A Search for Common Ground

A Theoretical Evaluation and Integration of Object Relations Theory, Self Psychology, the Psychoanalytic Interpersonal School, and the Atlanta School of Experiential Psychotherapy

Diane Hughes

A dissertation presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy
College of Arts and Sciences, Georgia State University, 1999

Committee: James L. Pate, Ph.D. (Chair) · Robin Morris, Ph.D. · Bernhard Kempler, Ph.D. · Walter A. Pieper, Ph.D.

Download the PDF

Abstract

Four theories of psychotherapy are extensively reviewed and analyzed: the psychoanalytic interpersonal theory, object relations theories, the theory of self psychology, and the theory of the Atlanta School of experiential psychotherapy. Criteria were taken from various philosophies of science to evaluate the conceptual assets and liabilities of each theory with special attention given to the heuristics or methods of gathering data in each theory and to the ontology of each theory which was defined as the assumptions made in each theory about essential human nature, the nature and development of psychopathology, the goals of treatment, the role of the therapist, and the mechanism of change. Theoretical concepts also were compared to results from psychotherapy process and outcome research to identify empirical support for or contradiction of these concepts. An extensive evaluation then was followed by a suggested integration of some concepts from each theory.

Acknowledgments

I acknowledge gratefully the help, support, and patience of my chair: Dr. James L. Pate.

I acknowledge gratefully the assistance, support, and reasonable voice of calmness provided by Dr. Robin Morris.

I acknowledge the concern and involvement of Dr. Bernhard Kempler who has been a source of good advice and support since my undergraduate days.

I similarly acknowledge the concern and support of Dr. Walter A. Pieper, with a special appreciation for his wonderful sense of humor and his willingness to be available when needed.

I acknowledge Dr. Cheryl Glickauf-Hughes for inspiring my interest in this topic and giving me advice about the best ways to proceed with the research.

I acknowledge Dr. Ray Craddick for his always dependable support in my academic career.

I thank my family for their patience with this seemingly interminable endeavor and their willingness to allow me to monopolize the computer.

Sincere thanks to Joey McWilliams and Michael Smith for their patience, kindness, and willingness to be available at night and on weekends to type and edit this document.

Finally, I acknowledge that my good friends, Dr. Mark Goldenthal and Dr. Deborah Liner, were probably quite correct when they tried to convince me that this dissertation topic was impossibly and absurdly grandiose.

Chapter 1: Introduction

Psychotherapy has been described as an art as well as a science. Strupp (1989) wrote that psychotherapy may always remain a practical art but that its operations should be susceptible to corrections and always should be derived from systematic empirical support. Indeed, most practitioners of psychotherapy have some theoretical basis for the interventions that they use in practice (Garfield & Kurtz, 1976; Johnson & Brems, 1991; Norcross & Prochaska, 1982, 1983; Smith, 1982; Zook & Walton, 1989). In a general way, psychotherapy might be defined as the systematic use of interventions based upon theoretical principles, empirical evidence, or both to help an individual or set of individuals change behaviors, thoughts, and/or feelings. A problem is that there is little agreement in psychotherapy about which particular theoretical system is correct or about which systems are most useful (Norcross & Saltzman, 1990; Norcross, Saltzman, & Giunta, 1990; Wolfe & Goldfried, 1988).

As a result of this lack of agreement, practitioners of psychotherapy have in recent years been moving in the direction of the eclectic use of psychotherapeutic techniques (Garfield & Kurtz, 1976; Norcross & Prochaska, 1982; Smith, 1982; Spett, 1983). Some practitioners have integrated two or more theoretical models and use techniques taken from these models. Other practitioners simply use whatever techniques seem useful while making little, if any, attempt to integrate the theoretical models from which the techniques are taken. Researchers have been inspired by the equivalent success rates of various models of psychotherapy to identify the common factors in models that may explain why different models are equally effective bases for therapy (Lambert & Bergin, 1992; Lambert, Shapiro, & Bergin, 1986; Orlinsky & Howard, 1986; Strupp & Hadley, 1979). Clinicians and researchers recently have shown interest in the integration of many theoretical models and in the search for common factors and common grounds that explain equivalent efficacy in different models.

In this chapter, there will be a review of the current trend among practitioners, theorists, and researchers toward the integration of theoretical orientations. There then will be a brief review of attempts to integrate the ideas of psychodynamic theories with behavioral and cognitive-behavioral models (Arnkoff, 1983; Beitman, Goldfried, & Norcross, 1989; Gill, 1984; Goldfried, 1985; Meichenbaum & Gilmore, 1984).

In the next section of this chapter, the theoretical approaches that are the focus of this study will be considered. The theoretical approaches include the psychoanalytic interpersonal approach (Sullivan, 1953, 1954, 1956, 1973), object relation theories (Fairbairn, 1952; Jacobson, 1964; Kernberg, 1976, 1980, 1984; Klein, 1964, 1975; Mahler, 1968, 1972; Mahler, Pine, & Bergman, 1975; Winnicott, 1958, 1964, 1968, 1989), self psychology (Kohut, 1971, 1977, 1984), and the Atlanta school of experiential psychotherapy (Whitaker & Malone, 1953, 1980).

The emphasis in this chapter will be a review of comprehensive attempts to compare, to criticize, and/or to integrate the ideas of these theoretical models (Gedo & Goldberg, 1972; Greenberg & Mitchell, 1983; Pine, 1990; Summers, 1994; Wolberg, 1988/1995). There also will be a brief review of more modest attempts to compare and to contrast specific ideas within each model (Akhter, 1989; Bacal, 1989; Brandshaft, 1989; Bromberg, 1989; Gedo, 1989; Kernberg, 1980; Shane & Shane, 1989; Stern, 1987). Finally, the chapter will be concluded with a statement of the problem to be considered in this research and with a rationale for the study.

The Movement Toward Integration of Theories in Psychotherapy

Researchers have found that between one-third and one-half of clinical and counseling psychologists described themselves as eclectic (Garfield & Kurtz, 1976; Norcross & Prochaska, 1982, 1983; Smith, 1982; Spett, 1983). Garfield and Kurtz (1977) found that psychologists who described themselves as eclectic claimed that no single theory was adequate to explain all psychopathology and that clinicians should select the therapeutic approach of greatest benefit to a given client. Norcross and Prochaska (1982) found that most clinicians who endorsed eclecticism as their primary orientation tended to be older and to have more clinical experience than clinicians who chose one of the major theoretical approaches, such as psychodynamic, behavioral, or humanistic.

In order to understand and to explain the increasing popularity of eclecticism, Norcross and Prochaska (1988) conducted a study in which clinicians who described themselves as eclectic were asked to discuss and to explain their eclecticism. The clinicians were asked to identify their eclectic viewpoint either as synthetic eclecticism or as technical or atheoretical eclecticism. Synthetic eclecticism was defined as conceptual synthesis of diverse theoretical systems including the identification of core ingredients shared by diverse theoretical approaches. Technical or atheoretical eclecticism was defined as the selection and use of treatment interventions from different theoretical systems based upon the demonstrated, empirical efficacy of the chosen techniques. It was found that the majority derived their eclectic conception from the use and integration of two or more theoretical systems and that the majority preferred to describe their approach as synthetic eclecticism rather than as atheoretical or technical eclecticism.

Zook and Walton (1989) provided further support for the growing popularity of eclecticism. Clinical and counseling psychologists were asked to choose from a list of major theoretical positions in which eclecticism was not an option. Subjects were asked to choose theories and techniques that they most frequently used and to rank their top three choices. Over 75% of the respondents chose at least two theoretical approaches, and over 61% selected three. The authors concluded that "a majority of the sample are eclectic to the extent that they do not subscribe exclusively to one orientation" (Zook & Walton, 1989, p. 25). It should be noted that the use of more than one theoretical approach by clinicians does not imply that these clinicians are practicing synthetic eclecticism with integration of the theories chosen. It might be assumed, however, that the choice of two or more theoretical positions does imply at least the use of what Norcross and Prochaska (1988) called technical eclecticism.

Zook and Walton (1989) found that clinical psychologists preferred psychodynamic and behavioral approaches and that the counseling psychologists preferred behavioral and humanistic approaches. Of those individuals who gave three orientations, more than 60% chose cognitive behaviorism as one of the three. Among the clinical psychologists who selected three orientations, more than 64% chose psychoanalysis or psychodynamics as one of the three. A humanistic orientation was the most preferred choice of counseling psychologists and was the fourth most popular choice of the clinical sample. Among the individuals who listed three orientations, more than 26% of the clinical psychologists and more than 54% of the counseling psychologists listed the humanistic orientation as one of their choices.

These studies reflect a growing trend in psychotherapy toward integration of theoretical orientations and a rapprochement between previously diverse clinical theories (Beitman et al., 1989; Norcross & Prochaska, 1988). Participants in a National Institute of Mental Health (NIMH) workshop on research in psychotherapy integration posited a theoretical progression in the field of psychotherapy that may be analogous to a social progression that proceeds from segregation to desegregation to integration (Wolfe & Goldfried, 1988). Norcross and Prochaska (1988) proposed that eclecticism represented a desegregation in the field of psychotherapy in which ideas and methods from diverse theoretical backgrounds are mixed and mingled.

Norcross and Prochaska (1988) hypothesized that psychotherapy now seems to be in transition from desegregation to integration with increasing efforts devoted to finding viable integrative principles for assimilating and accommodating the best ideas, principles, and paradigms from different theoretical systems. Similarly, Beitman et al. (1989), in giving an overview of the movement toward integration in psychotherapy, proposed that "integration denotes the conceptual synthesis of diverse theoretical systems . . . a superordinate umbrella, coherent theoretical gestalt, metatheoretical framework, or conceptually superior therapy" (p. 139).

Six interacting, mutually reinforcing factors have fostered, supported, and promoted the advancement of psychotherapy integration.

1. There has been a proliferation of articles in which the authors have stated that no single theory of psychotherapy has stood out from the others in validity or utility. (Goldfried, 1985; Kazdin & Boss, 1989; Lambert et al., 1986; Laborsky, Singer, & Laborsky, 1975; Orlinsky & Howard, 1986; Strupp, 1987, 1989; Wolfe & Goldfried, 1988).

2. There has been a growing consensus that no single theoretical approach is clinically adequate for all problems and patients (Adams, 1984; Beitman et al., 1989; Johnson & Brems, 1991; Kazdin & Bass, 1989; London, 1988; Norcross & Prochaska, 1988, 1983, 1982; Norcross & Saltzman, 1990; Norcross et al., 1990; Wolfe & Goldfried, 1988; Zook & Walton, 1989).

3. Research has shown an equality of outcomes among the different psychotherapies (Kazdin & Bass, 1989; Kiesler, 1973; Lambert & Bergin, 1992; Lambert et al., 1986; Luborsky et al., 1975; Orlinsky & Howard, 1986; Strupp, 1987; Strupp & Howard, 1992).

4. Equal outcomes lead logically to an inquiry about the common factors in disparate therapeutic traditions that result in similar outcomes regardless of the tradition in use (Beitman et al., 1989; Karasu, 1986; Orlinsky & Howard, 1986; Prochaska, 1979; Strupp, 1987; Strupp & Hadley, 1979; Wolfe & Goldfried, 1988).

5. Research in psychotherapy indicates that the most powerful determinants of therapeutic success lie in the personal qualities of the patient and the therapist and in the interaction between them and, thus, indicates that attention should be paid to personalities and interactions of the patient and the therapist rather than to theoretical predispositions (Beitman et al., 1989; Bergin & Lambert, 1978; Prochaska & Norcross, 1982).

6. Sociopolitical contingencies demand accountability from the practitioners of psychotherapy and promote the search for a unified psychotherapy paradigm to meet most efficiently the needs of the public (Beitman et al., 1989).

A pragmatic perspective has been offered by Adams (1984), who stated firmly and unequivocally that "a comprehensive theory of human behavior is a remote possibility and may be for many years in the future" (p. 93). Adams expressed the idea that, at this stage in the development of clinical psychology, theorists should look for large effects that are clinically rather than statistically significant. He wrote that the theoretical interests of clinical psychologists should shift from being "obsessively concerned with behavior change or therapeutic techniques" (p. 93) to theories about the development and causes of psychopathology.

Norcross and Saltzman (1990) opined that eclecticism or psychotherapy integration is more likely to increase in popularity than is any specific therapy system in the 1990s. They stated that, in addition to the studies already cited (Garfield & Kurtz, 1976; Smith, 1982; Norcross & Prochaska, 1982; Norcross & Prochaska, 1988; Wolfe & Goldfried, 1988; Zook & Walton, 1989), several interdisciplinary organizations committed to psychotherapy integration, including the Society of the Exploration of Psychotherapy Integration, have been formed in the last 10 years. Norcross and Saltzman (1990) emphasized the importance of discovering fundamental convergences and divergences in theories of psychotherapy and the concomitant demand to find and to use a common language for discussing psychopathology and therapeutic interventions.

Wolfe and Goldfried (1988), in reporting on the recommendations and conclusions of the NIMH workshop on research in psychotherapy integration, stated that the single most important step in advancing the goal of psychotherapy integration is the development of a solution to "the language barrier that separates therapy from therapy and researcher from clinician" (p. 450). It was suggested that common everyday English may be most acceptable to practitioners of different orientations. A second candidate for an integrative language was the language of cognitive psychology, which would have the advantage of maintaining a link between psychotherapy researchers and theorists and the broader discipline of experimental psychology.

Wolfe and Goldfried (1988) also reported that participants in the NIMH workshop strongly emphasized the importance of clarification of the different kinds of therapeutic alliances that are associated with different theoretical orientations. It also was recommended that investigations be conducted to determine whether different types of therapeutic tasks require different types of therapeutic alliances or whether a patient's personality or level of distress affects the role that the therapeutic alliance plays in mediating positive change in psychotherapy.

Integration of Psychodynamic Theories and Behavioral Orientations

An overview of recent trends in the integration of psychotherapeutic theories (Beitman et al., 1989) included several important claims. The authors stated that, in a very general way, the major theoretical orientations of psychotherapy have tended to focus on different aspects of individual functioning. Psychodynamic theories focus on awareness and the internal functioning of the mind with all its many cognitions both explicit (conscious) and implicit (unconscious). The humanistic and experiential theories have tended to focus on emotionality and affect, while behavior therapies have focused on action patterns.

According to Beitman et al. (1989), certain specific disagreements among theoretical orientations have been identified and debated as the integration movement has expanded. The authors concentrated, in much of the article, on the differences between psychodynamic and behavioral conceptions, particularly in regard to the differing conceptions of reality, the role of the unconscious, the importance of transference, and the goals of psychotherapy. In brief, the authors reviewed literature in which psychodynamic and behavioral positions were compared and took the stance that the differences between the two schools of thought were either insubstantial or a source of enrichment for both theoretical positions. For example, the authors conceded that the behavioral view of reality emphasizes realism and objectivity in contrast to the psychoanalytic perspective that emphasizes subjectivity and introspection. However, the authors stated that, in general, these weaknesses in philosophy "are precisely what makes psychotherapy integration interesting, in that it brings together the strengths of different orientations" (Beitman et al., 1989, p. 144). It should be noted that integration of theories also has the potential to combine their weaknesses and undermine their strengths. In some case, integration of differing theories of psychotherapy could possibly yield an integrated system less effective than the individual systems were before the integration.

In regard to the role of the unconscious and the importance of transference, Beitman et al. (1989) proposed that recent developments in both psychoanalytic and behavioral perspectives have made these points of conflict of minor importance. Psychodynamic theorists have accepted the importance of conscious thoughts, actions, and environmental factors, and the behavior therapists are integrating cognitive factors deliberately into their systems. Meichenbaum and Gilmore (1984) stated that theorists from all schools of psychotherapy must consider directly or indirectly the patient’s hypothesized cognitive structures.

Similarly, Beitman et al. (1989) stated that several authors have noted that there are definite commonalities in psychodynamic and cognitive behavioral approaches to the use of the therapeutic relationship (Arnkoff, 1983; Goldfried, 1985). From a cognitive behavioral philosophy, the therapeutic relationship offers a sample of the patient's relevant thoughts, emotions, and behavior, offering the therapist an opportunity for an "in vivo" intervention. Beitman et al. (1989) stated that the primary difference between psychodynamic and behavioral approaches to the therapeutic relationship appears to be in the relative emphasis placed on the relationship as an agent of change rather than on its importance or existence.

Integration of Different Models in the Psychoanalytic Domain

Within the psychoanalytic system, there has been a proliferation of theoretical positions "each with a distinct line of conceptual development and its own idiosyncratic language" (Greenberg & Mitchell, 1983, p. 1). In addition to differing languages and terminology, many of these psychodynamic models emphasize the importance of different intrapsychic structures, different developmental stages, and different developmental tasks. However, much of the conflict and controversy may be based on the choices each theorist made about what was important and what was not. Many apparent disagreements are not substantive but rather are different descriptions of similar phenomena or different perspectives about a common phenomenon as found in the classic metaphor of the blind men encountering an elephant (Greenberg & Mitchell, 1983).

In the following section, previous attempts to compare, to contrast, and to integrate those theoretical positions that have been identified as the focus of this research will be examined. The ways in which this study might make a unique contribution in the field of psychology may be determined. It should be noted that there have been other attempts to compare and to integrate psychotherapeutic theories that are beyond the scope of this study.

Attempts to Evaluate and to Compare Psychoanalytic Models

Several theorists have made attempts to review psychoanalytic models comprehensively for various purposes. Gedo and Goldberg (1973) attempted to integrate recent innovations with the traditional model. Greenberg and Mitchell (1983) examined the progression and evolution of the concept of object relations within differing theoretical models in the psychoanalytic domain. Wolberg (1988/1995) summarized and criticized all of the theoretical models that are discussed in this study. Pine (1990) wrote primarily about the clinical application of different psychoanalytic models. Summers (1994) used his review and analysis of object relations theories, self psychology, and the psychoanalytic interpersonal school to support his conclusions about the importance of object relations in the development of intrapsychic structure and in the treatment of psychopathology. Other authors have concentrated their comparison and critiques on specific theorists; for example, Aktar (1989) compared Kohut's model with the model presented by Kernberg.

Comparisons of Theories

A Comparison of the Interpersonal and Object Relations Models

Stern (1987) compared interpersonal psychoanalytic theory with object relations theories. He argued that the interpersonal analyst and the analyst operating from an object relations viewpoint are essentially participating in very similar treatment strategies but are using different language to describe very similar processes. Stern contended that the object relations theorist invents unnecessary and cumbersome abstract ideas such as projection, introjection, and transitional relatedness. Rather than analyzing hypothesized, abstract complex processes such as projective identification, Stern claimed that the interpersonal analyst is concerned with how one person treats another.

Stern (1987) also argued that the object relations analyst and the interpersonal analyst have different ideas about the role of the analyst and the mechanism of change. He stated that object relations oriented clinicians such as Winnicott (1958, 1965) and Kohut (1971, 1977) assign a major share of the mechanism of therapeutic change to the relation with the therapist. "The analyst provides the nutritive atmosphere within which the patient can resume growth at the point at which it was arrested earlier in life, and the analyst's role is to be of use to the patient" (Stern, 1987, p. 75).

In contrast, according to Stern (1987), the interpersonal analyst is not interested in providing any type of relation but rather in clarifying the relation that does develop. The mechanism of change for the interpersonal analyst lies in developing with the patient a joint interest in observing and understanding the therapeutic relationship. It should be noted that, despite Stern's (1987) inclusion of Kohut with object relations theorists, Kohut considered himself and is also considered in this study to have supported the theory of self psychology.

A Comparison of the Interpersonal Model with Self Psychology

Bromberg (1989), an interpersonal analyst like Stern (1987), offered criticisms of self psychology that are similar to Stern's criticism of object relations theories. Bromberg contended that the critical clinical difference between the interpersonal and the self psychology models is that the self psychology model proposes an analytic stance, the introspective/empathic stance, that is believed to be "correct" for all patients. By contrast, the interpersonal model is based on the idea that what is crucial to the analytic experience is the unique phenomenological experience of the interaction between the specific individuals participating in the analytic experience. Bromberg asserted, for example, that there is no need for the concept of empathic failure because the analytic process is always vacillating between empathy and anxiety, between understanding the patient and failing to understand.

Comparisons of Different Object Relations Theories

Greenberg and Mitchell (1983) made a deliberate effort to examine the concept of object relations as described and developed in different psychoanalytic theories including the theories of Freud, Sullivan, Klein, Fairbairn, Winnicott, Mahler, Jacobson, Kernberg, and Kohut. The authors stated early in the book that the central problem in all psychoanalytic theory has been to accommodate the importance of human relations within the context of Freud's original conceptual model based on drives as the basis for understanding all human behavior.

Greenberg and Mitchell (1983) explained that much of the complexity and diversity of current psychoanalytic thought could be clarified by using their approach of evaluating the strategy taken in each divergent theory to understanding object relations. They stated that "Every major psychoanalytic author has had to address himself to this issue, and his manner of resolving it determines the basic approach and sets the foundation for subsequent theorizing" (p. 4). Greenberg and Mitchell's (1983) ideas will be examined in greater depth in Chapters 5 and 8. It will be noted at this point that these authors proposed that the understanding and clarification of the concept of object relations have great value in any effort to achieve synthesis and integration of different theoretical positions.

Kernberg (1980) compared and contrasted his particular theoretical orientation with the positions taken by Klein (1964, 1975), Fairbairn (1952), Mahler (1968, 1972; Mahler et al., 1975), and Jacobson (1964). He specified the contributions made by each of these theorists as well as his disagreements with their theories. Kernberg’s acknowledgments and disagreements with other object relations theorists will be examined in Chapter 5.

Comparisons of Self Psychology and Object Relations Theories

Brandshaft (1989) stated that Kohut (1971, 1977, 1984) and Klein (1964, 1975) disagreed about several major theoretical concepts. First, Kohut stated that empathy or understanding the patient's subjective experience was of far greater importance than interpretation as emphasized by Klein. Second, Kohut disagreed with Klein about infantile destructiveness and aggression. He claimed that rage was a response to traumatic frustration rather than an innate, primary psychological given.

Brandshaft (1989) gave few specific comparisons of the ideas of Fairbairn (1952) and Kohut (1971, 1977, 1984) except to state that Kohut's ideas logically followed from Fairbairn's. He indicated that Fairbairn's ideas were refreshing and stimulating but limited Although Fairbairn recognized the need for gratification from early objects, he did not grasp the complexity of the developing self structure and its relation to object ties. Furthermore, he stated that Fairbairn, like Klein (1964, 1975), did not appreciate the importance of the empathic stance of the analyst.

Bacal (1989) stated, as did Kernberg (1980), that Winnicott (1958, 1965, 1971) did not develop a comprehensive theory of object relations and did not found a school of psychoanalytic thought. However, Bacal claimed that Winnicott's ideas were of major importance in the understanding of early child development. Bacal also contended that many of Winnicott's ideas anticipated ideas later elaborated in self psychology by Kohut (1971, 1977, 1984). Winnicott's concept of a subjective object is very similar to Kohut's concept of a "selfobject." Both Winnicott and Kohut also emphasized the importance of "mirroring," which both defined as a recognition of the individual's uniqueness and creative capacity. Winnicott also proposed that "good enough" mothering arose from the state of primary maternal preoccupation, which enhanced the mother's empathizing with her baby and meeting its needs. Bacal contended that this proposal is very similar to what self psychologists call optimal responsiveness to the infant's selfobject needs. Finally, Bacal stated that Kohut's conception of healthy and unhealthy development of the self is very similar to the ideas Winnicott proposed about the development of the true and the false self.

Akhtar (1989) offered a comparison of the theoretical positions of Kohut (1971, 1977, 1984) and Kernberg (1975, 1976, 1980, 1984). He stated that perhaps the most important theoretical difference between them concerns their positions on fundamental human nature.

For Kohut, man is born whole, full of potential, even happy and eager to joyfully actualize the blueprint of his destiny. If he is unhappy, it is because of environmental failure. All his conflicts are the end result of unfortunate, tragic disorganization caused by lack of parental empathy. For Kernberg, conflict is embedded in normal development. Lifelong struggle with intrapsychic and reality conflicts is unavoidable. There is no escape from aggression both from within and from outside. Life, comprising the constant reactivations of the infantile conflicts as well as renewed challenges posed by reality is, however, still interesting and possesses the potential for that greatest of human experiences, love. However, even love can never be totally free of early transferences. (pp. 356-357)

Shane and Shane (1989) compared and contrasted the theories of Kohut (1971, 1977, 1984) and Mahler (1968, Mahler et al., 1975) in the context of research on the normal development of infants. They identified one major theoretical difference between Kohut and Mahler and several that seem less important. The major difference of opinion concerns Mahler's emphasis on the importance of intrapsychic separation from the mother and the attainment of autonomy. Kohut contended that the caregiver supplies primitive selfobject functions that, through the process of maturation, are gradually taken over by the self through the development of self structure by means of transmuting internalization. Kohut postulated a continuing need for selfobject experiences throughout life.

Shane and Shane (1989) contended that infant research and observations indicate that Kohut's (1971, 1977, 1984) position is probably more accurate than is Mahler's (1968, Mahler et al., 1975) position, in that the world of the infant is a very social world and that the normal person from birth to death is never completely autonomous. The individual is "always reliant for self sustenance on the internal, psychological presence of an other or on the external presence of an other" (p. 408). Autonomy from and interdependence with others are both required for adequate adaptation, and such adaptation differs from one culture to another.

Shane and Shane (1989) also noted that Mahler (1968, Mahler et al., 1975) and Kohut (1971, 1977, 1984) differ in the emphasis given to the development of the self rather than an emphasis on the object relations. Finally, they stated that Mahler was faithful to the classical dual drive theory in which aggression is considered to be inborn and not dependent on environmental frustration for activation. Kohut argued that aggression is solely a reaction to frustration.

Efforts to Integrate Theories

Gedo and Goldberg (1973) made an attempt to integrate the traditional psychoanalytic model of the tripartite intrapsychic structure with the models that emphasize the importance of the development of self structure and/or object relations as part of the intrapsychic structure of the individuals as have been proposed by Jacobson (1964), Kohut (1971), Winnicott (1958), and Modell (1968). The authors particularly focused on Kohut's ideas and accepted his proposal that narcissism has a separate line of development from the libidinal development proposed in traditional analytic systems.

Gedo and Goldberg (1973) suggested an hierarchical model of the development of intrapsychic structure with different processes taking precedence at different developmental stages. Thus, the first intrapsychic task of infancy was described as the cognition of self and object. As the child matures, the next task was the development of a cohesive self, followed by the formation of the superego. The authors argued that the ideas of Kohut (1971), in particular, and, to a lesser degree, Jacobson (1964), Winnicott (1958), and Modell (1968) were important in understanding psychopathology that originated in infancy and very early childhood long before the development of the superego.

Greenberg and Mitchell (1983) were primarily concerned with fully explaining the development of what they labeled the "relational/structure model" (p. 402) of psychoanalytic thought. They compared and contrasted the relational/structure model with the “drive/structure model” (p. 402) and stated in their conclusions the strong suspicion that the two types of models could not be reconciled or integrated. The drive/structure model as originally developed by Freud is a theory that focuses on the human individual as separate from others. The primary focus in the relational/structure theories discussed by Greenberg and Mitchell was on understanding the particular vision of the role in human life of relations with other humans in each theoretical formulation considered. Thus, the authors thoroughly examined interpersonal psychoanalysis, object relations theories, and self psychology, but they did not try to achieve any theoretical integration.

Pine (1990) attempted to demonstrate that concepts from drive theory, ego psychology, object relations theory, and self psychology were all clinically useful in explaining different aspects of human function and experience. He stated that he was interested in the description and exploration of the domains of (a) the drives, urges, and wishes described in drive theory; (b) the defenses, adaptations, reality testing, and defects in the development of each described in ego psychology; (c) the relationships with significant others as experienced and as carried in memory with whatever attendant distortions such experiences and memories may entail as described in object relations theories; and (d) the subjective experience of self in relation to such phenomena as boundaries, esteem, and authenticity as described both in self psychology and in some object relations theories. He viewed interpersonal relations not as a separate theoretical system but as one of the domains, along with the intrapsychic, in which the other four human forces are manifested. He stated that each of these four aspects of human functioning have been the subject of serious and comprehensive theories.

Pine (1990) stated that he thought that the task of reconciling or integrating the four theories of drive, ego, object relations, and the self would be too cumbersome to undertake and possibly impossible to achieve. His focus was on examining these differing theories as alternative narrations or interpretations of any given individual's life story. He argued that the phenomena of each of these four theoretical systems become apparent in any given analysis and are identified by interpretation. Psychoanalysis, thus, has been enriched by the complex and multifaceted view of human functioning provided by all four systems. When keeping all of these perspectives in mind, the analyst views psychoanalysis as the psychology of conflict, repetition, and development with an understanding of developmental delays and aberrations as a part of the substance of psychopathology.

Pine (1990) provided a review of the development of psychoanalytic theories. The remainder of his book was focused upon the use of these different theoretical systems within clinical practice including a significant number of case studies and clinical examples in which the therapist used concepts from each theoretical system to understand and to help specific patients.

Summers (1994) provided a comprehensive examination and critique of all of the major object relations theorists including Fairbairn (1954), Klein (1964, 1975), Winnicott (1958, 1965, 1971) and Kernberg (1975, 1976, 1980, 1984). He also examined, compared, and contrasted the theories of Kohut (1971, 1977, 1984) and the psychoanalytic interpersonal school (Levenson, 1972, 1983; Sullivan, 1953, 1956, 1972). His summaries and critiques were quite specific and are best left to be considered in each of the later sections of this study.

It is of interest, however, that Summers (1994), in disagreement with Kernberg (1980) and Bacal (1989), argued that Winnicott (1958, 1965, 1971), indeed, had developed a complete and comprehensive theory of early childhood development based upon principles of the development of object relations. He also stated that, although Winnicott did not elucidate a comprehensive theory explaining psychopathology and its treatment, he nevertheless provided valuable insights into both.

Summer's (1994) conclusions were that the theoretical paradigm offered by object relations theories was a successful replacement for drive and ego psychology. He dismissed the efforts of Kernberg (1975, 1976, 1980, 1984) to integrate drive theory with object relations and rejected the importance of libidinal and aggressive drives.

Summers (1994) stated that the curative factor in psychoanalysis is analytic understanding, including its successes and failures, which creates and sustains the analytic relationship that provides the foundation for the creation of new psychological structure. The therapeutic efficacy of the analyst depends upon insight within the context of his or her ability to serve as the instrument of a new type of relationship for the patient without the expectations and frustrations of past relationships. The responses of the therapist thwart the patient's expectations based on past experiences and promote a new, more benign, and productive relationship. This new type of relation provided in the analytic setting is the context within which the patient can develop insight into previously unconscious expectations and patterns. The relation between patient and analyst was considered by Summers to be integral to the therapeutic action of psychoanalysis.

Summers (1994) concluded that the intrapsychic structure of the individual was the product of the experience of early attachments in the individual's history. Psychopathology is the result of defects derived from the absence or malignant transformation of such attachments with resulting problems in the internalization of healthy, self-sustaining attachments. The clinician's task was described by Summers (1994) as one of "connecting with the patient in a new way, contacting an unengaged part of the self, confronting the resulting annihilation anxiety, and thereby forming a new type of attachment that eventuates in more functional object relations structures" (p. 379).

Summary

As can be determined from this brief overview of the literature in which theorists have made attempts to compare or to integrate theories, the only truly comprehensive attempts are the ones made by Greenberg and Mitchell (1983), Wolberg (1988/1995), Pine (1990), and Summers (1994). In the other articles, the authors restricted themselves to examining and to comparing at most two or three different systems of thought. However, in all of these recent attempts to examine and to compare theoretical positions, the authors have suggested that some seemingly great theoretical differences are at least partly due to the previously mentioned idiosyncratic languages used in different theoretical systems. Furthermore, each attempt to compare and to integrate theoretical systems has led to greater clarity of the true underlying postulates of each position and identification of commonalities as well as differences.

Statement of the Problem

Any attempt to compare, to synthesize, or to integrate various theoretical systems is a service for all practicing clinicians (Adams, 1984; Beitman et al., 1989; Norcross & Prochaska, 1988; Norcross & Saltzman, 1990; Wolfe & Goldfried, 1988). This study is an attempt to achieve a comparison and synthesis of several theoretical positions that have some important theoretical commonalities. The theories of psychotherapy that are to be considered in this research are object relations theory, the psychoanalytic interpersonal school, self psychology, and the Atlanta school of experiential psychotherapy.

These theoretical schools of thought share three common fundamental assumptions. (a) The unconscious is profoundly involved in the development and resolution of psychopathology. (b) The relation between the patient and therapist is vital to the identification of both the healthy and pathological functioning of the patient, and what transpires in the relationship is an integral part of the treatment. (c) Psychopathology originates developmentally, usually from experiences in early childhood (Detrick & Detrick, 1989; Fairbairn, 1952; Felder, 1967; Felder & Weiss, 1991; Gantt, 1984; Greenberg & Mitchell, 1983; Jacobson, 1964; Kernberg, 1975, 1984, 1987; Klein, 1975; Kohut, 1971, 1977, 1984; Sullivan, 1953, 1956, 1972; Whitaker & Malone, 1953/1981; Winnicott, 1958, 1965, 1971, 1988, 1989; Wolf, 1988).

The Atlanta school of experiential psychotherapy is included in this investigation for two reasons. Experiential psychotherapy has been described as one of the more comprehensive and influential theoretical models of the process and dynamics of psychotherapy within what is generally considered to be the humanistic school of thought (Felder, 1967; Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981). The theory of experiential psychotherapy shares the three previously described commonalities with the psychoanalytic theories, that is, the importance of the unconscious, the emphasis on the therapeutic relationship, and the understanding of the developmental nature of psychopathology. The understanding of the developmental basis of psychopathology is very strongly implied in the theory of experiential psychotherapy, but the stages and processes of normal development were not specifically described nor discussed. Whitaker and Malone (1953/1981) and Felder and Weiss (1991) explicitly stated that psychopathology originates in childhood experience and is learned in the family of origin.

In this research, the objective was to identify and to clarify the underlying assumptions and postulates of object relations theory, self psychology, the psychoanalytic interpersonal school, and the Atlanta school of experiential psychotherapy. This study is the first attempt to integrate the theoretical position of the Atlanta school of experiential psychotherapy with psychoanalytic theory.

There are four major hypotheses in this investigation.

(1) Many apparent differences and disagreements among theoretical schools of thought are based on language.

(2) When disagreements are found to be substantive, there will be instances in which the empirical evidence of psychotherapy outcome research will provide support for one system or another.

(3) When psychotherapy outcome research does not clearly support one theoretical system over another, there will be instances in which the criteria for the evaluation of scientific theories from the philosophy of science will support the robustness of one theory over another.

(4) Some disagreements will remain unresolved and will provide ground for further research and investigation.

The general format of this inquiry will be as follows. There first will be an examination of some tenets of the philosophy of science concerning methods of theory evaluation. Particular attention will be given to theory evaluation in the science of psychology. There then will be an identification of the specific theoretical issues that will be considered in the comparison of theories in this investigation.

The psychoanalytic interpersonal theorists will be Sullivan (1953, 1956, 1972), Levenson (1972, 1983), Chrzanowski (1982), Anchin (1982), Kiesler (1982), and Stern (1987a, 1987b). Object relations theorists will include Klein (1964, 1975), Fairbairn (1952), Winnicott (1958, 1964, 1971, 1988, 1989), Jacobson (1964, 1971), Kernberg (1975, 1976, 1980, 1984), and Mahler (1968, Mahler et al., 1975). Experiential theorists will be Whitaker and Malone (1953/1980) and Felder and Weiss (1991). Theorists from self psychology will include Kohut (1971, 1977, 1984), Goldberg (1988), and Wolf (1988).

There will be an identification of the underlying assumptions of each theory and an identification of commonalities and differences among the theories. After identification of substantive differences, empirical findings of outcome research in psychotherapy will be examined in an effort to determine whether the data support some theoretical positions over others. Any substantive differences unresolved by empirical findings will be evaluated according to criteria identified in the philosophy of science for evaluating the robustness of scientific theories. Attempts will be made to integrate theoretical models. Finally, there will be a summary of findings and suggestions for further research.

Chapter 2: Methodology: Evaluation of Theories

In this research, theories of psychotherapy were evaluated and compared. Therefore, the philosophy of science, with particular attention given to current theories and ideas that can be used to evaluate scientific theories, will be reviewed. The philosophers of science to be considered are Kuhn (1970), Lakatos (1978), and Laudan (1977, 1984, 1990). The phenomonological approach in philosophy as a basis for understanding theories of psychology and psychotherapy (Dreyfus & Wakefield, 1988; Grunbaum, 1984; Sass, 1988; Woolfolk, Sass, & Messer, 1988) also will be discussed. Three other philosophies of science that may aid in understanding and evaluating theories of psychology and psychotherapy have been identified by psychologists. Adams (1984) discussed the value of empirical positivism in evaluating psychological theories; Manicas and Secord (1983) similarly discussed the applicability of the realist theory of science; and Mulaik (1984) proposed that pragmatism is useful in evaluating theories of psychology and psychotherapy. The recommendations of the National Institute of Mental Health workshop on research in psychotherapy integration (Wolf & Goldfried, 1988) will be described. The relevance of ideas from the philosophy of science to the evaluation of theories in psychology, psychotherapy, and psychoanalysis will be discussed. Finally, the methodology used in this research will be described explicitly.

Current Ideas in the Philosophy of Science

Gholson and Barker (1985) offered a complete review of the philosophy of science concerning the evaluation of scientific theory and applied these ideas to a consideration of theoretical progress in psychology. They examined the ideas of Kuhn (1970), Lakatos (1978), and Laudan (1984, 1990) and explained how these differing perspectives from the philosophy of science applied to the consideration of theoretical ideas in psychology.

Kuhn

Kuhn (1970) posited a theory of scientific revolutions in which he proposed that scientific paradigms are so fundamental to a scientific discipline that they are immune from empirical testing. Paradigms were defined as unique combinations of ontology, epistemology, and methodology. Paradigms determine the way scientists make sense of the world and are adhered to rigidly until the data and data-based philosophy force a cataclysmic destruction of one paradigm and its replacement with an entirely new paradigm that organizes the theory of the scientist and provides him or her with a new set of ideas about ontology, epistemology, and methodology. A classic example of Kuhn's theory of scientific revolutions was the replacement of the physics of Newton with the theory of relativity proposed by Einstein (Gholson & Barker, 1985).

Kuhn's (1970) ideas included the hypothesis that scientific revolutions are not related to the empirical validity of the data but to the social climate and sociology of the scientific community. Kuhn also concluded that scientific paradigms were incommensurable. There is no common basis for comparing one paradigm with another. This led to the idea that scientific ideas are not based in rational or empirical evaluations of data but instead are based on cultural and social evaluations of what constitutes the appropriate basis of scientific study.

Since the publication of Kuhn's (1970) Structure of Scientific Revolutions, some philosophers, scholars, and scientists have written extensively about what is called epistemic relativism, especially within the social sciences and the humanities. Essentially, many individuals interpreted Kuhn's ideas of paradigmatic revolutions within scientific disciplines as evidence that science can be regarded as a cult in which scientific beliefs are largely based upon the culture of the science that produced those beliefs. Some individuals further assumed that the beliefs of the culture of science are essentially based upon the beliefs of the society within which the scientists reside. Thus, any scientific theory is as accurate and relevant as any other because all scientific theories are entirely subjective (Adams, 1984; Gholson & Barker, 1985; Laudan, 1990; Manicas & Secord, 1983).

Philosophers who represent several different schools of thought oppose and repudiate the ideas of scientific relativism. Laudan (1990) wrote that many of his fellow philosophers of science define epistemic relativism as "one of those episodic cultural sillinesses that will wither and die of its own accord" (p. viii). Laudan criticized the relativistic ideas of the supposed incommensurability and underdetermination of scientific theories.

The displacement of the ideas that facts and evidence matter by the idea that everything boils down to subjective interests and perspectives is–second only to American political campaigns–the most prominent and pernicious manifestation of anti-intellectualism in our time. (p. x)

It also should be noted at this point that this theory of scientific relativism has been repudiated by Kuhn, himself. However, his ideas have been embraced and extended by other theorists, particularly in the social sciences, as support for a relativistic philosophy of science, which depends upon cultural and social support of scientific endeavors and accepts the incommensurability of paradigms (Gholson & Barker, 1985).

