long-term treatment of many extremely difficult, sometimes putatively “untreatable” cases, I am convinced that devoted therapists do significant prevention. We preempt psychotic breaks, prevent suicides and homicides, and keep people out of hospitals. (These critical effects of therapy go mostly undocumented; no one can prove that he or she prevented a calamity, and critics tend to argue that if one claims to have forestalled a psychotic break, the patient was not really at risk of psychosis in the first place.)
I share with many analysts the view that it is also useful to conceive of some people who may never become diagnosably psychotic as nevertheless living in a symbiotic–psychotic internal world or, in Klein’s (e.g., 1946) terms, in a consistently “paranoid–schizoid” state. They function, sometimes quite effectively, but they strike one as confused and deeply terrified, and their thinking feels disorganized or paranoid. One man I worked with, for example, told me with palpable dread that he would never return to a particular gym to exercise: “Three times someone has moved my things, so it’s obvious that I’m not wanted there.” Another used to switch topics abruptly whenever he was becoming very sad. I commented on this, and he said, “Oh yeah, I know I do that.” I asked him what his understanding of the pattern was, expecting him to say something like “I’m not ready to go there,” or “It hurts too much,” or “I don’t want to start crying.” But what he said, in a tone suggesting it was self-evident, was “Well, I can see I’m hurting you!” He saw sympathetic sadness on my face and could not imagine he was not damaging me.
To understand the subjective world of psychotic-level clients, one must first appreciate the defenses they tend to use. I will expand on these in Chapter 5; at this point I am simply listing them: withdrawal, denial, omnipotent control, primitive idealization and devaluation, primitive forms of projection and introjection, splitting, extreme dissociation, acting out, and somatization. These processes are preverbal and prerational; they protect one against a level of “nameless dread” (Bion, 1967) so overwhelming that even the frightening distortions that the defenses themselves may create are a lesser evil than that state of terror. As Fromm-Reichmann (1950) noted, people who struggle with psychosis have a core, immobilizing dread of their fantasied superhuman potential for destructiveness.
Second, people whose personalities are organized at an essentially psychotic level have grave difficulties with identity—so much so that they may not be fully sure that they exist, much less whether their existence is satisfying. They are deeply confused about who they are, and they usually struggle with such basic issues of self-definition as body concept, age, gender, and sexual orientation. “How do I know who I am?” or even “How do I know that I exist?” are not uncommon questions for psychotically organized people to ask in earnest. They cannot depend on a sense of continuity of identity in themselves and do not experience others as having continuity of self either: They live in fear of “malevolent transformations” (Sullivan, 1953) that will turn a trusted person abruptly into a sadistic persecutor. When asked to describe themselves or other important people in their lives, they tend to be vague, tangential, concrete, or observably distorting.
Often in rather subtle ways, one feels that a patient with an essentially psychotic personality is not anchored in reality. Although most of us have vestiges of magical beliefs (e.g., the idea that saying something positive will jinx a situation), careful investigation will reveal that such attitudes are not ego alien to psychotic-level individuals. They are often confused by and estranged from the assumptions about “reality” that are conventional within their culture. Although they may be preternaturally attuned to the underlying affect in any situation, they often do not know how to interpret its meaning and may assign highly self-referential significance to it.
For example, a very paranoid patient I worked with for a long time, whose sanity was often at risk, had an uncanny feel for my emotional state. She would read it accurately but then attach to her perception of it the primitive preoccupations she had about her own essential goodness or badness, as in “You look irritated. It must be because you think I’m a bad mother.” Or “You look bored. I must have offended you last week by leaving the session 5 minutes early.” It took her years to feel safe enough to tell me that was how she was interpreting my expressions, and several more years to transform the conviction “Evil people are going to kill me because they hate my lifestyle” into “I feel guilty about some aspects of my life.”
People with psychotic tendencies have trouble getting perspective on their psychological problems. They lack the “reflective functioning” that Fonagy and Target (1996) have identified as critical to cognitive maturation. This deficit may be related to the well-documented difficulties that schizophrenic people have with abstraction (Kasanin, 1944). Those whose mental health history has given them enough jargon to sound like good self-observers (e.g., “I know I tend to overreact” or even “My schizophrenia interferes with my judgment”) may reveal to a sensitive interviewer that in an effort to reduce anxiety they are compliantly parroting what they have been told about themselves. One patient of mine had had so many intakes at psychiatric hospitals during which she had been asked (in a mental status evaluation that helps determine whether the patient is capable of abstract thought) to give the meaning of the proverb “A bird in the hand is worth two in the bush” that she had asked an acquaintance what it meant and memorized the answer (she proudly offered this explanation when I commented in an interested way on the automatic quality of her response).
Early psychoanalytic formulations about the difficulties that psychotic people have in getting perspective on their realistic troubles stressed energic aspects of their dilemma; that is, they were expending so much energy fighting off existential terror that none was left to use in the service of coping with reality. Ego psychology models emphasized the psychotic person’s lack of internal differentiation between id, ego, and superego, and between observing and experiencing aspects of the ego. Students of psychosis influenced by interpersonal, object relations, and self psychology theories (e.g., Atwood, Orange, & Stolorow, 2002) have referred to boundary confusion between inside and outside experience, and to deficits in attachment that make it subjectively too dangerous for the psychotic person to enter the same assumptive world as the interviewer.
Recently, in light of fMRI studies showing similarities between effects of trauma on the developing brain and the biological abnormalities found in the brains of individuals diagnosed with schizophrenia, John Read and his colleagues (Read, Perry, Moskowitz, & Connolly, 2001) have argued for a traumatic etiology of schizophrenia. A full account of the lack of “observing ego” in psychotic-level clients probably includes all these perspectives as well as genetic, biochemical, and situational contributants. The critical thing for therapists to appreciate is that close to the surface in people with psychotic-level psychologies, one finds both mortal fear and dire confusion.
The nature of the primary conflict in people with a potential for psychosis is literally existential: life versus death, existence versus obliteration, safety versus terror. Their dreams are full of stark images of death and destruction. “To be or not to be” is their recurrent theme. Laing (1965) eloquently depicted them as suffering “ontological insecurity.” Psychoanalytically influenced studies of the families of schizophrenic people in the 1950s and 1960s consistently reported patterns of emotional communication in which the psychotic child received subtle messages to the effect that he or she was not a separate person but an extension of someone else (Bateson, Jackson, Haley, & Weakland, 1956; Lidz, 1973; Mischler & Waxier, 1968; Singer & Wynne, 1965a, 1965b). Although the discovery of the major tranquilizers has diverted attention from more strictly psychological investigations of psychotic processes, no one has yet presented evidence controverting the observation that the psychotic person is deeply unconvinced of his or her right to a separate existence, or may even be unfamiliar with the sense of existing at all.
Despite their unusual and even frightening aspects, patients in the psychotic range may induce a positive countertransference. This reaction differs a bit from warm countertransference reactions to neurotic-level clients: One may feel more subjective omnipotence, parental protectiveness, and deep soul-level empathy toward psychotic people than toward neurotic ones. The phrase “the lovable schizophrenic” was for a long time in vogue as an expression of the solicitous attitude that mental health personnel often feel toward their most severely troubled patients. (The implicit contrast group here, as I discuss below, is the borderline population.) Psychotic people are so desperate for respect and hope that they may be deferential and grateful to any therapist who does more than classify and medicate them. Their gratitude is naturally touching.
People with psychotic tendencies are particularly appreciative of sincerity. A recovered schizophrenic woman once told me she could forgive even serious failings in a therapist if she saw them as “honest mistakes.” Psychotic-level clients may also appreciate educative efforts and may respond with relief to the normalization or reframing of their preoccupations. These dispositions, along with their propensity for fusion and idealization, can make the therapist feel strong and benevolent. The downside of these patients’ poignant dependence on our care is the burden of psychological responsibility they inevitably impose. In fact, the countertransference with psychotic-level people is remarkably like normal maternal feelings toward infants under a year and a half: They are wonderful in their attachment and terrifying in their needs. They are not yet oppositional and irritating, but they also tax one’s resources to the limit. I should not work with a schizophrenic, a supervisor once told me, unless I was prepared to be eaten alive.
This “consuming” feature of their psychology is one reason that many therapists prefer not to work with individuals with schizophrenia and other psychoses. In addition, as Karon (1992) has noted, the access of psychotic patients to deeply upsetting realities that the rest of us would prefer to ignore is often too much for us. In particular, they see our flaws and limitations with stunning clarity. Other reasons for their relative unpopularity as patients despite their appealing qualities probably include therapists’ lack of adequate training in psychotherapy with psychotics (Karon, 2003; Silver, 2003), economic pressures that breed rationalizations about limited approaches or “management” instead of therapy (Whitaker, 2002), and personal dispositions not to work toward relatively modest treatment goals in contrast to what can be achieved with a neurotic-level person. But as I stress in the next chapter, it can be effective and rewarding to work with clients in the psychotic range if one is realistic about the nature of their psychological difficulties.