Lakatos

According to Gholson and Barker (1985), Lakatos (1978) repudiated and replaced the Kuhnian theory of paradigms with a philosophy that science involves an entity called a "research program" in which a succession of theories is postulated. These theories hold in common a shared commitment to a hard core of data and hypotheses. Any acceptable new theory must accommodate the data and successes of its predecessors and must explain the data that brought the earlier theory into question. In addition, a new theory also leads to new predictions that can be verified experimentally. A new theory is theoretically progressive if it leads to new predictions and is empirically progressive if some of the new predictions have empirical support.

Lakatos (1978) postulated that a new theory could be progressive in one era, could degenerate in another, and then could return to being progressive in later times. Lakatos assumed that the simultaneous existence of several research programs was the norm in the history of science. Although Kuhn hypothesized that the replacement of one paradigm with another did not indicate progress, Lakatos posited that the preference for any research program over another was based upon empirical progress in the preferred research program.

Laudan

Laudan (1977, 1984) replaced Lakatos's research programs with a super theoretical entity that he called a research tradition. A research tradition consists of a family of theories sharing a common ontology and methodology, both of which are capable of change as a research tradition evolves. Laudan also proposed a much wider variety of factors to be used in the evaluation and appraisal of theories. In addition to empirical factors, Laudan explicitly identified conceptual factors as important in theory evaluation, independent of a theory's experimental success or failure.

Lakatos (1978) claimed that there are core commitments in theories, which pass unchanged through successive theories. Laudan (1977) stated that this requirement is unrealistic for several reasons. First, it has proven impossible to locate core principles of the sort required by Lakatos in some of the crucial episodes in the history of physics. Second, in some research programs, the core commitments change over time. In learning theory for example, there have been clear examples of changing cores, the change from probabilistic models that predict incremental learning to models that predict all-or-none learning (Gholson & Barker, 1985).

Laudan (1977) observed that core principles of theory do not pass unchanged through successive theories in a program. Some continuity is helpful, but no element of a research tradition is so essential that it cannot be changed. For Laudan, the principles of the core are not fundamentally metaphysical. They can be modified in response to empirical testing.

Laudan (1977) also recognized a class of metaphysical propositions that are uniquely associated with a research tradition at any given time. The metaphysical commitments may change as a program develops. However, while particular commitments are in operation, researchers are inhibited from the construction of theories incompatible with the ontology of those commitments. These principles need not be and often are not stated explicitly. Indeed, these metaphysical ideas about the ontology of a research tradition may be identified sometimes only through extensive philosophical analysis.

Laudan (1977, 1984) also recognized the importance of conceptual factors in evaluating the success of a theory. A theory and the research tradition to which it belongs may have assets and liabilities that are independent of empirical measures of progress. Laudan identified several conceptual problems that may lead a theory into difficulty. First, he noted that degenerating programs frequently produce overly complex theories. Another conceptual problem, common in psychology, occurs when a theory contains and depends upon terminology that has been inadequately defined. An example is the work of Piaget in which central concepts such as equilibrium, assimilation, and organization have never been precisely defined, and Gholson and Barker (1985) noted that the work of Piaget seemed to have entered a period of stagnation.

According to Laudan (1990), philosophers of science who oppose the ideas of epistemic relativism are those who support empirical positivism, the realist ontology of science, or pragmatism. Philosophers within empirical positivism propose that a scientific theory may be taken to be accurate if it is supported by data that have been collected by using agreed upon and correct scientific methods (Adams, 1984; Laudan, 1983). Philosophers within the realist ontology of science posit the idea that models of science function successfully when they approximate in some way the structure of the object under consideration (Laudan, 1990; Manicas & Secord, 1983). According to Laudan (1990), the pragmatist makes the assumption that the meaning of scientific theories lies in their observable consequences and that progress is made in science as knowledge becomes more reliable. Mulaik (1984) wrote that the pragmatist argues that all claims to knowledge are subject to possible revision depending upon their success in organizing experience. These ideas will be examined in greater detail in subsequent sections of this chapter.

Phenomenonology or Ontological Hermeneutics

Some psychologists and philosophers have argued that human psychology can be understood best within the philosophical framework of existential phenomenology or ontological hermeneutics. Ontology is the branch of philosophy that studies "being" or "what is.” Hermeneutics may be defined as the science and methodology of interpretation. Thus, ontological hermeneutics may be defined as the study of possible interpretations of being (Dreyfus & Wakefield, 1988; Grunbaum, 1984; Sass, 1988; Woolfolk et al., 1988).

According to the concepts of ontological hermeneutics, when the subject matter of science is humanity rather than the observable physical universe, then the subject matter itself consists of meanings and intentions rather than nonintentional events. The human sphere is different from the domain explained by the physical sciences. There is a dichotomy between explanation and understanding. Science may explain nature but can only hope to understand human experience (Woolfolk et al., 1988).

Therefore, the social sciences must use methods for collecting data that are radically different from those used in the physical sciences. As hermeneutics have developed, followers of hermeneutics have decided that the fundamental data of the human sciences are meanings, intentions, plans, goals, and purposes (Woolfolk et al., 1988).

Fischer and Fischer (1983) claimed that phenomenology and existentialism arose as a reaction against twentieth century physicalism and against the social sciences’ modeling themselves after the physical sciences. They stated that, in recent years, most psychologists and clinicians have rejected the model of empirical positivism. Psychologists and clinicians, however, have not developed a philosophy of science that explicitly and systematically accounts for the differences of human subjects from other objects of nature. Many of the ideas of humanistic psychology have been absorbed into the practice of many clinicians without explicit revision and integration on a theoretical level.

Fischer and Fischer (1983) stated that phenomenology arose out of an effort to capture the unity and relational character of consciousness. The phenomenological method is characterized by an effort to describe and to understand the essence of a phenomenon and is used when the researcher seeks to understand rather than to demonstrate, to explain, or to assume. The phenomenological researcher attempts to suspend judgment and to relinquish preconceptions about the phenomenon under investigation and recognizes the fundamentally unique perspective of any individual's experience of the phenomenon.

Because human access to phenomena is always perspectival, the nature of reality always remains, at least partially, ambiguous. The phenomemological research method provides a framework for the social sciences that is deliberately and explicitly designed for the varieties of human experience and provides a viable alternative to the theoretical framework of the physical sciences (Fischer & Fischer, 1983).

The phenomemological research method is founded on the assumption that all knowledge is fundamentally co-determined by the "knower" and the object being "known." Because the perceptions of an object can never be completely exhausted, all knowledge is necessarily incomplete. When the subject matter under investigation is human perception, experience, consciousness, or behavior, its fundamentally relational structure leaves the phenomenological researcher no choice but to respect its essential ambiguity (Fischer & Fischer, 1983).

Phenomenological research yields results in the form of descriptions of the structures of the world as experienced by the subjects of the investigation. Such descriptions of structures disrupt preconceptions but, ironically, often seem upon reflection to be familiar as if these structures were already known. Finally, "phenomenological research is appropriate when we want to know the whatness of a phenomenon, that is, the way people live it" (Fischer & Fischer, 1983, p. 501).

Fischer and Fischer (1983) concluded that phenomenological psychology offers a framework for understanding the unique aspects of human experience and a methodology for conducting research into usually unstudied phenomena. What is required next is an agenda of studies designed to study systematically the body of knowledge on usual and unusual ways of being, on growth phenomena, and on what actually occurs in the process of psychotherapy. The authors wrote:

When more of these “whatness” studies become available, then the agenda will call for systematic comparison and integration with the psychotherapy research that had been conducted within the natural science tradition. (p. 503)

Sass (1988) wrote that ontological hermeneutics is a philosophical framework congruent with the unique characteristics of human experience and free of the positivism, mechanism, and reductionism of the nineteenth century physical sciences. He stated that the phenomenological approach is an attempt to discover the domain of a body of knowledge about human experience. This claim about the essential validity of internal experience constitutes a philosophical position that can be described as subjectivism. Human reality is founded upon the inner experiences of the individual, which are considered to be the foundation for human understanding of the world.

Sass (1988) stated that human beings are constituted subjectively with their own self interpretations. For the most part, these interpretations are not unique, freely chosen, or consciously recognized because they are derived from and embedded in the determining facts of language, culture, and history and are so pervasive as to be almost invisible.

Empirical Positivism

Adams (1984) proposed that the major purpose of theory in science is to provide explanations of phenomena, to organize facts in a scientific field, and to stimulate research which generates new knowledge that may lead to the enrichment of the theory that generated the research. Adams noted that the recalcitrance of psychologists in adhering to unfounded clinical theories is largely due to the inadequacy of traditional research methods in investigating the issues of concern to clinicians. However, he wrote that the possibility that theories can be refuted by data is a major prerequisite for the development of science.

Adams (1984) claimed that several major steps in the founding of an accurate and valid philosophy of science in the field of psychology would change clinical psychology from a collection of cults to a system of scientific inquiry and application. First, there would be an attempt to adhere to a scientific approach to phenomena either through a return to empirical positivism or to the more recent realistic model of science as described by Manicas and Secord (1983). He stated that a critical attitude toward theories would provide assistance in eliminating misconceptions and ignorance.

Second, a major goal would be the development of methods to evaluate theories. Psychology, particularly clinical psychology, is in great need of developing classification systems and measurement procedures that are independent of theoretical orientations. Third, there would be the establishment of an objective evaluation of various theories and/or techniques to determine which hypotheses are most feasible. Fourth, there is a need to abandon the hope that a comprehensive theory of human behavior and experience is possible in the near future. At the present time, theories that can explain specific phenomena and can be evaluated with evidence about these phenomena are needed (Adams, 1984).

Fifth, Adams (1984) noted that statistically and mathematically substantiated studies have done little to advance knowledge in the domain of clinical psychology. He wrote that, at this stage of development, clinical psychology would be better served by searching for effects that are clinically significant rather than statistically significant. Finally, Adams concluded that, for clinical psychology, the interest of the scientist should shift from an interest in therapeutic technique or behavioral change to an interest in theories of development. According to Adams, effective intervention into an individual's problems must be based on an understanding and knowledge of the etiology and nature of the specific problems presented rather than a belief in the efficacy of a technique or a group of techniques described by behavior therapists or psychoanalysts as a universal palliative of all types of human misery.

The Realist Theory of Science

Manicas and Secord (1983) proposed that a realist theory of science has been developed in recent decades as an alternative theoretical framework that avoided the deficiencies of empirical positivism and the relativist paradigmatic conceptions. The authors wrote that this new conception has profound value and implications for the social sciences.

The scientific theory discussed by Manicas and Secord (1983) has been called the realist theory of science, transcendental realism, and fallibilist realism. Supporters of this theory argue that all knowledge is a social and historical product. Epistemologically, there can be nothing known to which our ideas and theories can accurately and specifically correspond. However, it is still precisely the aim of science to invent theories that attempt to represent the world. The practices of science generate their own rational criteria, which are used to accept or to reject theory. It is possible for these criteria to be rational because, from a realist viewpoint, there is a world that exists independently of our "cognizing experience" (p. 401). Because our theories are a construction of the world based on experience, the theories may be wrong, but they have a realist basis. The authors stated that one must be a realist ontologically to be a fallibilist epistemologically.

The realist position rejects the Humean analysis of causation and lawfulness. Instead of the usual characterization of R_i = f(S_i) in which R is the dependent variable that is some function of the independent variable S, the realist position is that "if S_i, then ceteris paribus, R_i necessarily acts in a certain fashion because of its nature" (Manicas & Secord, 1983, p. 402). Scientific laws in this non-Humean framework are not about events or classes of events connected by laws or probabilities but are about the causal properties of structures that exist and operate in the world.

The important distinction in the above paragraph lies in the phrase or clause "ceteris paribus," which means literally other things being equal. In the realist position, there must be a theory about the causal properties of the structure that describes conditions for the ceteris paribus clause to be satisfied. However, from the realist point of view, the world is such a stratified complex of structural properties and structured processes that it is unusual to be able to specify the conditions of ceteris paribus. It is not possible to know that all other things are equal (Manicas & Secord, 1983).

Manicas and Secord (1983) wrote that from the basis of the realist approach there are few scientific laws that may be used to predict events. The purpose of scientific activity is not prediction but explanation. From the view of empirical positivism, "the world is a determined concatenation of contingent events; for the realist it is a contingent concatenation of real structures" (p. 403). The past may be explained, but the future is not determined because the complexly related structures and systems of the world are constantly in a state of flux and reconfiguration.

Manicas and Secord (1983) offered the opinion that the realist view of science provides a uniquely appropriate framework for the science of psychology. They wrote that specific behaviors like most events in the world cannot be explained as a manifestation of any psychological law. Behaviors of individuals are events that can be explained only in terms of a wide variety of interacting systems and structures, physical, biological, psychological, and social. “Identification of structures and their dynamics can only be accomplished by the multilevel application of imaginative theory that simultaneously guides observation, analysis, and experiment” (Manicas & Secord, 1983, p. 405).

They also wrote that consciousness is best studied from the realist position. They stated that humanistically oriented psychologists have been correct in insisting that the study of consciousness constitutes a different level of stratification from the physical sciences because, although the phenomena of consciousness are different from natural phenomena occurring in the physical environment, they are nonetheless real phenomena within the mind. The authors argued that it is possible to discover explanatory and causal mechanisms that operate within the realm of the study of consciousness (Manicas & Secord, 1983).

Manicas and Secord (1983) wrote that the hermeneutic science of individual persons is the study of a particular concrete person, his or her history, and particular patterns of behavior. However, as a scientific effort, it also requires that the researcher use whatever special knowledge may be available regarding implicated psychological structures and mechanisms that may have been operating in the individual's biography and current circumstances. The hermeneutic inquiry as a science also has a responsibility to the public to establish the evidential credibility of its accounts of human experience.

The hermeneutic approach must also distinguish between the explanatory, diagnostic, and therapeutic aspects of such science. As an explanatory science, the hermeneutic approach is very similar to the study of history. When history seeks to explain a particular event, it usually offers an account that traces and then connects the complicated conjunctions of motivated acts with their intended and unintended consequences within a specific arena of space and time. The account usually takes the form of a narrative, and the explanation requires the historian to become immersed in the situation and to communicate to the audience not only what happened but what it was like to have been there. "The appeal of a psychoanalytic account (whether valid or not ) lies in just this feature" (Manicas & Secord, 1983, p. 411).

Philosophy of science most often has been the work of philosophers rather than scientists. Scientists generate, from their daily work in scientific research, their own criteria for validity. The realist theory of science may be more useful to the working scientist than either empirical positivism or Kuhnian relativism. It stipulates that scientists are grappling with entities that may not be directly observable but are nonetheless real. Scientists also realize that there are no universal methods for reasoning from observations to generalizations to theoretical propositions. Furthermore, "even in the confirmatory phase, spelling out a theory (even if in mathematical language) is a slippery and hazardous process" (Manicas & Secord, 1983, p. 412).

Manicas and Secord (1983) concluded that a final bonus of the realist theory is that it provides a means of distinguishing between the tasks of the scientist and the clinician. The scientist practices science by creating at least partially closed systems of understanding. The clinician uses the discoveries of science in order to bring about changes in the world and also employs a great deal of knowledge that extends beyond what has been provided by science.

Leary (1984) offered a critique of the article by Manicas and Secord (1983). Leary stated that there are a variety of forms of realism in the philosophy of science and that each of these varieties poses a set of unique problems. Transcendental realism presupposes ontological realism and epistemological relativism. In accordance with this position, scientific concepts are best interpreted as constructive approximations rather than as real knowledge. Scientific concepts and laws are best understood as constructions that must be changed over time as part of a continual quest to make cognitive categories consonant with experience.

Leary (1984) concluded that it is instructive to think of reality as a set of interacting levels of stratification, a set of open systems, and a set of interrelated structures. It is important, however, to keep in mind that each such conceptualization is inadequate and incomplete. Every metaphor or model for representing the world is a statement of approximal similarity rather than one of absolute identity. Leary wrote that it is important not to overlook the ways in which the world is not a set of structures or to fail to specify the differences between cognitive structures, linguistic structures, structures of behavior, social structures, and physical structures. It is of paramount importance to avoid overgeneralization and simplification or the elaboration of metaphor into a larger belief system.

Pragmatism

Mulaik (1984) wrote that the theory of science known as pragmatism is superior to the theories of realism in understanding theories of psychology. Mulaik stated that both ontological and epistemological views in pragmatism are fallibilistic. Pragmatists assert that all claims to knowledge, including those that seem indubitable, are founded on various tacit assumptions that are often not testable in advance of experience. From the pragmatist viewpoint, an hypothesis is accepted as long as it works given whatever grounds are used to decide what works. The grounds themselves are subject to revision on the basis of new experience. Epistemological fallibilism requires that all ideas about reality be regarded as hypothetical, tentative, and subject to revision.

Recommendations of the National Institute of Mental Health Workshop

Wolfe and Goldfried (1988) reported the discussion and recommendations from an NIMH workshop on the research on psychotherapy integration. Over 20 recommendations, including some specific proposals for pilot studies to investigate various aspects of psychotherapy integration, were made. What will be examined here are those recommendations that seem to have some applicability to an attempt to evaluate object relations, self psychology, the psychoanalytic interpersonal school, and the Atlanta school of experiential psychotherapy.

The first recommendation was that conceptual development and empirical investigation need to proceed in tandem. One of the goals in the present study was to examine empirical data from developmental psychology and outcome and process studies in psychotherapy to determine whether the data support the assumptions made in theories.

The second recommendation was that conceptual and research efforts be made to investigate how practicing therapists of different orientations actually behave in therapy sessions before there is an attempt to integrate the systems. It was suggested that it might be helpful to determine whether therapists of different orientations actually use different interventions or whether there are similarities in therapist’s behavior across theoretical interventions. In the seventh recommendation, it was suggested that researchers need to elucidate common and unique factors among the various schools of psychotherapy. Additional goals, in this current evaluation of several theories of psychotherapy, are to investigate the conceptual claims of the particular theories and to elucidate common and unique factors among these schools.

The ninth recommendation was that research is needed to clarify the different kinds of therapeutic alliances that are associated with different therapy orientations, and in the eleventh recommendation, it was suggested that research is needed to determine whether a patient's personality style or level of distress affects the role the therapeutic alliance can play in mediating positive changes in therapy. As was discussed in Chapter 1, all the schools of therapy considered in this study have the relationship between the therapist and the patient and the effect of the therapeutic alliance on the successful outcome of treatment as primary concerns. In this study, outcome and process research findings concerning relationship factors will be examined to ascertain whether the findings provide support for theoretical assumptions about relationship factors.

In the twelfth recommendation, it was resolved that the single most important step in advancing psychotherapy integration is the development of a solution to the language barrier that separates therapy from therapy and researcher from clinician. In the thirteenth recommendation, it was resolved that the research language and the theoretical language that are capable of incorporating and translating the concepts of different orientations while being neutral to all of them must be developed.

One of the goals in this study was to examine the languages used in the various theories and to attempt to compare the concepts proposed in the theories. Whenever possible, jargon will be redefined into common English, and it is possible that some of the differences between approaches will be found to be semantic.

Relation of the Philosophies of Science to Psychotherapy and Psychotherapy Research

Sass (1988) stated that, from a hermeneutic viewpoint, he found something "trivializing, perhaps ultimately condescending and even countertherapeutic" (p. 255) in the approach to psychotherapy of Carl Rogers or the "humanistic" psychoanalyst Heinz Kohut. He argued that he found false or condescending the benign and selfless acceptance and validation of the patient's reported experience claimed by Roger's unconditional positive regard and Kohut's empathic introspection. He stated that a true hermeneutic approach imposes authoritatively the reality of the therapist upon the subjective reality of the patient. However, a true hermeneutic approach would leave open the possibility of a more probing and challenging therapeutic dialogue, which might confront more directly the traditional issues of values and the perception of "objective" truth than the empathic, reflective dialogue seemingly proposed by Rogers and Kohut.

Sass (1988) claimed that the framework of ontological hermeneutics provided an enriching alternative to the traditional views of humanistic psychology, behaviorism, and psychoanalysis. The emphasis on human freedom as a source of optimism in humanistic psychology or as a source of anguish in some forms of what he calls "pop existentialism" (p. 262) would necessarily be tempered by a greater concern and respect for the biological, cultural, and historical contexts in which human life is embedded. These inescapable contexts of human existence need not be seen as obstacles to self fulfillment but as the ground and being from which self fulfillment can be derived.

Respect for the social and biological context of human experience also differentiates a hermeneutic approach from behavior therapy, which Sass (1988) described as surprisingly humanistic in its belief and advocacy of self determination and self control. Finally, Sass challenged the assumption of some psychoanalytic thought that the truth of experience is preexisting and something for the patient to discover. Instead, he stated that insight should be described as an exploratory, interpretive dialogue in which both patient and therapist explore their mutual habitual preconceptions in order to illuminate the meaning of the patient's actions and reported experiences.

Sass (1988) concluded with the statement that hermeneutics seeks not objectivism or neutrality but the kind of illumination that comes from image and metaphors. Hermeneutics has the potential to encourage in the psychologist,

an ironic and self-critical, but by no means despairing, awareness of both the value and danger of presuppositions–and with this, a realization that though knowledge can never be value-free, it is not naive to seek truth. (Sass, 1988, p. 263)

Michels (1985) wrote that the models used by psychotherapists to understand psychic reality include a model in which there is an outer world of real events and an inner world of unconscious fantasies. In this model, the task of the therapist is to help the patient learn to distinguish between reality and unconscious fantasies.

Another common model, which may seem similar but is slightly different, contains the hypothesis that there is an external reality that is "true" and that the psychic reality of a patient is a distortion or misrepresentation of the truth. According to this model, the role of the therapist is to help the patient correct the distortion, to see reality more clearly such that the psychic reality of the patient becomes more like external reality so that adaptation improves (Michels, 1985).

Michels (1985) wrote that a third model is much more in harmony with current philosophical notions of reality. This model proposes that only the world of subjective experience is knowable and that all other versions of reality are derivative and are, thus, abstractions or projections from subjective experience. In this model, the task of the therapist is to help the patient broaden the range of subjective experience to incorporate repressed or split off aspects of experience so that the patient can successfully know and integrate his or her mental life. This particular model seems similar to a phenomenological point of view.

Michels (1985) proposed a fourth model of psychic reality, which suggests that there are many frameworks for understanding the subjective world and that psychoanalytic thought offers a uniquely valuable paradigm for organizing human experience. The task of the therapist in this model is to help the patient construct a new psychoanalytic reality out of the psychic realities brought to the therapeutic experience with the expectation that this new construction of psychic reality will have a special value in integrating experiences in life and in therapy. Michels stated that this particular viewpoint does not claim that psychoanalytic reality is the best or only way to organize experience but claims that the psychoanalytic framework of reality is particularly valuable in the psychoanalytic situation.

Gholson and Barker (1985) gave a rather lengthy explanation of the current status of various philosophical positions in the philosophy of science and their relation to psychology. Their main purpose was to account for the progress of scientific disciplines while avoiding the problems associated with radical incommensurability. They stated that both Lakatos (1978) and Laudan (1977, 1984) offered accounts for scientific progress that avoided the pitfalls of scientific relativism.

Gholson and Barker (1985) noted that the current argument in favor of a realist philosophy of science (Manicas & Secord, 1983) is that there is a relation between the observational and theoretical vocabulary that is strong enough to support ontological claims using the theoretical vocabulary. In contrast, instrumentalists claim that theory is only a means of systematically connecting observations and that the concepts of a theory do not correspond to real entities.

Lakatos (1978) disagreed that any real conflict need exist between realists and instrumentalists. He suggested that research programs evolve from an initial state resembling instrumentalism to a mature state that resembles realism. An important part of the heuristic of a program consists of recommendations for the incorporation of new features, which are known to be absent from the initial theory but are thought to be required for real world representations (Gholson & Barker, 1985).

Mature theories of a research program offer only candidates for reality. Multiple competing research programs are the norm in science. To the extent that these programs have different core commitments, they offer incompatible representations of reality. Laudan (1977) discussed this issue and distinguished research traditions by their differing ontologies and heuristics. Cases in which a single research program dominates a science leading to unanimity of opinion on the existence of fundamental entities are rare indeed. In the more usual case, research programs compete, attacking the fundamental entities of rival research traditions and defending their own, leading to lively centers of conflict (Gholson & Barker, 1985).

Laudan's (1977) concept of a research tradition with its malleable core and clusters of theories sharing an ontology and heuristic can account for much scientific progress that has been observed in recent decades, particularly in psychology. Gholson and Barker (1985) gave the specific example of the history of competition between conditioning and cognitive programs in the study of learning. An equally valuable concept from Laudan (1977) is his recognition that the conceptual assets and liabilities of a research tradition are involved in the appraisal of scientific progress, independent of experimental factors. The role of conceptual factors awaits further analysis, and much work remains to be done in elucidating the nature of the guiding heuristic of research traditions, but at least, it seems clear that growth and development in psychology may be explained and evaluated by using the frameworks described by Lakatos and Laudan (Gholson & Barker, 1985).

Russell (1986) argued that the empirical positivist position is not an appropriate framework for psychology and psychotherapy. Recently, the philosophers of science have stressed the need to use methods of research and epistemological criteria commensurate with the type of phenomena under study. Furthermore, in the new research in the philosophy of science, the quest for a unification of knowledge not only has come into disrepute but also can be judged in some instances as harmful to the advancement of science. Recent trends in the philosophy of science suggest that it is the proliferation of and conflicts among theories, practices, metasciences, and classificatory definitions that lead to scientific progress.

Russell (1986), therefore, concluded that attempts to integrate different schools of psychotherapy would be counterproductive. Development of different frameworks for the investigation and practice of psychotherapy should proceed as usual within the constraints imposed by the principles of the framework itself. However, these frameworks and principles should constantly be challenged by empirical and theoretical criticisms emanating from alternative frameworks. This cycle of conviction and criticism can result in a better understanding of the scope and limits of each approach to psychotherapy while allowing each approach to maintain its heuristic and ontological autonomy.

In summary, the criteria for evaluation of the robust, explanatory nature of scientific theory in the investigation of clinical phenomena are (a) the identification and understanding of patterns of human experience (Fischer & Fischer, 1983); (b) the extent to which the theoretical model explains the phenomena under investigation (Manicas & Secord, 1983); (c) the extent to which theoretical formulations describe what works (Mulaik, 1984); (d) the extent to which theoretical formulations are clinically significant (Adams, 1984); (e) the extent to which new theories accommodate the data and successes of previous theories and explain new data not integrated into previous theories (Lakatos, 1978); (f) the simplicity of the theory; (g) the rigorous, well-defined terminology of the theory; and (h) the extent to which the theory stimulates growth, progress, and competition in theoretical development (Laudan, 1977, 1984).

Method

The method in this study will consist of an examination of the literature of object relations, self psychology, the psychoanalytic interpersonal school, and the Atlanta school of experiential psychotherapy. The concepts of each theorist will be described and when possible reformulated in common language. It is hoped that some differences will be semantic. The theories also will be evaluated from the perspectives of the philosophy of science with special attention paid to their heuristics, the method of investigation used to gather the data upon which the theory is based, and ontology, which will be defined as the assumptions about the essential nature of humans, the nature and development of psychopathology, the goals of treatment, the role of the therapist, the importance of the relationship, and the mechanism of change. Theoretical ideas and hypotheses will be compared to research findings in psychotherapy studies. The criteria from the philosophies of science for theory evaluation will be used to identify the conceptual assets and liabilities of each theory. Theoretical ideas and hypotheses will be compared to research findings in psychotherapy studies.

Chapter 3: Outcome and Process Research in Psychotherapy

As soon as Freud announced that he had found a uniquely effective treatment for psychological problems, questions were asked about its effectiveness and efficacy. Thus, the birth of the talking treatment was also the birth of research on psychotherapy process and outcome. People in treatment centers associated with psychoanalytic training institutes, first in Berlin and later in London, Chicago, and Topeka, started to collect systematic data on treatment results in the 1920s. These studies may be regarded as the precursors of modern psychotherapy research (Strupp & Howard, 1992).

Strupp and Howard (1992) contended that the modern era of psychotherapy outcome research generally is dated from Eysenck's (1952) broad attack on all forms of psychotherapy. Eysenck (1952) reviewed two studies of the spontaneous improvement of psychologically distressed individuals and compared their improvement rate with the improvement rate of individuals who had been treated in psychotherapy. He concluded that the effectiveness of psychotherapy was unproven.

According to Strupp and Howard (1992), the discipline responded to Eysenck's (1952) criticisms with a variety of attacks on his data base, methodology, and motives and with an explosion of research activity on the effectiveness of psychotherapy. Lambert and Lauper (1980) cited more than 4,000 publications of psychotherapy research published through 1978. The research on psychotherapy has been reviewed and analyzed by various authors who all concluded that psychotherapy helps individuals achieve goals and overcome psychopathology faster than an individual's natural healing process and from supportive elements in the environment (Bergin & Lambert, 1978; Lambert & Bergin, 1992; Lambert, Shapiro, & Bergin, 1986; Luborsky et al., 1975; Smith, Glass, & Miller, 1980).

Smith et al. (1980) found no studies in which the untreated group had a better outcome than the treated group. They examined 475 controlled outcome studies. They performed a statistical meta-analysis in which they calculated an effect size (ES) based on the difference of the average score of the treatment group and untreated group divided by the standard deviation of the control group. These calculations allowed different scales to be standardized and transformed to a common scale. The researchers reported a mean ES for psychotherapy of .85 which means that the average treated person had an outcome that was equal to or better than 85% of the untreated individuals. Criteria for inclusion in their analysis was that each study had to have an untreated control group.

Strupp (1992) noted that the researchers "were elated by this statistical-scientific finding, and the efficacy of psychotherapy seemed firmly established" (p. 312). Most of the research examined by Smith et al. (1980) consisted of studies of the following types of treatment: (a) analytic therapy and many of its psychodynamic variations, (b) client centered therapy, (c) rational-emotive therapy, (d) systematic desensitization, (e) behavior modification, and (f) cognitive behavior therapies. There were no significant differences in the outcomes of these various types of therapy.

Similarly, although the superiority of one psychotherapeutic technique over another was sometimes demonstrated in individual studies, in comprehensive reviews of the outcome literature, the reviewers usually suggested that different treatments differed very little in the outcomes they produced (Kazdin & Bass, 1989; Lambert & Bergin, 1992). As clinicians and researchers became apprised of the findings of equivalent effects across different models of treatment, they were inspired to search for common mechanisms of change within the practice of different theoretical models (Lambert & Bergin, 1992).

Kazdin and Bass (1989) have questioned the validity of the majority of past comparative studies and have argued that these studies may lack the statistical power to identify differences in treatment outcome among types of therapy. They argued that many previous studies simply might have lacked subject populations of sufficient size to detect differences that in fact may exist. They challenged researchers in the future to design studies with sample sizes substantially larger than the sample sizes of previous studies.

However, the possibility also exists that there are common factors in the different therapies that lead to common outcomes. Researchers have given much attention to the identification of common factors in different treatment approaches, which has led to a proliferation of research on psychotherapy process. Some of the factors that have been identified include therapist characteristics, patient characteristics, the structure of the treatment (scheduling, fees, timeliness, and duration), types of treatment interventions, and behaviors of the patient and therapist in the treatment hour (Lambert & Bergin, 1992; Lambert et al., 1986; Orlinsky, Growe, & Parks, 1994; Orlinsky & Howard, 1986; Strupp, 1987, 1989; Strupp, Butler, & Rosser, 1988; Strupp & Hadley, 1979).

In several studies, the results have been that relationship factors have considerable impact on success in psychotherapy. Researchers have examined personality characteristics of both the therapists and the patient and the correlation of those factors with success in treatment. Lambert and Bergin (1992) stated that despite the methodological issues that can be raised, one research finding remains clear: "Relationship factors predict, if not cause, outcome" (p. 373). Strupp (1989) concluded that, in present times, most researchers and practitioners typically view "patient and therapist as a complex two-person system whose actions and reactions to one another are of coequal importance and concern" (p. 719).

Lambert and Bergin (1992) stated that there has been considerable agreement in studies in which the researcher asked clients what was most helpful to them in their therapy. What clients emphasized were the personal qualities of the therapist (e.g., sensitivity, honesty, and gentleness) rather than specific technical interventions. Both Lambert et al. (1986) and Beutler, Crago, and Arizmendi (1986) reviewed numerous studies in which personal qualities of the therapist were rated by either the patient or an observer of the therapeutic process. In the majority of cases, understanding and acceptance on the part of the therapist were highly correlated with successful outcome.

Researchers who compared behavioral and insight oriented therapies found that 70% of successful clients listed as extremely important or very important items such as: (a) the personality of the therapist, (b) help given to them by the therapist in understanding their problems, (c) encouragement to face their fears, (d) being able to talk to an understanding person, and (e) the help given by the therapist in achieving greater self-understanding. These results were obtained from a 32-item questionnaire four months after completing therapy. The questionnaire included items that were specific to behavior therapy techniques and dynamic therapy techniques as well as items that were thought to be important for both therapies (Sloane, Staples, Cristol, Yorkston, & Whipple, 1975).

Lambert and Bergin (1992) emphasized that relationship factors include the qualities of the patient as well as the qualities of the therapist. Attributes of the patient may play an important role in establishing the quality of the therapeutic relationship and, thus, the outcome of therapy. Several studies have been done in which researchers have examined the effect of patient characteristics on therapeutic outcome.

Strupp (1980a, 1980b, 1980c, 1980d) reported a series of four studies in which two patients were treated by one therapist in time limited psychotherapy. In each study, one case was assessed as a successful outcome, and the other was judged a treatment failure. These reports were part of a larger study in which extensive outcome measures were obtained and in which an analysis of patient-therapist interactions was preformed during the process of psychotherapy. In all reports, the patients who had successful outcomes were more willing and able to develop a meaningful relationship with the therapist than were the patients who did not do well. Patients who were identified as treatment failures did not relate well to the therapist, had a tendency to keep interactions on a superficial level, and/or had more hostile interactions with the therapist than did successful patients.

In Strupp's (1980a, 1980b, 1980c, 1980d) analysis, contributions of the therapists remained relatively constant throughout therapy with each of the two patients, and the differences in outcome were attributed to differences in the personalities and characteristics of the patients. Strupp, however, did not rule out the possibility that poor outcomes with less functional patients could just as well be attributed to failure on the part of therapist to adapt techniques and therapeutic styles to the more difficult patients. In particular, Strupp (1980d) and Strupp et al. (1988) hypothesized that poor outcomes with chronically angry and/or negative clients might be attributed to the therapist's failure to handle his or her countertransference reactions.

Meta-analysis of Psychotherapy Research

Orlinsky and Howard (1986) reviewed 1100 research findings concerning the relation of outcome in psychotherapy to various aspects of therapeutic process and made 34 conclusions about the relationships between process and outcome variables. Orlinsky, Grawe, and Parks (1994) built upon the foundation of the earlier report by Orlinsky and Howard and examined 2,354 separate research findings. Orlinsky et al. stated unequivocally that the relation of various aspects of process to outcome have been so well replicated across numerous studies that they can be accepted as established facts. The authors also identified those process-outcome links that are robust in the sense that such links have been measured from multiple process perspectives (e.g., links that have been measured by therapist, patient, independent observers or raters, and/or by using psychometric scales).

Orlinsky et al. (1994) defined a “finding” as a “methodologically independent observation of a relationship between process and outcome variables” (p. 271). The research studies examined by Orlinsky and his colleagues included some which yielded only one or two findings and others that yielded a dozen or more process-outcome findings. Orlinsky et al. stated that conclusions in their meta-analysis were drawn only from well-replicated findings based on multiple data bases.

The Therapeutic Contract

Provisions of the therapeutic contract showed no consistent relation to outcome. Effective therapy can be conducted in different formats, with different schedules, and for varied terms and fee arrangements. However, Orlinsky et al. (1994) emphasized the need for more extensive and in depth examination of contractual provision and assessment of interaction effects such as a possible interaction between frequency of sessions and the level of patient impairment.

Comparisons of group, individual, couple, and family therapy showed no significant difference in outcomes among those types of treatment. Nearly 75% of the research failed to demonstrate any consistent main effect indicating an advantage for individual, group, or any other format of therapy. It should be noted, however, that such a conclusion can only be applied to patients who willingly remain in the format to which they were assigned (Orlinsky et al., 1986).

An examination of the effect of frequency of therapy sessions showed that in 70% of the studies there was no difference in outcome between meeting once weekly and other more or less frequent schedules. There were only 25 studies, however, and the participants in many of the studies were either psychiatric inpatients or participants in very structured behavioral programs. Orlinsky et al. (1994) stated that frequency of treatment might be more effectively studied in interaction with other variables (e.g., level of patient impairment).

Researchers had compared the effectiveness of open-ended and time-limited therapeutic contracts for the duration of treatment in few studies. Orlinsky et al. (1994) found that the findings were inconsistent with an almost equal division among positive, negative, and null findings. The authors stated that patients in time-limited therapy may actually attend more therapy sessions than patients in open-ended therapy, which has a median length of only five or six sessions.

There were nine studies in which the relationship between payment of a fee and outcome in psychotherapy was examined. In none of the studies was an attempt made to measure the actual psychological cost of treatment (work time missed, relation of fee to financial resources, etc.). However, in none of the studies was negative outcome associated with the payment of a fee, and a positive outcome associated with payment was indicated in several (Orlinsky et al., 1994).

Orlinsky et al. (1994) identified several variables as facets of implementation of the therapeutic contract. The first was called expectational clarity and goal consensus which was described as the achievement of a consensus between patient and therapist as to what may be expected in treatment. The second was called patient role preparation in which the therapist provides the patient with some type of instruction and informs him or her of what to expect from treatment and how he or she might participate effectively. The third was the patient’s verbal activity. The fourth was the therapist’s verbal activity. The fifth was called therapist’s skillfulness. The sixth was called patient suitability. The seventh was identified as procedures for termination. The eighth was called stability of treatment arrangements. Expectational clarity and goal consensus, patient role preparation, patient verbal activity, therapist skillfulness, and patient suitability for treatment were consistently associated with a positive therapeutic outcome.

There were 35 findings in which expectational clarity and goal consensus were associated consistently with positive outcomes. Of 42 findings on role preparation of patients, there were 24 in which significantly better outcomes were obtained with patients who received some form of early role preparation than for patients who were given no role preparation. No study showed a significant negative effect. In general, role preparation produces better outcome more often than not and does no harm (Orlinsky et al., 1994). Several of the studies also contained data that were interpreted by the researchers to indicate that some patients with a generally poor prognosis, such as patients with low socioeconomic status, benefited more from psychotherapy when given some preparation than patients who were not (Orlinsky & Howard, 1986).

In 11 studies in which the amount of speech produced by the patient was examined, 7 showed a significant positive correlation between the verbal activity of the patient and success in therapy. No study showed a negative relation between the amount spoken by the patient and outcome. Thus, the authors concluded, not surprisingly, that it is important for patients to talk.

The amount of verbal activity of the therapist seemed to have a more complex relation to outcome than did the patient’s verbal activity. In 34 findings concerning the amount of talking by the therapist, 13 were significantly positive, 19 showed no significant relationship, and 3 were significantly negative. Although 55% of the findings were null, almost 40% of the total consisted of a significant positive relation between the therapist’s verbal activity and the outcome. One study, which showed a significant positive effect for therapist’s verbal activity with some patients, also showed that good outcome was associated with silence on the part of the therapist with patients who had low levels of pretreatment disturbance. There were also several significant negative findings, suggesting that under some circumstances therapists may talk too much (Orlinsky et al., 1994).

Therapist’s skillfulness and patient suitability for the type of treatment in which the patient is placed were found to be related positively to outcomes in treatment. Of 36 findings on the relation of therapist’s skillfulness to outcome, 68% were significantly positive with many of the ratings provided by patients and independent external raters. Similarly, 68% of 40 findings indicated a significant positive association of outcome with judgments of patient’s suitability for the treatment in which they were placed Most of the judgments of suitability were made by independent raters (Orlinsky et al., 1994).

Appropriate and proper termination procedures (e.g., planned, agreed upon, and discussed termination) were shown to have a positive association with outcomes in 32% of 19 findings. The remaining findings were null. These results suggest that the proper handling of termination may be important in some cases. Finally, four studies with six experimental results provided preliminary evidence that stability in treatment arrangements may be an important factor in positive outcome (Orlinsky et al., 1994).

Therapeutic Interventions

The actual process of psychotherapy consists of a series of overlapping interactions between the patient and the therapist. Patients must present their problems to the therapist. The therapist must interpret and understand these presentations within the context of practical and theoretical knowledge, which then must be used to design and to implement a program of intervention. The patient's response and/or cooperation with the interventions, in turn, affects the progress and outcome of the treatment (Orlinsky et al., 1994). Patient participation has been shown to be vital both in respect to problem presentation and to active cooperation. Affective arousal of the patient also seems to be important. Focus on life problems and core personal relations was shown to contribute to positive outcome. The therapeutic interventions most consistently associated with positive outcome in treatment were interpretation, confrontation, and what Orlinsky et al. described as experiential confrontation.

Patients’ focusing on life problems was shown to be positively associated with outcome in 7 of 11 findings. Other studies provided preliminary evidence that patients' focusing on core personal relationships (e.g., families of origin) was associated with improvement in individual psychodynamic treatment, particularly as judged by external raters. By contrast, patients' focusing on here-and-now involvement in the therapeutic interaction has little impact on outcome (Orlinsky et al., 1994).

Studies of the focus of attention of the therapist have not yet produced findings that are robust and unequivocal. Therapists' focusing on patients' problems was often (53% of 19 findings) associated with positive outcome except in cases in which problem focus was disruptive to the patient's defenses, as was demonstrated in 2 studies with negative findings. In 9 of 18 findings, focusing on patients' affect was positively associated with outcome, but again 2 negative findings indicated circumstances in which a focus on affect was contraindicated.

Therapists' focus on patients’ here-and-now involvement in sessions was more often negatively associated with outcome. Focusing on core personal relationships and transference issues was shown to have a significant negative effect in 2 of 27 findings, and only 10 had a significant positive association with outcome (Orlinsky et al., 1994).

Orlinsky and Howard (1986) summarized research findings on the relationship between therapeutic techniques and outcome in psychotherapy. In 22 findings regarding the influence of interpretation, 11 showed a significant positive effect, 8 showed no effect, and 3 showed a negative effect (2 of the 3 negative findings were found in using interpretation with psychotic or borderline patients in a study that also showed positive effects of interpretation with neurotic patients). The updated findings presented by Orlinsky et al. (1994) were that 63% of 38 findings indicated a significant positive association with outcome. The authors concluded that interpretation was a potentially powerful intervention to be used only under the therapeutic conditions that indicated that there would be a positive influence. The authors did not specify the conditions that made it possible to use interpretation effectively, although the analysis made by Orlinsky and Howard (1986) indicated that interpretation may be contraindicated with psychotic and borderline patients.

Orlinsky and Howard (1986) defined confrontation as a therapeutic technique used to heighten self awareness. They emphasized that therapeutic confrontation carries no implication of hostility or criticism but is an attempt to provide the patient with self experience. In 7 studies of the relation between confrontation and outcome, all 7 showed a significant positive correlation between confrontation and therapeutic outcome. The authors concluded that the evidence suggests that confrontation may be a potent form of therapeutic intervention.

Orlinsky et al. (1994) grouped the findings on confrontation from the 1986 study by Orlinsky and Howard with a new category that the authors labeled experiential confrontation. Experiential confrontation was not defined by the authors, but examples, including Gestalt two-chair dialogues, abreaction, in-vivo desensitization (flooding), and therapists emphasizing feeling to deepen the patients' affective experiences, were given. It would seem that experiential confrontation might be defined as techniques that heighten self awareness while the patient is in a state of affective arousal. Among the 22 findings (from 11 studies) examined by Orlinsky et al., 70% had a significant positive association between experiential confrontation and outcome. The only negative effect was observed in a psychiatric inpatient setting.

The techniques of exploration, support, reflection, and clarification showed little positive or negative correlation with outcome. The authors concluded that, although these techniques may occasionally be helpful and are rarely harmful, they are not powerful techniques. Advice giving was examined in 5 studies, which showed that it was more likely to be unhelpful or harmful than beneficial. The therapist’s self disclosure was rarely associated with outcome, and when it was, the impact seemed as likely to be negative as positive (Orlinsky et al., 1994; Orlinsky & Howard, 1986).

The cognitive, behavioral, and emotional participation of the patient would seem to be a vital ingredient for successful treatment. In fact, 69% of nearly 50 findings were significant positive associations of patient cooperation with outcomes. When the patient's positive affective response was considered, all of nine findings were significantly associated with positive outcome.

However, a negative affective response did not necessarily predict poor outcomes. In 46 studies of negative affective response, 35% were significantly associated with poor outcome, 39% showed no significant association, and 26% showed that arousal of negative affect was positively associated with outcome. The authors commented that successful outcome may reflect circumstances in which the understanding of negative affect leads to relief and positive feelings. This suggested that further research is urgently needed (Orlinsky et al., 1994). Orlinsky and Howard (1986) stated that, according to several studies, negative affect expressed in early treatment sessions was predictive of positive outcomes.

Patient’s self-exploration was found to have no association with outcome in 67% of 79 findings but still showed a significant positive association in 30% of the findings. The patient’s expressiveness was positively associated with outcome in 63% of 51 findings. Preliminary investigations suggest that the patient’s identification with or internalization of the therapist may have a positive impact (Orlinsky et al., 1994).

The Therapeutic Alliance

The therapeutic bond or alliance has been studied more extensively than any other aspect of the therapeutic process. Orlinsky et al. (1994) summarized 132 findings on the overall therapeutic bond, which was significantly and positively associated with outcome in 66% of them. Specific aspects of the therapeutic bond also have been examined extensively.

In 37 findings on the impact of the variables identified as therapist engagement and detachment, 21 showed a significant positive impact of engagement, but none showed a beneficial effect of detachment. The positive impact of therapist engagement was most positively associated with outcome when judged by the patient. Eighteen findings on the impact of the variable identified as therapist confidence and unsureness were examined. Therapist confidence had a significant positive association with outcome in 59% of the 27 findings. Therapist confidence showed its greatest impact on outcome when judged by nonparticipant observers (Orlinsky et al., 1994).

The authors examined 46 findings in which the effects of therapist collaboration and directiveness or permissiveness on outcome were compared. The authors also examined 42 findings in which the effects of patient collaboration and dependency on outcome were compared. In 64% of these findings, positive outcome was significantly related to the therapist's encouragement of patient initiative and the patients' assumption of an active role in resolving problems. None of the studies favored a dependent or controlling style of relating for the patient, although a few showed positive outcome with a directive therapeutic style (Orlinsky et al., 1994; Orlinsky & Howard, 1986).

Orlinsky et al. (1994) stated that there is very strong evidence that therapist empathy has a powerful correlation with positive therapeutic outcome when empathy is measured by the patients or when observer ratings of empathy are compared to objective outcome measures. The authors made the point that the patient's perspective often seemed to be the most discriminating with respect to therapist process variables and that the therapist's perspective was most discriminating with patient process variables.

Orlinsky et al. (1994) discussed 115 findings on the impact of the therapist’s empathic understanding, with particular attention paid to whether the empathy was evaluated by the patient, the therapist, or an independent observer. Overall, in 54% of the findings, a significant positive association between the therapist's empathic understanding and outcome was shown, and none showed a negative association between empathic understanding and outcome. When therapist empathy was reported by the patient, in 72% of 47 findings, a significant positive effect on outcome was shown. When therapists rated their own empathy, in 13 of 23 findings (56%), a positive relation with outcome was shown. When independent observers rated therapist empathy and outcome was based on evaluation by therapists or observers, only 5 of 13 findings (38%) had a positive correlation of empathy and outcome. On the other hand, when outcome was measured by objective indexes and tests, the therapist's empathy as rated by nonparticipant observers had a significant positive effect in 26 of 45 findings.

Research on the variable called therapist expressiveness was insufficient to draw conclusions, but the authors noted that ratings of the therapist's voice quality showed promise in predicting outcome. Finally, 42 process-outcome findings showed a clear pattern linking communicative attunement to positive outcome particularly when process was evaluated by patient ratings or objective indexes and outcome was evaluated by patients and therapists (Orlinsky et al., 1994).

The aspect of the therapeutic bond that has been the most extensively studied is therapist affirmation (e.g., acceptance, nonpossessive warmth, and positive regard). In 154 process-outcome findings, 56% were positive from all evaluative perspectives, and 65% were positive when therapist affirmation was evaluated by the patient. The authors suggested that the most important task for future researchers was to determine the conditions and circumstances in which therapist affirmation plays a major role in determining outcome (Orlinsky et al., 1994).

Patient affirmation toward the therapist (as indicated by declarations of respect or liking made by the patient to the therapist) has been less often studied, but 59 findings demonstrated that it is more consistently positively associated with outcome than is therapist affirmation toward the patient (69% versus 56%). The authors commented that patient affirmation may be a result rather than a cause of progress but could still be regarded as a clinically important sign that treatment is going well. Finally, reciprocal affirmation between patient and therapist showed a very consistent association with outcome; 78% of 32 findings were significantly positive (Orlinsky et al., 1994).

Self-Relatedness

Self-relatedness refers to a person’s characteristic style of relating to himself or herself. It concerns the way internal cognitive and affective processes are experienced and evaluated and the way ideas, feelings, and impulses are controlled. Individuals may be receptive and flexible or guarded and constrained. The former tend to be regarded as open, and the latter tend to be regarded as defensive.

Patient's openness has been demonstrated to be consistently related to outcome in 45 findings. In 80% of the studies, significant positive association between patient openness and outcome was shown.

Therapist self-congruence (genuineness) has been the aspect of therapist self-relatedness that has been studied as a contributing variable in the therapeutic process most often. Orlinsky et al. (1994) presented 60 research findings about the variable identified as therapist genuineness. The mass of the findings indicated that genuineness had an occasional but not consistent positive association with outcome. In only 38% of the 60 findings was this variable found to have a positive association with outcome, but the authors suggested that there may be some conditions under which therapist genuineness makes an important contribution to therapeutic success.

Therapeutic realizations

Therapeutic realizations is a term used by Orlinsky et al. (1994) to describe what the authors called “positive impacts” experienced by patients during specific therapy sessions. The term “positive impacts within sessions” was not well defined by the authors but seems to be a term used by them to indicate problem solving, insight, behavior change, or cathartic experiences that occur within a session. Findings of therapeutic realizations within sessions showed a consistent positive association with outcome. Of 79 findings, 67% had a positive association. This level of positive association was consistent throughout, particularly when judged by the independent raters and by the patients themselves.

Treatment Duration

Finally, Orlinsky et al. (1994) examined 156 findings concerning the relation between length of therapy and outcome. The authors concluded that the evidence rather consistently indicated that patients who have more therapy get more benefit from therapy than those with less therapy. The authors stated that in approximately 64% of the findings there was a significant positive relation between duration of therapy and outcome. The few studies that showed a negative relation of outcome and length of therapy could be explained by special circumstances. For example, one of the studies with a negative finding was a long-term retrospective analysis of hospitalized childhood schizophrenics, all of whom had a very long duration of treatment. Within this context, it seems reasonable that the most disturbed and intractable cases would be reported as having the most treatment and the least improvement. Another study with a negative correlation between duration and outcome of treatment was a poorly controlled study in which patients were not assigned randomly to the treatments of different lengths (Orlinsky & Howard, 1986).

Howard, Kopta, Krause, and Orlinsky (1986) did a meta-analysis of over 2400 cases in studies in which detailed data on outcome at various times was provided. The authors demonstrated that improvement in therapy is a linear function of the logarithm of the number of sessions. Thus, improvement is proportionately greater in earlier sessions than in later sessions, with the most rapid improvement occurring in the first 6 to 12 months. Improvement then increased much more slowly over the duration of treatment presumably as the patient's less tractable, perhaps more characterological, problems became the focus of treatment.

Conclusions

Research on the outcome of psychotherapy across a variety of analyses in numerous studies indicates that psychotherapy is effective in helping individuals resolve problems and alleviate psychopathology. However, no significant differences in effectiveness among various types of psychotherapy with different theoretical orientations have been found (Bergin & Lambert, 1978; Howard et al., 1986; Lambert & Bergin, 1992; Lambert et al., 1986; Luborsky et al., 1975; Orlinsky et al., 1994; Orlinsky & Howard, 1986; Smith et al., 1980). This conclusion has led researchers to conduct numerous studies in which the process of psychotherapy has been examined in an attempt to find common factors in the various types of psychotherapy that would explain their success (Lambert & Bergin, 1992; Lambert et al., 1986; Orlinsky et al., 1994; Orlinsky & Howard, 1986; Strupp, 1980a, 1980b, 1980c, 1980d, 1987, 1989; Strupp et al., 1988; Strupp & Hadley, 1979).

One of the most comprehensive reviews of the outcome research was done by Orlinsky et al. (1994). A summary of the therapeutic conditions that are most highly correlated with outcome included: (a) goal consensus and expectational clarity, (b) role preparation of the patient, (c) verbal activity and the establishment of a collaborative relationship between patient and therapist in which the therapist encouraged patient initiative and the patient assumed an active role in resolving his or her problems, (d) patient’s suitability for treatment, and (e) therapist skill. The relations of patient suitability and therapist skill to outcome were particularly robust with consistent findings across the various process perspectives (e.g., with the process criteria measured by patient, therapist, independent rater, or objective scores).

In an examination of therapeutic operations, Orlinsky et al. (1994) found that experiential confrontation and interpretation were associated with a positive outcome. The authors also found that the patient’s focus on life problems and core personal relationships had a significant positive relation to outcome. Evidence was also strong across process perspectives for positive outcome associated with patient cooperation and positive affective arousal.

The strongest evidence linking process to outcome concerned the therapeutic bond or alliance. Orlinsky et al. (1994) stated that, in more than 1,000 process-outcome findings, many significant positive associations with successful outcome were obtained for: (a) overall global quality of the therapeutic bond, (b) therapist engagement, (c) therapist confidence, (d) patient collaboration, (e) patient expressiveness, (f) therapist empathic understanding, (g) therapist affirmation of the patient, (h) patient affirmation of the therapist, and (i) reciprocal affirmation. The correlation between the therapeutic relationship and outcome was especially powerful when the quality of the relationship was evaluated by the patient.

Orlinsky et al. (1994) found strong evidence linking outcome to patient openness. The authors commented that this particular finding probably should be grouped with other patient variables, such as suitability for treatment, cooperation, collaboration, expressiveness, and affirmation. They stated that, when these variables are considered together, the data indicate the critical importance of the patient's contribution to treatment and, thus, the importance of the therapists’ assisting patients in making a constructive contribution to his or her own treatment.

Finally, many studies showed that improvement was significantly associated with length of time in treatment. Positive effects have been documented for brief psychotherapy, but in general, process-outcome research and follow-up outcome data have suggested that longer treatment duration is generally (though not linearly) associated with better outcome than short treatment duration (Orlinsky et al., 1994). In another study in which researchers performed a meta-analysis of many research findings, it was shown that progress and improvement in psychotherapy are proportionately greater in early stages of therapy, particularly in the first six months to a year, than in later stages. Progress then continues but at a much slower rate (Howard et al., 1986).

Orlinsky et al. (1994) wrote that 11 process-outcome variables were found to be very robustly linked to outcome in that significant associations were found across both process and outcome assessments. These variables were: (a) patient suitability, (b) patient cooperativeness, (c) global therapeutic bond, (d) patient contribution to the bond, (e) patient interactive collaboration, (f) patient expressiveness, (g) patient affirmation of the therapist, (h) reciprocal affirmation, (i) patient openness, (j) therapeutic realizations, and (k) treatment duration.

Orlinsky et al. (1994) concluded that the quality of the patient's participation in therapy stands out as the most important determinant of outcome. The therapeutic bond or alliance, especially as understood by the patient, was shown to be of particular importance. One vital contribution of the therapist to a successful therapeutic outcome was the achievement of an empathic, affirmative, collaborative engagement with the client. Another significant activity of the therapist was the skillful, confident use of therapeutic interventions such as experiential confrontation and interpretation. The authors asserted that the consistency of process-outcome relations verified by hundreds of empirical findings over more than 40 years of research has established those findings as facts about some of the conditions and operations that lead to successful therapeutic outcome.

Garfield (1992) discussed the problems of past research in psychotherapy process and outcome. He particularly noted the problems associated with evaluations made solely by therapist or patient either of whom may have various biases that would preclude an objective assessment. This particular problem was certainly noted in the review of the literature by Orlinsky and Howard (1986). Garfield (1986) suggested that more accurate assessments of psychotherapy might be achieved by using a tripartite system of evaluation with assessments drawn from therapist, patient, and independent observers.

Garfield (1986) described various improvements that could be made in psychotherapy research and ethical and pragmatic problems that arise in psychotherapy research. Some of the issues raised include the differences between experimental conditions and actual clinical practice, the use of control groups, measures and criteria of outcome, the logistics of follow up studies, the difficulties in studying negative effects, and various statistical and other methodological issues.

Lambert and Hill (1994) provided a critique of current research methods for assessing psychotherapy outcome and process. The authors identified multiple processes that may have an impact on outcome and multiple strategies for measuring outcomes and processes. One important point made in their article was that there is a definite need for research on how different individuals respond to different process events in treatment (e.g., what types of patients respond well to interpretation).

A thorough examination or comprehensive consideration of the literature that contains criticisms of current psychotherapy research and suggestions, with more than 4000 studies identified by Lambert and Lauper (1980), is beyond the scope of this paper. The purpose of this brief review has been to examine what has been discovered about the effectiveness of psychotherapy and to identify those activities and components in psychotherapy that are associated with outcome. It is to be hoped that some of the factors that have been established as being related to outcome will be of help in evaluating the theories to be examined in the remainder of this paper.

It is of particular interest that many of the authors who reviewed psychotherapy research emphasized the importance of relationship factors in the success of psychotherapy (Beutler et al., 1986; Lambert & Bergin, 1992; Orlinsky et al., 1994; Orlinsky & Howard, 1986; Sloane et al., 1975; Strupp, 1980a, 1980b, 1980c, 1980d, 1987, 1989). In the theoretical orientations chosen for examination in this paper, the psychoanalytic interpersonal school, object relations theory, self psychology, and the Atlanta school of experiential psychotherapy, the theorists have emphasized the importance of relationship factors as a curative element in psychotherapy. It remains to be seen in the chapters that follow how each theorist's ideas of the use of relationship is supported by the empirical findings and how other hypotheses and postulates of these theories are supported or disconfirmed by research findings.

Chapter 4: The Psychoanalytic Interpersonal School

Harry Stack Sullivan developed the interpersonal approach to psychotherapy and psychoanalysis. According to Chapman (1976), Sullivan was the most original and influential American-born psychiatrist. His ideas have gradually permeated American psychiatry and are spreading to a worldwide professional group. Sullivan has been acknowledged repeatedly as one of the first major theorists to offer an innovative, comprehensive system of thought as an alternative to classical Freudian psychoanalytic ideas (Bromberg, 1989; Chapman, 1976; Chrzanowski, 1982; Greenberg & Mitchell, 1983; Kiesler, 1982; Levenson, 1983; Mullahy, 1970). Psychoanalytic interpersonal theory is often simply called the interpersonal school and will be referred to as such in this chapter.

During his lifetime, Sullivan exerted his influence in the field of psychiatry through teaching and lectures and rarely through published writing. Since his death in 1949, his students and associates have collected and published a large number of his seminars and lectures, which had been recorded. Most of his fully developed ideas are now in print (Chapman, 1976; Greenberg & Mitchell, 1983; Mullahy, 1970). An overview of the major ideas of Sullivan will be presented in this chapter, and then, current ideas and concepts in the psychoanalytic interpersonal school (Bromberg, 1989; Cashdan, 1982; Chapman, 1976; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Mullahy, 1970; Stern, 1987; Sullivan, 1953, 1954, 1956, 1972) will be presented. The heuristic and ontology of the interpersonal school, as derived from the discussion of the major ideas of Sullivan and his followers, will be summarized.

The Interpersonal Theory of Harry Stack Sullivan

Basic Concepts

Sullivan’s (1953) theories were drawn from the interpersonal situations in which persons manifest mental health or mental disorders. He defined psychiatry as an expanding science concerned with events or processes in which the psychiatrist participates while observing. This notion of the psychiatrist as a participant-observer who is, of necessity, a part of and an influence on the process of psychotherapy was one of Sullivan's (1953, 1954) important ideas. The concept of the therapist as a participant-observer will be explored in more detail later in the discussion of the treatment methods of interpersonal psychiatry.

Sullivan (1953, 1954, 1956, 1972) built his theories of psychiatry on the study of personality characteristics that could be directly observed in the context of interpersonal relations. Although he offered some speculations about internal processes that could be inferred, he was more concerned with what could be directly seen, heard, or felt by the therapist and/or what was reported by the patient.

Sullivan (1953) wrote that personality consists of the characteristic ways in which a person deals with other people in interpersonal relationships. "Personality is the relatively enduring patterns of recurrent interpersonal situations which characterize a human life" (p. 111). He stated that personality was formed by interpersonal relationships, particularly close ones during the individual's entire lifetime. Personality never becomes totally fixed and rigid although the influences upon personality of the important relationships of infancy, childhood, and adolescence are particularly powerful and enduring.

Sullivan (1953) also wrote that personality functioning operates according to a principle that he called the tendency toward health. In individuals, there is a basic tendency toward emotional health and sound interpersonal functioning. Barring environmental pressures, people tend to seek opportunities to grow in healthy ways and to develop satisfying interpersonal relationships.

An idea of great importance to Sullivan (1953, 1954, 1956) was the one-genus hypothesis, "everyone is much more simply human than otherwise" (1953, p. 32). The differences between an individual who is psychiatrically ill and one who is emotionally healthy are differences in degree rather than in nature. Sullivan (1953) stated that anomalous interpersonal situations are a function of differences in the relative maturity of the persons involved.

The role of anxiety in the formation and maintenance of personality was a vital component in Sullivan's (1953) theories. Sullivan used the term anxiety to describe almost all unpleasant emotions, varying in degree from mild discomfort to absolute panic. Anxiety was considered to be always interpersonal in origin. He stated that anxiety originating from past experience interfered with the individual's capacity to improve subsequent relations. Anxiety, thus, had a tendency to bind a person in unhealthy interpersonal patterns. Sullivan (1953) stated that, “Anxiety as a phenomenon of relatively adult life can often be explained plausibly as anticipated unfavorable appraisal of one's current activity by someone whose opinion is significant” (p. 113).

Sullivan (1953, 1954, 1956) defined security as a state of relaxed comfort in which the individual feels no apprehension, self-doubt, inadequacy, or any kind of emotional discomfort. He thought that individuals at all times attempt to arrange their interpersonal lives so as to maximize the experience of security. Interpersonal devices called security operations are used by individuals to abolish anxiety and to establish security.

Security operations were described as either healthy or unhealthy depending on the cost of their use to the individual. Security operations include sublimation, selective inattention, and “as if” operations. Sublimation was defined as the suppression of anxiety generating behavior and replacement with a socially acceptable behavior that, at least partially, satisfies the motivation that led to the impulse for the anxiety generating behavior. Selective inattention was described as the failure to observe stressful or disturbing aspects of interpersonal relations. As-if operations were said to occur when the individual acts as if one's self or some significant other were a different kind of person in an interpersonal relationship. Unhealthy security operations produce the interpersonal maladjustments that constitute psychiatric symptoms and illnesses (Sullivan, 1953, 1956).

Sullivan (1953) also used the concept of dynamisms to explain human behavior patterns. Dynamisms were described as relatively enduring patterns of energy transformations that characterize the interpersonal relationships and emotional functioning of a person. A dynamism originates in a biological source of energy such as hunger. This energy is expressed and directed toward culmination in an interpersonal event (e.g., the hungry infant's cries summon the mother, with whom the infant has an interpersonal experience). Dynamisms may be healthy or unhealthy. Sullivan (1953, 1954, 1956) often linked dynamisms to diagnostic labels such as obsessional dynamisms or paranoid dynamisms. The concept of dynamism emphasizes the unity of internal and interpersonal functioning.

Sullivan's (1953) concept of the self system was that of a system composed of all the security operations used by a person to defend against anxiety and to seek security. Both the self system and security operations are used to achieve close satisfying relationships. The self system will be discussed more thoroughly in the discussion of developmental issues.

Sullivan (1953) hypothesized that personality and behavior develop in response to experience and that the human organism interprets behavior from three modes of organization, which depended upon the maturity of the organism. The prototaxic mode is used in infancy. It is primitive, limited, irrational, and emotional. Within the prototaxic mode, perceptions are global and unidimensional (e.g., the perception of the mothering one is separated into a good and bad mother). The parataxic mode develops in early childhood and is a way of thinking in which connections are made between events, but these connections are not necessarily logical or reasonable. For example, the child in the parataxic mode of thinking may decide that some normal emotions or behaviors are unacceptable because of parental responses. The syntaxic mode of thinking develops throughout later childhood and onward and is a mode of organizing experience that develops realistic and logical connections between events and experiences. The underlying basis of the syntaxic mode is consensual validation, which was described as agreements reached between or among people about connections between thoughts, feelings, and experiences. Consensual validation is the tool used by individuals who have achieved the syntaxic level of thinking to achieve realistic appraisals of the interpersonal environment.

Sullivan (1953) was more interested in awareness and unawareness than in the abstract concept of the unconscious. He stated that one could directly observe that an individual has varying degrees of awareness about what he or she is actually doing and about the reasons for the behavior. Anxiety is often the cause of unawareness, especially unawareness about interpersonal relations. An individual who is unaware of the nature of his or her interpersonal relations learns nothing from them. Unawareness of unhealthy interpersonal relationships leads to the repetition of unhealthy behaviors and can result in deteriorating relationships and new problems.

Personality Development

Sullivan (1953) divided personality development into heuristic stages, a division that he stated was convenient for the organization of thought. These stages were not identified by specific ages but rather by the development of interpersonal skills and/or needs. The ages at which these stages begin and end differ from individual to individual, and the transitions are gradual rather than abrupt. The stages identified by Sullivan (1953) were infancy, childhood, the juvenile period, preadolescence, adolescence, and adulthood. Sullivan seemed to have been interested primarily in personality development in the first five stages. In his lectures that have been collected and printed, he avoided discussion of continuing personality development in adulthood except for that which might occur in psychotherapy (Sullivan, 1953, 1954, 1956, 1972).

Infancy. Sullivan (1953) defined the period of infancy as existing from birth until the baby starts to use language coherently. He described infancy as a very important period of psychological development, which lays the groundwork for many important developments in interpersonal relationships, dynamisms, security operations, and the emerging self system. From the time of birth, the infant is flooded with sensations beyond his or her capacity for understanding, leaving the infant relatively helpless in the face of threats to his or her well being as well as in procuring the satisfaction of needs.

Sullivan (1953) stated that hunger is usually the source of the infant's most acute discomfort and that food is the primary need. Crying is the signal by which the infant announces discomfort and what Sullivan called the tension of needs. Sullivan (1953) stated, as a theorem, that, “The observed activity of the infant arising from the tension of needs induces tension in the mothering one which tension is experienced as tenderness and as an impulsion to activities toward the relief of the infant's needs” (p. 39). When the infant is approached with emotionally comfortable tenderness, the infant experiences satiation of needs and is provided with the bedrock of interpersonal security.

On the other hand, Sullivan (1953) stated that the mothering one may approach the crying infant with discomfort and anxiety. He introduced a second theorem. "The tension of anxiety, when present in the mothering one, induces anxiety in the infant" (p. 41). In this circumstance, the infant is provided with a foundation for discomfort about closeness with others. Sullivan was aware that from day to day, even from hour to hour, the mothering one may approach the infant in differing states of mind and with differing balances of tenderness and anxiety. What is learned by the infant depends upon the preponderance of one state over the other. Conditions in which needs are not satisfied lead to the dynamism of apathy. Inescapable or prolonged anxiety leads to the dynamism of somnolent detachment, a lethargic state of noninvolvement in the world.

Sullivan (1953) stated that the view of the world that the infant carries into childhood and beyond is determined by the emotional atmosphere that prevailed in the infant's introduction to material objects in the context of close interpersonal relationships. In the stage of infancy, the child develops what Sullivan called rudimentary personifications of the mothering one. It should be noted that Sullivan recognized that the mothering one may not even be the same person at each time that the infant's needs are met. The mothering one, upon occasion, may be another individual such as an older sibling, the father, or a nanny. Nevertheless, as infancy progresses, the infant develops a prevailing personification of either a good mother/nipple or a bad mother/nipple. The good mother personification is one of a consistent, emotionally comfortable mothering one who provides tenderness. The infant who develops a good mother personification will, in general, in later life approach interpersonal relationships with confident, pleasant anticipation. When the infant is approached 50% to 60% of the time with anxiety producing discomfort, the infant will develop a personification of the mothering one as a bad mother, and interpersonal relationships will be anticipated with discomfort and anxiety.

Sullivan (1953) stated that the infant fuses the concepts of good and bad mother into a single composite of mother in the later stages of infancy. Concepts formed later in life of father, siblings, peers, authority figures, and others are influenced by patterns formed in the mother-child relationship.

Sullivan (1953) also hypothesized that the rudimentary aspects of the self system develop in the later stages of infancy. The infant evolves three concepts of his or her self: the good-me, the bad-me, and the not-me. The good-me is emotionally comfortable. Sullivan stated that the good-me is the beginning personification that organizes experiences in which satisfactions have been enhanced by rewarding increments of tenderness. "Good-me, as it ultimately develops, is the ordinary topic of discussion about 'I'" (p. 162). This is the product of an infancy in which most of the contacts between self and other have been in an atmosphere of anxiety free tenderness. The message conveyed is that the infant is a worthwhile, esteemed individual.

When the infant is cared for in a predominantly anxious and/or rejecting manner, he or she develops a self system that is conceived as a bad-me. Bad-me is the beginning personification that organizes experience in which increasing degrees of anxiety are associated with the interpersonal experiences with the mothering one. The self concept that develops in the overly anxious interpersonal experience with the mothering one is the bad-me self concept in which the self is assumed to be inadequate, worthless, and troublesome. The bad-me is anxiety ridden, and interpersonal relationships are approached with anxiety, shame, and feelings of inferiority (Sullivan, 1953).

The rudimentary personification of not-me comes from the experience of intense anxiety. These are organizations of experiences marked by what Sullivan described as "uncanny emotions" (1953, p. 163). Such emotions include ones that are experienced later in life as awe, horror, loathing, or dread. Such experiences associated with extreme anxiety can not be clear or useful guides for understanding interpersonal experience. Thus, the experiences of intense anxiety marked by uncanny emotion persist throughout life as "primitive, unelaborated, parataxic symbols" (1953, p. 163). The not-me personification encapsulates the most extreme kinds of emotional distress and personality disintegration. The not-me component of the self system was considered by Sullivan to be practically indescribable in ordinary communicative terms.

In late infancy and childhood, the child fuses the three concepts of the self. The nature of the final concept depends upon the strength of the three components. The dominant concept will prevail most of the time, and most people establish a stable equilibrium with the good-me as the dominant component. Emotionally disturbing events may trigger the emergence of the bad-me. Under extreme stress the not-me may be dominant. Emotionally healthy relationships strengthen the role of the good-me and weaken the less desirable senses of the self.

Childhood. Childhood begins when articulate speech emerges and extends until the child develops strong interpersonal needs for association with children of the same age outside the family. Personality continues to develop within the close interpersonal relations of the family. Constructive relations with the father or other family members can help the child develop strengths that were not learned in infancy. Although destructive relationships may undermine earlier accomplishments, Sullivan (1953) stated that later interpersonal relationships were more likely to repair harm than to erode strengths.

During childhood, the individual slowly starts to develop syntaxic thinking. The basic process of syntaxic thinking is consensual validation. In consensual validation, a person arrives at a healthy consensus or agreement with one or more people about some aspect of thoughts, feelings, or experiences. This consensus is validated further by repeated interpersonal experiences with continued consensual validation (Sullivan, 1953).

The juvenile period. The juvenile period begins when the child develops strong needs for interpersonal experiences with peers outside the family. The juvenile period offers continuing interpersonal experiences that may have profound therapeutic effects in correcting earlier unhealthy experiences. Sullivan (1953) stated that the primary contributions to growth in the juvenile era are the experience of social subordination and the experience of social accommodation.

The experience of social subordination is provided to the child by his or her exposure to a variety of authority figures, such as teachers, recreational directors, older children, and school crossing guards. These authority figures allow the child to have new experiences of success and failure with concomitant influences on the personality and the understanding of interpersonal experience. The experiences with new authority figures also allow the child the opportunity to compare his or her parents with other authority figures and to compare the interpersonal experiences with these authority figures with the interpersonal experiences with the parents (Sullivan, 1953).

Social accommodation refers to the experiences the child has with same age peers. The child learns that there are great varieties of individual differences among people. As the child compares himself or herself to others, acceptance of his or her own individuality is achieved. The child also has the opportunity to refine skills of competition and

compromise. Finally, the association with both authority figures and peers allows opportunities for consensual validation that were not available in the family.

Preadolescence. Sullivan (1953) hypothesized that, during preadolescence, there is a growing need for intimacy or a strong need for close friendship with a same sex peer. Sullivan called this type of friend a chum. He claimed that this capacity for intimacy with a chum carried over into adolescence and was transformed into intimacy with a member of the opposite sex. The establishment of intimacy with a chum also provided opportunities for interpersonal growth and the development of new interpersonal strengths. Intimate friendship enhances the capacity to compromise and to cooperate. It also provides the individual with the opportunity to analyze the self through the eyes of another and to correct autistic, fantastic, or distorted views of the self. Finally, within intimate friendship, there is an opportunity for consensual validation of personal worth (Sullivan, 1953).

Adolescence. Sullivan (1953) postulated that the basic organizing force of adolescence is lust, defined as “the felt component of integrating tendencies pertaining to the genital zone of interaction, seeking the satisfaction of cumulatively augmented sentience culminating in orgasm” (p. 263). Sullivan (1953) stated that there was no necessarily close relation between lust and the need for intimacy. He thought that much of the complexities and difficulties experienced in adolescence and later stages of life depended on the ability of the individual to distinguish among the needs for personal security and freedom from anxiety, the need for intimacy or close collaboration with another person, and the need for lustful satisfaction, which is connected with genital activity in pursuit of orgasm.

Sullivan (1953) stated that, during adolescence, lust impels a person to seek intimacy with an individual of the opposite sex. However, the early adolescent needs three or four years to develop interpersonal skills and comfort with the opposite sex before the individual is able to achieve intimacy. Intimacy was defined by Sullivan as a state of being in which the well-being of another person is as important to the individual as his or her own well-being. Lust was defined as the last of the major dynamisms that motivate and direct human emotional processes and personality development. When the lust dynamism and the drive to achieve intimacy are combined during adolescence, they form a powerful force in molding a person's interpersonal functioning.

Sullivan (1953) thought that the achievement of a two-person heterosexual relationship in which each person finds comfortable expression of the lust dynamism and the need for intimacy provided valuable opportunities for correcting unhealthy interpersonal expectations. Such an intense relation provides an opportunity to correct interpersonal defects formed earlier in life and can lead to the development of improved capacities for forming healthy, comfortable relationships with people in general. Adolescence is, in general, an opportunity for spontaneous therapeutic growth.

The Processes and Goals of Psychotherapy

Sullivan (1954) defined psychotherapy as an interpersonal process in which one person, designated an expert in interpersonal relations and emotional functioning, helps another person resolve problems. The therapist is a participant-observer engaged in an interpersonal relationship in which he or she is alertly observing the interaction while simultaneously influencing the nature of the relationship through his or her participation in it. When a therapist observes a patient, the patient's behavior and emotions are necessarily altered.

Sullivan (1954) stated that the therapist should assume an active role in the verbal interchange with the patient by asking questions and making comments. Of particular interest is Sullivan’s two-person “interpersonal field” composed of therapist and patient, which is a sample of the patient's interpersonal life that is available for direct scrutiny.

Sullivan (1954) identified five basic processes composing psychotherapy. First, recent interpersonal relations of importance to the patient should be examined. Second, he explored the patient's relationships with significant people in his or her past. Third, Sullivan thought that it was important to pay close attention to any immediate interpersonal crises in the patient's life. Fourth, he discussed the patient's anticipation of future interpersonal relationships. Finally, there was a continuing exploration of the patient-therapist relationship.

Sullivan (1954) wrote that the behavior of the therapist varied from patient to patient and from interview to interview depending on the individual patient and the emotional condition of the patient. In some interviews, the therapist might be silent most of the time. In others, he or she may be actively engaged in dialogue.

Sullivan (1954) stressed that the therapist should never assume that he or she knows what the patient is talking about until the therapist and the patient have thoroughly explored the subject. He stated that the therapist should ask questions to discover exactly what happened and what it meant to the patient. He also commented that the therapist should be aware of the patients’s facial expressions, intonations, and gestures because much was communicated through nonverbal means.

Sullivan (1954) stated that emotional health is achieved to the extent that a person becomes aware of what occurs in interpersonal relationships. Thus, a major task in psychotherapy is to identify the patient's use of security operations to avoid awareness of painful aspects of past and current interpersonal relationships and to identify and to attempt to relieve the anxiety that causes the security operations.

One security operation to be examined is the use of selective inattention in the patient's interpersonal relationships in his or her life outside of therapy. In addition, the therapist should make a systematic, deliberate attempt to identify and to stop the use of selective inattention within the therapeutic relationship. The therapist, thus, helps the patient to include in his or her experiences interpersonal experiences that previously have been excluded (Sullivan, 1954).

Another security operation that often occurs in the therapeutic relationship is the use of the “as if” operation in which the patient begins to treat the therapist as if the therapist were someone important to the patient in his or her past life. The use of this particular operation provides patient and therapist with important information about the origins of maladaptive behavior in important interpersonal relationships in the patient’s history. Sullivan (1954) called this behavior a parataxic distortion and said that it was important for the therapist to treat this behavior as material for the therapeutic process. The task of the observing therapist is to eliminate gradually the imaginary parataxic therapist by helping the patient realize the interpersonal origins of the distortion and how that interpersonal situation was disturbing to the patient.

Sullivan (1954) declared that therapy, to a large extent, consists of broadening the patient's experience by bringing into the focus of attention many experiences and behaviors of which the patient had previously been unaware. This lack of awareness had made it impossible to resolve his or her problems healthily. Sullivan (1954) stated that emotional problems occur when an individual has been restrained from the full use of his or her interpersonal capacities. In accordance with his concepts of the tendency toward health and the one-genus hypothesis, Sullivan stated that an individual spontaneously moves toward emotional health when obstacles to progress are removed. The principal task of the therapist is to remove obstacles by increasing awareness.

Sullivan (1954) said that awareness and understanding of interpersonal experience can develop only when formulated in communication with another person or through articulate, reflective thinking. Planned, systematic validation is a central aspect of the therapeutic process, which validates understanding and syntaxic thinking. An understanding of interpersonal relations that is consensually validated eliminates parataxic distortions and allows the patient to abandon unhealthy security operations.

Finally, Sullivan (1954) stated that a person achieves emotional health to the extent that he or she becomes aware of his or her interpersonal experience. With some individuals, interpersonal difficulties are so deeply ingrained that the person cannot develop awareness. Any approach to awareness arouses more anxiety than the individual can tolerate and, in some cases, may lead to disintegration of the personality. Sullivan warned that, in such cases, the usual goals of psychotherapy can be dangerous and should be modified or curtailed.

Current Interpretations in Interpersonal Psychotherapy

Kiesler (1982), in agreement with Sullivan (1953, 1954, 1972) on most underlying assumptions of the interpersonal theory of psychotherapy, emphasized the importance of interaction in understanding the individual. What each individual thinks and feels about his or her self is inextricably interwoven with what that individual believes others think of him or her. The way in which the client interacts with the therapist mirrors his or her interactions in other interpersonal relationships. The task of the therapist is to attend to, to identify, and to assess the client's characteristic interpersonal style. The goal of the therapy is for the therapist and client to identify, to clarify, and to establish alternative, flexible ways of interacting interpersonally. Kiesler (1982) placed additional emphasis on the importance of nonverbal communication of the client and the importance of the therapist's assessing and understanding nonverbal behavior. Kiesler also emphasized the importance of the therapist's own emotional response in understanding the client's maladaptive interpersonal style. He cautioned that it is a priority for the therapist not to respond in a complementary fashion to the patient's characteristic style. The therapist may break the unhealthy transactional style by responding in a neutral way to the client's characteristic provocations.