Characteristics of Borderline Personality Organization
One of the most striking features of people with borderline personality organization is their use of primitive defenses. Because they rely on such archaic and global operations as denial, projective identification, and splitting, when they are regressed they can be hard to distinguish from psychotic patients. An important difference between borderline and psychotic people, though, is that when a therapist confronts a borderline patient on using a primitive mode of experiencing, the patient will show at least a temporary responsiveness. When the therapist makes a similar comment to a psychotically organized person, he or she will likely become further agitated.
As an illustration, consider the defense of primitive devaluation. Being devalued is a familiar and painful experience to any therapist. Devaluation is an unconscious strategy that is often intended to preserve self-esteem, but which does so at the expense of learning. An effort to address that defense might go something like “You certainly love to cherish all my defects. Maybe that protects you from admitting that you might need my help. Perhaps you would be feeling ‘one down’ or ashamed if you weren’t always putting me down, and you’re trying to avoid that feeling.” A borderline patient might scorn such an interpretation, or grudgingly admit it, or receive it silently, but in any event, he or she would give some indications of reduced anxiety. A psychotic person would react with increased anxiety, since to someone in existential terror, devaluation of the power of the therapist may be the only psychological means by which he or she feels protected from obliteration. The therapist’s discussing it as if it were optional would be extremely frightening.
Borderline patients are in some ways similar to and in others different from psychotic people on the dimension of identity integration. Their experience of self is likely to be full of inconsistency and discontinuity. When asked to describe their personalities, they may, like psychotic-level patients, be at a loss. And when asked to describe important people in their lives, they may respond with anything but three-dimensional, evocative descriptions of recognizable human beings. “My mother? She’s just a regular mother, I guess” is a typical response. They often give global, dismissive descriptions such as “An alcoholic. That’s all.” Unlike patients with psychosis, they rarely sound concrete or tangential to the point of being bizarre, but they do tend to dismiss the therapist’s interest in the complexities of themselves and others. Fonagy (2000) writes that borderline clients are insecurely attached and lack the “reflective function” that finds meaning in their own behavior and that of others. They cannot “mentalize”; that is, they cannot appreciate the separate subjectivities of other people. In philosophical terms, they lack a theory of mind.
Clients in the borderline range may become hostile when confronted with the limited continuity of their identity. One of my patients flew into a full-blown fury at a questionnaire she was given as a standard intake procedure in a clinic. It had a sentence-completion section in which the client was asked to fill in blanks like “I am the kind of person who ____________.” “How can anybody know what to do with this shit?” she raged. (Some years and countless sessions later she mused, “Now I could fill in that form. I wonder why I went ballistic about it.”) In general, borderline patients have trouble with affect tolerance and regulation, and quickly go to anger in situations where others might feel shame or envy or sadness or some other more nuanced affect.
In two ways, the relation of borderline patients to their own identity is different from that of psychotic people. First, the sense of inconsistency and discontinuity that people with borderline organization suffer lacks the degree of existential terror of the schizophrenic. Borderline patients may have identity confusion, but they know they exist. Second, people with psychotic tendencies are much less likely than borderline patients to react with hostility to questions about identity of self and others. They are too worried about losing their sense of ongoing being, consistent or not, to resent the interviewer’s focus on that problem.
Despite these distinctions, both borderline and psychotic people, unlike neurotics, rely heavily on primitive defenses and suffer a basic defect in the sense of self. The dimension of experience on which the two groups differ substantially is reality testing. Borderline clients, when interviewed thoughtfully, demonstrate an appreciation of reality no matter how crazy or florid their symptoms look. It used to be standard psychiatric practice to assess the degree of the patient’s “insight into illness” in order to discriminate between psychotic and nonpsychotic states. Because a borderline patient may relentlessly deny psychopathology yet still show a level of discrimination about what is real or conventional that distinguishes him or her from a psychotic peer, Kernberg (1975) proposed that “adequacy of reality testing” be substituted for that criterion.