Chrzanowski (1982) wrote that Sullivan's (1953, 1954, 1956, 1972) formulations were intentionally open-ended and incomplete and that, since his death in 1949, valuable modifications, extensions, and innovations have arisen in interpersonal psychotherapy. One of the major changes introduced by Chrzanowski (1982) is the replacement of the concept of the therapist as a participant-observer with the concept of therapist and patient in what he calls "relational participation" (p. 36). Within the context of relational participation, the patient and therapist engage in what Chrzanowski identified as a "collaborative inquiry" (p. 37). For Chrzanowski, both relational participation and collaborative inquiry imply a more egalitarian relationship between client and therapist than was connoted by the participant-observer relationship in which the therapist is an expert.

Chrzanowski (1982) also criticized Sullivan (1953) for placing too much emphasis on the role of anxiety in the development of the self. He hypothesized that there was more underlying structure and detail to an individual's self system than can be explained by the individual’s efforts to avoid anxiety or the mechanisms used to avoid anxiety . One of his comments was that qualities such as integrity, compassion, courage, temperament, and leadership cannot be sufficiently explained as interpersonal or transactional phenomena.

Chrzanowski (1982) emphasized the mutual contributions made to the therapeutic process by both the client and the therapist. The therapeutic process identified by him is a collaborative pattern in which the therapist is not necessarily an expert at identifying the pattern. The primary supplier of information is the patient, but the exploration of the information or data emerging is a mutual process of the patient and therapist. Both are involved in a mutual examination of how patterns of the patient’s interpersonal relationships are manifested in the therapist-patient interaction. The problems of the patient are no longer self-perpetuating psychopathologies but are phenomenological events involving patient and therapist. The patient’s perception of the therapist can be just as important to the therapeutic process as the therapist's perception of the patient.

Wachtel (1982) maintained the position that therapeutic activity in interpersonal psychotherapy appropriately can include much more active interventions than had been proposed by Sullivan (1954). Wachtel described how active interventions used in such therapeutic techniques as assertiveness training and role-playing can enhance the patient's awareness about characteristic interpersonal interactions and possibilities for new styles of relating to others.

Cashdan (1982) defined psychopathology as behavior that is composed of maladaptive strategies that exploit others in interpersonal relationships. He defined these strategies as subtle and as persisting over long periods of time. He identified four primary maladaptive strategies: dependency, martyr, sexuality, and power strategies. He stated that each represents a stylistic mode of structuring close relationships with each having a set of unique communications and metacommunications.

Cashdan (1982) stated that dependency strategies are perhaps the most prevalent and commonly occur in, but are not restricted to, clinical depression. With dependency strategies, the individual uses helplessness as a means of forming and maintaining personal relationships. Persons who employ dependency strategies spend an inordinate amount of time asking for directions and opinions and manipulating others into making decisions for them. The metacommunication of dependency strategies is "I can't live without you" (p. 217).

The martyr strategies are founded on duty and sacrifice. Persons who employ martyr strategies tend to be obsessed with doing things for others, often despite the annoyance of the others. The metacommunication of the martyr strategy is "You owe me" (Cashdan, 1982, p. 217).

Sexuality strategies rely on eroticism as a basis for a relationship. The primary mode of interaction is seductiveness. The person who uses sexuality strategies uses sex to buy love, to relieve boredom, and to impress others. The metacommunication is “I’m only valuable as a sexual object” (Cashdan, 1982).

Power strategies are interpersonal control maneuvers that are based on fear and intimidation. Persons who use power strategies seem unable to share responsibility or to take suggestions. The metacommunication is "You can't survive without me" (Cashdan, 1982, p. 218).

Cashdan's (1982) description of the process of psychotherapy is similar to what has been described previously by Sullivan (1954) and Kiesler (1982). The therapist and patient establish a therapeutic relationship. Maladaptive patterns of relating emerge in the therapeutic relationship as the patient engages in his or her characteristic maladaptive strategies in his or her behavior with the therapists. The therapist identifies the maladaptive strategies and clarifies them for the client. The client and therapist then enter a phase of therapy in which the major consideration is the development of new interpersonal patterns of behavior to replace the maladaptive strategies. Cashdan noted that it is important for the therapist to respond to new behaviors in a supportive, affirming manner. Cashdan emphasized that the encouragement and support from the therapist are less important to the client than is the offering of an impersonal mirror that shows the client how his or her behavior affects others–in this case, the therapist.

Levenson (1983) wrote persuasively about the superiority of the interpersonal approach of psychoanalysis to classical Freudian approaches. He stated that psychoanalytic inquiry should be rooted in the matrix of the real experience of the patient and the therapist in which a successful outcome is one in which the patient develops interpersonal competencies based on semiotic skills that make it possible to distinguish the nuances of human interaction. Semiotic skills were defined by Levenson as skills in the entire range of symbolic exchange, including verbal language, customs of society, rhetorical devices, and nonverbal cues. Levenson essentially dismissed the importance of relinquishing infantile distortions as an outcome of psychotherapy.

Levenson (1983) proposed that psychological difficulties arise because of the complexities involved in understanding nuances of social experience, especially as mediated by language. Meaning is context-dependent, and understanding messages depends on grasping the nuances of the interpersonal experience. People run into difficulties "because they are tangled in an elusive web of omissions, simulacrums, and misrepresentations" (p. 45).

Levenson (1983) emphasized that one of the first important steps in psychotherapy is the establishment of the frame that is the definition of limits in the therapeutic context. The frame includes physical constraints such as limited contact, appointment times, frequency of sessions, fees, cancellations, and vacations. The frame also includes a conceptual delineation of the expectations and constraints of the patient-therapist interaction. A mutual contract is established to maintain the therapeutic relationship and to examine what happens between patient and therapist.

Levenson (1983) suggested that an additional aspect of the psychotherapeutic frame is the capacity of the patient and therapist to engage with one another in the spirit of "structured play" (p. 59). Levenson went on to state:

From the interpersonal perspective, the playground of psychotherapy is a situation of augmented and clarified semiotic message rendered unique by the framing and permitting patient and therapist an opportunity to examine the layering of semiotic experience first-hand. (p. 61)

According to Levenson (1983), in interpersonal analysis, transference develops in a subtle, ongoing dialogic discourse between the two participants, even when the therapist is totally silent. The interpersonal therapist does not even attempt to achieve neutrality but instead strives toward authenticity. The interpersonal effort is to be one's self with all of one's imperfections and shortcomings, taking full responsibility for one's own thoughts, feelings, and behavior. If the therapist is wrong, it can not be blamed on the patient's "resistance" or "borderline pathology" (p. 104).

Levenson (1983) hypothesized that if the patient were placed in the safety and security of the therapeutic playground, he or she would learn that interaction with another imperfect person makes it possible to acquire the skills needed to discriminate and to identify the nuances of others' communications and to influence others with his or her own communications. Maturity emerges as a by-product of a sense of power and control of one's own life. The patient learns to trust his or her own judgment.

Levenson (1983) stated that one cannot avoid the observation that people resist change. Providing a corrective emotional experience, abreaction of feeling, direct guidance, and many other therapeutic strategies and techniques may help a patient but will not necessarily induce change. In order for change to be possible, the patient must invest and participate in the very system of human experience that confuses and mystifies the patient.

Stern (1987) stated that, from an interpersonal view, a therapist cannot decide what kind of relationship to provide for a patient or what function to serve. The interpersonal therapist simply finds himself or herself in a relationship serving a function, and the task of the therapist is to develop with the patient a joint interest in observing and understanding the relationship without having this effort converted into a new example of the characteristic interpersonal problems of the patient. The interpersonal therapist understands that a lucid and accurate description of the analytic relationship leads to a mutative influence. When the analyst and patient concern themselves with identifying and understanding the pattern of the relation in which they have involved themselves, through characteristic patterns, they are no longer able to continue the pattern, and a new experience occurs.

The Heuristic or Guide to Investigation

The interpersonal therapist's interest is in information about personality characteristics that can by directly observed in interpersonal relationships. The primary sources of information about the patient are the behaviors that can be directly seen or heard or observed by the therapist or what the patient reports about his or her experience. Although there was some speculation about internal processes, the main focus was on observable or reported feelings, thoughts, and behavior. The interpersonal therapist is equally concerned with the information communicated by both verbal and nonverbal behaviors (Bromberg, 1989; Chapman, 1976; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Mullahy, 1970; Stern, 1987; Sullivan, 1953, 1954, 1956, 1972).

Ontology

Essential Human Nature

As was discussed in the section on basic concepts, Sullivan (1953) stated that individuals have a basic tendency toward emotional health and sound interpersonal functioning. This tendency toward health persists throughout the life of the individual regardless of the nature of earlier experiences. If family relations are unhealthy, the child seeks to learn healthy interpersonal skills from authority figures at school or from peers on the playground (Sullivan, 1953).

Sullivan (1953, 1954, 1956) also developed what he called "the one-genus hypothesis" (1953, p. 32) according to which the differences between an individual who is emotionally healthy and one who has a psychiatric illness are differences of degree rather than of nature. Thus, Sullivan’s conception of essential human nature was one that he described as fairly homogenous with all humans basically motivated by the desire and capacity for healthy, satisfying interpersonal relationships.

The Development and Nature of Psychopathology

As was discussed previously in the section on basic concepts, Sullivan (1953) stated that all psychopathology develops in response to anxiety in interpersonal relations. Security operations, dynamisms, and the development of a self system were described in detail as were the prototaxic, parataxic, and syntaxic modes of thinking. Sullivan’s (1953) stages of personality development also were discussed. Infancy is a particularly important period of development, forming a foundation of expectations about interpersonal relationships. The rudimentary beginnings of the self system were described as originating in infancy with the development of three concepts of the self: the good-me, the bad-me, and the not-me. As the child develops, the three concepts of the self are fused with dominance of one of the three selves being dependent upon the nature of previous interpersonal experiences.

Personality continues to develop throughout childhood, the juvenile period, preadolescence, and adolescence with opportunities for healthy interpersonal relationships that may correct earlier unhealthy experiences. However, a predominance of anxious uncomfortable interpersonal experiences in any stage of development may lead the individual to develop unhealthy dynamisms and security operations that exist outside of awareness (Sullivan, 1953).

Cashdan (1982) defined psychopathology as subtle, maladaptive behaviors that exploit others in interpersonal relationships. He identified four primary pathological strategies: dependency strategies, martyr strategies, sexuality strategies, and power strategies.

In general, most contemporary interpersonal theorists, in agreement with Sullivan (1953, 1954, 1956), have not been concerned particularly with labeling or classifying psychopathology. The psychopathology of an individual is interpreted as characteristic of that person's interpersonal style, which is based on past interpersonal experience. How the individual feels about and presents himself or herself is dependent upon what the individual thinks that others think of him or her (Bromberg, 1989; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987).

The Role of the Therapist

To most interpersonal theorists, the therapist is a participant-observer engaged in the interpersonal therapeutic relationship in which the therapist alertly observes the relationship while simultaneously affecting the nature of the relationship by participating in it. The relationship between therapist and patient is a sample of the patient's interpersonal life that is available for direct observation (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1954).

Sullivan (1954) stated that it is important for the therapist to take an active role in collecting information, particularly in the beginning stages of psychotherapy. He stated that it is important to explore current and past important interpersonal relationships as well as to continue to explore the patient-therapist relationship. It is the responsibility of the therapist to identify security operations, dynamisms, and parataxic distortions. As the patient's awareness increases and unhealthy security operations are abandoned, the patient engages the therapist in consensual validation of healthy, satisfying interpersonal experiences. Later interpersonal theorists have emphasized the role of the therapist in providing the patient with a mirror with which the patient can identify the effects of his or her behavior upon others (Bromberg, 1989; Cashdan, 1982; Kiesler, 1982).

Most interpersonal theorists assume that both patient and therapist are involved in a mutual, collaborative effort to understand the interpersonal dynamics of the patient and, often, the origins and meaning of those dynamics. Accordingly, the most important contributions made by the therapists are interventions that enhance the patient's awareness of the self (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1954; Wachtel, 1982). The basic question to be answered is "what's going on around here?" (Levenson, 1983, p. ix). It should be noted, however, that several interpersonal theorists have commented that some patients have such deeply ingrained interpersonal difficulties that the usual procedures of psychotherapy must be modified or curtailed (Bromberg, 1989; Sullivan, 1954).

The Mechanism of Change

The mechanism of change in interpersonal psychotherapy can be defined as an increase in awareness of the patient's habitual maladaptive interpersonal patterns of behavior accompanied by an increased awareness of the possibilities for healthy, satisfying interpersonal experience. This increase in awareness is the result of collaborative exploration of the interpersonal experiences of the patient, including the interpersonal experience with the therapist. The therapist often takes an active role in identifying unhealthy interpersonal patterns and in providing the patient with an opportunity for a new interpersonal experience that does not repeat the unhealthy patterns of the past (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1954; Wachtel, 1982).

In addition, Wachtel (1982) proposed that active therapeutic interventions such as role-playing and assertiveness training could enhance awareness of the possibilities for new ways of relating to others. Levenson (1983) emphasized the importance of the therapist’s maintaining authenticity and taking full responsibility for personal imperfections and shortcomings in order to demonstrate the full spectrum of interpersonal interaction. However, the primary mechanism of change advocated by the interpersonal theorists was described by Stern (1987) as a therapeutic collaboration in which the analyst and patient concern themselves with identifying and understanding the dynamics of the relationship in which they have involved themselves based upon characteristic patterns of interacting. Through the identification and understanding of these characteristic modes of participating in interpersonal relationships, the pattern can be broken, and a new experience can occur. Thus, the primary mechanisms of change identified by interpersonal theorists are interpretation, insight, and understanding attained through utilization of the therapeutic relationship.

The Goals of Treatment

The goals of treatment in the psychoanalytic interpersonal school are basically to provide the patient with a therapeutic relationship within which the patient can use his or her tendency toward health to correct interpersonal deficits learned earlier in life and to develop improved capacities for forming healthy interpersonal relations with others. The collaboration between patient and therapist in exploring and interrupting the characteristic unhealthy patterns of relating increases the patient's awareness of his or her dynamisms and security operations. Increasing awareness breaks the patterns of previous unhealthy modes of interaction. The patient then can explore new and healthy modes of interpersonal interaction with the therapist, using consensual validation to support previously disrupted growth. The major goal of the therapeutic endeavor is to increase the patient's awareness of characteristic interpersonal interactions and of possibilities for new styles of relating to others (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1954; Wachtel, 1982).

Critique of the Psychoanalytic Interpersonal School

Wolberg (1995) criticized Sullivan (1953, 1954), in particular, and the interpersonal school, in general, for its focus on what is apparent and observable in interpersonal interaction while discrediting the importance of the inaccessible and unobservable within the human psyche. He was particularly critical of the tendency of Sullivan and the interpersonalists to dismiss the importance of infantile sexual drives.

Chapter 5: Object Relations Theories

Within psychoanalytic thought, object relations is the term used to describe the interactions of an individual with both people in the external world and the images of people that are established internally. The word "object" was used originally (Freud, 1905) to identify the source of gratification for a drive. The word "object" reflected the idea that the source of gratification could be a person, a part of a person such as a thumb or breast, or an inanimate object such as a blanket or a bottle. In object relations theory, however, the object is a person or the internal representation of a person in the intrapsychic experience of an individual (Greenberg & Mitchell, 1983; Pine, 1990).

Theories about object relations have been largely concerned with internal mental representations of self and others. These representations are bound together by emotions, memories, and expectancies that have an influence on current functioning, particularly interactions with others. Object relations theorists generally agree that people who have been important to an individual, particularly the significant others of childhood, become internalized as representations within the mind. These representations may be understood as serving a loose anticipatory image of what is to be expected from people in the real world. They are closely intertwined with the individual's experiences of who he or she is. They may serve as critical and demanding persecutors or as sources of internal security and comfort (Greenberg & Mitchell, 1983; Pine, 1990).

This idea of an internalized representational "map" of what is to be expected of and experienced by self and others leads to several difficult and interesting questions.

How do the characteristics of internal objects relate to those of "real" people, past and present? Is the internal object a representation of the individual's perception of a total relationship with another person or of specific aspects and characteristics of the other? What are the circumstances in which such images become internalized and what is the mechanism by which they are established as part of the individual's inner world? What is the connection between these internal representations and subsequent relations with real others in the external world? How do internal objects function within mental life? Are there different types of internal objects? Do different circumstances and mechanisms of internalization lead to different kinds of internal objects? (Greenberg & Mitchell, 1983, p. 11)

Different theorists have answered these and other questions in different ways. Some theorists have examined the development of object relations as a process to be considered separately from theories concerning drive and defense but without repudiating classical Freudian drive theory (Klein, 1964, 1975; Winnicott, 1958, 1964, 1965, 1989). Some object relations theorists have claimed that their theory of object relations replaced classical Freudian drive theory (Fairbairn, 1954). Other theorists have attempted with varying degrees of success to integrate object relations with drive theory (Jacobson, 1964; Kernberg, 1976, 1980, 1984; Mahler, Pine, & Bergman, 1975).

The focus in this chapter is on an examination of the major theories of object relations within psychoanalytic thought. The group of psychoanalytic theorists to be considered includes Klein (1964, 1975), Fairbairn (1954), Winnicott (1958, 1964, 1968, 1989), Jacobson (1964, 1971), Mahler (1968, 1972; Mahler et al., 1975), and Kernberg (1976, 1986, 1987). These theorists and their theories have been presented roughly in chronological order. Each theorist’s most important ideas will be presented with a discussion of basic concepts, personality development, and the processes and goals of psychotherapy. At the end of the section about each theorist, there will be a brief summary of the heuristics and of the ontology of each theory.

Melanie Klein

Klein (1964, 1975) made several major contributions to psychoanalytic thought. She was the first major theorist to emphasize the formation of internal intrapsychic objects and to create the concept of an internal world of objects. She described the formation of an individual’s character as being derived from the inherent object-relatedness of the libidinal and aggressive drives. She hypothesized that ego and superego development were products of the internalization of object relations (Greenberg & Mitchell, 1983; Hughes, 1989; Summers, 1994). Greenberg and Mitchell (1983) stated that for Klein “internal object relations constitute the basic substructure of experience and the stuff of the entire personality” (p. 150).

Summers (1994) also stated that Klein (1964, 1975) made valuable contributions to the understanding of psychopathology in her descriptions of primitive defenses. He stated that “her notions of splitting, projective and introjective cycles, and the mechanisms of denial, omnipotence, and especially, projective identification to be of inestimable value in the treatment of severe character pathology” (Summer, 1994, p. 121).

Basic Concepts

One of the central concepts in Klein's (1964, 1975) theories about the mental life of babies is an emphasis on innate destructive impulses, which differ in strength and pervasiveness across individuals. According to Klein (1975), hostility, resentment, greed, and envy are present at birth and exist in individuals to varying degrees as internal factors of personality that have an important impact on personality development. These "destructive impulses" (1975, p. 31) were thought by Klein to have their origins in an inherent death instinct, which aroused in the infant a fear of annihilation. Klein identified this innate death instinct as "the primary cause of anxiety" (1975, p. 28).

Klein (1975) hypothesized that the newborn infant feels unconsciously as if every discomfort has been inflicted by hostile forces. Thus, the birth experience itself and the adjustment to the postnatal environment give rise in the newborn to anxiety of a persecutory nature. Whenever discomfort occurs, persecutory anxiety is aroused within the infant. Frustration, discomfort, and pain, experienced as persecution, enter into the infant's feelings about the mother.

Klein (1975) also hypothesized that comfort given to the infant in the form of warmth, loving touch, and the gratification of being fed would relieve anxiety and induce "happier emotions" (p. 248). Such comfort makes possible the infant's first loving relation with a person or an "object" (p. 248). Klein (1975) wrote that when the infant's needs are met and love and understanding are experienced through the mother's handling of the baby, the infant feels, "an unconscious oneness that is based on the unconscious of the mother and of the child being in close relation to each other" (p. 248). These feelings provide the foundation for the fundamental relation of the infant's life, the relation to the mother.

In the first few months, the mother represents to the child the whole of the external world. In the infant's mind both good and bad come from her, leading to a twofold attitude toward the mother even under the best possible conditions. Thus, both the capacity for love and a sense of persecution have deep roots in the infant's earliest mental processes. The mother upon whom the infant depends for the gratification of all needs is also the target of the destructive impulses of resentment, rage, and envy (Klein, 1975).

Klein (1975) further hypothesized that the infant has from birth "an ego which has some rudiments of integration and cohesion and progresses increasingly in that direction" (p. 71). The ego was described by her as the organized part of the self that directs all activities and establishes and maintains the relation to the external world. At birth, the ego has the important task of defending itself against anxiety primarily through the use of processes Klein identified as introjection and projection.

Introjection was defined as a process in which the outer world, its impact, the situations the infant lives through, and the objects encountered are taken into the self and become a part of the infant's inner life. Projection was described as a process in which the feelings and impulses of the child are attributed to people or parts of people in the environment. A special type of projection which Klein called projective identification was described as a process in which the child splits the self and projects one part of the self onto another (Klein, 1975).

Klein (1975) proposed that the processes of introjection and projection operated on the level of what she called unconscious fantasias. Unconscious fantasias were described as the constant activity of the mind that occurs on deep unconscious levels and accompanies every impulse experienced by the infant. "Fantasias—becoming more elaborate and referring to a wider variety of objects and situations—continue throughout development and accompany all activities; they never stop playing a great part in mental life" (Klein, 1975, p. 251).

Klein (1975) also hypothesized that object relations exist from the beginning of life with the original relations being formed with part-objects (i.e., parts of the mother's body such as the breast). Both self and object are split in the early months of life which creates a twofold relation to the first object.

Among the destructive impulses at birth, which vary in intensity from one individual to another, Klein (1975) was particularly interested in greed and envy. Klein described greed as an insatiable craving exceeding what the subject needs and what the object is able and willing to give. Greed is an impulse to devour the breast and absorb it totally through "destructive introjection" (p. 181).

Envy, on the other hand, was defined by Klein (1975) as a desire to spoil and to destroy an all-powerful good object. Klein stated that "the first object to be envied is the feeding breast, for the infant feels that it possesses everything he desires and that it has an unlimited flow of milk, and love which the breast keeps for its own gratification" (p. 183).

Personality Development

Klein (1975) labeled the first three or four months of life "the paranoid-schizoid position" (p. 61). As was previously described, Klein hypothesized an innate death instinct as the primary internal source of anxiety. In addition, the newborn experiences the pain and discomfort of birth and the loss of the intra-uterine state as a hostile attack which forms an external source of anxiety. Persecutory anxiety, thus, enters into the infant's relations with objects from the beginning.

The first external object experienced by the child is a part object, the breast. To the extent that the child feels persecutory anxiety, the breast appears to be hostile. Accordingly, the infant turns his or her own rage and other destructive impulses against the breast while also projecting the destructive impulses onto the breast. The emotions of the young infant are characteristically extreme and powerful. The frustrating and bad breast, thus, is felt to be a terrifying persecutor that is hated. In the infant's destructive and sadistic fantasias, the breast is bitten, torn, devoured, and annihilated (Klein, 1975). This persecutory hostile breast is also introjected, intensifying the fear and anxiety within the infant.

There is therefore a constant fluctuation between the fear of internal and external bad objects, between the death instinct acting within and deflected. Here we see one important aspect of the interaction—from the beginning of life—between projection and introjection. External dangers are experienced in the light of internal dangers and are therefore intensified The very fact that the struggle has, to some extent, been externalized relieves anxiety. Externalization of internal danger-situations is one of the ego's earliest methods of defense against anxiety and remains fundamental in development. (Klein, 1975, pp. 31-32)

Klein (1975) wrote that the life instinct is also innate, activated from birth and is attached to the good, gratifying breast. The good gratifying breast is, then, also introjected, reinforcing the power of the life instinct within the infant. The good internalized breast "forms a vital part of the ego and its preservation becomes an imperative need" (p. 32). The good breast is felt to be an idealized, perfect, and inexhaustible source of gratification. The idealization of the good breast makes it a powerful counterpart to the destructive persecuting breast. This idealization is, thus, a defense against anxiety.

Klein (1975) stated that destructive impulses such as greed, rage, and envy are the most powerful impulses in the early months of life. Dominant defenses at this stage are splitting, omnipotence, idealization, denial, and control of internal and external objects. These defenses are primitive but are in congruence with the intensity of early feelings and the limited capacity of the infant's ego to bear acute anxiety. They allow the infant to achieve a relative and temporary security predominantly by keeping the persecutory object separated from the good object.

During the paranoid-schizoid position, anxiety is mainly of a paranoid nature, and the defenses are primarily schizoid. If a predominance of destructive impulses, particularly an envious spoiling of the good object, interferes with the establishment of a separate good object, there is no basis for a fully developed and integrated adult personality. Klein (1975) suggested that this disturbance in normal development forms the basis of schizophrenia.

Klein (1975) hypothesized that, following the paranoid-schizoid position, the next stage of development is the infantile depressive position which occurs during the first four to six months of life. During this time, the infant's relation to external objects grows more differentiated. As the ego develops, the range of gratifications, the interests, and the ability to express emotions and communication increase. Fantasias become more elaborate and differentiated.

During this stage of development, the infant begins to perceive and introject the mother as a person, a whole object. Processes of integration and synthesis result in less separation between love and hate and between good and bad objects. The infant realizes that the destructive impulses are directed to the same object (the mother) who is the source of love and gratification. In this stage, anxiety is of a depressive nature accompanied by guilt induced by the perceived danger to the object of the destructive impulses (Klein, 1975).

The internalized mother is felt by the infant to be injured, suffering, and in danger of annihilation. There is an identification with the injured object that is accompanied by a drive to make reparation and by attempts to inhibit aggressive impulses. The infant feels that destructive impulses and fantasias are directed against the complete person of the loved object. Guilt about these destructive impulses is accompanied by urges to repair, to preserve, or to revive the loved, injured object (Klein, 1975).

If the infant is able to introject a reassuring, uninjured external object, his or her internal world improves. As the infant re-introjects again and again a realistic and reassuring external world and establishes within the self complete and uninjured objects, essential developments take place in the organization of ego and superego. Bad internal objects are mitigated by the strength of good internal objects, which allows the ego to assimilate progressively a constructive superego. Furthermore, full identification with a good, uninjured object promotes the feeling that the self possesses goodness of its own (Klein, 1975).

If the infant is faced with a multitude of anxiety-provoking situations and is unable to internalize a good uninjured object, the ego may enter a state of denying the fact that the object is loved at all. The result may be an increase in persecutory anxiety and a regression to the paranoid-schizoid position (Klein, 1975).

Klein (1975) hypothesized that working through the depressive position results from an assimilation of and identification with a good object. This assimilation of a good object is the foundation for a capacity for love and devotion to people, values, and causes. A major derivative of a capacity for love is the capacity to feel gratitude, and a capacity for gratitude is the foundation for generosity. "Inner wealth derives from having assimilated the good object so that the individual becomes able to share its gifts with others" (Klein, 1975, p. 189).

Klein (1975) hypothesized that the Oedipus complex developed concurrently with the depressive position. The successful resolution of the Oedipus complex depends upon the successful resolution of the depressive position. If the infant is successful in internalizing the mother as a good object and overcoming destructive impulses of envy and greed, then the child can resolve the jealousy and rivalry of the Oedipus complex. The successful resolution of the Oedipus complex also results in a loving relation with the father, who becomes the second externalized object.

Klein (1975) suggested that working through the envy, rivalry, and jealousy of the Oedipus complex starts in infancy and extends into the first years of childhood, normally coming to an end at about five years of age. Klein's (1964, 1975) theories were almost exclusively concerned with development of personality and mental life in the first year of life.

The Processes and Goals of Psychotherapy

Klein (1975) stated that the most important task of psychotherapy was the analysis and resolution of the conflicts and anxieties experienced in the first year of life. Both persecutory and depressive anxieties would be aroused in transference with the internal introjections of early childhood, good and bad, being projected onto the analyst. It was only by analyzing the negative as well as the positive transference that anxiety could be reduced at its roots in early childhood.

As persecutory and depressive anxieties are interpreted and reduced in the analysis, the patient ceases to separate the analyst in the transference into different representations of the various different internal introjections of the significant others of early development. The patient achieves a greater synthesis among various aspects of the analyst, accompanied by a greater synthesis of the various aspects of the superego. Good objects are established in the mind as the split between persecutory and ideal objects is diminished and as the ego achieves integration (Klein, 1975).

Klein (1975) developed the idea that, in the analysis, some of what is revived in the patient is what she calls "memories in feeling" (p. 234). The idea is that memories of early infancy exist only in emotions that may be activated and experienced in transference. In the course of the revival of these feelings, it may become possible for the patient to develop a different attitude toward early experiences.

Klein (1975) also pointed out that some analyses end in failure because of excessive envy on the part of the patient. Patients who suffer from excessive envy (which Klein thought was largely constitutional and innate) are unable to accept the relief, comfort, and hope brought by a good interpretation. The envious patient spoils and ruins the good interpretations, leading to slow progress or failure of the analysis.

In other cases, the interpretation of envy and other destructive impulses in the patient can lead to appropriate depression, guilt, and urges to make reparation. The working through of the mourning associated with the depressive situation can lead to the establishment of an internal good object and to the integration of the ego. Envy is diminished, and the patient's capacity for enjoyment and gratitude increases.

Heuristic

Klein (1964, 1975) based her theoretical model upon a combination of her observations of small children in analysis and upon the transference responses of adult patients in analysis. From these observations, Klein developed complex interpretations of the meanings of children’s play and adult transference responses and formulated her theories about the mental life and development of intrapsychic structure in infants.

Ontology or the Essential Concepts of the Theory

Essential Human Nature

Klein (1964, 1975) stated that the individual had innate libidinal and aggressive instincts or drives. Innate destructive impulses of rage, hostility, resentment, greed, and envy emanate from the inborn death instinct and differ in intensity and pervasiveness from individual to individual. The infant was also considered to be born with a capacity to love which could be developed by warmth, loving touch, and gratification provided by the mother. Severe psychopathology was described as the result of an innate, constitutional hostility or destructive envy which seeks to destroy the good object regardless of the consequences to the self.

The Nature And Development Of Psychopathology

Klein (1975) proposed that the infant functioned from a paranoid-schizoid position in the first few months of life. Persecutory or paranoid anxiety are, thus, part of the infant's relations with objects from birth. Establishing a separate good object is the foundation of early ego development and, thus, is the basis of a fully developed and integrated adult personality. The disturbance in the normal development of an internalized good object caused by the preponderance of destructive impulses is the foundation for the development of schizophrenia (Klein, 1975). In normal development, the infant emerges from the paranoid-schizoid position and enters the depressive position. During this stage, the infant begins to perceive the mother as a person, a whole object, and realizes that the destructive impulses are directed at the same object who is the source of love and gratification. Anxiety becomes depressive in nature and is accompanied by guilt (Klein, 1975). Normal development is based on the infant's experience of a reassuring, uninjured external object. Full identification with a good, uninjured object promotes the feeling of the self possessing goodness of its own.

In summary, Klein (1964, 1975) most often described the development of psychopathology as based upon the destructive impulses of the infant interfering with the infant's absorption and internalization of experiences with good, gratifying objects. There is very little discussion of the effect on the child of depressed, frustrating, or otherwise inadequate parents.

The Role Of The Therapist

Klein (1975) thought the most important task of the therapist was the interpretation of the primitive conflicts and anxieties of the first year of life as these issues and feelings are aroused in the transference. As the therapist interprets the persecutory and depressive anxieties, these anxieties are reduced.

Klein (1975) did not describe directly the process of treatment as an opportunity for the patient to recapitulate and to repair the deficiencies of early development, although her description of successful therapeutic progress was obviously quite similar to her description of normal early development. Klein maintained that interpretation was the only mechanism of change and did not hypothesize any curative properties of the therapeutic relationship.

It is also the responsibility of the therapist to help the patient understand that feelings that arise in the analytic setting may represent or be the only memories the patient has from infancy. The revival and understanding of these feelings may help the patient to a deeper understanding of early experiences and accompanying changes in attitude and/or internal structure (Klein, 1975).

The therapist's interpretation of envy and other destructive impulses may lead to appropriate depression, guilt, and urges to make reparation. Working through the mourning of the depressive situation promotes the establishment of an internal good object and integration of the ego. Envy is diminished and the capacity for enjoyment and gratitude increases (Klein, 1975).

The Mechanism Of Change

Klein (1975) stated that interpretation, particularly interpretation of primitive affect states aroused in the transference, was the mechanism of change. Interpretation of both negative and positive transference reduces anxiety and promotes a greater synthesis of various aspects of the therapist. As the split between persecutory and ideal objects is diminished, good objects are established in the mind, and the ego achieves integration.

The Goals Of Treatment

Klein (1975) proposed that the goals of treatment are to reduce persecutory and depressive anxiety through interpretation and analysis. The patient is also encouraged to understand how feelings aroused in the transference may be memories of feelings in early infancy, thus giving the patient insight and understanding of early experience. The basic goals of these interpretations are to help the patient establish internalized good objects and integration of the ego.

W. R. D. Fairbairn: An Object-Relations Theory of the Personality

W. R. D. Fairbairn (1954) developed a theory in which he hypothesized that all human behavior and experience are based fundamentally on relations with others. Psychopathology is a result of disturbed internalized object relations that are a reflection of actual disturbances in relations with others. He stated that libido, the primary motivational energy of individuals, was essentially object seeking, and he called his theory an object relations theory of the personality.

Basic Concepts

Fairbairn (1954) said that, from infancy to maturity, the individual is primarily motivated by desires to secure and to maintain satisfactory and gratifying relations with others. He explicitly stated that aggression is not an innate human instinct. He thought aggression was a response of the individual to frustration or deprivation.

Fairbairn (1954) developed a unique theory of endopsychic structure based upon the internalization of interpersonal experience. In his conception, the infant internalizes the first object experienced. This internalized object is split into three objects: (a) an overly exciting object, which is repressed and made unconscious, (b) a frustrating object, which is also repressed and unconscious, and (c) the "nucleus of the original object shorn of its over-exciting and over-frustrating elements" (p. 178), which assumes the status of an idealized object and is accepted by the ego and not subjected to repression.

This splitting of the internalized object is accompanied by a concomitant splitting of the ego. The libidinal ego is a repressed, unconscious split off part of the ego tied to the exciting object. The anti-libidinal ego, also called the internal saboteur, is a repressed, unconscious, split off part of the ego connected to the frustrating object. Both the exciting and frustrating objects are bad objects. The exciting object promises and entices but does not gratify needs or wishes. The frustrating object simply withholds gratification. The anti-libidinal ego is the repository of hatred and rage which is directed both at the exciting object for its false promises and at the libidinal ego for its naive hopefulness that promises will be fulfilled (Fairbairn, 1954).

The central ego with its ties to a good, accepted object remains partly conscious and partly unconscious. The good, accepted object of the central ego forms the nucleus of the ego-ideal. The degree of psychopathology from which any individual suffers depends upon the amount of central ego available for real and potentially gratifying relations with others (Fairbairn, 1954).

Fairbairn (1954) explained in detail how the various internal object relations available to any individual contributed to various types of psychopathology. He wrote that obsessional, paranoid, hysterical, and phobic symptoms operate as defenses against the emergence of either schizoid or depressive tendencies. Thus, according to Fairbairn, psychopathology originates in personalities that have a preponderance of either schizoid or depressive tendencies.

Fairbairn (1954) wrote that conditions underlying the development of schizoid tendencies appear to arise in very early infancy in what he called "the pre-ambivalent oral phase . . . in which the individual feels that his love is bad because it appears destructive towards his libidinal objects" (p. 25). Essentially, Fairbairn (1954) hypothesized that when the infant experiences frustration, the child feels unloved and as if the love offered by the child to the mother is not valued or accepted. Repeated frustration leads to the feelings that what the child has to offer is poisonous and destructive.

The fundamental schizoid phenomenon is the presence of splits in the ego; and it would take a bold man to claim that his ego was so perfectly integrated as to be incapable of revealing any evidence of splitting at the deepest level. (Fairbairn, 1954, p. 8)

Fairbairn (1954) wrote that depressive tendencies arose from the late oral stage in which the basic conflict faced by the individual is the problem of how to love the object without destroying it with hate. Unlike the individual with schizoid tendencies, the individual with underlying depressive tendencies is spared "the devastating experience of feeling that his love is bad" (p. 53). The depressive individual feels inherently that his or her love is good and, thus, is readily able to establish libidinal contacts with others. As long as these libidinal relationships are satisfactory, the depressive individual remains undisturbed. However, the depressive individual is not able to achieve the step in development in which the object becomes differentiated so that hatred can be directed toward the rejected object while maintaining a loving relation with an accepted object. Thus, the depressive individual experiences any frustration in object relationships as functionally equivalent to loss of the object and cannot handle conflict or frustration in a relationship. The depressive individual fears that the expression of aggressive feelings will lead to the loss of the object. Loss of the object is the essential trauma that provokes the depressive state.

Fairbairn (1954) assumed that in early childhood all object relationships are based upon identification and that all children encounter bad objects which are internalized and repressed. He wrote that the development of delinquent, neurotic, psychotic, or normal characteristics in any individual depends upon (a) the extent to which bad objects have been internalized in the unconscious and the extent of their badness, (b) the extent to which the ego is identified with externalized bad objects, and (c) the nature and strength of the defenses that protect the ego from these objects.

Fairbairn (1954) further hypothesized that the internalization of good objects allowed the formation of a superego. The superego is capable of making judgments about goodness and badness from a moral position, which allowed the formation of what Fairbairn (1954) called "the moral defense" (p. 66). In the moral defense, the individual transforms the unconditional badness from the libidinal viewpoint to conditional badness from a moral viewpoint. The unconscious identification with what were originally unconditionally bad (libidinally unsatisfying) objects can be transformed by the individual into conditional badness accompanied by feelings of guilt and unworthiness. The belief in conditional or moral badness contains within it the hope of redemption, of becoming morally good.

Fairbairn (1954) stated, "It is better to be a sinner in a world ruled by God than to live in a world ruled by the Devil" (pp. 66-67). There is a sense of security and a hope of redemption in believing that the world is basically good. If the world itself is bad, all hope is lost.

Personality Development

Fairbairn (1954) theorized that ego development based upon object relations is essentially a process in which infantile dependence upon the object is gradually replaced by mature dependence upon objects. This process includes the abandonment of the original object relationship based upon identification and the adoption of an object relationship based upon differentiation of the object from the self. He regarded the process of ego development as having three stages: (a) a stage of infantile dependence, (b) a transitional stage, and (c) a stage of mature dependence.

The stage of infantile dependence is characterized by what Fairbairn called "an attitude of taking" (p. 39). The infant is totally helpless and dependent upon the object. When faced with deprivation and frustration, the infant uses primitive defenses to cope with his or her own aggression, which is aroused by the experience of frustration and deprivation. The mother is split into two objects, one satisfying and one unsatisfying. Through the process of internalization, the unsatisfying object then is removed from the outer reality, in which the infant has no hope of control, and is placed in the infant's inner reality, in which there is the possibility of control of it as an internal object (Fairbairn, 1954).

The transitional phase of ego development is the period in which differentiation and separation between self and object takes place. The great conflict of the transitional stage is the conflict between a progressive urge to surrender the infantile identification with the object in order to separate from it and a regressive urge to maintain the infantile identification. Satisfactory development in the transitional stage is characterized by the replacement of object relationships based upon identification with the object with relationships with a differentiated object.

The stage of mature dependence was described by Fairbairn (1954) as a state of healthy interdependence with others. The individual in a state of mature dependence is concerned with giving to others and with the exchange of gratification rather than the preoccupation with taking that characterizes the stage of infantile dependence. Mature dependence "is characterized by a capacity on the part of a differentiated individual for co-operative relationship with differentiated objects" (p. 145).

Fairbairn (1954) emphasized that the successful development from infantile to mature dependence was based upon the child’s receiving conclusive assurance that he or she is genuinely loved by the parents and that the parents genuinely accept the love given by the child. Such assurance allows the child to depend safely upon real objects and relinquish infantile dependency. All forms of psychopathology "represent relationships with internalized objects, to which the individual is compelled to turn in default of a satisfactory relationship with objects in the outer world" (p. 40).

The Processes and Goals of Psychotherapy

Fairbairn (1954) proposed that the release of bad objects from the unconscious was the primary goal of the psychotherapist. The analyst was required to become established as a sufficiently good object within a satisfactory transference in order to allow the patient to release bad objects safely.

Fairbairn (1954) made many recommendations for psychotherapy. (a) Interpretations should be made in terms of object relationships including relationships with internalized objects. (b) Libidinal strivings should be represented to the patient as motivated by strivings for object love and as therefore basically good. (c) Bad or aggressive feelings should be interpreted as responses to bad objects. (d) Guilt should be interpreted as a response to the internalization of bad objects. (e) Interpretations of the patient's aggression should be made cautiously with the attempt to identify the libidinal needs underlying the aggression.

Fairbairn (1954) stated that dreams are not wish fulfillments but are dramatizations of the patient's internal reality. The situations depicted in dreams represent the parts of the ego and the internalized objects and the relations between them. Dreams, thus, offered fertile ground for the interpretation of endopsychic structure. Fairbairn (1954) thought that resistance in psychotherapy was a result of both the terror of facing the release of internalized bad objects and the libidinal attachment to these internalized objects.

The Heuristic

Fairbairn's (1954) guide to investigation was not clearly explained in his writing but seems to have been based upon his experience with patients and upon his interpretation of their intrapsychic experience. He used clinical material taken from his patients to introduce and to illustrate his concepts. In his work with large numbers of soldiers returning from World War II, Fairbairn stated that the size of the patient population composed of individuals who had been removed from their normal environment and separated from their love-objects provided him with ample opportunities for testing his conclusions about normal human dependency needs. According to Hughes (1989), his observations of these patients convinced him that his hypotheses had been supported and confirmed. However, it also should be noted that the data he used to develop his theories of childhood development were clinical material taken from adult patients (Hughes, 1989).

Ontology or the Essential Concepts of the Theory

Essential Human Nature

Fairbairn (1954) believed that, from infancy to maturity, the primary motivation of the individual is to maintain secure and satisfactory interpersonal relations with others. He defined libido as the primary motivational energy and postulated that it was essentially object seeking. Fairbairn hypothesized that aggression was not an innate human instinct but rather was a response to frustration or deprivation. The child is predisposed from birth to internalize experiences, both good and bad, with objects in the environment. Successful development involves a progression from infantile dependence to mature dependence. The attainment of mature dependence was based upon the child’s having a preponderance of good experiences with real objects (usually parents) that communicate genuine love to the child and genuinely accept the love given by the child.

One interesting idea proposed by Fairbairn (1954) concerning essential human nature was about the tendency of individuals to form a "moral defense" (p. 66) as was previously described. When experiences with objects are bad (emotionally and libidinally unsatisfying), the child may simply internalize a general sense of feeling and being bad in a world of objects that are bad. The moral defense is an attempt made by the child to transform this sense of unconditional badness to a belief that there have been specific misbehaviors. The child's belief in his or her conditional, moral badness is accompanied by feelings of guilt and unworthiness. However, the belief in moral badness contains the hope of redemption and forgiveness, of becoming good. If the child is inadequately loved because love is not available, there is no hope. Fairbairn believed that the human individual needs the sense of hope of believing that the world is basically good.

The Nature and Development of Psychopathology

Fairbairn (1954) stated that psychopathology was a result of disturbed internalized object relations that were a reflection of actual disturbances in relations with others. All pathological symptoms operate as defenses against the emergence of either schizoid or depressive tendencies. The nature and etiology of both schizoid tendencies and depressive tendencies were previously discussed. When an infant experiences frustration, the child feels unloved as if the love offered by the child to the mother is not valued or accepted. Repeated frustration in early infancy leads to schizoid tendencies in which the child comes to believe that what he or she has to offer to others is poisonous and destructive. Depressive tendencies arise in the late oral stage. Unlike schizoid individuals, depressive individuals feel their love is good and are readily able to establish libidinal contacts with others. As long as these relationships are satisfactory, the depressive individual is undisturbed.

Fairbairn (1954) hypothesized that the conflict over separation from the object in the transitional stage leads to the development of paranoid, hysterical, phobic, or obsessional symptoms. Paranoid symptoms reflect an internal state of acceptance of the internalized object and a rejection of the externalized object. Hysterical symptoms are based upon rejection of the internalized object with an overvaluation of the accepted externalized object. The individual with phobic symptoms demonstrates a conflict between flight to an externalized accepted object and flight from an externalized rejected object. Obsessional symptoms reflect a conflict in which both the accepted and rejected object are internalized and the individual is torn between retention and expulsion of internal contents. All forms of psychopathology reflect the individual's involvement in internalized objects that are turned to when there is a failure to find satisfactory relationships in the real world (Fairbairn, 1954).

The Role of the Therapist

Fairbairn (1954) stated that the therapist must become established as a good object for the patient, within a satisfactory transference, to allow the patient to release bad objects from the unconscious. Dreams were interpreted as dramatizations of the patient's internal reality, representing the parts of the ego, the internalized objects, and the relationships among them.

Fairbairn (1954) commented that the therapist should be sympathetic to resistance, which represents both the terror of releasing internalized bad objects and the reluctance to give up the libidinal attachments to these bad objects. In general, Fairbairn's approach to psychotherapy was presented as an approach that was supportive of and sympathetic with guidelines for interpretation that focus on the patient's essential goodness and strivings for object love.

The Mechanism of Change and the Goals of Treatment

Fairbairn (1954) did not make completely clear how change is accomplished in psychotherapy. However, he did make it clear that progress in therapy required that the therapist become established as a good object within a satisfactory transference. He also seemed to think that interpretations, as described in the previous section, had a mutational effect. The primary goal of treatment, which one might presume is the most powerful mechanism of change, was the release of bad objects from the unconscious in order to strengthen the central ego, which is needed for the development of real, satisfying relations with others.

D. W. Winnicott

D. W. Winnicott has been described as an extremely innovative and influential contributor to the development of psychoanalytic theory and practice. Greenberg and Mitchell (1983) stated that his theories provided “an intricate, subtle, and often powerfully poetic account” (p. 188) of the development of the self in the context of the relation between child and mother. Winnicott (1964) wrote that:

I once risked the remark, "There is no such thing as a baby" — meaning that if you set out to describe a baby, you will find you are describing a baby and someone. A baby cannot exist alone, but is essentially part of a relationship. (p. 88)

According to Summers (1994), Winnicott, himself, often stated that his primary contribution to the psychoanalytic theory of development lay entirely in his understanding of preoedipal phases.

Basic Concepts

Winnicott (1965) proposed that the human infant starts life with an inherited potential but in a state of helplessness and absolute dependence. This state of helplessness and dependence is psychological as well as physical. Winnicott (1969) interpreted the infant as beginning life in a state of unintegration, unable to separate "a NOT-ME from what is ME" (p. 254).

Winnicott (1958, 1965, 1989) described many psychological processes experienced and performed by the mother, which facilitate the growth and maturation of the infant. Before the birth of the infant and for the first few months of life, Winnicott stated that a mother entered a state he called "primary maternal preoccupation" (1958, p. 302). Primary maternal preoccupation was described as a mental state of "heightened sensitivity" (1958, p. 302), in which interest is withdrawn from the outside world in order to be exclusively preoccupied with the infant. This condition was described by Winnicott (1958) as a "normal illness" (p. 303).

During the early stages of an infant's life, the mother organizes the experiences of the infant and provides a "holding environment" (Winnicott, 1965, p. 54), which contains the experiences of the infant. The mother also is responsible for bringing the world to the child. Winnicott (1958) hypothesized that the excited infant hallucinates the object most suitable to meet his or her need. Thus, the hungry infant hallucinates a breast. Ideally, at the moment of hallucination, the attentive mother presents the infant with the object that is hallucinated. The infant then believes that he or she created the needed object.

The infant comes to the breast when excited and ready to hallucinate something fit to be attacked. At that moment the actual nipple appears and he is able to feel it was the nipple that he hallucinated. So his ideas are enriched by actual details of sight, feel, and smell, and next time this material is used in the hallucination. In this way he starts to build up a capacity to conjure up what is actually available. The mother has to go on giving the infant this type of experience. (1958, pp. 152-153)

When the infant's hallucination matches the object presented, the infant experiences feelings of omnipotence (i.e., hallucinatory omnipotence). Winnicott (1958) suggested that these feelings of omnipotence are the bases for the healthy development of the self. The mother's empathic attunement to the baby's needs allows the timing of presentation to be accurate. The juxtaposition of the infant's hallucination and the mother's presentation as a repeated experience allows the development of the child's contact with external reality along with feelings of control and power in the satisfaction of needs from the provisions of external reality.

Winnicott (1989) also proposed that, in addition to providing a holding environment and an empathic response to the infant's needs, the mother functions as a mirror. As the mother reflects the infant's experiences, the child develops a capacity to experience and to integrate the self. Winnicott wrote that the first mirror is the mother's face and that one of the functions of the mother and the parents and of the family is to provide a mirror, figuratively speaking, in which the child can see himself or herself. In order to use the parents and family as a mirror, the child must recognize that there is permission to be whatever he or she is, accepted completely without evaluation or pressure to change.

Winnicott (1965) also distinguished two roles of the mother that he called the "object mother" and the "environment mother" (p. 75). The object mother is the mother as object or owner of the part object that may satisfy the infant's urgent needs. The environment mother is the person who wards off the unpredictable and who actively provides care in handling and in general management. It is the environment mother who is reliably present, fosters the capacity to be alone, and provides protection from impingement.

Protection from impingement was described as including both the maternal empathic attunement to provide for needs of the infant in an excited state and the undemanding maternal presence in the infant's quiescent state. If the mother does not respond when needed to provide the hallucinatory creation or if the mother disrupts the formless, unintegrated state of going on being, the child experiences an impingement on his or her personal existence (Winnicott, 1958).

The capacity to be alone develops in the infant as a result of the reliable presence of the mother. The mother's undemanding presence when the infant is not making demands or experiencing needs allows the infant to experience needlessness and unintegration, a state of "going-on-being" (Winnicott, 1965, p. 33) out of which needs and spontaneous gestures can emerge. The mother's undemanding presence makes it possible for the infant to experience comfortable solitude. A capacity for comfortable solitude was interpreted by Winnicott as a basis for the development of a stable, personal self. Winnicott considered the capacity to be alone one of the most important signs of maturity in emotional development. The capacity to be alone only comes into existence if the child has sufficient experiences of being alone, as an infant and small child, in the presence of mother. The basis of the capacity to be alone is, thus, a paradox; it is based on the experience of being alone although someone else is present.

When an infant experiences numerous impingements upon his or her personal existence, there is a fragmentation of experience. The infant becomes attuned to the claims and to the needs of others and loses touch with his or her own spontaneous needs and gestures. Winnicott (1965) described the resulting intrapsychic situation as a split between a "true self" (p. 144) and a "false self" (p. 144). The false self was described as a structure based upon reactions to and compliance with environmental demands. The false self becomes compulsively attuned to the needs and requests of others while the true self, the source of spontaneous needs and gestures, goes into hiding. The false self uses intellectual functions to anticipate and to react to environmental impingements which leads to overactive cognitive functioning and a separation of the mind from emotional functioning.

Winnicott (1958) proposed that, as the infant matured, the mother's heightened sensitivity and attunement should and does decrease in response to the baby's growing abilities to communicate. The child, thus, confronts the reality of the world he or she does not control. Winnicott (1958) called this process the mother's "graduated failure of adaptation" (p. 246) and claimed that it is essential to the infant's development of separation, differentiation, and normal ego functions.

Correspondingly, Winnicott (1958) hypothesized that the child develops a capacity for concern as he or she makes the transition from infantile omnipotence to realistic perception. Similar to Klein's (1975) depressive position, Winnicott proposed that the baby reaches a depressive crisis when the realization occurs that the object mother, who is used for the fulfillment of needs, is the same person as the environment mother, who provides the holding environment. The infant responds to this realization with feelings of deep concern.

Winnicott (1958) suggested that the mother had two crucial responsibilities during the period when the capacity for concern is developed. The first is to demonstrate her capacity to survive. The second is to provide the baby with an opportunity to make reparation that allows the child to relieve guilt and to express his or her concern.

The transitional object was a concept proposed by Winnicott (1958) to describe an object that the child uses as an hallucinatory, symbolic replacement of the omnipotently controlled mother of early infancy. The special characteristics of the transitional object are listed below.

1. The infant assumes rights over the object, and significant others in the environment agree to this assumption.

2 The object is affectionately cuddled as well as excitedly loved and mutilated.

3. The object must never change, unless changed by the infant.

4. The object must survive instinctual loving and also hating.

5. The object must seem to the infant to give warmth, to move, to have texture, or to do something that seems to show that it has vitality or reality of its own.

6. The object comes from without from the point of view of others in the environment but not so from the point of view of the baby. However, it does not come from within. It is not an hallucination.

7. In the course of years, the object loses its symbolic importance. It is not forgotten, and it is not mourned. It loses meaning because the transitional phenomena have become diffused, have become spread out over the whole intermediate territory between what is perceived as inner psychic reality and what is perceived as reality shared with others.

Transitional objects help the baby make the shift from an immersion in the subjective world of internal experience to an acceptance of external reality with separate and independent others. Winnicott (1958) further proposed that transitional experiencing remains part of healthy intrapsychic functioning into adulthood. In childhood, transitional experience exists in play. In the adult, transitional experiences exist in the capacity to play with fantasies, ideas, and possibilities. Transitional experience was interpreted by Winnicott to be the source of creativity.

The Processes and Goals of Psychotherapy

Winnicott (1958) proposed that the analytic setting provides elements that were missing in the patient's childhood environment and, thus, that the analytic setting fills early developmental needs. The analyst and the analytic setting provide a "holding environment" (p. 168) for the patient. If the patient perceives the analyst as reliable, attentive, responsive, and indestructible, then the patient uses regression to return to points in childhood when environmental failures disrupted development.

Winnicott (1965) proposed that the "tendency to regression in a patient is now seen as a part of the capacity of the individual to bring about self-cure" (p. 128). Winnicott (1972) also stated that the advantage of a regression is that it is an opportunity for correction of experiences in the past history of the patient. Winnicott further stated that, whenever the analyst understands the patient in a deep way and shows this understanding by providing a correct and well-timed interpretation, the analyst is "in fact holding the patient" (1965, p.192). Both analyst and patient are taking part in a relationship in which the patient is in some degree regressed and dependent.

Winnicott (1958, 1965) described the responsibilities of the therapist as similar to the responsibilities of the mother. He wrote that the analyst should be reliably present and preoccupied with the patient. He also proposed that, just as the mother gradually reduced her adaptation to the needs of the infant, there were limits to the analyst's capacity for adaptation to the patient's needs that should emerge over time. The failures of the analyst were in themselves potentially therapeutic. Just as the maturing infant can accept the good-enough mother, the maturing patient can allow for the failures of the good-enough analyst.

Winnicott (1958, 1965) defined mental health as the integrity and spontaneity of the true self. Psychopathology resulted from the blockage of the spontaneous growth and development of the true self. The goal and purpose of psychotherapy was defined as the provision of a facilitating environment in which the true self could emerge and resume its growth.

The Heuristic

Winnicott (1958, 1964, 1965, 1971, 1989) was a practicing pediatrician as well as a psychoanalyst. His theoretical formulations were based upon direct observation of mothers and children and their interaction, his attempts to imagine the intrapsychic experience of mothers and infants, and his attempts to integrate these observations and mental constructions with psychoanalytic ideas and development. Most of his theory was based upon his attempts to integrate psychoanalytic ideas about intrapsychic processes with his direct observation of mother and child in their interpersonal interaction (Greenberg & Mitchell, 1983; Hughes, 1989; Pine, 1990).

Winnicott (1958, 1964, 1965, 1971, 1989) stated that the human infant is endowed with the innate capacity to develop a stable sense of self, concern for loved ones, and creativity given an appropriately supportive and appropriate response from mother and family. Winnicott also believed that the child was naturally endowed with aggression, which he thought originated in the child's innate needs to move, to grow, and to explore. Winnicott explicitly repudiated a belief in an innate death instinct and thought that aggression was only manifested as anger or destructiveness under conditions of frustration. Perhaps the most important point concerning innate human nature made by Winnicott was about the individual’s need for an optimal, good-enough human environment to stimulate the optimal development of the child's inherited potential.

The Nature And Development Of Psychopathology

Winnicott’s (1958, 1964, 1965) primary contributions to the ideas about the nature and development of psychopathology were his ideas about the development of a false self. As described previously, the infant depends upon empathic, responsive, maternal attunement to gratify needs and an undemanding maternal presence, when the infant is in a quiescent state.

Numerous impingements upon the personal existence of the infant have the effect of causing the infant to become attuned to the claims and needs of others while losing touch with his or her own spontaneous needs and gestures. As a result, the infant develops a split between a true self and a false self. The true self, the source of spontaneous needs and gestures, goes into hiding while the false self uses intellectual functions compulsively to anticipate and to respond to the needs and requests of others. The use of intellectual functions in the service of the false self leads to overactive cognitive functioning and a separation of mental from emotional functioning.

The Role Of The Therapist

Winnicott's (1958, 1965, 1971, 1989) ideas about the curative factors in psychoanalysis were discussed in many different essays about different aspects of the therapeutic relationship. Again, he never organized his ideas into a coherent, logically explained model of the processes and goals of treatment. However, he suggested several important ideas about the nature of the therapeutic relationship, the mechanisms of change, and the goals of treatment.

The Mechanism Of Change And The Goals Of Treatment

Winnicott (1965) wrote that the manner in which the analytic setting provides missing parental functions and fills early developmental needs for reliable, indestructible, nonimpinging responsiveness encouraged regression. Regression was used to search for the missing interpersonal experience needed for healthy development and growth. Regression was viewed by Winnicott as part of an individual's capacity to cure the self of developmental damage. If the appropriate facilitating environment is provided, the patient can regress to the point at which the true self was hidden, can recover it, and can continue its growth.

The mechanism of change was, thus, defined by Winnicott as the provision of the optimally responsive and facilitative therapeutic relationship and setting. Interpretation plays a role in communicating the therapist's understanding to the patient, but the more important curative factor is interpersonal. The other vital mechanism for change is the patient's capacity for regression in order to correct developmental deficits and to resume normal, healthy growth. The goal of treatment is, thus, regression in the service of finding the true self and resuming growth.

Edith Jacobson

Jacobson's (1964, 1971) theoretical contributions to psychoanalytic thought were remarkable attempts to integrate and to synthesize ideas about the development of identity, object relations, the ego-superego systems, and the mutual influences upon one another of each of these intrapsychic structures. She was thorough and meticulous in her examination of the literature concerning the development of identity, object relations, and the ego-superego systems. Her own original contributions arose from her attempts to explain and to integrate her many years of clinical and personal experience with psychoanalytic theory about intrapsychic structure and its development.

Jacobson (1964, 1971) wrote in a style that was meticulous and complex. Greenberg and Mitchell (1983) commented that her prose was "dense, almost impenetrable" (p. 306). Kernberg (1987) stated that the study of her theories is difficult because of the complexity of her ideas. Therefore, the attempt to summarize her ideas and to classify them into the categories of basic concepts, personality development, and the processes and goals of psychotherapy cannot reflect the richness and intricacy of her actual presentation.

After a careful examination of Jacobson's (1964, 1971) writings, it seemed appropriate to abandon the attempt to separate her ideas about personality development from basic concepts. Her ideas about the development of object relations, self-representations, and other intrapsychic structures were the basic concepts of her theory.

Personality Development

The undifferentiated infant. The earliest infantile period precedes the infant's awareness of self as differentiated from the environment. The infant is born with undifferentiated instinctual energy. The first intrapsychic developments are based upon experiences of tension and relief, of frustration and gratification. The infant’s psychic apparatus matures in the first weeks of life and becomes capable of retaining memory traces of pleasurable and unpleasurable experiences. These memory traces become the first images formed of self and object, which remain undifferentiated and fused. The young infant cannot discriminate between pleasurable sensations and the objects from which they are derived (Jacobson, 1964).

Repeated unpleasurable experiences of frustration and separation from the loved objects provide the basis of the first awareness of separateness and the development of fantasies of total incorporation and merger with the gratifying object. Jacobson proposed that these wishful fantasies of being one with the mother are the foundation of object relations and of all types of identification. The hungry infant's longing for food, libidinal gratifications, and physical merging with the mother is also the origin of the first, primitive type of identification, an identification achieved by a refusion of self and object images. The refusion of self and object images is accompanied by a temporary weakening of the perceptive functions and, thus, by a return from the level of beginning ego formation to an earlier, less differentiated state.

Jacobson (1964) proposed that this type of fused self-object identification plays a dominant role in the mental life of the baby and preoedipal child and, in addition, has a useful function in the mature intrapsychic organization of the adult. In the adult, the subtle empathic understanding of others is based upon the temporary use of fused self-object identification. The temporary fused identifications of the adult do not weaken the boundaries between images of self and others but instead "coexist and collaborate with mature personal relations and firmly established ego and superego identification" (p. 41).

Differentiation of self and object. During the second year of life, as the child becomes increasingly mobile and as perceptual processes mature, the infant starts the gradual process of differentiation of the self representation from the object representation. During this period, “Multiple, rapidly changing and not yet clearly distinguished part images of love objects and body part images are formed and linked up with memory traces of past pleasure-unpleasure experiences and become vested with libidinal and aggressive forces” (Jacobson, 1964, p. 53).

Jacobson (1964) further proposed that, during this period of early differentiation of self and object, the self and object images undergo temporary fusion, separate, and join again. There is a tendency during this time to invest the aggressive drive into one fused bad self-object representation that is completely separate from the libidinally invested fused representation. The behavior of the child in this period may at times be erratic, with displays of submissive clinging alternating with aggressive struggles to control the love object.

Object constancy and identity. Throughout the second and third year of life, the child continues the process of differentiation. Objects in the environment are distinguished from one another, and as the child develops a toleration for ambiguity, there is an integration of good and bad object representations into total object representations. The development of whole object representations fosters and supports the integration of good and bad self representations into total self representations. Object constancy is established, and the child begins to develop a sense of identity (Jacobson, 1964).

As the ego matures during this phase of individuation, the child develops ambitions for independent and realistic achievements. The desire for fusion with the love objects diminishes and is replaced by wishes for realistic likeness with them through the process of selective identification. The development of total self and object representations and the development of selective identifications evolve concomitantly and "exercise a mutually beneficial influence on each other" (Jacobson, 1964, p. 65). Jacobson (1964) proposed that, as the child develops the capacity to experience the totality of other persons and of the self, the child can tolerate perceptions of the differences between self and others and that similarities are discovered, accepted, desired, and acquired.

Jacobson (1964) wrote that the relation between the parents and the child was the most influential factor in the development of object constancy and a stable identity. The love the child receives from the parents promotes the investment of libidinal energy in both the object and the self. Parental love is the best foundation for the development of object and self constancy, of healthy relations, and of lasting identifications and, thus, for normal ego and superego formation.

Jacobson (1964) stated that the frustrations, prohibitions, and demands of the parents also make significant contributions to the process of individuation. These frustrations, prohibitions, and demands gradually teach the child to relinquish the infantile magical expectation of total support, protection, and wish fulfillment from the object. Experiences of deprivations, disappointments, and hurt arouse feelings of ambivalence. In the spirit of taking in what is liked and spitting out what is disliked, this ambivalence encourages the child to turn aggression toward the frustrating objects and to turn the libido towards the self. Frustrations, demands, and restrictions, within normal bounds, reinforce the process of discovery and distinction of objects and self. These frustrations, demands, and restrictions force the child to identify and to use internal resources and stimulate progressive forms of identification with the parents, which open the road to realistic independent achievements. "Enhancing the narcissistic endowment of his ego, they promote the eventual establishment of ego and superego autonomy" (p. 56).

The development of the autonomous ego and superego. The next phase of development discussed by Jacobson (1964) is the Oedipal phase of development. During the Oedipal phase, sexual identity is formed, the sexual identity of others is recognized, and both are incorporated into the self and object representations. Sexual prohibitions and castration anxiety stimulate drive neutralization and the development of autonomous ego functions.

Jacobson (1964) proposed that the growing complexity of identification and autonomous ego functions in the Oedipal period finally culminates in the development of the superego. Superego formation is based upon the identifications with parental demands and expectations and involves the internalization of general ethical and moral standards. Ideal self and object representations are merged into the ego ideal, which is incorporated into the superego. Castration anxiety is transformed into fear of guilt, which will be experienced if superego directives are violated.

The development of the autonomous superego structure is of enormous benefit to the individual's entire intrapsychic experience. Superego autonomy provides freedom from the demands of early precursors of the superego which are based on the internalization of punitive, frustrating self and object representations and on the internalization and identification with aggrandized idealizations of the object. The integration of good and bad object relations occurring in the ego permits the dissolution of the earliest sadistic and later idealized superego forerunners. The emerging autonomous superego is based upon the internalization of realistic demands and prohibitions of the parents (Jacobson, 1964).

The superego protects self-esteem in that it is more concerned with the degree of internal harmony between the moral codes and ego manifestations than with external success or failure. The superego can, thus, provide a stable and enduring libidinal investment in the representations and activities of the self, which is not easily disrupted by experiences of rejection, frustration, and failure. "In summary, the superego introduces a safety device of the highest order, which protects the self from dangerous internal instinctual stimuli, from dangerous external stimuli, and hence from narcissistic harm" (Jacobson, 1964, pp. 132-133).

In contrast, Jacobson (1964) described the psychopathology that results from an archaic superego structure made up of the earlier sadistic and idealized superego forerunners. The pathological archaic superego produces self-punitive behavior and creates recurring depressive mood deviations, but the ego is unable to ward off and to sublimate forbidden strivings. The archaic superego is punitive but seems to have little capacity to guide. Individuals with this type of archaic superego may alternate impulsive acting out with self-punitive depressions or may suffer from chronic, recurring depression.

The development and maturation of an autonomous superego stimulates the development of sublimations, of general autonomous ego activity, and stable object relations. Jacobson (1964) proposed that sublimations occur when part of the libido previously invested in love objects undergoes partial neutralization as a result of resolutions of Oedipal conflicts. This libidinal energy is then available for other external, animate, and inanimate objects, which reflect ego interests. This shift from infantile dependence and investment in the love object to libidinal investment in ego activity promotes libidinal investment in the self.

The development and maturation of the superego, the ego, and object relations was demonstrated by the complex, manifold, subtle emotions that develop concurrently. Jacobson (1964) contrasted the rich, complicated emotions of the postoedipal child with the preponderance of autistic-narcissistic and clinging-dependent attitude toward the object displayed by the preoedipal child. She also compared the affects of the autistic-schizoid personality type with the affects of people with rich, complex object relations. The latter display manifold and subtle shades of feeling with warm and vivid emotional qualities, which indicate the predominance of object libido. In contrast, the range of feelings in the former is limited mainly to certain types of affects, such as cold hostility, anxiety, hurt, humiliation, shame or pride, security or insecurity, high or low self esteem, grandeur or inferiority, and guilt. These individuals experience extreme and/or polar affects but not the broad range of subtle and complicated emotions between these extremes. The affective coldness and rigidity of autistic-schizoid individuals appear to point to the prominent part of inhibited but unamalgamated aggression and the prevalence of self-directed emotional discharge processes.

The resolution of the Oedipal period and the development of the superego were described by Jacobson (1964) as a developmental stage when all developmental processes are given an enormous impetus. Large amounts of psychic energy are liberated and utilized for rapid progress in the development of social, intellectual, cultural, and physical activities. The child enters the latency period.

Latency. Jacobson (1964) described the latency period as one in which there is continuing development of the ego and superego with increasingly focused, cognitively differentiated affects and demands. She stated, in most cases, that ego and superego formation is supported and stimulated by starting school, an environment potentially rich in new opportunities for sublimation, new and different types of relations, and the possibility of new identifications with teachers and peers.

Latency-age children often develop strong identifications with peers of the same age and sex. The experience of belonging to a group was described by Jacobson (1964) as an opportunity to reinforce the sense of personal identity and also to make possible the development and acceptance of common group standards—"ethical, social, physical, and intellectual" (p. 138).

Jacobson (1964) also noted that upon entering school some children who have been raised in confused or unstable home environments become painfully aware of contradictions between the school world and the world at home with regard to sexual, social, ethical, and intellectual attitudes, goals, and standards. She remarked that some adult patients recalled states of loneliness coinciding with the beginning of latency either at home or at school or both. The history of such patients reveals that the internalization of confusing parental attitudes at an early infantile stage may result in lasting contradictions in a person's unconscious and conscious set of values. Such contradictions may disturb the establishment of stable personal relations with sufficient object constancy and, thus, of consistent superego and ego identifications. The child may also be predisposed to identity problems that may gain a dangerous momentum during adolescence and may extend into the life of the adult.

Adolescence. Jacobson (1964) described adolescence as a turbulent time with violent mood swings and recurring states of depression, which involve both severe guilt conflicts and nagging feelings of shame and self consciousness. The adolescent also manifests fluctuations in his or her feelings towards others and towards the self, accompanied by disturbances in identity. The adolescent at one point may feel a joyous relatedness to the world, to art, or to nature and at other times may suffer from despairing feelings of loneliness, isolation, and meaninglessness.

Jacobson (1964) explained these trials of adolescence at some length. She described adolescence as "life between a saddening farewell to childhood . . . and a gradual, anxious-hopeful passing over many barriers through the gates which permit entrance to the as yet unknown country of adulthood" (p. 161). The adolescent must separate emotionally from the profoundly important familial attachments of childhood. Preparing to leave home, he or she must develop new and different personal and social relations and find new interests, sublimations, values, and goals that will give direction to his or her life as an adult (Jacobson, 1964).

Jacobson (1964) proposed that these profound changes of adolescence require "a drastic overhauling of the entire psychic organization" (p. 161). Her discussion of these processes focused on the remodeling of the adolescent's ego and superego and on interrelationship of this remodeling with the development of feelings of identity, object relations, and identifications.

Jacobson (1964) noted that the extensive psychobiological changes in the adolescent and their accompanying arousal of instinctual impulses have a significant impact on identity feelings and identity formation. The arousal of sexual feelings may induce intense guilty and anxious conflicts. The changes in the body as sexual maturation progresses may produce confusion in the adolescent's sense of identity. Jacobson (1964) remarked that the adolescent's instinctual development shows how, "in climbing up the tortuous ladder to adulthood” (p. 168), the adolescent seems to experience at every new step anxiety, confusion, disorganization, and a return to infantile positions, followed by propulsion and reorganization at more advanced and more adult levels. The adolescent's tendency toward recurring temporary regressions in all areas and systems obviously is a result of the powerful assault of instinctual forces on the ego. The growing and changing ego must cope with the task of finding new instinctual controls and new avenues of discharge that will help the adolescent gain the optimal and socially permissible degree of emotional freedom needed to build adult sexual and personal relationships.

Jacobson (1964) added that the breaking of Oedipal ties, the establishment of new object relations, and the reorganization of intrapsychic structure in adolescence cannot succeed if these processes completely deplete and eradicate the libidinal investments and identifications of the past. The new process merely reduces the investments and identifications of the past, displacing them onto new objects, changing their qualities, and subordinating them to new attachments and partially to new identifications. New, remodeled, or reorganized intrapsychic structures cannot be developed and integrated unless they develop organically from those of the past.

One of the most important tasks of the adolescent is the gradual development of new identifications with the parents as sexually active individuals. The accomplishment of this identification paves the way to the adolescent’s recognizing his or her right to sexual and other adult activities. These identifications, which were unacceptable in the past, can only be achieved if the superego and ego become reconstructed and consolidated and reach new levels of strength, autonomy, and maturity.

This reconstruction of ego and superego is not easily accomplished. The adolescent, in struggling partially to lift repressions, suffers severe sexual, narcissistic, and ambivalence conflicts. The ego experiences pressures from id and superego and may alternately yield to the superego or actively rebel against it and join forces with the id. During these internal struggles, the adolescent's behavior may vacillate between sexual and aggressive acting out with narcissistic inflation and periods of repentance, of strictly abstinent moral behavior with feelings of guilt, shame, and inferiority (Jacobson, 1964).

Jacobson (1964) hypothesized that psychophysiological sexual development, accompanied by rapid body changes and mental growth, creates an excess of psychic energy. This energy liberates sexual and aggressive impulses and narcissistic strivings. As the adolescent disengages from infantile love objects, he or she passes through a prolonged stage of overinvolvement with narcissistic aims and preoccupations at the temporary cost of truly object-directed goals.

Jacobson (1964) further proposed that the adolescent struggles with intense shame and inferiority conflicts accompanied by painful guilt conflicts. The adolescent's vacillations in self esteem originate both in moral conflicts and in more primitive narcissistic conflicts. The adolescent struggles with the discrepancies,

between images of the grown-up, powerful, glamorous, brilliant, or sophisticated person he wants to be and sometimes believes himself to be, and the undeniable aspect of the physically and mentally immature, unstable, half-baked creature between two worlds, which he actually is. (p. 181)

The combination of moral conflicts with feelings of shame and inferiority is the basis of the adolescent's fluctuations in self esteem and sense of identity. As the adolescent's ego and superego mature and identification progresses, these conflicts and painful feelings subside (Jacobson, 1964).

In late adolescence, there occurs a preoccupation with forming a philosophy of life based upon opinions, ideas, and ideals concerning everything from superficial pleasures, manners, and appearances to serious ethical and intellectual problems. Jacobson (1964) called this process the development of a Weltanschauung, an interpretation of the world that includes values, ideals, ethics, opinions about nature and culture, and opinions about “sexual, social, racial, national, religious, political, and general intellectual problems" (p. 182).

The development of such a philosophy of life in late adolescence is an indication of the influence of the ego over both superego and id. The evolution of the Weltanschauung is based on the identifications with realistic parental figures who grant, within limits, freedom of emotions, thought, and action. However, as the adolescent matures, he or she will seek the company of others and will join social, political, cultural, religious, or professional groups that stimulate thinking and foster the development of new ideas and ideals (Jacobson, 1964).

As the individual in late adolescence joins groups of peers, new and potentially valuable personal and group identifications develop on an advanced basis. These identifications ultimately support the reintegration of the superego and the development of ego defenses, which rein in the excesses acquired in adolescence. Furthermore, the liberation and maturation of thinking bring about a reduction in the role of identification. The final establishment of ego and superego autonomy brings about an increasing freedom from external influences and from internal pressures from past experience.

Adulthood. Jacobson (1964) stated that, after adolescence, ego development proceeds on the basis of individual, independent, critical, and self-critical judgment. Ego and superego autonomy and identity formation depend upon the successful modification, stabilization, and integration of the object relations and identifications formed with persons in the individual's past. These well-integrated and stabilized intrapsychic structures continue to influence the individual throughout life. Identifications with parents may particularly influence experience and behavior as a spouse and a parent.

The adult also continues to be influenced by group relations and identifications but to a lesser extent than in adolescence. In adulthood, most individuals' interests are shaped and invested in vocational, personal, and family life. The successful modification and stabilization of the superego in adolescence provides regulation of ego functions, ego goals, and object relations and protects the maintenance of normal identity feelings in the adult. The regulation and protection of the autonomous, well-functioning superego provides the basis for flexible well-directed emotional, sexual, social, and vocational development (Jacobson, 1964).

The Processes and Goals of Psychotherapy

Jacobson (1964, 1971) wrote very little that was specifically concerned with the description of the processes and goals of psychotherapy. She illustrated many of her expositions on development and psychopathology with extended clinical case histories. Greenberg and Mitchell (1983) commented that these case histories reflected her remarkable personal warmth and sensitivity. Jacobson (1971) also specifically wrote about transference problems in the psychoanalytic treatment of severely depressed patients.

Jacobson (1971) stated that most patients with severe depression suffer from borderline conditions characterized by "ego distortions and superego defects, disturbances in their object relations, and a pathology of affects beyond what we find in ordinary neurotics" (p. 285). She thought that such patients required years of slow, patient, consistent treatment. She commented that such patients usually established an immediate, intense rapport with the therapist that is a prerequisite for success in treatment. In the course of the treatment, "the analyst inevitably becomes the central love object and the center of the depressive conflict" (p. 286).

Jacobson (1971) described the characteristic phases of treatment as follows: (a) the initial, deceptive transference success; (b) a period of hidden, negative transference with corresponding negative therapeutic reactions (i.e., more severe states of depression); (c) a stage of dangerous, introjective defenses and narcissistic retreat; and (d) the phase of gradual, constructive conflict resolution. In other words, the patient starts with warm hopeful feelings toward the therapist, who will magically cure all ills. Over time, the patient becomes disillusioned and disappointed in the power of the therapist and reacts with depression and anxiety. The patients typically will go on to express hostility either overtly or covertly to the disappointing therapist, and various symptoms, which reflect a narcissistic retreat from the therapist, may appear. Jacobson emphasized the need for the therapist to remain calm, interested, respectful, and understanding while providing interpretations of the transference phenomena. If both the therapist and patient remain engaged in the therapeutic process, the potential exists for the resolution of deep unconscious conflicts.

Jacobson (1971) also stressed the need for "a continuous, subtle, empathic tie between the analyst and his depressive patients" (p. 299). She recommended that the therapist should neither let an empty silence develop nor talk too much but, instead, should provide a spontaneous and flexible adjustment to the mood of the patient. She cautioned that the therapist should neither give too much nor too little and that warm understanding and respect should not be confused with over-kindness and reassurance.

Jacobson (1971) recommended that, during the periods of narcissistic withdrawal, the therapist should demonstrate an active interest in the daily activities of the patients, especially their sublimations. There are also times when supportive interventions and deviations from traditional psychoanalytic techniques might be necessary. In working with severely depressed patients, she warned that there would be periods in which the patients might experience the analyst's attitude and interpretations as seductive promises, severe rejection, lack of understanding, or sadistic punishments. Any of these experiences may increase the patient's insatiable demands, frustration, ambivalence, and depression. "At critical moments, the analyst must be prepared to respond either with a spontaneous gesture of kindness or even with a brief expression of anger, which may carry the patient over an especially dangerous depressive phase" (p. 300).

Jacobson (1971) warned that any deliberate show of emotional response requires extremely careful self-scrutiny and self-control in the analyst. She emphasized the need for interpretation of the positive transference from the beginning of treatment with warnings of the possibility of subsequent disillusionment. If critical transference situations then arise that require the use of special emotional countermeasures on the part of the therapist, the countermeasures can be explained on the basis of previous interpretations and warnings.

Jacobson (1971) also advised against premature exploration and interpretation of deep, preoedipal material. She warned that such material may be introduced by the patient early in treatment but that such early introduction is usually reflective of primitive, regressed primary process associations that are not amenable to integration with more mature levels of functioning. Jacobson thought that treatment should involve a slow and precise analysis of ego-superego and transference conflicts that will gradually prepare the patient to tolerate and to work through deep, preoedipal fantasies. Jacobson (1971) continued that some patients are never able to tolerate the emergence of preoedipal fantasies and impulses. However, "the most thorough and lasting therapeutic results were achieved in cases where this deep fantasy material could be fully revived, understood, and worked through" (p. 301).

The Heuristic

Jacobson's (1964, 1971) guide to investigation was based upon her attempts to integrate and to synthesize ideas about the development of identity, object relations, the ego-superego systems, and the mutual influences these intrapsychic structures have on each other. She integrated her personal and clinical experience with an in-depth analysis of psychoanalytic theory.

Ontology or the Essential Concepts of the Theory

Essential Human Nature

Jacobson (1964, 1971) theorized that, in normal human development, complex intrapsychic structures evolve when given appropriate environmental support. When the child’s needs are consistently and appropriately gratified, the infant gradually develops object constancy and a stable sense of identity. Stable love provided by the parents promotes the child’s investment of libidinal energy in both the object and the self.

As parental love promoted the development of libidinally invested self and object representations, it also promotes lasting identifications, all of which contribute to normal ego and superego development. The ego of the child also develops with accompanying ambitions for realistic achievements. (Jacobson, 1964).

Increasing complexity of selective identifications and of autonomous ego functions culminates in the development of the superego. Superego formation is based upon identification with parental expectations and demands with the internalization of general ethical and moral standards. The emerging, autonomous superego combines signal fears, self-critical, and self-rewarding functions with guiding, inspiring inner principles. “Signal fears” is a term used in psychoanalytic literature to describe fear or anxiety that arises as a signal indicating a dangerous situation. The autonomous superego also protects self-esteem in that it is more concerned with internal harmony between superego principles and ego striving than with actual external success and failure. The superego, thus, provides a stable libidinal investment in the representation and activities of the self that is not easily disrupted by failure, rejection, or frustration, thus protecting the self from narcissistic injury (Jacobson, 1964).