To make a differential diagnosis between borderline and psychotic levels of organization, Kernberg (1984) advises investigating the person’s appreciation of conventional notions of reality by picking out some unusual feature of his or her self-presentation, commenting on it and asking if the patient is aware that others might find that feature peculiar (e.g., “I notice that you have a tattoo on your cheek that says ‘Death!’ Can you understand how that might seem unusual to me or others?”). The borderline person will acknowledge that the feature is unconventional and that outsiders might not understand its significance. The psychotic person is likely to become frightened and confused because the sense that he or she is not understood is deeply disturbing. These differing reactions, which Kernberg and his coworkers (e.g., Kernberg, Yeomans, Clarkin, & Levy, 2008) have explored both clinically and via empirical research, may be viewed as support for psychoanalytic assumptions about the centrality of separation–individuation issues for people with borderline pathology as contrasted with unconscious deficits in self–other differentiation in psychosis.
The capacity of someone at the borderline level to observe his or her own pathology—at least the aspects of it that impress an external observer—is quite limited. People with borderline psychologies come to therapy for complaints such as panic attacks or depression or illnesses that a physician has insisted are related to “stress,” or they arrive at the therapist’s office at the urging of an acquaintance or family member, but they rarely come with the agenda of changing their personalities in directions that outsiders readily see as advantageous. Even in recent years, when they are apt to know they “have BPD” and can endorse the DSM criteria for diagnosing it, they still lack a sense of what it would be like to be different. Having never had any other kind of character, they have little emotional basis for knowing how it would feel to have identity integration, mature defenses, the capacity to defer gratification, a tolerance for ambivalence and ambiguity, or an ability to regulate affects. They just want to stop hurting or to get some critic off their back.
In nonregressed states, because their reality testing is fine and because they may present themselves in ways that compel our empathy, they do not look particularly “sick.” Sometimes it is only after therapy has proceeded for a while that one realizes that a given patient has a borderline structure. Usually the first clue is that interventions that the therapist intends to be helpful are received as attacks. In other words, the therapist keeps assuming a capacity for reflective functioning that the patient mostly lacks. (In older language, the therapist is trying to talk with an observing ego, something the client cannot access, especially when upset.) The patient knows only that some aspect of the self is being criticized. The therapist keeps trying to forge the kind of alliance that is possible with neurotic-level patients and keeps coming to grief in the effort.
Eventually, one learns that one must first just weather the affective storms that seem to keep raging, while trying to behave in ways that the patient will experience as different from whatever influences have shaped such a troubled and help-resistant person. Only after therapy has brought about some structural change will the patient be different enough to begin to understand what the therapist is trying to work toward. This may take a long time—sometimes 2 years in my experience—but it is of comfort that in the meantime, the most disabling borderline behaviors may disappear. Clarkin and Levy (2003) describe significant symptom reduction after 1 year of transference-focused therapy. Still, the work will typically have been tumultuous and frustrating to both parties.
Masterson (1976) has vividly depicted, and others with different viewpoints report similar observations, how borderline clients seem caught in a dilemma: When they feel close to another person, they panic because they fear engulfment and total control; when they are alone, they feel traumatically abandoned. This central conflict of their emotional experience results in their going back and forth in relationships, including the therapy relationship, in which neither closeness nor distance is comfortable. Living with such a basic conflict, one that does not respond immediately to interpretive efforts, is exhausting for borderline patients, their friends, their families, and their therapists. They are famous among emergency psychiatric service workers, at whose door they frequently appear talking suicide, for manifesting “help seeking–help rejecting behavior.”
Masterson saw borderline patients as fixated at the rapprochement subphase of the separation–individuation process (Mahler, 1972b), when the child has attained some autonomy yet still needs reassurance that a caregiver remains available and powerful. This drama unfolds around age 2, when children typically alternate between rejecting mother’s help (“I can do it myself!”) and dissolving in tears at her knees. Masterson (1976) believed that borderline patients have had mothers who discouraged them from separating in the first place or neglected them when they needed to regress after attaining some independence. Whether or not his ideas about etiology are correct, his observations about the borderline person’s entrapment in dilemmas of separation and individuation help make sense of the changing, demanding, and often confusing qualities of borderline patients.
Transferences in borderline clients tend to be strong, unambivalent, and resistant to ordinary kinds of intervention. The therapist may be perceived as all good or all bad. If a well-intentioned but clinically naive therapist tries to interpret transference as one would with a neurotic person (e.g., “Perhaps what you’re feeling toward me is something you felt toward your father”), he or she will find that no relief or helpful sense of insight follows; in fact, often the client will simply agree that the therapist is actually behaving like the earlier object. Also, it is not uncommon for a borderline person in one state of mind to perceive the therapist as godlike in power and virtue, and in another (which may appear a day later) as weak and contemptible.
Not surprisingly, countertransference reactions with borderline clients tend to be strong and upsetting. Even when positive (e.g., dominated by fantasies of rescuing the devastated patient), they may have a disturbing, consuming quality. Analysts in hospital settings (Gabbard, 1986; Kernberg, 1981) have noted that with some borderline patients, staff tend to be either oversolicitous (seeing them as deprived, weak, and in need of extra love to grow) or punitive (seeing them as demanding, manipulative, and in need of limits). Inpatient personnel frequently find themselves divided into opposing camps when treatment plans for borderline clients are discussed (Gunderson, 1984; Main, 1957). Outpatient practitioners may move internally between one position and the other, mirroring each side of the client’s conflict at different times. It is not unusual for the therapist to feel like the exasperated mother of a 2-year-old who will not accept help yet collapses in frustration without it.
SUMMARY
This chapter has given a cursory overview of evolving efforts to describe different realms of character organization. From Kraepelinian distinctions between the sane and the insane, through early psychoanalytic conceptions of symptom versus character neuroses, to taxonomies that emphasize either neurotic-level, borderline, or psychotic-level structure, to characterizing clients in terms of attachment pattern and traumatic influences, therapists have sought to account for the varying reactions of their individual clients to their efforts to be of help. I have argued that the assessment of a person’s central preoccupation (security, autonomy, or identity), characteristic experience of anxiety (annihilation anxiety; separation anxiety; or more specific fears of punishment, injury, and loss of control), primary developmental conflict (symbiotic, separation–individuation, or oedipal), object relational capacities (monadic, dyadic, or triadic), and sense of self (overwhelmed, embattled, or responsible) constitutes one useful dimension of psychoanalytic diagnosis.
SUGGESTIONS FOR FURTHER READING
Phyllis and Robert Tyson (1990) have made a helpful synthesis of traditional psychoanalytic developmental theory through the late 20th century. Two classic books by Gertrude and Rubin Blanck (1979, 1986) have sections on the connection between development and diagnosis. Clinicians who treat children will find Stanley Greenspan’s Developmentally Based Psychotherapy (1997) useful. For a contemporary book connecting recent developmental research with clinical practice, especially with borderline clients, I recommend Affect Regulation, Mentalization, and the Development of the Self (Fonagy et al., 2002), a comprehensive tome that is thankfully available in paperback. For a recent, readable self psychologically influenced account of psychological development, I suggest Russell Meares’s Intimacy and Alienation: Memory, Trauma, and Personal Being (2002).
For a classical exegesis of the difference between neurotic symptom and neurotic character, the chapter on “Character Disorders” in Fenichel’s The Psychoanalytic Theory of Neurosis (1945) is the standard. More recently, Josephs (1992) and Akhtar (1992) have published integrative books that pursue at a more advanced level some of the characterological issues introduced here. For a study in the Kleinian tradition of the clinical implications of different levels of development, Steiner’s Psychic Retreats (1993) is brilliant but may be difficult for beginning therapists.
For classic analytic articles about personality organization, New York University Press has put out fine collections of papers on character neurosis (Lax, 1989), psychosis (Buckley, 1988), and borderline conditions (M. H. Stone, 1986). For a phenomenological appreciation of psychosis, Laing’s The Divided Self (1965) remains unmatched. Eigen’s The Psychotic Core (1986) is difficult but rewarding. Elyn Saks’s (2008) memoir of living with schizophrenia gives a moving yet witty close-up of psychotic experience and also of the potential for individuals with psychotic dynamics, when given good medical and psychological care, to live rich, generative lives.
The literature on borderline conditions is so abundant and diverse as to be overwhelming, but recent contributions by Kernberg and his colleagues (e.g., Clarkin, Yeomans, & Kernberg, 2006) and Fonagy and his colleagues (Bateman & Fonagy, 2004) usefully consider classical formulations in light of recent research and connect their ideas to treatment. For a readable book that values a categorical rather than dimensional definition of borderline psychology and has synthesized a vast amount of research in the tradition of John Gunderson, I recommend Paris’s Treatment of Borderline Personality Disorder (2008).
Since the first edition of this book there has been an explosion of clinical and empirical literature on attachment. The struggles of borderline patients have been described in Wallin’s Attachment in Psychotherapy (2007) and in Mikulincer and Shaver’s Attachment in Adulthood (2007) in terms of severe attachment anxiety. For application of trauma research and theory to the experience of patients who are diagnosed as borderline, Judith Herman’s Trauma and Recovery (1992) is probably the best place to start. See also the suggestions at the end of Chapter 15.