Jacobson (1964) described adulthood as a time when development is based upon ego and superego autonomy and proceeds on the basis of independent judgment. The well-integrated intrapsychic structures influence the individual throughout life, particularly as a spouse and parent. In adulthood, most individual’s interests and concerns are influenced by vocational, personal, and family life. The stabilization of the superego in adolescence provides the regulation of ego functions, ego goals, object relations, and sense of personal identity. The protection of the autonomous superego provides the basis for flexible, satisfying emotional, sexual, social, and vocational growth (Jacobson, 1964).

It can be understood that Jacobson (1964) hypothesized that the human individual is endowed with the capacity to develop empathy, autonomy, productivity, good social relations, values, and a variety of positive human attributes given an appropriately responsive human environment. Essential human nature was defined by Jacobson as inherently positive and productive if the individual had been loved and appreciated in the formative stages of development.

The Nature And Development Of Psychopathology

Jacobson (1964) proposed that all severe psychopathology, which includes affective disorders, borderline states, and overt psychoses, is derived from disturbances in self and object representations. Both normal and pathological development are based on the evolution of intrapsychic self and object representations. The crucial factors distinguishing normal development from the development of psychopathology and in determining the severity of psychopathology is the extent to which representations of self and object are stable, realistic, separate, and articulated at times of developmental disappointments.

If disappointments are early and harsh, occurring before the consolidation, differentiation, and instinctual investment of the self and object representations, these disappointments and frustrations arouse aggression in the child. The child who has not yet differentiated self from object cannot turn aggression against the object without a corresponding turning of aggression against the self, leading to a devaluation of both self and object representations. The result is a merger of overly idealized, unattainable self and object images accompanied by merged, hated self and object representations that are devalued in an exaggerated, unrealistic manner. Under these conditions, affects are severely limited, large amounts of psychic energy are required for defensive operations, and ego development and superego development are arrested or disturbed (Jacobson, 1964, 1971).

Jacobson (1964) described the functioning of the autistic-schizoid personality as an example of a primitive level of emotional development. The autistic-schizoid personality was described as having a limited range of feelings: hostility, anxiety, hurt, humiliation, and alternating states of shame and pride, security and insecurity, and grandiosity and inferiority. The limited range of affects of the autistic-schizoid personality reflects the predominance internally of unmodulated aggression and a prevalence of self-directed emotional discharge processes.

Serious disappointments in a more advances preoedipal stage may result in the unattainable, overly idealized, merged self and object representations becoming established as the ego ideal. The individual may then develop an archaic superego composed of this unattainable ego ideal combined with the aggressively invested, devalued, hated, merged self and object images that represent fantastic, irrational prohibitions and demands (Jacobson, 1964). Jacobson (1971) thought that most patients with severe depression suffered from borderline conditions with a pathology of affects, ego distortions, superego defects, and disturbances in their object relations. Early, preoedipal interpersonal experiences are the primary source of the development of psychopathology.

The Role of the Therapist, the Mechanism of Change, and the Goals of Treatment

Jacobson (1964, 1971) wrote very little that was specifically concerned with a theoretical model of psychotherapy. Most of her discussion of treatment was in the context of her treatment of severely disturbed patients with whom she advised modifications of the traditional psychoanalytic technique. Jacobson's (1971) discussion of these modifications focused on the management of transference problems in severely disturbed patients, particularly severely depressed patients.

The primary mechanism of change in therapy was through insight and interpretation that led to the understanding and repair of intrapsychic structure. The warmth and sensitivity with which the transference was managed primarily served the purpose of maintaining the therapeutic alliance so as to allow the development of insight, understanding, and conflict resolution through interpretation and insight. The goal of treatment was the eventual uncovering, understanding, and resolution of deep unconscious conflicts (Jacobson, 1971).

Margaret Mahler

Margaret Mahler, like Winnicott, was a pediatrician before becoming a psychoanalyst. Her ideas, which mainly concern the developmental stages in the first few years of life, were based upon her observations of the behavior of babies and children (Greenberg & Mitchell, 1983; Mahler, 1968, 1972; Mahler et al., 1975). Mahler's ideas had their genesis in the study of infantile psychosis and later were elaborated and refined in a longitudinal study of normal infants and mothers. These ideas were concerned with the inferred intrapsychic process of the infant in the development of separation-individuation. The separation-individuation process was described by Mahler et al. (1975) as:

the establishment of a sense of separateness from, and relation to, a world of reality, particularly with regard to experiences of one's own body and to the principal representative of the world as the infant experiences it, the primary love object. (p. 3)

The research described by Mahler et al. (1975) included a three-year pilot study followed by a formal research project lasting four years. The research involved the observation of normal mother-infant pairs in which the babies varied in age from 4 months to 36 months. The pilot study included 17 children of 16 mothers. In the formal research study, 21 children of 13 mothers were observed in a naturalistic setting (a community center for mothers and infants with play rooms for different age groups) by trained observers who kept records of mother-child interactions. In the formal research study, more data were collected over a longer period of time than in the pilot study. The data included numerous reports from participant-observers, interviews with mothers and fathers, reports of home visits, developmental testing of the children, and personality testing of the mothers.

There were two basic hypotheses in the study. First, in human development, there is a normal intrapsychic separation-individuation process, preceded by a normal symbiotic phase. Second, in certain predisposed but rare cases of development, the child has a spurt of physical maturation accompanied by a lack of emotional readiness for functioning separately from the mother, which gives rise to panic. "It is this panic that causes ego fragmentation and thus results in the clinical picture of symbiotic infantile psychosis" (Mahler et al., 1975, p. 13).

The setting for the research described by Mahler et al. (1975) was specifically designed to allow for the observation of the infant's active experimentation with separation and return to the mother and of the infant's reaction to passive separations from the mother. From these observations, the authors developed descriptions of the normal phases and subphases of separation-individuation in the first three years of life and identified "innumerable degrees and forms of partial failure of the separation-individuation process" (p. 13).

Mahler's (Mahler, 1968, 1972; Mahler et al., 1975) research and theories are primarily concerned with processes and vicissitudes in the psychological development of the infant and small child. She made some observations about the possibility of correlations between developmental problems and later adult psychopathology, but the discovery of the origins of adult disorders was not the focus of her research. She wrote little about psychotherapy, and therefore, there will not be a section concerning her ideas about the processes and goals of psychotherapy.

Basic concepts

Separation and individuation were considered to be complementary developments in the child. Separation was described as the child's emergence from symbiotic fusion with the mother. Individuation was described as the achievement of cognitive, perceptual, and affective functions, which the child comes to identify as his or her own individual characteristics. Separation and individuation processes may develop congruently or divergently with a developmental lag or precocity in one or the other (Mahler et al., 1975).

Adaptation was identified as the process by which the child's personality develops in response to the style and personality of the mother. "The infant takes shape in harmony and counterpoint to the mother's ways and style—whether she provides a healthy or pathological object for such adaptation" (Mahler et al., 1975, p. 5).

The object relationship was thought to be of primary importance in the development of separation and individuation. The object relationship develops out of infantile symbiosis and through the subphases of separation-individuation. The intellectual and emotional experience of separate individuality was described as a precondition of a healthy object relationship. An inability to enter or an inability to leave the mother-infant symbiosis was a basis for infantile psychosis (Mahler et al., 1975).

Symbiosis and separation were defined as intrapsychic conditions. Symbiosis is "a primitive cognitive-affective life wherein the differentiation between self and mother has not taken place" (Mahler et al., 1975, p. 8). Separation is the intrapsychic achievement of a sense of separateness from the mother, which gradually leads to clear intrapsychic representations of the self as distinguished from representations of the object world.

The hypotheses developed and described by Mahler et al. (1975) were inferences about intrapsychic processes based upon observation of infant behavior, particularly motor behavior. The researchers made the assumption that, in the first years of life, motor behavior is correlated with intrapsychic events because "the motor and kinesthetic pathways are the principal expressive, defensive, and discharge pathways available to the infant" (p. 15).

Essentially, the emphasis in Mahler's ideas about normal separation-individuation was that the process is necessary to the development of a normal, healthy intrapsychic functioning. It is through the normal process of separation-individuation that an individual develops appropriate intrapsychic representation of self and object. Normal separation-individuation is also a crucial prerequisite for the development of a sense of being an individual entity, a feeling of wholeness, and "a sense of human identity" (Mahler et al., 1975, p. 11).

Personality development

The phases and subphases of the normal child's developmental experience have been called by Mahler et al. (1975) the "psychological birth of the human infant," which is, in fact, the title of their book. These phases and subphases are described sequentially, starting with the forerunners to the separation-individuation process, the normal autistic, and the normal symbiotic phases. The separation-individuation process itself is divided into four subphases: (a) differentiation, (b) practicing, (c) rapprochement, and (d) the consolidation of identity and the beginnings of libidinal object constancy.

The normal autistic phase. This was the term used by Mahler et al. (1975) to describe the first few weeks of life. In those weeks, the newborn spends most of the time sleeping or half-sleeping. The infant wakes to a state of arousal principally when hungry or when other needs cause crying and sleeps again when needs are satisfied. The primary task of the autistic phase is the achievement of physiological equilibrium, which allows further physiological growth and development in the new environment outside the uterus (Mahler et al., 1975).

The infant in the autistic phase is concerned only with satisfaction of its needs and "seems to be in a state of primitive hallucinatory disorientation in which need satisfaction seems to belong to his own 'unconditional,' omnipotent, autistic orbit" (Mahler et al., 1975, p. 42). During the first few weeks of life, the infant is in a state of absolute primary narcissism. This is followed by a state of beginning awareness that need satisfaction comes from somewhere outside the self. This awareness marks the beginning of the symbiotic phase.

Mahler et al. (1975) pointed out that, while the infant in the autistic phase is not actively involved or invested in external stimuli, this does not imply a total lack of responsiveness to external stimuli. They stated that young infants have clearly been shown to have periods of "alert inactivity" (p. 43) in which the infant is most likely to be responsive to external stimuli. The authors pointed out that it is the infant's fleeting responsivity to external stimuli that makes possible the transition from the normal autistic to the symbiotic phase.

The normal symbiotic phase. This phase was described as beginning in the second month of life, during which time the infant starts to behave and to function as though he or she and the mother were an omnipotent system: "a dual unity within one common boundary" (Mahler et al., 1975, p. 44). The term "symbiosis" in this context of human development was intended to be a metaphor which describes a "state of undifferentiation, of fusion with mother, in which the 'I' is not yet differentiated from the 'not-I' and in which inside and outside are only gradually coming to be sensed as different" (Mahler et al., 1975, p. 44).

The essential feature of the symbiotic phase is an hallucinatory fusion with the representation of the mother, particularly the hallucinatory idea of a common boundary between two physiologically separate individuals. The human infant lacks the equipment and ability for self-preservation. Survival of the infant and the development of the rudimentary ego requires the emotional rapport of the mother's care. According to Mahler et al., (1975) "It is within this matrix of physiological and sociobiological dependency on the mother that the structural differentiation takes place which leads to the individual's organization for adaptation: the functioning ego" (p. 45).

During the normal symbiotic phase of development, the infant's needs are repeatedly gratified by the ministrations of the mother. Similarly, the infant rids his or her body of unpleasurable tensions by urinating, defecating, sneezing, spitting, and vomiting. The effects of these experiences over time help the infant to differentiate between experiences that have a pleasurable/good quality and those that have a painful/bad quality (Mahler et al., 1975).

Mahler et al. (1975) wrote that "the infant's inner experiences form the core of the self" (p. 47). The infant's experience of external stimuli contributes to the formation of the boundary between self and outer world. These two kinds of experiences form intrapsychic structures, which provide the framework for self-orientation within the external world. The intrapsychic structures that are established in the symbiotic phase provide a framework to which all experience must be related before there are clear, whole representations of the self and the object world in the ego. The mother provides an auxiliary ego for the infant (e.g., she organizes the infant's experience, providing optimal stimulation, soothing and comforting, and generally providing psychological supplies the infant cannot provide for himself or herself) and is the symbiotic organizer which allows the psychological birth of individuation.

Separation and individuation: The first subphase: Differentiation. Starting at four or five months of age, the infant begins for the first time to seem consistently alert whenever awake. Mahler et al. (1975) called this new look of alertness "hatching" (p. 54). The alertness is soon followed by new exploring behaviors. The baby pulls at the mother's hair and clothes. Instead of molding to the mother's body as the symbiotic infant does, the baby pulls away, often scanning back and forth between objects in the environment and the mother's face. During differentiation, the infant begins to realize that he or she is in a separate body from the mother. As motor skills develop, the baby will not only pull away from the mother's body but will also slide out of her lap, although tending to stay close to her feet.

As the infant learns to discriminate between mother and other people in the environment, there is also the appearance of what has been called "stranger anxiety" (Mahler et al., 1975, p. 56). The research observations of Mahler et al. (1975) showed tremendous variations in the timing, quantity, and quality of stranger anxiety. Some children, particularly those who had been observed to have had a pleasurable and harmonious symbiotic phase, showed more curiosity than anxiety about strangers.

The Second Subphase: Practicing. Mahler et al. (1975) separated the practicing subphase into two parts: (a) early practicing, in which the infant is first physically able to move away from the mother by crawling, paddling, and climbing, and (b) the practicing period proper, which is characterized by free upright locomotion. Early practicing, which starts at about 10 months of age, overlaps the differentiation subphase.

Mahler et al. (1975) described three psychological developments that foster the child's individuation and awareness of separation: (a) rapid body differentiation from the mother, (b) the establishment of a specific emotional bond with the mother, and (c) the growth of autonomous ego functions in close proximity to the mother.

In early practicing, the infant may move away from the mother, but the mother still serves as an anchoring center for the infant's world. Interest in the mother and in her continuous availability takes precedence over interest in the rest of the world. However, the infant's interest in the mother does "spill over" (Mahler et al., 1975, p. 65) onto inanimate objects, particularly objects provided by her, one of which may become a transitional object of the type described by Winnicott (1965).

In the practicing subphase proper which follows early practicing, the child achieves autonomous, upright locomotion. During this subphase, the child seems intoxicated with his or her new abilities. The main characteristic of this period is "the child's great narcissistic investment in his own functions, it’s own body, as well as objects and objectives of his expanding reality" (Mahler et al., 1975, p. 71). The child often seems impervious to knocks, falls, and other frustrations.

During the practicing subphase, the mother continues to be the home base, which provides "refueling through physical contact" (Mahler et al., 1975, p. 69). The toddlers at this stage were observed to become "low-keyed" (Mahler et al., 1975, p. 74) when aware that the mother was absent from the room. They would slow down, lose interest in their surroundings, and appeared to be internally preoccupied. This low-keyed state would end with the reunion with the mother, sometimes accompanied by a brief crying spell as if to release the tension accumulated in the mother's absence.

In the practicing subphase, it seems to be important to the process of separation-individuation for the mother to relinquish the symbiotic bonds and to allow and to enjoy the child's increasing capacity to operate at a distance from her. Mahler et al. (1975) wrote:

The expectation and confidence that mother exudes when she feels that the child is now able to "make it" out there seems to be an important trigger for the child's own feeling of safety and perhaps also the initial encouragement for his exchanging some of his magic omnipotence for pleasure in his own autonomy and his developing self-esteem. (p. 74)

The Third Subphase: Rapprochement. Around the middle of the second year of life, the child comes to the realization that he or she is in reality a very small person in a very big world. The sense of omnipotence and imperviousness to frustration fades, and there is an increase in separation anxiety.

As the toddler's awareness of separateness grows—stimulated by his maturationally acquired ability to move away physically from his mother and by his cognitive growth—he seems to have an increased need, a wish for mother to share with him every one of his new skills and experiences, as well as a great need for the object's love. (Mahler et al., 1975, pp. 76-77)

Characteristic behaviors of the rapprochement subphase include what Mahler et al. (1975) called "shadowing of mother," which was defined as "the child's incessant watching of and following every move of the mother" (p. 74) and playing a game that involved darting away from the mother with the expectation of being chased and caught in her arms. These behaviors seem to reflect the child's ambivalence about his or her increasing autonomy. The child fears the loss of the mother's love with the increasing separateness but also fears reengulfment into symbiosis.

The increasing awareness of the mother as a separate person with needs and desires that do not necessarily coincide with those of the child leads to what Mahler et al. (1975) called the rapprochement crisis. The rapprochement crisis is a difficult, painful time in which the child alternates intense neediness and clinging to the mother with equally intense negativity and struggling with her.

Mahler et al. (1975) wrote that the defense mechanism of splitting is at a height during the rapprochement crisis. The child protects the loving relationship with the good object by completely separating it from the bad object at which all aggression is directed.

The successful resolution of the rapprochement crisis seemed to involve increasing individuation and the achievement of an "optimal distance" (Mahler et al., 1975, p. 101) from the mother. The individuating child demonstrates a growing interest in socialization with others, including the father, and an increasing capacity for empathy. Mahler et al. (1975) hypothesized that these developments were facilitated by the growth in language skills, the increasing capacity for internalization, and progress in the use of play to express wishes and fantasies. Naming objects seems to give the child a greater sense of control in the environment. Processes of internalization allow the child to develop identifications with his or her good mother or father.

The successful resolution of the rapprochement crisis also, of course, depends upon the mother's response to the child. The mother who longs for a return to symbiosis and who rewards the periods of needy clinging or the mother who rewards the independent child and rejects dependency are each responding inappropriately to the legitimate needs in this subphase. Mahler et al. (1975) wrote that any psychopathology that develops from disturbances in the rapprochement subphase will depend upon the individual circumstances of each child. However, she also pointed out that two mechanisms of the rapprochement crisis, coercion and splitting, are also characteristic of adult borderline transference.

The Fourth Subphase: The phase of libidinal object constancy. The primary tasks of the fourth subphase are the establishment of a sense of identity and individuality and the attainment of object constancy. The attainment of object constancy involves the maintenance of the representation of the absent love object and "the unifying of the 'good' and 'bad' object into one whole representation" (Mahler et al., 1975, p. 110).

The establishment of object constancy was described by Mahler et al. (1975) as a gradual process which is largely influenced by emotional availability of the mother at earlier phases and subphases. If the child has trust and confidence that his or her needs will be met, he or she has a secure basis for developing object constancy.

Mahler et al. (1975) proposed that the principal conditions for mental health depend heavily on "the ability of the child to retain or restore his self-esteem in the context of relative libidinal object constancy" (p. 118). Both inner structures, a unified self image and libidinal object constancy, have their beginnings in the third year of life but are only the beginning of an ongoing developmental process.

The fourth subphase is a time when there is rapid development of complex cognitive functions, verbal communication, fantasy production, and reality testing. Observations about the real world become detailed, and there is an increasing interest in playmates and in adults other than the mother. A sense of time and an increased capacity to tolerate delays of gratification and to endure separation begin to develop. Under optimal conditions, the child seems to form and to hold onto an internal image of the mother in her absence.

In summary, Mahler et al. (1975) concluded on the basis of their observations that "the drive for and toward individuation in the normal human infant, is an innate, powerful given which, although it may be muted by protracted interference, does manifest itself all along the separation-individuation process" (p. 206). Normal development leading to a sense of identity and object constancy and providing a basis for mental health is determined by (a) the individual child's endowment, (b) the early mother-child interaction and relationship, and (c) crucial events in the child's growing up process "by positive and negative experiential factors which impinge upon the exquisitely pliable makeup of the child's individual psyche" (p. 201).

The Heuristic

Mahler’s (1968, 1972, Mahler et al., 1975) guide to investigation and theory development was the direct observation of babies and young children both in treatment and in a longitudinal study of interaction between normal infants and mothers. Most of her theoretical formulations concerned the normal development of intrapsychic structure in infancy, as was inferred from observation. She did not specifically discuss issues relating to the process or goals of psychotherapy.

Ontology or the Essential Concepts of the Theory

Mahler (1968, 1972, Mahler et al. 1975) was primarily concerned with the vicissitudes of normal development rather than with the development or treatment of psychopathology. Thus, this summary of Mahler's theory will only contain her ideas about essential human nature.

Essential Human Nature

Mahler (1968, 1972, Mahler et al., 1975) thought that the human infant had an innate drive to separate and to individuate intrapsychically from the mother after a normal stage of emotional and cognitive symbiosis. Symbiosis was used as a metaphor by Mahler to describe the early infant experience of an hallucinatory sense of fusion between the intrapsychic representations of the self and the mother. Mahler thought that, in normal human development, the infant is predisposed to evolve from an initial state of autistic self-absorption to a symbiotic merger and then through the stages of separation and individuation.

Separation and individuation were considered to be separate but complementary developments. Separation is the child's emergence from symbiotic fusion with the mother. The intrapsychic sense of separateness from the mother provides the foundation for clear intrapsychic representations of the self as distinguished from representations of objects. Individuation is the achievement of cognitive, perceptual, and affective attributes identified by the child as characteristic of his or her identity. The child's personality is shaped by the style and personality of the mother and by the quality of her responsiveness in each stage of development. The mother's responses to the developing child optimally support the unfolding of the child's innate predisposition to the development of common and universal intrapsychic structures and the individual child's unique endowment of temperament and personality.

Mahler et al. (1975) stated that the human child has an innate and powerful drive toward autonomy and individuation. Normal development leads to a sense of personal identity and object constancy. Normal development is dependent upon the inherited temperamental characteristics of the child, crucial events in the child's experience, and most importantly on the mother-child interaction and relation.

Otto Kernberg

Kernberg's (1976, 1980, 1984) theoretical contributions to psychoanalytic thought were based originally upon his attempts to understand and to explain the intrapsychic dynamics and behavior of patients who had been described in the psychoanalytic literature as "borderline" (Kernberg, 1980, p. 3). Kernberg (1980) wrote that, in the 1950s, borderline patients were described as patients with ego weakness that was manifested in acting out and severely regressive transference reactions. Borderline patients' symptoms were variously described as "a primitive kind of megalomania, intense aggression, magical thinking, severe mood swings, and a striking tendency to perceive significant others as all good or all bad" (pp. 3-4).

Kernberg (1980) explained that borderline patients could not be understood or explained within the context of the usual stages of libidinal development. Kernberg further stated that analysis of defense-impulse constellations in the borderline patient seldom clarified the existence of specific intrapsychic conflicts within the tripartite structure of id, ego, and superego. Similarly, there was no agreement among the theorists regarding treatment of borderline patients. Some psychoanalysts had abandoned psychoanalysis completely with these patients in favor of an essentially supportive approach aimed at strengthening ego functions. Others used modifications of strictly psychoanalytic techniques. Kernberg (1980) observed that questions about the most appropriate kind of treatment "could not be resolved as long as clinical observations remained unintegrated with theoretical formulations" (pp. 4-5).

The psychotherapy research project of the Menninger Foundation (Kernberg et al., 1972) was the first systematic, detailed study of the treatment of borderline patients over several years. Kernberg (1980) noticed a resemblance between normal behaviors of infants and toddlers described by Mahler (Mahler, 1972; Mahler et al., 1975) in the stages of separation/individuation and the pathological behavior of borderline patients described in the psychotherapy research project of the Menninger Foundation. In response, he developed both a theory of the dynamic characteristics and intrapsychic structure of borderline patients and a model of treatment specifically geared to patients with borderline personality organization.

These theories about the dynamics, structure, and treatment of the borderline patient led Kernberg (1976, 1980, 1984) to develop a theory of psychological development and psychodynamic intrapsychic structure based upon objects relations theory. His theory of development and intrapsychic structure covers and explains the continuum of psychological functioning from psychotic conditions to normal functioning with treatment recommendations for different levels of psychopathology.

Basic Concepts

Kernberg's (1976, 1980) initial emphasis was upon the distinction of neurotic, borderline, and psychotic conditions in patients. It should be noted that Kernberg's (1987) description and understanding of patients with what he termed borderline personality organization does not correspond with the diagnostic entity of the borderline personality disorder (American Psychiatric Association, 1987). Kernberg's (1980) conception of borderline personality organization included borderline personality disorder, most cases of infantile and narcissistic personality disorders, practically all schizoid, paranoid, and hypomanic personality disorders, and all "as if" and antisocial personality disorders.

Differentiation of neurotic, psychotic, and borderline patients. Kernberg (1976, 1980) noted that the neurotic patient demonstrates ego defenses based upon repression and other advanced defensive operations such as reaction formation, isolation, undoing, intellectualization, and rationalization. The ego defenses of the neurotic protect the ego from intrapsychic conflict by expelling from the conscious ego drives that are unacceptable to an overly rigid superego.

In individuals with borderline personality organization, splitting and related defensive mechanisms such as primitive idealization, primitive projection and projective identification, denial, omnipotence, and devaluation are exercised. These primitive defenses protect the ego from conflicts by dissociating contradictory experiences of the self and of significant others. As long as these contradictory ego states are separately activated and kept apart from each other, the ego is protected, and anxiety is prevented or controlled (Kernberg, 1980). These defenses of separating contradictory ego states protect the borderline patient from intrapsychic conflicts, but the cost is a weakening of the ego with reduced adaptability and flexibility (Kernberg, 1980).

Similar defensive operations can be observed in psychotic patients and are used to protect these patients from disintegration of the boundaries between self and object. However, when an interpretation of splitting mechanisms is given to psychotic patients, it brings about a deterioration in their functioning. With borderline patients, interpretation of splitting mechanisms leads to an improvement in functioning and some integration of the ego. Improvement or deterioration in response to interpretation of splitting mechanisms contributes to the differentiation of borderline from psychotic organization (Kernberg, 1980).

Kernberg (1976) defined splitting of the ego as "the lifelong co-existence of two implicitly conscious contradictory dispositions which did not influence each other" (p. 21). Each of these ego states, in treatment, is represented by "a full-fledged transference paradigm, a highly developed regressive transference reaction in which a specific internalized object relation was activated in the transference" (p. 21). Splitting represents a mutual denial of independent sectors of the psychic life with alternating "ego states" (p. 20) of repetitive, temporarily ego syntonic, compartmentalized psychic manifestations.

Kernberg (1980) made the observation that the borderline patient seems to have an ego organization that is different from that of a neurotic patient and with a similarity in defense mechanisms to psychotic patients. He, thus, suggested the hypothesis that borderline patients occupy a developmentally intermediate position between psychotic and neurotic patients. It was certainly clear that the borderline patient's ego has a different level of organization from that of either neurotic or psychotic individuals. The question was whether borderline conditions were a fixation at or a regression to a normal developmental stage or an abnormal deviation. Furthermore, if borderline conditions are a fixation at and/or a regression to a normal stage of development, at what period in life did this stage of development occur, and what conditions produced fixation and/or regression?

Object relations of the borderline patient. Kernberg (1980) hypothesized that careful examination of the internalized object relations of borderline patients, particularly the characteristic transference phenomena, might yield insight into the origins of borderline conditions. He theorized that the stages of development of internalized object relations in the borderline patient reflect the earliest structures of the psychic apparatus. Separate units of self-representations, object representation, and the affects linking them are the basic substructures of these early developmental stages which gradually evolve into more complex structures (e.g., the real-self, ideal-self, real-object and ideal-object representations).

Patients with borderline personality organization typically display identity diffusion, a poorly integrated concept of the self and significant others. Identity diffusion is manifested by chronic feelings of emptiness, contradictory self-perceptions, contradictory behavior that the patient cannot integrate in an emotionally meaningful way, and a shallow, impoverished perception of others. In contrast, patients with a neurotic personality structure have a solid sense of self and a deep understanding of significant others (Kernberg, 1980).

Kernberg (1987) further noted a significant difference between psychotic and borderline patients in their transference reactions despite the fact that psychotic patients also display identity diffusion. Psychotic patients tend to develop transference in which fusion or merger phenomena occur. They feel an absence of any boundary between themselves and the therapist, as if a common identity were shared by the patient and the therapist. Such transference fusion phenomena are usually accompanied by a loss of reality testing.

Borderline patients, even when experiencing transference psychosis, maintain a boundary between themselves and the therapist. The patient maintains a sense of being different from the therapist at all times. The transference of the borderline patient is one in which it seems as if the patient creates with the therapist an acting out of some aspect of a dissociated, unintegrated part object relation. For example, in one session the borderline patient may act toward the therapist as if the patient were a needy hungry baby and as if the therapist were the frustrating, cold, and distant mother, but during the next session, the patient may behave as if he or she were the cold and critical parent and the therapist were an inadequate, unpleasing child. Kernberg (1980) stated that, in the transference of the borderline patient, "There is a repetitive, alternating, sometimes quite chaotic 'interchange' of personality attributes partly projected and partly enacted by the patient, but never a true experience of fusion or merger" (p. 11).

Contributions of Mahler and Jacobson to Kernberg's theories. Kernberg (1980) hypothesized that the observations of the identity diffusion of the borderline patient and the radical differences in the transference phenomena between borderline and psychotic patients could be integrated with the theoretical formulations about early object relations contributed by Jacobson (1964) and Mahler (1972, Mahler et al., 1975).

Kernberg (1976) gave credit to Jacobson (1964) for describing the processes of internalization of object relations, the importance of differentiation of self and object, and the influence of object relations on the development of intrapsychic structure. Kernberg (1980) also claimed that his understanding of borderline dynamics was assisted by Jacobson's hypothesis that in borderline adolescents there is "a relation between a lack of integration in ego and superego and the persistence of nonintegrated internalized object relations as unintegrated components of ego and superego structure" (p. 10).

Kernberg (1980) was further inspired by Mahler's (1972, Mahler et al., 1975) descriptions and analysis of the differences between symbiotic psychosis and the abnormal outcome of the rapprochement subphase. Combining Mahler's and Jacobson's ideas with his own observations, Kernberg (1987) concluded that the early ego has to accomplish two tasks in rapid succession; it must differentiate self-representations from object representations, and it must integrate libidinally and aggressively invested self and object representations. The psychotic individual fails to achieve the first task. A pathological fusion or refusion of self and object representations results in a failure in the differentiation of ego boundaries and, therefore, in the differentiation of self from nonself. In contrast, with the borderline personality organization, differentiation of self from object representations has occurred to a sufficient degree to permit the establishment of ego boundaries and a differentiation of self from others.

The second task is the task of integrating self and object representations invested with libido with their corresponding self and object representations invested with aggressive drive derivatives. Borderline patients fail to accomplish this task because of a pathological dominance of aggression. The resulting lack of synthesis of contradictory self and object representations interferes with the integration of the self-concept and with establishing object constancy or "total" object relations.

Kernberg (1980) proposed that an integrated self-concept and object constancy are required for a stable ego identity. Object constancy was defined as the integration of good and bad (e.g., libidinally invested and aggressively invested) object representations into total object representations. Stable ego identity is a crucial factor in the foundation of an integrated and flexible ego and also influences the development of sophisticated superego functions.

Characteristics of the borderline patient. In the borderline patient, the intensity of the aggression linked with the all-bad object relation and the defensive idealization of the all-good object relation make integration impossible. The borderline individual is unable to experience consciously the coexistence of the split all good and all bad representations of self and significant others because the implicit danger to the good object relations would trigger unbearable anxiety and guilt. Splitting protects the ego while sacrificing integration (Kernberg, 1980).

In consequence, the borderline patient develops several related symptoms. The borderline individual displays a chronic overdependence on external objects. Contradictions between exaggerated, all-good, ideal object representations and sadistic, all-bad object representations interfere with superego integration. The lack of superego integration makes the borderline individual prone to projection of cruel superego forerunners (e.g., similar to those described by Jacobson, 1964 as explained earlier in this chapter) onto others, creating paranoid thoughts and feelings. The lack of integration of object representations and self-concept interferes with the capacity for empathy and understanding others (Kernberg, 1980).

Kernberg (1976, 1980) concluded that the pathology of internalized object relations of the borderline patient indicated a regression to and/or fixation at a stage of development later than symbiosis and preceding object constancy. Specifically, the intact reality testing, the identity diffusion, and the dominance of a defensive constellation centered around splitting indicate that the borderline structure is a product of a pathological resolution of the rapprochement subphase of separation-individuation.

Personality Development

Kernberg (1976) proposed a general theory of the normal and pathological development of object relations that explained (a) the basic units of internalized object relations composed of a self and object representation and the affects linking them, (b) the five basic stages in the differentiation and integration of object relations, (c) the relation between failure in these developments and the origins of various types of psychopathology, and (d) the implications of the sequences of stages for general developments of intrapsychic structure.

Stage one: Normal autism. This stage of primary undifferentiation, precedes the development of the normal symbiotic relation with the mother. This stage lasts through the first month of life, and throughout it, there is a gradual buildup of the normal, primary, undifferentiated self-object representation (Kernberg, 1976).

Stage two: Normal symbiosis. This stage of primary undifferentiated self-object representations is the stage in which pleasurable, gratifying experiences with the mother are consolidated in a good, fused self-object image. This good self-object constellation becomes the nucleus of the self-system of the ego and the basic organizer of the early ego. Experiences that are frustrating or painful are integrated into a bad self-object constellation. These two primary intrapsychic structures are organized separately under different affective conditions. This stage extends from the second month of life to around six to eight months of age (Kernberg, 1976). Pathological fixation or regression at this stage results in the loss of ego boundary differentiation, which is characteristic of symbiotic psychosis of childhood, most types of adult schizophrenia, and depressive psychosis (Kernberg, 1976).

In the context of the development of the undifferentiated self-object representations, affects gradually become differentiated. Pleasurable affects evolve into specific pleasures of oral satiation, excitement of the erotogenic zones, gratification of exploratory behavior, and most importantly of interpersonal experiences. Painful affects evolve into anxiety, fear, and rage (Kernberg, 1976).

Gratification and limited frustration activate attention and motivate learning, which contributes to the gradual differentiation of self representations from object representations. This gradual differentiation is fostered by the maturation of autonomous ego functions such as perception and memory. Excessive frustration or deprivation creates a state of generalized anxiety in the infant. The disorganizing effect of anxiety interferes with the process of differentiation of self and object representations (Kernberg, 1976).

The developmental series of good self-object representations becomes the intrapsychic structures invested with libido. The series of bad self-object representations becomes those invested with aggression. Kernberg (1976) proposed that "from a clinical point of view, one might say that the evolving affect states and affect dispositions actualize, respectively, libidinal and aggressive drive derivatives" (p. 64).

Stage Three: Differentiation of self-representation from object-representation. This stage begins with the completion of the differentiation of the self-representation from the object-representation within the core of good self-object representations. This stage includes the later differentiation of self and object representations within the core bad self-object representation. The stage ends when the child has integrated good and bad self-representations into an integrated self concept and integrated good and bad object representations into a total object representation, which represents the achievement of object constancy. This stage begins around the eighth month of life and ends around the thirty-sixth month (Kernberg, 1976).

During this stage, the defense mechanism of splitting is used to protect the ideal, good relation with the mother from the potential destructiveness of the bad self or object representations. Because there is not an integrated concept of either self or object, this stage is characterized by the experience of "part object relations" (Kernberg, 1976, p. 65), defined as being composed of (a) a representation of the self, (b) a representation of the object in interaction with the self-representation, and (c) an affective state, usually of a strong quality such as rage, fear, or idealized love. Pathological fixation at or regression to this stage of development is the basis for borderline personality organization.

Stage four: Integration of self representations and object representations and development of higher level intrapsychic object relations derived structures. This stage begins in the latter part of the third year and lasts through the entire Oedipal period. There is a continuing development of the integration of libidinally and aggressively invested self representations into a unified self system and integration of libidinally and aggressively invested object representations into whole object representations. The ego and superego are consolidated as definite unified intrapsychic structures (Kernberg, 1976).

Pathological conflict at this stage typically occurs between the ego and "a relatively well-integrated but excessively strict and punitive superego" (Kernberg, 1976, p. 67). The psychopathology that results from problems at this stage is usually represented by the neuroses and higher level character disorders, particularly the hysterical, obsessive-compulsive, and depressive-masochistic. Kernberg (1976) also hypothesized that the narcissistic personality is characterized by an abnormal condensation of the new intrapsychic structures appearing at this stage of development.

Kernberg (1976) proposed that the structure of narcissistic personalities is characterized by (a) a pathological condensation of real self, ideal self, and ideal object structures, (b) repression and/or dissociation of "bad" self-representations, (c) generalized devaluations of object representations, and (d) blurring of normal ego-superego boundaries. These structures result in the development of a grandiose self embedded in a defensive organization similar to that of borderline personality organization.

Kernberg (1976) proposed that the integration of affectively opposite self-representations provides the basis for the depressive position described by Klein (1974) and the stage that Winnicott (1958) described as a phase in which feelings of guilt and concern appear. Integration of opposite self representations produces a general deepening and broadening of affective potential including the capacity for guilt feelings.

In contrast to the new, more realistic self and object representations developing in this phase, the child also develops representations of an ideal self and an ideal object that reflects, in fantasy, the lost ideal state of the all good self and object representations. The condensation of the ideal self and object constitute the nucleus of the ego ideal, which contributes to the development and integration of the superego which begins in this stage (Kernberg, 1976).

The earliest forerunners of the superego are the internalization of "fantastically hostile, highly unrealistic object-images reflecting 'expelled,' projected and reintrojected 'bad' self-object representations" (Kernberg, 1976, p. 71). These images probably originate in the primitive efforts of the infant to protect the good relation with the idealized mother by turning aggressively invested images of her against himself or herself. The amount of frustration and aggression experienced by the infant determines the strength and dominance of these sadistic superego forerunners.

During stage four, as the ego ideal develops, it is integrated with the sadistic superego forerunners. The superego, thus, repeats the process occurring in the ego, the integration of libidinally and aggressively invested object relations. This integration modulates the fantastic, exaggerated nature of both the primitive idealization of the early ego ideal and the punitive hostility of the sadistic forerunners. The modulation of these superego nuclei leads to a decrease in projective defenses, which, in turn, lead to the next level of superego structure, the internalization of realistic parental demands and prohibitions (Kernberg, 1976).

Stage five: Consolidation of superego and ego integration begins with the completion of the integration of the superego. The integrated superego fosters further integration and growth of ego identity. Ego identity continues to evolve by means of an ongoing reshaping of self and object representations on the basis of real experiences with others. The more integrated the self-representations, the more self-perception in any particular situation corresponds to the total reality of the person's interactions with others. The more integrated the object representations, the greater the capacity for realistic appreciation of others and reshaping one's internal representations on the basis of such realistic appraisals (Kernberg, 1976).

A harmonious world of internalized object representations provides love, confirmation, support, and guidance within the object relations system of the ego. This internal world gives depth to interaction with others. "The intrapsychic and interpersonal worlds relate to and reinforce each other" (Kernberg, 1976, p. 73). Kernberg (1976) proposed that "Internalized object relations may be considered a crossroad where instinct and the social system meet and contribute crucially to the development of the personality of the individual" (p. 59). Criteria for mental health and normality include (a) the depth and stability of internal relations with others, (b) the tolerance of ambivalence toward loved objects, (c) the capacity to tolerate guilt and separation and to work through depressive crises, (d) the extent to which the self concept is integrated, and (e) the extent to which behavior patterns correspond to the self-concept.

Kernberg (1976) gave no timetable for the developments of stage five, but the implications of the continuing reshaping of self and object representations in light of real experiences with others suggest that stage five processes of superego and ego integration could continue throughout life. Individuation includes the gradual replacement of primitive introjections and identifications with partial subliminatory identifications fitting into the overall concept of self. Emotional maturity corresponds with the capacity to discriminate subtle aspects of one's own self and of other people and in an increasing selectivity in accepting and internalizing the qualities of other people. Mature friendships are based on such selectivity and the capacity of combining love with independence and emotional objectivity.

The Processes and Goals of Psychotherapy

Kernberg (1976, 1977, 1978, 1979, 1980, 1984) has written extensively about psychoanalytic treatment of the borderline personality. He has discussed the defining parameters of a spectrum of psychoanalytic psychotherapies, the differences in approach among these therapies, the differences in effect of techniques on different types of patients, and special problems that arise in the treatment of borderline patients, including the management of transference and countertransference.

Kernberg (1984) defined psychoanalytic technique as the "(1) Consistent adherence to a position of technical neutrality; (2) consistent use of interpretation as a technical tool; and (3) facilitation of the development of a full-fledged transference neurosis and of its psychoanalytic resolution solely by interpretation" (p. 101). Technical neutrality depends on the therapist's maintaining an empathic attitude even when faced with regressive aggression and on the therapist's cognitive capacity to integrate the fragmented transference reaction expressed by the patient (Kernberg, 1984).

Kernberg (1984) wrote that psychoanalysis, expressive psychoanalytic psychotherapy, and supportive psychoanalytic psychotherapy can each be differentiated from one another on the basis of modifications of those three defining parameters. The type of therapy he recommended for borderline patients was expressive psychotherapy, which is an interpretive, uncovering approach.

Psychoanalysis. Kernberg (1984) claimed that psychoanalysis is the treatment of choice for patients with neurotic personality organization, hysterical, obsessive-compulsive, and depressive-masochistic disorders, patients with narcissistic personalities who are not functioning on an overt borderline level, and for some patients with a mixture of infantile and hysterical personality structures. In psychoanalysis, with patients with good ego strength, technical neutrality fosters transference regression and permits interpretation by not gratifying transference. Transference analysis permits its gradual resolution, and interpretation reduces defenses, allowing the emergence of repressed material (Kernberg, 1980).

Expressive psychotherapy. In expressive psychotherapy, technical neutrality provides the basis for transference interpretation and resolution. In the treatment of borderline personality, technical neutrality is often interfered with or limited by the necessity of helping to provide structure in the external life of patients who cannot function autonomously. Interpretation of these deviations from technical neutrality may permit a return to a neutral stance (1984).

In working with the borderline patient, interpretation increases ego strength by resolving primitive defenses. However, with the borderline patient, the therapist must carefully explore the patient's understanding of the therapist's interpretations. Thus, in expressive psychotherapy, clarification usually takes precedence over interpretation (Kernberg, 1984).

Transference interpretation in expressive psychotherapy focuses on the immediate realities of the patient's life. The interpretation of primitive transference gradually leads to the integration of part object relations into total object relations. Kernberg (1984) recommended that (a) the transference of borderline patients, particularly the negative transference, be interpreted only in the here and now without attempts at genetic reconstruction, (b) primitive defenses should be interpreted as soon as they enter the transference (their interpretation strengthens the ego and brings about structural intrapsychic change), and (c) the less primitive, modulated aspects of positive transference should not be interpreted because they support the development of the therapeutic alliance.

The general rule Kernberg (1984) recommended is that interpretation should proceed from surface to depth. In other words, interpretations should begin with defenses obviously displayed in the treatment and should proceed very gradually to interpretations linking historical events to present day functioning. The therapist should tell the patient about his or her observations, in order to stimulate the patient to integrate these observations, and should interpret only when it is clear that the patient cannot achieve insight on his or her own. Clarification of reality and interpretation in depth should be integrated as often as possible.

Kernberg (1984) proposed that the therapist's rational cognitive ability is as important as are warmth and empathy. He commented that "the therapist's cognitive formulations strengthen or broaden the patient's integration of affects and internalized object relations" (p. 19).

However, Kernberg (1984) also commented that, with the borderline patient, "The nonspecific ‘real’ human relationship reflected in the therapeutic alliance may constitute an important corrective emotional experience" (p. 121). Such a positive interaction, which often goes far beyond anything the patient has previously experienced, is normally gratifying. The real relation with the therapist may allow the patient to experience parental functions never experienced before.

This is not to say that the therapist must gratify needs or abandon technical neutrality to provide a supportive function. Kernberg (1984) emphasized that demands of the patient for the therapist to fulfill the function of a parent must be carefully analyzed and understood. What strengthens the ego of the patient is the achievement of genuine insight into the pathological reactions, impulses, and defenses that were activated by early traumas and frustrations rather than gratifications in the here and now of what was denied in the past. Kernberg (1986) commented that "authentic insight is a combination of intellectual and emotional understanding of the deeper sources of one's psychic experiences, accompanied by concern and an urge to change the pathological aspects of those experiences" (p. 122).

Supportive psychotherapy. Kernberg (1984) stated that expressive psychotherapy is almost always the most appropriate treatment for patients with borderline personality organization. However, in some cases, patients who are unwilling or unable to tolerate the demands of interpretive treatment may be treated in supportive psychotherapy.

Kernberg (1984) defined supportive psychotherapy as a treatment in which mostly suggestion and environmental intervention, some clarification and abreaction, and no interpretation are used. The use of suggestion and environmental intervention eliminates technical neutrality, and transference is not interpreted.

The basic technique of supportive psychotherapy is the exploration of primitive defenses in the here-and-now in order to foster a better adaptation to reality by making the patient aware of the disorganizing effects of these defensive operations. Transference, especially negative transference, is not interpreted but is examined in the context of the reality of the treatment situation and is utilized for the clarification of related interpersonal problems (Kernberg, 1984).

In supportive psychotherapy, the mechanisms of change are described by Kernberg (1984) as follows. First, the patient's ego functions are activated and strengthened, particularly in the areas of reality testing, self-awareness, and the toleration and integration of contradictory affects, experiences, and behaviors. The examination of primitive defenses, even with analytic resolution, decreases their control and strengthens the ego. On a deeper level, the activation of primitive object relations in transference permits some "consolidation of basic trust of the libidinally invested self and object representations and should permit some partial identification with the therapist" (Kernberg, 1984, p. 164).

The Heuristic

Kernberg's (1976, 1980, 1984) original guide to investigation was an attempt to understand and to explain the intrapsychic development and dynamics and the resulting behavior of patients whom he described as having borderline personality organization. Kernberg decided that questions concerning the appropriate treatment of patients with borderline personality organization could only be answered and resolved if clinical observations were integrated with theoretical formulations concerning the development of intrapsychic structure. Accordingly, Kernberg attempted to integrate clinical data about borderline patients from the psychotherapy research project of the Menninger Foundation (Kernberg et al., 1972) with the theories about the development of intrapsychic structure presented by previous object relations theorists.

Ontology or the Essential Concepts of the Theory

Essential Human Nature

Kernberg (1976, 1980, 1984) was primarily concerned with understanding the dynamics and etiology of borderline personality organization of patients. In his study of borderline patients, Kernberg noted behaviors, particularly in their transference responses, that seemed to reflect internalized object relations similar to the primitive object relations hypothesized by Mahler (1972, Mahler et al. 1975) in the stages of separation and individuation. His understanding of borderline pathology was also influenced by Jacobson's (1964) ideas about the influence of object relations on the development of intrapsychic structure. Using the phases of development described by Mahler (1972; Mahler et al., 1975) and Jacobson's (1964) ideas about the construction of intrapsychic structure, Kernberg developed a general theory of the normal and pathological development of object relations.

In regard to essential human nature, Kernberg (1976) wrote that the development of internalized object relations represented the result of interaction between the genetic endowment of the individual and the environment. Kernberg stated that the origins of pathology could be based in the innate characteristics of the individual such as excessive aggression or dependency or from the lack of optimal responsiveness of the environment or probably most often from an interaction between the two.

Essential human nature as described by Kernberg (1976) seemed to be a nature that evolves intrapsychically in response to interpersonal experience. A harmonious world of intrapsychic object representations provides love, confirmation, support, and guidance within the object relations system of the ego. The internal world gives depth and clarity to interactions in the external world. Intrapsychic and interpersonal worlds of the individual mutually support, influence, reinforce, and modify each other.

The Nature and Development of Psychopathology

Kernberg (1976, 1980), as has been explained, was particularly interested in identifying the differences among neurotic, borderline, and psychotic patients. He was concerned with explaining both dynamics and etiology. Many of Kernberg's ideas about the differing dynamics of neurotic, borderline, and psychotic patients were based upon meticulous observations of their use of ego defenses, particularly in transference reactions. The characteristic use of defenses of each of these groups of patients inspired Kernberg to form hypotheses about differences in intrapsychic structure. He was particularly intrigued by the qualitative differences in interpersonal behavior that indicated striking differences in internalized intrapsychic object relations.

The neurotic patient. The interpersonal relations of neurotic patients indicate that these individuals have a solid sense of self and a deep understanding of significant others, demonstrating an integrated self representation and object constancy in their intrapsychic object relations. Pathological symptoms and character formation result from conflict between the ego and an excessively strict and punitive but relatively well-integrated superego. Kernberg (1976) hypothesized that neurotic psychopathology occurred during the fourth stage of development.

The borderline patient. The interpersonal behavior of borderline patients indicates an underlying identity diffusion with a poorly integrated concept of self and significant others. In transference (as well as in other significant relationships), the borderline individual often seems to be acting out a dissociated, unintegrated, part object relation, behaving as if both patient and therapist were limited to particular aspects of personality linked by a powerful affect. There is a repetitive, alternating projection and enactment of personality attributes by the patient.

The borderline patient is typically not aware of the contradictory and often chaotic nature of the manifestation of part object relations in interpersonal relations with others. The defense mechanism of splitting allows a mutual denial of independent sectors of intrapsychic life. The repetitive manifestations of part object relations represent compartmentalized intrapsychic experience. This compartmentalization allows the existence of contradictory ego states which enter consciousness separately without influencing one another. Interpretation of splitting mechanisms to borderline patients leads to an improvement in functioning and in ego integration.

Kernberg (1976) proposed that borderline personality organization is based on a pathological fixation or regression at the third stage. There is no integrated concept of either self or other in this stage, which leads to the experience of part object relations that are characteristic of borderline dynamics.

The psychotic patient. The defense mechanisms of the psychotic patient are similar to those used by the borderline patient. Psychotic patients use primitive defenses to protect themselves from disintegration of the boundaries between self and object unlike the borderline patient who uses them to protect the intrapsychic representations of the good object relations from the aggressively invested object relation. The interpretation of splitting mechanisms to psychotic patients brings about a deterioration of their functioning.

The psychotic patient has suffered a failure in the differentiation of ego boundaries, of self from nonself. Patients with psychotic personality organization include patients with symbiotic psychosis of childhood, most types of schizophrenia, and depressive psychosis. Psychotic disorders represent a regression or fixation at the stage of normal symbiosis (Kernberg, 1976, 1984).

The Role Of Therapist

Kernberg(1984) proposed that all psychotherapy requires at minimum that the therapist express warmth and empathy. However, the therapist's capacity for empathy must include not only an intuitive awareness of the patient's conscious emotional experiences but also the ability to empathize with the actively split off or dissociated emotions and attitudes that the patient cannot tolerate as being a part of the patient's experience of his or her self.

Kernberg (1984) proposed that the therapist's rational, cognitive ability is as important as warmth and empathy because the cognitive formulations provided by the therapist are the foundation for the patient's integration of affects and intrapsychic object relations. It is important that the therapist not attempt to gratify needs or to provide a supportive function. The patient's wish or demand for the therapist to fulfill the function of a parent must be analyzed by the therapist and understood by the patient.

The role of the therapist depends upon the type of treatment utilized. The recommended treatment depends on the ego strength and intrapsychic organization of the patient. In psychoanalysis, the therapist maintains strict technical neutrality and uses interpretation of defenses and transference. In expressive psychotherapy, the therapist maintains technical neutrality and uses interpretation of primitive defenses and transference reaction with the extensive clarification of the interpretations. In supportive psychotherapy, the therapist uses suggestion and environmental interventions which eliminate technical neutrality but also uses the identification and exploration of primitive defenses and transference reactions to strengthen ego functions and object relations.

The Mechanism Of Change

The mechanism of change identified by Kernberg (1984) also depends upon the type of treatment used with the patient. In psychoanalysis, the basic mechanisms of change are the analysis and resolution of transference neurosis and the interpretation of defense. The interpretation of defense reduces the use of defensive operations and allows the emergence of repressed intrapsychic conflict, which then can be resolved.

In expressive psychotherapy with borderline patients, Kernberg (1984) claimed that the most important mechanism of change was the achievement of genuine insight into the pathological reactions, impulses, and defenses that were activated by early traumas and frustrations. Authentic insight was defined by Kernberg as a combination of intellectual and emotional understanding of the source of psychic experience accompanied by concern and the urge to change the pathological aspects of psychic experience.

In the treatment of the borderline patient, Kernberg (1984) acknowledged the possibility of a corrective emotional experience existing within the nonspecific, real human relationship of the therapeutic alliance. The positive therapeutic relationship may provide the patient with normal gratifications that have never before been experienced. The genuine concern, warmth, empathy, and attentiveness of the therapist may provide parental functions that were never before offered to the patient. These corrective emotional experiences are regarded by Kernberg as providing the foundations for a deepening therapeutic alliance within which the patient can achieve the insight that is the most important vehicle for therapeutic change.

Kernberg (1984) described the mechanisms of change in supportive psychotherapy as follows. (a) The patient's ego functions are activated and strengthened particularly in the areas of reality testing, self-awareness, and the toleration and integration of contradictory affects, experiences, and behaviors. (b) The identification and examination of primitive defenses decreases their control and strengthens the ego. (c) On a deeper level, the activation and identification of primitive object relations permits some consolidation of trust in the libidinally invested self and object representations and fosters partial identification with the benign concern and empathy of the therapist.

The Goals of Psychotherapy

As can be discerned from the descriptions of the mechanisms of change, the goals of treatment in each type of therapy described by Kernberg (1984) are the strengthening of the ego and/or the resolution of intrapsychic conflict. In psychoanalysis, the ego of the patient is strengthened as the overly strict and punitive superego is modulated into a more reasonable stance. As the transference neurosis and defenses are interpreted, intrapsychic conflict emerges into consciousness and can be resolved. The processes of resolution of intrapsychic conflict and ego strengthening complement each other and stimulate growth (Kernberg, 1984).

In expressive psychotherapy, the goals of treatment of the borderline patient are to strengthen the ego by resolving primitive defenses and to encourage the development of integrated self and object representations through the interpretation of part object relations manifested in the transference. The development of integrated self and object representations is the basis of mature total object relations which, in turn, strengthen the ego and foster the development of modulated, realistic superego functions.

In supportive psychotherapy, the goals of the therapy are primarily to strengthen ego and object relations. Improved ego and object relations, in turn, provide the foundation for improved interpersonal relationships and environmental functioning.

Critiques of Object Relations Theorists

Wolberg (1988/1995) made several criticisms of object relations theories. First, he stated that the concept of an innate need for attachment to an object means exactly the same thing as an instinctual libidinal drive. Second, he criticized Fairbairn (1954) and other object relations theorists for repudiating the concept of instinctual aggression and stated that neurophysiologists have found subcortical centers for aggression (no reference was cited). Third, and perhaps, most importantly, he criticized the hypotheses developed by object relations theorists concerning normal development because these hypotheses were based on the observations of borderline and psychotic patients rather than on the observation of normal infants. He questioned the validity of making assumptions about the internal experiences of normal infants based upon regressive experiences of very disturbed patients.

Wolberg (1988/1995) gave credit to object relations theorists for original and ingenious ideas that have "challenged the sanctity of some existing metapsychological credentials" (p. 300) thus opening a way to a reassessment of their value. He also gave credit for the clinical usefulness of the ideas of Fairbairn (1954), Winnicott (1965), Mahler, (Mahler et al., 1975), Jacobson (1964), and Kernberg (1976, 1980).

Wolberg (1988/1995) considered Klein (1964, 1975) separately from other object relations theorists. He stated that her theories provided a basis for present day ego analysis and object relations theory. He gave her credit for proposing an environmental basis (rather than a biological basis) for the development of the superego, the impact of intrapsychic conflict between interjected good and bad objects, the influence on object relations of projections from the objects, and the importance of aggressive instincts in pathological development.

Wolberg (1988/1995) completely dismissed Klein's (1964, 1975) concepts of phylogenetically determined innate images of genitals, breasts, and the primal scene and her ideas about incorporative and projective symbolic thought processes in infancy. He also indicated that Klein's idea that all psychopathology was based on distortions developed in the first year of life is generally considered to be too circumscribed and limiting. He stated that most clinicians rejected Klein's ideas about pregenital Oedipal conflict and about the influence and impact of an innate death instinct.

Kernberg (1980) acknowledged that Klein (1964, 1975) provided important contributions to the understanding of borderline and psychotic conditions and to the understanding of the early interrelations and variations in the manifestation of the aggressive instinct and the development of defense mechanisms and object relations. He gave her credit for the first major development of object relations theory in psychoanalysis, but he also criticized her heavy emphasis on the influence of instinctual drives.

Kernberg (1980) identified two paradoxical issues that permeate Kleinian theory. First, although she stressed the importance of object relations, Klein's (1964, 1975) emphasis on instinct minimizes the importance of psychosocial determinants of intrapsychic conflict. Second, while emphasizing the importance of instincts, Kleinian authors have been little concerned with important developments in contemporary biology and ethology. Kernberg (1980) specifically criticized the Kleinian concepts of an inborn death instinct and of the death instinct as the earliest determinant of anxiety (Klein, 1964, 1975). He also criticized her assumptions about the innate knowledge of infants and was particularly critical of some of the techniques advocated for use in the process of psychotherapy. He stated that Kleinian theorists recommend the use of premature, "deep" interpretations of unconscious fantasy and transference manifestations. He argued that these premature "deep" interpretations interfere with the interpretation and resolution of ego defenses and create the risk of indoctrinating patients intellectually rather than allowing deep unconscious material to emerge naturally.

Kernberg (1980) stated that Fairbairn (1952) stands out among the psychoanalytic theoreticians who have made object relations a major focus of their theories because of his consistent effort to formulate a developmental model based upon the internalization of object relations. He gave credit to Fairbairn for first formulating the theoretical concept that pure drives cannot be found in clinical situations. Fairbairn (1952) proposed that what does occur in the clinical setting is an activation of affects that reflect drives that emerge in the context of internalized object relations. The patient's internalized object relations are reenacted in the transference. He also stated that Fairbairn's proposal that the internal world of object relations starts as a dyadic, internalized relation between what he called a self-component and an object-component was an impressive contribution.

Kernberg (1980) had three specific criticisms of Fairbairn (1952). First, he criticized Fairbairn for regarding aggression as the effect of frustrating object relations rather than as an inborn disposition evoked by frustrating object relations. Second, he thought that there was no need for Fairbairn to replace the traditional tripartite structure of id, ego, and super-ego with a tripartite structure composed of conscious and unconscious three way splits of the ego with each split connected to an appropriate object representation. Essentially, Kernberg (1985) asserted that the various internal self (ego) and object representations hypothesized by Fairbairn were contained within the traditional tripartite structure. Finally, Kernberg (1980) stated that a major problem with Fairbairn's (1952) theory was his ideas about psychological development in the first few months of life. Fairbairn seemed to hypothesize that self and object were differentiated even in the earliest months of life. Kernberg stated that Jacobson’s (1964) and Mahler's (1968, Mahler et al., 1975) concept of earliest experience as reflecting a fused self-object representation presented a more accurate picture of early development than Fairbairn's hypothesis of early differentiation.

Kernberg (1980) was, in general, appreciative of the ideas of Jacobson (1964) and Mahler (1968, Mahler et al., 1975). He stated that Jacobson had the only comprehensive psychoanalytic object relations theory that adequately accounted for the early development of affects, instincts, object relations, and defense mechanisms and that linked these developments with the development of the tripartite structure. He seemed particularly impressed with Jacobson's ideas about the development of the superego and the relation between faulty super-ego development and depression.

Kernberg (1980) gave credit to Mahler (1968, 1972, Mahler et al., 1975) for identifying the earliest stages of development and for substantiating them with evidence. He stated his particular appreciation for her identification of the substages of separation-individuation. He stated that Mahler made significant contributions to the understanding of early development and its relation to the formation of intrapsychic structure and the development of psychopathology.

Summary of Object Relations Theorists

Klein's (1964, 1975) theories contributed significantly to establishing the importance of early object relations in normal and pathological development, which Kernberg (1980) claimed is a generally accepted psychoanalytic concept at this time. Klein also made valuable contributions with her subsequent formulations about introjection and projection and, particularly, with her identification and description of the primitive defense mechanisms of projective identification and splitting. Her ideas about part object relations have been developed further, as have her ideas about the paranoid-schizoid position and the depressive position.

Klein's (1975) idea that memories of early childhood exist only in feelings that are revived in transference is similar to the formulations of Fairbairn (1954) and Kernberg (1976, 1980, 1984) that the feelings that arise in the transference are linked to the early internalized object relations and can be used to help the patient understand intrapsychic experience. Kernberg, however, cautioned against the use of these transference phenomena to reconstruct childhood experience until late in treatment when the development of ego strength allows the modulation of early bizarre and fantastic experience.

However, most of the object relation theorists reviewed in this chapter disagree with many of Klein's (1964, 1975) ideas. Mahler et al. (1975), Kernberg (1976, 1980), and Jacobson (1964) all clearly disagree with her ideas that growth and development in the first year of life encompassed the depressive position and the origins of the Oedipal conflict. Kernberg (1976) explicitly stated his disagreement with the idea that the child reaches the depressive position within the first year of life.

There is also little agreement with Klein's (1964, 1975) emphasis on instinctual aggression and the innate death instinct. Fairbairn (1954) stated that he did not consider aggression to be an instinct but only a response to frustration. Winnicott (1989) specifically repudiated Klein's concept of envy of the good breast, and although he emphasized the importance of the mother surviving the aggression of the child, he interpreted aggression as originating in the child's innate needs to move, to grow, and to explore. Winnicott (1989) explicitly stated his disagreement with the idea of an innate death instinct and thought that aggression is manifested as anger or destructiveness only under conditions of frustration. Jacobson (1964) and Kernberg (1976) expressed the idea that individuals have an innate capacity for libido and aggression, which develops with the object relations.

Fairbairn (1954) made what Kernberg (1980) calls an impressive contribution in proposing that the internal world of object relations starts as a dyadic internalized relation between a self component and an object component. Kernberg (1980) also stated that Fairbairn's idea that, in clinical situations, affects are activated only in the context of internalized object relations reenacted in transference was a unique and original contribution. Both of these concepts were also recognized and explored by Jacobson (1964).

There has been no adoption of Fairbairn's unique theory of intrapsychic structure. However, his descriptions of the intrapsychic structure based entirely on object relations is not unlike the ideas of Jacobson (1964) and Kernberg (1976, 1980) about the development of the tripartite system of ego, superego, and id based upon the maturation of object relations as Kernberg stated in his critique of Fairbairn’s theories.

Both Kernberg (1980) and Greenberg and Mitchell (1983) stated that Winnicott (1955, 1964, 1965, 1989) did not formulate a fully integrated developmental theory. However, he developed a great many interesting and useful ideas, which have been used by subsequent theorists. Mahler (1972), Jacobson (1964), and Kernberg (1976) all expressed agreement with Winnicott's idea that life begins in a state of undifferentiation. There was similar agreement concerning Winnicott's ideas about primary maternal preoccupation, the feelings of omnipotence in infancy, transitional objects, and the good-enough mother.

Kernberg (1980, 1984) agreed with Winnicott (1958) that one of the functions of the therapist is to provide a holding environment. He also agreed that the "holding" (Kernberg, 1984, p. 103) was provided through empathic understanding and concern. Kernberg (1984) placed more emphasis than Winnicott on the cognitive and interpretative functions of the therapist, and Winnicott emphasized the ways in which the analytic setting fulfills early, unmet, developmental needs. However, Winnicott (1972) also stated that the therapist provides symbolic holding to the patient through an accurate and well-timed interpretation that demonstrates that the patient is deeply understood.

The similarities in the ideas and theories of Mahler (1972, Mahler et al., 1975), Jacobson (1964, 1971), and Kernberg (1976, 1980, 1986) are numerous. Both Mahler and Kernberg acknowledged the influence of Jacobson on the development of their ideas.

Kernberg's (1976) entire theory of personality development with stages of normal autism, normal symbiosis, differentiation of self from object representations, integration of self representations and object relations, and consolidation of superego and ego integration was based on Mahler's developmental stages of normal autism, normal symbiosis, and the subphases of separation-individuation. Kernberg's (1976) understanding of borderline character dynamics was partially based upon Mahler's recognition of the similarities of splitting and coercion of the child at the rapprochement subphase of separation-individuation and the transference reactions of the borderline patient.

Jacobson's (1964, 1971) formulations about the initial fused self-object identification was adopted by Mahler (1972, Mahler et al., 1975) and Kernberg (1976, 1980) as were her ideas about the gradual process of differentiation, object constancy, and the development of a stable identity. Kernberg (1976, 1980) also adopted and elaborated Jacobson's formulations about the three stages of superego development with sadistic forerunners, the development of the ego ideal, and the final internalization of realistic parental demands.

There is general agreement among all these theorists that the child progresses normally from stages of undifferentiated internalization and identification to stages of differentiation and separation in the development of stable internal object relations (Fairbairn, 1954; Jacobson, 1964; Kernberg, 1976, 1980; Klein, 1975; Mahler, 1968, 1972; Mahler et al., 1975, Winnicott, 1958, 1964, 1968, 1989). Finally, in all of these authors, there was a profound appreciation of the interaction and mutually influencing development of interpersonal relationships and internal objects relations. These object relations theorists all emphasized the importance of relational factors on maturation. Healthy adult independence was interpreted by them as built upon the optimal gratification of dependency needs in infancy and early childhood (which included manageable experiences of frustration). From Fairbairn's (1964) concept of mature dependence to Kernberg's (1976) comments "that the intrapsychic and interpersonal worlds relate to and reinforce each other" (p. 73), there is general agreement among these theorists that the achievement of stable healthy interpersonal relationships is based upon a stable healthy intrapsychic self and object world, which, in turn, is based upon either appropriately gratifying relationships in childhood or the successful outcome of psychotherapy.

Chapter 6: The Atlanta School of Experiential Psychotherapy

Whitaker and Malone (1953/1981) wrote in 1953 that they were not interested in developing "a new school of psychotherapy" (p. viii), but in the introduction to the 1981 edition, they stated that the ideas they presented did form the basis for the development of "a school of psychotherapy" (p. xxvii). Both Whitaker and Malone (1953/1981) and Felder and Weiss (1991) indicated that thousands of psychotherapists identify themselves as experiential psychotherapists. Felder and Weiss (1991) stated that these experiential psychotherapists may include therapists who also describe themselves as Gestalt therapists, family therapists, psychoanalysts, client-centered therapists, or clinicians who use behavior modification. Experiential psychotherapy was described not as a set of techniques but as an approach to the world, a philosophy, that can be applied to any type of psychotherapy.

Experiential psychotherapy was developed from the philosophical basis of existential phenomenology (Felder & Weiss, 1991; Gantt, 1984). Gantt (1984) indicated that the existential-phenomenological philosophy provided the rational for the emphasis on the therapeutic relationship and the experience of the patient in experiential psychotherapy. Gantt (1984) described the focus of experiential psychotherapy as producing intrapsychic change, which is only possible within the context of the experience of an interpersonal relationship. However, it should be noted that Whitaker and Malone (1953/1981) provided very few references to previous publications. They wrote as if the majority of their ideas were based on personal experiences as patients and with patients in psychotherapy and were not based on previous ideas presented by others. They provided a bibliography at the end of their book, but there are no reference citations in the text. Of the theorists who are discussed in the present report, the only one cited in their bibliography is Sullivan (1947).

Whitaker and Malone (1953/1981) described the focus of their theoretical perspective as being on those processes and experiences in psychotherapy that promote intrapsychic integration and maturity. They hypothesized that increases in intrapsychic maturity occur only through emotional experiences that alter intrapsychic structures. The authors did not specifically describe or define the intrapsychic structures involved. Their use of intrapsychic structure seems to mean internal experience, especially the internal experience of well-being and wholeness rather than feeling unhappy and conflicted. They stated that there are also techniques and processes in psychiatric treatment that promote the social adjustment and interpersonal skills of the patient. However, Whitaker and Malone (1953/1981) stated that they were more interested in describing the processes of psychotherapy that catalyze intrapsychic growth than those that promote social adjustment. They viewed social adjustment as learning behaviors that assist in the individual's social functioning, which may or may not contribute to internal feelings of wholeness and well being.

Whitaker and Malone (1953/1981), Gantt (1984), and Felder and Weiss (1991) each described basic and fundamental concepts or assumptions upon which is built the perspective of experiential psychotherapy. These authors also stated strongly that the etiology of psychopathology lies in childhood experience in the individual's family of origin. These statements will be described and reviewed, but it should be noted that none of the authors presented a comprehensive theory of personality development. Finally, the primary emphasis in Whitaker and Malone (1953/1981) and Felder and Weiss (1991) was a thorough description of the processes and goals of psychotherapy.

Basic Concepts

Felder and Weiss (1991) described three guiding principles that define and form the basis of experiential psychotherapy. (a) The unconscious is naturally oriented toward growth and wellness and can be relied upon as an active therapeutic agent. (b) All pathology and all healing of pathology occur within the context of relationships. (c) The primary dynamics of the therapeutic relationship reside within the therapist, whose subjective involvement in the therapeutic relationship is central to the dynamics of the psychotherapy.

The unconscious. Felder and Weiss (1991) suggested that the unconscious represents the most truly authentic aspect of the individual, a core of "ultimate wisdom" (p. 3), which contains all an individual needs to be healthy and to grow. The unconscious was described as the wellspring of health and vitality.

Felder and Weiss (1991) hypothesized that because the unconscious is naturally oriented toward growth and health the patient is the individual best suited to direct the therapeutic experience. The therapist's role was described as a catalyst and a participant in a naturally occurring birthing process of the patient's unactuated potential. The therapeutic process is best directed by the patient's subjective unconscious experience. Regression on the part of the patient is understood by the experiential therapist as an effort to access the unconscious, which contains the early childhood experiences that led to the development of psychopathology. Regression, thus, allows the patient to use the unconscious to create a corrective emotional experience with the therapist, freeing intrapsychic energy that was bound in pathological intrapsychic structures of the patient.

Whitaker and Malone (1953/1981) compared the functioning of the unconscious in psychopathology to the functioning of the immune system and regenerative processes in physical disease. Just as the physician who specializes in surgery or internal medicine depends upon the inherent repair mechanisms of the human organism in the treatment of physical disease, the psychotherapist must depend upon the unconscious to assist in the treatment of psychopathology.

Whitaker and Malone (1953/1981) described the adaptive and curative properties of unconscious processes as follows. They postulated that, in the course of psychotherapy, unconscious material would emerge within the context of the therapeutic relationship. As the patient moves deeply into a symbolic and fantastic relation with the therapist, unconscious attitudes and beliefs emerge, and the therapist has the opportunity to provide the patient with responses that counter the pathological responses and adaptations the patient experienced in childhood. The new experience with the therapist mitigates the pathological effect of past experience on the organization of current living.

Whitaker and Malone (1953/1981) hypothesized that as the therapy progresses the patient becomes more and more deeply related to the therapist on a symbolic, unconscious level without referring to real relationships in the real world outside the therapeutic interview. This process of movement in the therapeutic relationship to deeper and deeper unconscious, symbolic relatedness between therapist and patient will be fully explored in the examination of the processes and goals of psychotherapy.

Whitaker and Malone (1953/1981) also wrote that the unconscious processes that guide psychological functioning work most efficiently when processes remain at or return to an unconscious level of functioning. Thus, they said that the unconscious beliefs, attitudes, and ideas that emerge in the process of psychotherapy are re-repressed during the termination of psychotherapy. Just as the heart beats and pumps blood without conscious thought directing it, human psychological processes "operate with maximal integrative effect when they operate unconsciously" (p. 56). This re-repression is very different from what is commonly meant by the defense mechanism of repression which is based on fear, guilt, or anxiety. Re-repression was defined by the authors as a return to the unconscious sphere of what was intended to remain in the unconscious sphere. Many emotions and beliefs such as those involved in intimate relationships function most efficiently when they remain unconscious. A phenomenon observed by Whitaker and Malone is that many patients seem to forget the deep, intense, symbolic relationship that was shared with the therapist. The therapeutic relationship is used to repair unconscious deficits and conflicts, and then, both the repairs and the memory of repairs are returned to the unconscious.

Gantt (1984) stated that unconscious processes in psychotherapy were used by the patient to recreate with the therapist early negative life experiences with significant others, usually parents. Transferential aspects of the relationship created by unconscious processes then are used by the therapist to provide a corrective emotional experience, by what the author called the process of reparenting, mainly through the therapist's use of unconscious, primary process relating to the patient. It is the responsibility of the therapist to upset and to change the unconscious experiences of relating that the patient attempts to recreate in the relationship with the therapist.

Gantt (1984) stated that the structure of therapy fosters regression in which the patient unconsciously provokes the therapist to repeat the pathological responses of the parents to the infant and child who became the patient. It is the responsibility of the therapist both to empathize with and to understand the genetic roots of the patient's transference and to provide a different experience. In this transferential exchange, the historical meanings of the patient's experience are understood, and simultaneously, other possible experiences become available. As the past is brought into the present, the future can offer possibilities for change. Unconscious self-expression from the patient is invited, valued, and supported by the therapist. The patient is led to redefine and to reorient the experience of the self because the response from the therapist differs from responses experienced in the past. The techniques used by the therapist will be further explained and examined in the section on the processes and goals of psychotherapy.

Both Gantt (1984) and Whitaker and Malone (1953/1981) emphasized the fact that nonverbal and unconscious communications between therapist and patient were potentially more therapeutic than rational, conscious exchanges because nonverbal communications more honestly reflect deep emotional states of being than do rational communications. Whitaker and Malone stated that it is impossible to describe adequately the type of communication that occurs in the deep symbolic stages of psychotherapy. They compared it to the communication between mother and child when the mother is holding the child, singing a lullaby, or telling a story. They also said that it is a type of communication that is similar to the communication between lovers involving "massive proprioceptive and exteroceptive reactions on the part of each person to subliminal stimuli from the other" (p. 91). On a more specific level, Whitaker and Malone (1953/1981) advised the therapist to be attentive to slips of the tongue, tone of voice, gestures, postures, and "the whole gamut of what is ordinarily thought of as symbolic slips of the patient in therapy" (pp. 203-204).

In summary, Whitaker and Malone (1953/1981), Gantt (1984), and Felder and Weiss (1991) emphasized the importance of the unconscious in the process of psychotherapy. Within the process of psychotherapy, the rules of rationality and normal social discourse are suspended in order to allow the emergence of fantasy and spontaneity. The patient recreates experiences from infancy and childhood and is given a new experience of acceptance, empathy, and understanding. The regression allows access to the unconscious, and the corrective experience allows energy to be freed and promotes new intrapsychic growth in the patient.

Relatedness. Felder and Weiss (1991) stated that a major task in experiential psychotherapy is to provide symbolically the relational experiences that have been absent in the patient's previous life experience. The primary interpersonal context in previous life experience is the family of origin. It is within the family of origin that each individual learns whether or not to trust, how to love, how to deal with conflict and aggression, how much intimacy is comfortable, and most other values about how to live. Felder and Weiss (1991) wrote that the patient typically comes to therapy to break the patterns learned in the family of origin and to learn to live a more satisfying life.

Felder and Weiss (1991) stated that the task of the therapist is to provide an interpersonal atmosphere in which the ordinarily suppressed and repressed feelings of the patient are drawn toward expression. The therapist then provides a "corrective" experience in response to the patient’s expressions. The authors commented that it is particularly important for the therapist to enjoy the patient, which facilitates the development in the patient of a sense of self worth.

Similarly, Whitaker and Malone (1953/1981) defined psychotherapy as "a special, isolated relationship, within which deep symbolic needs are gratified" (p. 186). They described the patient entering therapy as a person looking for an escape from an internal struggle. The therapist is imagined by the patient to have the magical omnipotence of the parents of childhood. According to Whitaker and Malone (1953/1981), the process of psychotherapy as a means of intrapsychic change starts when the patient includes the therapist in his or her intrapsychic family.

Whitaker and Malone (1953/1981) also stated that the patient, in turn, must become a part of the intrapsychic dynamics of the therapist. The therapeutic relationship specifically involves a relation between the unconscious dynamics of each of the participants. The therapist, however, has major responsibilities within this relationship. The therapist must have a greater capacity than the patient for symbolic experience and unconscious functioning. The therapist must have resolved the major portion of his or her own infantile transference needs, usually through the experience of having participated in therapy him- or herself. The therapist must communicate with the patient from a mature and healthy perspective. Whitaker and Malone stated specifically that the therapist should not need or demand any gratification from the patient, "possibly not even the gratification of getting him cured" (p. 163).

Whitaker and Malone (1953/1981) stated that given the exposure to a mature and adequate psychotherapist, the patient is able to access and to relive his or her infantile familial experience. The patient relates to the therapist in great depth such that the relationship with the therapist actually replaces the internalized, intrapsychic family of origin. The patient enacts the repetitive and compulsive patterns of infantile experiences in the relation with the therapist. The maturity and therapeutic experience of the therapist allows the therapist to function as a more adequate parent than the parents in the family of origin. Thus, the repetitive reliving of infantile experience results in a different and more satisfactory outcome for the patient as the therapist provides gratification which facilitates growth.

Whitaker and Malone (1953/1981) hypothesized that the therapist, at the beginning of treatment, serves as a symbolic screen for the projective repetition of earlier parental experience. However, the therapist as a real person is more than and different from the constellation of the patient's projections. Due to the wholeness and maturity of the therapist as a person, the patient is able to construct "a more adequate and, finally, a truly satisfying internal parent" (p. 94).

Whitaker and Malone (1953/1981) suggested that the discrepancies between the patient's projection and the maturity of the real person of the therapist may provide the most important dynamic in dissolving the neurotic process. Whitaker and Malone (1953/1981) proposed that, as the repetition compulsion in the transference is altered by the unexpected and mature responses of the therapist, the relationship between the therapist and the patient can enter into an even deeper level of fantastic relating. The patient has the opportunity to regress to a pattern of relating that is unconsciously directed "toward the ultimate and complete satisfaction of those infantile needs which underlie self love (narcissism) and antecede love of others (transference)" (p. 94).

Thus, Whitaker and Malone (1953/1981) proposed that the patient can achieve the gratification of infantile needs through the therapeutic relationship. The authors did not specify which infantile needs can be gratified or how these needs are gratified in the therapeutic relationship. It might be assumed, however, from the context that the infantile need with which the authors are primarily concerned is the need for emotional dependence upon a concerned, interested, caring therapist who symbolizes a loving parent. The gratification of infantile needs provides the basis for the dissolution of the repetitive compulsiveness of neurotic processes. The dissolution of the neurotic processes and the growth of the patient are based upon the patient's experience within the therapeutic relationship regardless of whether or not the patient consciously understands the process. The patient may or may not understand that actual child-parent gratification has occurred at a deep and meaningful level, but nevertheless, "this gratification constitutes the effective force which shatters the neurotic process in the patient" (p. 94).

Gantt (1984) stated that, in experiential psychotherapy, the goal of intrapsychic change is only possible within the context of an interpersonal relationship. The relationship between the therapist and patient provides a current experience that helps to mitigate the pathological effects of past relationships. It is the mutual psychological influence in an intimate dyad that is the basis of all character formation and character change.

In summary, in experiential psychotherapy, it is said that psychopathology develops within the context of interpersonal relationships and can only be treated within the context of a deeply meaningful interpersonal relationship. The patient brings to the therapeutic relationship the unconscious expectation that previous pathological interactions will be repeated. The task of the therapist is to interrupt the repetition of pathological patterns of interacting and to provide the patient with the gratification of unmet infantile needs. These new experiences within an intense, intimate relationship serve to disrupt neurotic patterns of behavior and to impel the patient to move in the direction of growth and maturity (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/198).

The psychotherapist's use of the self. As has been discussed previously, Whitaker and Malone (1953/1981) stated that the successful outcome of psychotherapy is largely dependent upon "the willingness of the mature therapist to respond as a whole person on a subjective level to the patient" (p. 51). Whitaker and Malone assumed that, regardless of the theoretical orientation or techniques used by the therapist, success with any particular patient is based upon what they call the "therapeutic capacity" (p. 61) of the therapist. The authors did not define "therapeutic capacity.” They stated that the therapist develops maturity and therapeutic capacity through participating in personal psychotherapy but also through training, supervision, consultation with colleagues, and continuing participation in the treatment of patients.

Whitaker and Malone (1953/1981) assumed that, as the therapist involves himself or herself on an unconscious level in the relation with the patient, unresolved emotional issues of the therapist may be aroused. Thus, continuing participation in the treatment of patients allows the therapist to recognize and to resolve his or her own infantile fantasies.

Whitaker and Malone (1953/1981) made it a premise of their theory that any impasse or pathology in the therapeutic process is primarily caused by pathology in the intrapsychic functioning of the therapist. Whitaker and Malone assumed that impasses in psychotherapy usually occurred when "patient-vectors" of the therapist interfere with the therapist's ability to respond therapeutically to the "growth-needs of the patient" (p. 162). Patient-vectors were defined by Whitaker and Malone (1953/1981) as demands that immature emotional needs be met with an emotional response similar to the response of an adequate parent to a needy child. Therapist-vectors, on the other hand, were defined as mature emotional responses to the needs of a patient that are similar to the responses of a good parent to an immature child.

Whitaker and Malone (1953/1981) suggested that an impasse in the process of psychotherapy is rooted in the therapist's denial of his or her own patient vectors in the therapeutic relationship. They stated that many such impasses can be resolved if the therapist admits responsibility for the impasse to the patient. In other cases, the resolution of impasse can only be achieved through supervision, consultation, or the therapist's return to personal psychotherapy.

Felder and Weiss (1991) stated that the subjective involvement of the therapist in the therapeutic relationship is the "primary determinant of the dynamics of the psychotherapy" (p. 25). They stated that the unconscious of the psychotherapist should be regarded as naturally oriented toward growth and wellness and, thus, is the source of ultimate wisdom, which can be used as a powerful therapeutic agent.

Felder and Weiss (1991) suggested four ways in which the therapist can make use of the self in the process of psychotherapy. (a) The therapist may tell the patient about those feelings that are provoked by the patient. Felder and Weiss particularly emphasized the value of communicating feelings of anger and frustration, which may be an appropriate and authentic response to the patient's behavior. (b) The therapist may report feelings aroused by the patient's experience, which may awaken the patient's similar and previously denied feelings. (c) The therapist may report any associations that occur internally to stories, incidents, anecdotes, songs, or parables. (d) The therapist may report dreams he or she has had about the patient, which may, in turn, inspire the patient to remember and to report dreams to the therapist.

Felder and Weiss (1991) warned that a frequently made and dangerous misconception about experiential psychotherapy is that the therapist operates entirely out of the unconscious and reports this unconscious experience to the patient without censorship. The therapist communicates feelings or internal experience only if the therapist has decided that the patient could derive therapeutic benefit from such communication of experience. However, the authors indicated that the best guide to the decision about possible therapeutic benefit to the patient is the unconscious of the therapist.

Felder and Weiss (1991) described a therapeutic impasse as occurring when therapeutic movement is halted. Like Whitaker and Malone (1953/1981), Felder and Weiss postulated that an impasse in the progress of psychotherapy most often is caused by the pathological dynamics of the therapist. They listed common causes of an impasse. (a) The therapist becomes overly responsible for resolving the problems of the patient. The message conveyed to the patient is that the patient is incapable of solving his or her problems. (b) The therapist takes grandiose pleasure in the positive transference and his or her heroic role in the patient's life. (c) The therapist sets goals for the patient instead of allowing the patient to establish goals. (d) The therapist passively hides or suppresses aggressive feelings toward the patient in order to avoid conflict. (e) The therapist blindly follows rules learned in training without questioning the applicability of the rules to a particular therapeutic relationship. (f) The therapist assumes that negative feelings expressed by the patient are simply manifestations of transference without examining the possibility that the patient may be responding appropriately to inadvertent or unconscious behavior of the therapist.

Felder and Weiss (1991) advised that several steps may be taken to resolve a therapeutic impasse. Sometimes a discussion with the patient is all that is necessary to resolve an impasse. Sometimes introspection on the part of the therapist may succeed in identifying and correcting the internal dynamics of the therapist. If neither of these possibilities succeeds, the impasse can be resolved in either formal or informal consultation or supervision. The final step beyond the therapist’s seeking supervision is to invite a consultant to intervene in the actual therapy interviews. The message given to the patient if this step is taken is that there is a problem in the relationship with which both parties need help to resolve.

Gantt (1984) proposed that, early in the therapeutic process, the therapeutic relationship is based upon the transference of the patient of early childhood experience into the relationship with the therapist. At this point in the relationship, the primary responsibility of the therapist is to recognize the repetition of earlier experience in the therapeutic relationship and to break the patterns of pathological functioning by not responding as expected. The therapist both joins and counters the patient's experience. When pathological behavior emerges in therapy sessions the task of the therapist is to be available emotionally, to be honestly a self, to value experience, and to behave ethically. These characteristics of the therapist encourage the patient to relinquish pathological functioning and to communicate with greater maturity and authenticity.

Gantt (1984) stated that, from the experiential perspective, all patients have disorders of the self. However, the self is an entity to be discovered rather than repaired in therapy. The therapist's honest use of self stimulates the patient's discovery of a sense of self. The therapist's honest and authentic use of self also provides the patient with a role model for the development and expression of a sense of self. As the patient discovers a sense of self, the self of the patient relates emotionally to the self of the therapist. The therapeutic relationship based upon transference is gradually replaced by existential relating between one authentic self and another.

In summary, the successful outcome of experiential psychotherapy was described by Whitaker and Malone (1953/1981), Felder and Weiss (1991), and Gantt (1984) as being strongly influenced by the therapist's mature and careful expression of his or her personal experiences in the relationship with the patient. This communication of personal experience in the therapeutic relationship is the way in which therapists may express their own sense of self. The therapist's use of self expression may then encourage the patient to discover and to express the self of the patient.

The Nature and Development of Psychopathology

Whitaker and Malone (1953/1981) specifically stated that personality "evolves out of the matrix of mother-child symbiosis" (p. 19). Psychopathology was described as a disruption of the normal growth of the individual. This disruption of normal growth most frequently occurs in the family of origin. The primary symptoms of psychopathology are described by the authors as compulsive repetition of relationship patterns learned in infantile experiences with recurring unsatisfactory outcomes.

Whitaker and Malone (1953/1981) stated that healthy ego development and personality structure are based upon the organization of emotions out of disorganized affect. Organization of affect is directly dependent upon the interpersonal experiences of the child, particularly in the parent-child relationship. If the child or infant does not receive parental or alternative interpersonal support for the organization of affect, the child is left with the disorganized and disorganizing experience of anxiety. The authors were not specific either about how the parents helped to organize affect or about how the organization of affect contributes to the development of intrapsychic structure.

When interpersonal relationships that support the organization of affect are not available for the child, the child is unable to organize affect on his or her own and is also unable to develop mature intrapsychic structures of ego and personality. The result is an inadequate, infantile adult who is overly dependent on interpersonal relationships. When adult interpersonal relationships collapse, this adult suffers from intense anxiety and possibly disintegration of the ego. The process of psychotherapy is essentially an effort to provide a structure of mature, autonomous personality through the intrapsychic organization of affect in the patient "as a result of the adequacy of the interpersonal organization of affect between patient and therapist" (Whitaker & Malone, 1953/1981, p. 121).

Whitaker and Malone (1953/1981) also proposed that aggression initially arises in infants as an organization of anxiety directed toward the goal satisfaction of physical and psychological needs. The core of healthy aggression is the effort to obtain the satisfaction of needs and the alleviation of anxiety from another participant in an interpersonal relationship. Negative aggression, which seeks the destruction of the object against which it is directed, arises because the individual is threatened with the continued frustration of needs. Positive aggression, on the other hand, can lead to the satisfaction of needs and can become the basis for more sophisticated emotions including "the varied expressions of love, e.g., parental feelings, sexual response, respect, tenderness, sympathy, and other more complex sentiments" (p. 133). When the individual either in childhood or as a psychotherapy patient has the interpersonal experience of organizing positive aggression and love out of the same interpersonal matrix, the opportunity exists to resolve the basic human problem of ambivalence, to understand that one may appropriately feel aggression and love within the same interpersonal matrix.

Felder and Weiss (1991) also described psychopathology as originating in early experience. As was previously paraphrased, they stated that,

The primary interpersonal context of our lives is the family of origin. The family is where we learn to love (or whether to allow ourselves to love), how to deal with or avoid conflict, how close or distant we will feel in relationships, and most of our other values about how to live. (p. 16)

Felder and Weiss (1991) described the family dynamics which produce psychopathology as families in which the parents have unmet childhood needs. These parents marry in order to get these needs satisfied, and when marriage fails to provide gratification, the needs are projected onto the children. The children are in a sense recruited to become the providers and gratifiers of their parents' unmet needs. In turn, these children experience deprivation. Felder and Weiss (1991) stated that an important basis of a sense of self worth and self-esteem is simply based upon the experience of having been enjoyed by parents.

Felder and Weiss (1991) defined psychopathology as a restricted capacity for experience. Whitaker and Malone (1953/1981) similarly defined psychopathology as non-experience, as the inability to live fully and authentically in the present and to experience it. Patients who have depression limit and depreciate experience. Those who have anxiety avoid or override experience. Those who have hysteria and obsessions ignore and deny experience. Patients who have compulsions replace experience, and those who are psychotic manufacture experience that is different from the common experience of others. The common underlying dynamic is either a learned fear of experience or perhaps an unlearned, biologically determined inability to experience. There was no explanation of what developmental and genetic factors determine the nature and style of the type of psychopathology developed by any given individual.

Whitaker and Malone (1953/1981) hypothesized that developmental and familial (e.g., learned) factors determine the specific nature of the style in which experience is rejected or denied. In the 1981 introduction to The Roots of Psychotherapy, the authors proposed that a major weakness of the presentation of the theories of experiential psychotherapy is the failure to explore and to explain the development and nature of the different categories of psychopathology.

In summary, experiential psychotherapy theorists implicitly and explicitly stated that the basis of psychopathology lies in the childhood experiences of the patient and in what is learned in the family of origin. There was some discussion, particularly by Whitaker and Malone (1953/1981), of the relation between the organization or disorganized affect (anxiety) and the development of secure, mature, intrapsychic structures of ego and personality. It was also made clear that, in the process of psychotherapy, success in therapy is predicated upon the regression of the patient to a childlike state in which the therapist symbolically provides the organizing experiences and gratifications not provided in the family of origin. However, in the theories of experiential psychotherapy, there is not a thorough examination of the developmental processes involved in the infantile and childhood experiences that result in the development of either healthy or pathological functioning.

The Processes and Goals of Psychotherapy

Whitaker and Malone (1953/1981) described in detail the stages of psychotherapy. In general, the stages were divided into three phases: (a) the pre-symbolic, (b) the symbolic, and (c) the post-symbolic or ending phase. The task in the pre-symbolic phase is to assist the patient into movement into the symbolic phase. The symbolic phase is what the authors described as the heart of the therapeutic process in which deep unconscious or transference needs of the patient arise and are resolved. The post-symbolic phase involves the transition of the patient out of therapy and back into the culture. Whitaker and Malone (1953/1981) hypothesized that progress continues after termination of the therapeutic interviews as the patient completes the task of integrating the gains of therapy into his or her real life functioning (Whitaker & Malone, 1953/1981).

The pre-symbolic phase has two substages, the anamnestic stage and the casting stage. The symbolic phase is composed of three substages, the competitive, the regressive, and the core stages. The post-symbolic phase has two substages, testing and withdrawal. Termination is the concluding process of all the stages preceding it. Each of these will be described.

The anamnestic stage. In the anamnestic stage, the patient enters the therapeutic relationship usually with the goal of receiving help with specific problems rather than with the goal of achieving a pervasive modification of his or her personality. During these initial interviews, the therapist, in turn, talks with the patient on an adult level and explains the realities of the therapeutic relationship, what might be expected in therapy, and what therapy involves (Whitaker & Malone, 1953/1981).

The patient in the anamnestic stages interprets the therapist as "the Doctor" (Whitaker & Malone, 1953/1981, p. 106). The therapist at this time interprets the patient as an adult or a pseudo-adult living behind a neurotic facade. The therapist talks to the patient as an adult but clearly conveys the expectation that the patient will bring more and more of the immature self to subsequent interviews.

The casting stage. In the casting stage, the patient gradually begins to perceive that, in addition to help with the presenting the problem, there is also the possibility that the therapeutic relationship may offer gratifications not obtained in earlier relationships. The patient may begin to see the therapist symbolically as an older sibling and the self as a younger sibling. Consequently, the patient may become competitive at this stage, boasting about his or her real and social accomplishments and questioning the therapist's professional competence (Whitaker & Malone, 1953/1981).

In this stage, the therapist continues to interpret himself or herself as a professional and as a therapist but begins to respond to the patient as an adolescent. The therapist begins to identify the possibility of gratifying the symbolic needs of the patient and "the possibility of a deeper participation which will satisfy his own needs" (Whitaker & Malone, 1953/1981, p. 107). The symbolic stages of the therapy start to emerge.

The competitive stage. The competitive stage is the beginning of the symbolic phase of the therapy. In this stage, the therapist begins to develop an unconscious parental responsibility for the patient. The patient becomes a representation of the therapist's own "child-self" (Whitaker & Malone, 1953/1981, p. 107), while both therapist and patient begin to form an internal representation of the therapist as a "parent-person" (Whitaker & Malone, 1953/1981, p. 107).

As the patient begins to interpret the therapist as a parent-person, the patient may begin to ask for advice about solving real-life problems outside the therapy, ask for information about diagnosis and psychodynamics, and ask the therapist to evaluate and to assess the patient's behavior. The therapist must avoid responding to the content of the patient's demands for advice, direction, and evaluation. If the therapist responds to the patient at this stage with advice and guidance, the patient may feel encouraged to depend on the actual, real person of the therapist for help with current day to day problems. This type of reliance on the therapist for help with every day life could impede the development of a deep symbolic dependency on the therapist that facilitates the resolution of problems based on unmet childhood needs. The therapist must also accept the transferential needs of the patient for a warm and close emotional relation (Whitaker & Malone, 1953/1981).

The refusal of the therapist to advise, to reassure, or to evaluate the patient serves the purpose of stimulating the patient to use the relationship at a more deeply symbolic level. The therapist denies all the social aspects of the therapeutic relationship and protects the isolation of the relationship from the patient's outside life. During this time, the patient's anxiety usually increases, but Whitaker and Malone (1953/1981) stated that continuing progress depends upon increasing anxiety that serves to propel the patient into deeper stages of regression.

The regressive stage. During the regressive stage, the therapeutic process is approaching the core stage. The patient feels increasingly secure in the relation with the therapist, who now is conceived of as the "Primordial-Parent" (Whitaker & Malone, 1953/1981, p. 108), while the patient's internal representation of the self is that of a "child" (p. 108). The therapist, in turn, has an internal sense of self as the primordial parent and interprets the patient as the "child-self" (p. 108) of the therapist. The patient's communications are often overtly childlike, and the therapist's response is often nonverbal but pervasively accepting.

The core stage. In the core stage, both patient and therapist interpret themselves and each other congruently as "the therapist as Parent and the patient as Child-self" (Whitaker & Malone, 1953/1981, p. 108). This is the stage of greatest therapeutic depth. The patient becomes deeply immersed in the fantastic, symbolic relation with the therapist. The therapist is responsible for maintaining a balance between the symbolic and the real. The therapist must enter symbolically into the fantastic transferential relation presented by the patient. However, the therapist is also responsible for maintaining a secure hold on reality from which both patient and therapist depart into the symbolic experience and to which both must return.

It is during the core stage of psychotherapy that the patient relives his or her early infantile experiences in which needs were not gratified in the family of origin. What Whitaker and Malone (1953/1981) described (without offering a definition or explanation) as the therapist's “maturity” and “therapeutic potential” make it possible for the therapist to break the patterns of the neurotic process and to provide the patient with the material to construct a satisfying internal parent. The therapeutic relationship allows the patient to achieve the gratification of deep infantile needs (Whitaker & Malone, 1953/1981).

As was mentioned previously, Whitaker and Malone (1953/1981) said that the patient's anxiety propelled movement into the core stage of the psychotherapy. This anxiety was described as the positive anxiety that occurs when the patient perceives that he or she is in a therapeutic relationship that offers the possibility of reorganizing the personality. The therapeutic relationship offers a new opportunity to organize the disorganizing affect of anxiety. As the patient uses the interpersonal relationship to organize affect intrapsychically, the patient also achieves a more efficient and mature organization of ego and personality. As the patient achieves greater maturity and security, there is movement out of the core stage of psychotherapy and into the beginning of the postsymbolic phase of therapy.

The testing stage. The testing stage of therapy is similar to the process of separation between the mature adolescent and the parent. The therapist continues to function symbolically as a parent-person, but the patient starts to function as a potential adult with increasing maturity. The patient is less and less invested in the therapist as a symbolic parent, and there is growing development of the patient's self concept as a mature adult (Whitaker & Malone, 1953/1981).

The withdrawal stage. In the withdrawal stage, the patient is increasingly aware of himself or herself as a mature adult and a unique individual. The patient rejects the therapist in the role of symbolic parent and accepts the therapist as a fellow adult and a professional therapist. Both Whitaker and Malone (1953/1981) and Felder and Weiss (1991) agreed that the termination of the therapeutic relationship is a significant and decisive factor in determining the success of the psychotherapy.

Termination. In order for the patient to achieve true autonomy and maturity as an adult, there must be a clear separation from, and ending, with the therapist. The growth achieved in psychotherapy can be assimilated best and claimed by the patient as his or her own after he or she has successfully disengaged from the therapist (Felder & Weiss, 1991; Whitaker & Malone, 1953/1981).

Felder and Weiss (1991) stated that the tasks of termination include (a) the expression of any feelings previously not reported, particularly the expression of anger, (b) a review of the progress made in the treatment, which contributes to the cognitive integration of the patient and facilitates the patient's return to everyday reality, and (c) a mutual communication between therapist and patient of the experience, including what has and has not been accomplished (Felder & Weiss, 1991). Whitaker and Malone (1953/1981) and Felder and Weiss (1991) pointed out that, during the termination phase of therapy, the patient may request a real, adult relationship, perhaps even a continuing social relationship after the termination. It is strongly recommended that such a request be denied so as to precipitate the patient into a true ending with the therapist, thus facilitating the patient's maturation, which should include increasing the gratification of emotional ideals in the day to day life of social reality.

Felder and Weiss (1991) classified termination into seven distinct types.

(1) The ideal termination is one in which the patient is unknown to the therapist outside the therapeutic relationship. The patient is not seen or heard from unless he or she returns for additional therapy.

(2) The satisfactory termination is one in which the patient is known by the therapist in a setting outside the psychotherapy (i.e., an acquaintance or a fellow professional). The task is to ensure an adequate therapist-patient termination without significant alteration in the pre-existing relationship.

(3) The administrative termination is one in which a therapeutic impasse has occurred, and all efforts at resolution have failed. The authors recommended that the therapist take full responsibility for the failure in treatment and refrain from attempting to make a referral.

(4) The forced termination is one which is initiated by the therapist, usually because of normal life changes (i.e., a new job, relocation). The authors described this type of termination as a violation of an implicit contract that psychotherapy will be continued to a natural conclusion. It is recommended that the patient be given as much notice as possible and encouraged to express all feelings aroused by the termination. It is also recommended that the patient be allowed to reestablish the therapeutic relationship if it is possible for the therapist to become available again.

(5) The avoided termination is one in which the patient makes progress and then disappears by not making any further appointments. The authors suggested that, in these circumstances, the therapist has an obligation to request that the patient return for the process of termination in order to conclude the psychotherapy.

(7) The schizophrenic negative termination is one in which the patient, usually a schizophrenic patient, leaves with a negative attitude, often with hostility and anger. These patients may leave therapy in a rage despite the fact that they have made therapeutic progress and their lives have significantly improved.

Whitaker and Malone (1953/1981) and Felder and Weiss (1991) observed that therapeutic progress continues after termination of the psychotherapy relationship. The patient has the opportunity to assimilate or to reject what was offered by the therapist. What was gained in the process of psychotherapy becomes integrated and adaptively re-repressed into the unconscious.

Techniques in psychotherapy. Whitaker and Malone (1953/1981) defined techniques in psychotherapy as the deliberate use of an interpersonal communication to catalyze spontaneous, deep affect in the therapeutic relationship. In the beginning stages of psychotherapy, the therapist may use techniques to foster movement to the symbolic phases of therapy. One important technique is the deliberate isolation of the therapeutic relationship from any interruptions or intrusions. The therapist does not answer the phone or discuss his or her personal life. The focus is on the feelings and experiences of the patient in the interview.

Whitaker and Malone (1953/1981) also suggested that it was particularly important for the therapist to avoid involvement in the patient's decisions about life in the day to day world outside of therapy. If the therapist does make decisions for the patient about employment, relationships, or any other aspects of the management of ordinary life, the therapist may encourage an excessive dependence for the magical gratification of real needs, thereby precluding the therapeutic task of satisfying deeper symbolic needs.

Whitaker and Malone (1953/1981) also recommended that, in the beginning stages of therapy, the reinterpretation of the patient's behavior in terms of the developing therapeutic relationship helps facilitate a focus on the relationship. Similarly, the verbal interpretation of nonverbal behavior facilitates a focus on unconscious communication.

One technique strongly recommended by Whitaker and Malone (1953/1981) is the use of silence throughout the process of psychotherapy. Silence tends to stimulate the kind of positive anxiety that propels the patient onward. Silence also stimulates the patient's fantasy associations to his or her own verbalizations.

Finally, Whitaker and Malone (1953/1981) recommended the use of physical contact with the patient, particularly in the core stages of the therapy. They stated that they have found that bottle feeding, physical rocking, holding during a crying spell, and spanking stimulate in patient and therapist a reproduction of the mother-child relationship. However, the authors' discussion of physical contact was somewhat confusing. At one point, they stated that physical contact should be considered to be,

auxiliary techniques which compensate for the therapist's inadequacy by inducing the deep affect which the patient seeks for the satisfaction of his infantile and dependent needs. Theoretically, the authors are convinced that these regressive and core satisfactions can be provided without any props or auxiliary techniques since the process is essentially an intra-psychic one. Technical implementation may only reflect certain immaturities in the intra-psychic functioning of the therapist. (p. 211)

On the other hand, Whitaker and Malone (1953/1981) included physical contact as one of the techniques that is useful throughout the process of psychotherapy. They stated that physical contact such as rocking or holding "brings to the therapeutic relationship certain proprioceptive and sensory modalities which make a significant contribution to the therapeutic process itself" (p. 225). They commented that, in using such physical contact, the therapist is assuming an active maternal role in the patient's regression to infantile experience.

It should be noted in regard to these conflicting statements about the therapeutic benefit of physical contact that in later publications about experiential psychotherapy, the symbolic nature of the therapeutic relationship was emphasized. Felder (1967) stated that the patient tries to make the therapist into a "second-chance parent" (p. 101) and expresses in the treatment many feelings from the relationships in the family of origin. He warned that the therapist could only be a better “parent” on a symbolic level.

Similarly, in the 1981 introduction, Whitaker and Malone (1953/1981) specifically stated that it is probably impossible for the therapist to serve as a replacement for the parents in the family of origin. Gantt (1984) stated that the more recent assumption in experiential psychotherapy is that the therapist can offer reparation but not replacement. In the core stage of the therapy, the therapist becomes the significant other with whom the patient is passionately and nonrationally involved. The patient then can define and orient the self in terms of the relationship with the therapist.

The goals of psychotherapy. Whitaker and Malone (1953/1981) stated that the emphasis in experiential psychotherapy is on synthesis and integration rather than analysis and insight. The understanding of the patient or the therapist about the genetic basis of the problems brought to the therapy is not as important as the development of the patient's capacity to function as a person integrated with the self and with the surrounding culture. It is neither relevant nor useful for a patient to understand that an inability to express aggression is based upon infantile fears and guilt if the patient is still unable to express aggression. On the other hand, the experience of being aggressive toward a parental figure (e.g., the therapist) who does not punish or reject the patient may lead to an increased capacity to express aggression even if the significance of the experience is not cognitively understood. Thus, experience is defined by the authors as a more powerful agent of change than is insight or analysis.

One of the most important goals of psychotherapy is the overcoming of emotional inertia. The patient leaves therapy with the understanding that the achievement of maturity is a lifelong struggle. The experience of relating to another individual on a primitive level teaches the patient to expand the limits of personality and to be more accepting of emotional needs whether those needs are infantile, adolescent, or adult. The patient gains the capacity to demand, to obtain, and to participate in new experience (Whitaker & Malone, 1953/1981).

Whitaker and Malone (1953/1981) summarized the successful outcome of psychotherapy as follows. (a) The patient learns to accept fantasies and unconscious experiences because fantasies and unconscious experiences have been expressed within a relationship with the therapist, who implicitly participates in them and thereby accepts them. (b) The patient develops a greater continuity between conscious and unconscious functioning with the awareness that one can be receptive to fantasy while maintaining the capacity to function in reality. (c) The patient develops the capacity to seek gratification of fundamental biological and emotional needs including the ability to accept and to use reality to obtain gratification and growth in life experiences. As the patient leaves the symbolic relationship with the therapist, the patient invests his or her needs and emotions in relationships outside the therapist's office and uses these relationships to continue the struggle for growth and maturity.

Felder and Weiss (1991) proposed that psychotherapy comes to its natural termination point when the patient is not only growing but is also aware that growth will continue after the therapy is finished. The role of the experiential psychotherapist was described as an assistant to growth. The therapist helps the patient ignore cultural customs that are limiting to growth and obtain from the culture that which is healthy and fulfilling. For the most part, Felder and Weiss avoided discussing general goals for psychotherapy. They emphasized that it is of major importance for the patient's goals to be the only important goals in the therapy. If the therapist imposes goals on the patient, an impasse in psychotherapy could result.

The Heuristic or Guide to Investigation

Whitaker and Malone (1953/1981) stated explicitly that their purpose was to identify and to arrive at some understanding of the essential process common to all forms of psychotherapy. However, the basis of their theories about the process of psychotherapy seems to have been their own personal and subjective experience as therapists and/or patients. There were virtually no attempts to compare or to integrate their ideas with the ideas of other theorists, and there were no reference citations in the text (although there was a list of references). There also were no clinical examples. Malone (1981) stated that the absence of clinical material was a "serious deficiency" (p. xxviii). He commented that concrete clinical material might have lent meaning and weight to some vague and nebulous concepts.

Felder and Weiss (1991) similarly stated that the concepts of experiential psychotherapy represent truths present in all psychotherapies and all human experience. The source of information was described by them as phenomenological, based upon the personal and subjective experience of the therapist. Felder and Weiss used clinical examples frequently to clarify the concepts presented by them, including a few verbatim transcripts of segments of clinical interviews. They also illustrated the theoretical material with descriptions of clinical experiences of therapists in clinical supervision. There were many references cited in the text, but the ideas of other theorists were usually mentioned briefly without exploration or clarification of similarities and differences.

Thus, it must be assumed that the theoretical model of experiential psychotherapy has been based upon the personal and subjective experience that these theorists have had in their participation in psychotherapy. Both Whitaker and Malone (1953/1981) and Felder and Weiss (1991) indicated that the concepts upon which the model is based were discussed thoroughly and were developed with professional colleagues before reaching their final form.

Ontology

Essential Human Nature

Whitaker and Malone (1953/1981) and Felder and Weiss (1991) proposed that the human unconscious is naturally oriented toward growth and wellness. The unconscious is a core of internal wisdom that contains all the individual needs to grow and to be healthy. Individuals are normally predisposed to healthy and satisfying relationships with others. Psychopathology was defined as a disruption of normal growth caused by unsatisfactory relationships in early life.

Thus, the experiential psychotherapists stated that the human individual is innately predisposed to a natural growth process. Given support and gratification in interpersonal relationships, the individual will develop into a healthy, well-functioning adult who participates in and contributes to interpersonal relationships. The unconscious is thought to be a source of inner wisdom and is so valued as such that, in experiential psychotherapy, the unconscious of the therapist is regarded as vital in choosing how to respond to the patient. It is the unconscious of the therapist that is the best guide to decisions about when to participate silently, to intervene actively, to share experiences, to share dreams, or to express feelings aroused by the patient. The unconscious of the patient provides the direction of the therapy. The unconscious of the therapist determines the most appropriate responses to the patient in order to facilitate continuing movement and growth.

The Nature and Development of Psychopathology

Psychopathology is thought to originate in early childhood experience within the family of origin. Childhood experiences that lead to the development of psychopathology were primarily described as experiences of deprivation. The developing child requires the support of parents to organize emotions and to develop intrapsychic structures of ego and personality. If the parents do not provide gratification of normal childhood needs, the normal growth of the individual is disrupted. Disruption of normal growth is the basis for psychopathology (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

According to the experiential theorists, psychopathology is thought to be a result of deficiencies in the interpersonal experiences of the child. However, the theoretical formulations of experiential psychotherapy do not provide a model of the conditions required for normal development. There was no description of normal stages of development, the types of support and gratification required for normal growth, ways in which intrapsychic structure is developed in interpersonal experience either in childhood or in treatment, or the origins of different types of psychopathology (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

The Role of the Therapist

The therapist provides the patient with a secure environment isolated from the demands and expectations of normal day-to-day life. Within this environment, the therapist responds to the patient with silence, verbal interpretation of nonverbal behavior, interpretation of the meaning of the patient's behavior in terms of the therapeutic relationship, and a firm refusal to advise or to guide the patient about how to direct everyday life outside the therapy. The responses of the therapist encourage the emergence of fantasy and the development of regression in which the patient recreates experiences from infancy and childhood. The therapist responds to fantasy and regression with acceptance, empathy, and understanding (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

One of the most important tasks of the therapist is the gratification of deep infantile needs when the patient is in a profound regression to early experience (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981). Felder and Weiss (1991) stated that the therapist becomes a second-chance parent by focusing on the identification and gratification of the patient's needs, by enjoying the patient, and by giving the honest, authentic, emotional responses of therapist to the patient when therapeutically appropriate.

Finally, in regard to the role of the therapist in experiential psychotherapy, an impasse in therapy is regarded as always originating in pathological dynamics of the therapist. The rationale for this assumption is never clearly explained by either Whitaker and Malone (1953/1981) or Felder and Weiss (1991), although both sets of authors described numerous situations in which the therapist's pathology may impede progress in therapy. Perhaps their assumption is that the inherent, innate tendency toward healthy growth on the part of the patient could only be disrupted by inappropriate responses of the therapist.

It also should be noted that the techniques and behaviors of the therapist that display enjoyment, warmth, authenticity, and maturity were not clearly described or explained. Similarly, how the patient uses the responses of the therapist to change intrapsychic structure or organize affect was never discussed or explained . Finally, no guidelines were offered as to how the therapist might assess his or her maturity, authenticity, wholeness as a person, or therapeutic capacity.

The Mechanism of Change

Whitaker and Malone (1953/1981), Gantt (1984), and Felder and Weiss (1991) agreed that the unconscious of both patient and therapist guide the process of psychotherapy. They emphasized the importance of the relation between the patient and therapist as providing the interpersonal experience that mitigates and relieves the psychopathology that was assumed to have originated in infantile interpersonal experience. They agreed that the therapist's use of self and expression of experience are vital to success in psychotherapy.

The role played by the unconscious of the therapist is not clearly explained in the experiential theories. However, it may be inferred that the unconscious of the therapist guides the therapist in choosing how to respond therapeutically to the patient. It is the unconscious of the therapist that chooses to be silent and accepting, thus encouraging the patient's continuing expression. It is the unconscious of the therapist that chooses when it is therapeutic to report the therapist's experience, particularly when to report feelings aroused in the therapist within the interview. The unconscious of the therapist is certainly involved when the therapist dreams about the patient, and the unconscious of the therapist must decide whether or not it would be helpful to the patient to report such dreams. It should be noted that the experiential theorists stated that silent acceptance and participation are the choices most often made to facilitate therapeutic progress (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

The relationship between patient and therapist evolves from an isolated, protected relationship to one that is profoundly symbolic and fantastic. As the patient becomes deeply involved in the symbolic relationship, it is the responsibility of the therapist to join with the fantasy of the patient while staying firmly grounded in the world of reality. As has been discussed previously, the patient then has the opportunity to recreate and to relive infantile experiences with more satisfying results in the relationship with the therapist than were achieved in relationships with the family of origin. The intimate dyads of interpersonal relationships in the family of origin provide the basis for pathological character formation. The intimate dyad of the therapeutic relationship provides the opportunity for character change. The interpersonal context in the therapeutic relationship in which the therapist provides the patient with the gratification of infantile needs enables the patient to relinquish pathological functioning and attain maturity (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

In general, it was recommended that the therapist's expression of his or her mature, authentic responses to the patient plays a vital role in the process of treatment. The patient attempts to recreate within the therapeutic relationship the pathological interactions of the family of origin. The therapist's response as a whole individual is not congruent with the patient's expectations and serves to break the pattern of pathological interactions. Furthermore, the therapist's use of self by the expression of authentic responses to the patient provides a role model and an example of the expression of self (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

The Development of Psychopathology

Psychopathology is thought to originate in early childhood experience within the family of origin. The childhood experiences that lead to the development of psychopathology are primarily experiences of deprivation. The developing child requires the support of parents to organize emotions and to develop intrapsychic structures of ego and personality. When the parents are unavailable to provide support for the child's interpersonal development or to provide the gratification of normal childhood needs, there is disruption of normal growth of the individual. The disruption of normal growth is the basis for the development of psychopathology (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

The symptoms of psychopathology are compulsive repetitions of the relationship patterns experienced in childhood in which needs were not gratified and support was not provided. Due to the nature of psychopathology, the individual repeats the frustrating patterns of interpersonal interaction in such a way that the outcome continues to be unsatisfactory. The therapeutic relationship provides the opportunity for the therapist to interrupt the process and to provide the patient with interpersonal support and gratification. When the individual is provided with the support and gratification needed, the natural growth process resumes, and the individual achieves normal healthy development (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

From the experiential viewpoint, psychopathology is thought to be created by deficiencies in the interpersonal experiences of the individual. However, these theories do not include a specification of the conditions that are required to provide normal development. No theories were developed to describe normal stages of development or the types of support and gratification required to provide the foundation for normal development (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

The Goals of Psychotherapy

Within the symbolic relationship with the therapist, the patient recreates and relives frustrating interpersonal experiences of childhood. The therapist breaks the pattern of the repetition of pathological patterns of interacting and provides the patient with the gratification of previously unmet infantile needs. As infantile needs are gratified, the anxiety of the patient diminishes, and intrapsychic energy is available to the patient. The patient uses the energy made available and the interpersonal experience with the therapist to achieve a more efficient and mature intrapsychic organization of affect, ego, and personality (Whitaker & Malone, 1953/1981).

In conclusion, the theory of experiential psychotherapy defines psychopathology as an interruption of natural growth toward maturity that occurs because of deficits in the interpersonal experience of the individual in childhood. These deficits in interpersonal experience lead the individual to develop pathological, repetitious patterns of interpersonal interaction in which the frustrating interpersonal experiences of childhood are reexperienced. These patterns of interpersonal interaction are recreated by the individual in the therapeutic relation with the unconscious expectation of continuing frustration and the hope of gratification of unmet infantile needs. The therapist interrupts the repetitious patterns of interacting by responding to the patient with emotional warmth, maturity, and personal authenticity. The responses of the therapist provide the patient with the gratification of unmet interpersonal needs and the process of growth is resumed. As the patient achieves intrapsychic security and growth toward maturity, he or she ends the therapeutic relationship with acceptance and encouragement from the therapist. Growth toward maturity continues throughout life (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

A Critique of the Atlanta School of Experiential Psychotherapy

Wolberg (1988/1995) summarized the ideas of Whitaker and Malone (1953/1980) and offered several criticisms. First, he commented that the greater activity on the part of the therapist than in a more traditional neutral psychoanalytic approach might tend to foster reparative and goal limited growth in treatment rather than a true reconstruction of character and intrapsychic structure. The passive approach used in the traditional psychoanalytic approaches is thought to promote the patient's development of his or her own resources by using them to appreciate and to understand transferential distortions and by using these insights to develop different ways of relating to others. He also criticized the minimization of insight as a constructive therapeutic tool.

Finally, Wolberg (1988/1995) stated that the strongest criticism of experiential psychotherapy was the advocacy of interactions between therapist and patient that are characterized by spontaneous physical and symbolic communication, such as physical contact, long periods of silence, and the communication of joint fantasies. He stated that he understood that the emerging irrational experience was presumed to uncover the core of problems but warned that such interactions are potentially dangerous.

The dangers pointed out by Wolberg (1988/1995) in the kind of symbolic and fantastic therapeutic relationship advocated by Whitaker and Malone (1953/1980) were as follows. (a) Although the patient might enjoy and derive gratification and relief from the experience of the therapist playing the role of a symbolic, primordial parent, the depth and permanence of any change in the patient is questionable. (b) A therapeutic milieu that encourages the symbolic behavior of the therapist based on the guidance of his or her own unconscious can create license for the therapist to use the therapeutic relationship for personal, emotional catharsis and acting out.

Finally, Wolberg (1988/1995) noted that some patients may tolerate a therapist's unconventional behavior because of feelings of protectiveness or sympathy. Such patients might take on the role of "helper" to the therapist and attempt to fulfill the therapist’s "neurotic needs for control and dominant status" (p. 318). Wolberg noted that these criticisms about the dangers of inappropriate behavior on the part of the therapist were not directed to all therapists who practice the type of experiential style recommended by Whitaker and Malone (1953/1980) but only to those whose personal problems lead them to interpret the recommendations of spontaneous and authentic behaviors as permission for the undisciplined expression of feelings and impulses.

Chapter 7: Self Psychology

Self psychology has been developed as a major modification of psychoanalytic theory and treatment with an emphasis on the development of the self as a supraordinate entity separate from and guiding the tripartite structure of ego, superego, and id. The focus of treatment in self psychology is on the treatment of disorders of the self, which are assumed to have developed due to deficiencies in intrapsychic experience. Self psychology theorists have been primarily concerned with understanding the types of psychopathology based upon narcissistic injuries and deficiencies rather than psychopathology based upon intrapsychic conflict among the demands of id, ego, and superego. Self psychology theorists assume that if an individual is given adequate intrapsychic experience to promote the development of a cohesive self, he or she will not develop neurotic psychopathology based on intrapsychic conflict. Intrapsychic conflict of a pathological nature would only develop within the individual when his or her normal childhood self becomes fragmented or weakened (Goldberg, 1988; Jackson, 1991; Kohut, 1977, 1984; Kohut & Wolf, 1978; Nicholson, 1991; Rowe & Isaac, 1989; Wolf, 1988, 1994).

Kohut (1971, 1977, 1984) originated and developed the theory of self-psychology. Kohut (1971) had observed and reflected at length upon the analytic experiences of many severely disturbed patients whose symptoms and pathology had previously been considered untreatable in analysis. Kohut argued that these patients could be considered to be a diagnostic entity, the narcissistic personality disorder, with a comprehensible developmental origin. He further proposed some modifications of analytic technique based upon theoretical formulations that should be used with this specific group of patients. At this point, he proposed that his ideas were an addition to and expansion of the ideas of classical psychoanalytic theories.

In his later work, Kohut (1977, 1984) proposed that his observations and understanding of narcissistic patients had generated "a psychology of the self" (1977, p. xiii), which offered a new and comprehensive system of understanding intrapsychic development, psychopathology, and the treatment of psychopathology. Kohut hypothesized that the self is the center of intrapsychic activity. A strong cohesive sense of self is vital to the healthy, productive functioning of the individual and provides creativity and a zest for life.

The development of a strong, cohesive self was proposed to be highly dependent on the empathic responsiveness of parents who are experienced intrapsychically as "selfobjects" (Kohut, 1977, p. xiii). The internalization of selfobject functions occurs naturally in a largely empathic environment through optimal, incremental frustrations and disappointments in the empathic responses of the selfobjects. Gross, pervasive failures of empathy and responsiveness are experienced by the developing self as narcissistic injuries, which may lead to disorders of the self if another source of empathic responsiveness cannot be used to develop compensatory structures (Kohut, 1977, 1984).

Kohut (1977, 1984) proposed that a strongly empathic approach on the part of the analyst is required in the treatment of individuals suffering from disorders of the self. The initial need in treatment of the patient suffering a disorder of the self was understanding based upon empathy. The self of the patient is calmed and strengthened by the understanding communicated by the therapist. When a sufficiently cohesive self-structure has developed, the understanding communicated by the therapist is accompanied by explanations of both the intrapsychic dynamics and the genetic origins of the deficient self.

Theorists writing about the treatment approach of self psychology have emphasized repeatedly that the empathy and understanding provided by the therapist is not to be considered a gratification of infantile needs nor a corrective emotional experience as described by Alexander, French, and Bacon (1946). The empathy, understanding, and explaining are considered to be the analytic process of interpretation that allows the patient to understand the deficiencies in the self-structure and to rebuild it (Goldberg 1988; Kohut 1977, 1984; Rowe & Isaac 1989; Wolf 1988, 1994).

There is no doubt that Kohut and his followers have been widely read and discussed. Their ideas about the treatment of narcissistic personality disorders in particular have been helpful and informative to a large number of clinicians, and the ideas of psychoanalytic self psychology have been debated extensively. Self psychology is a theoretical model in which theorists have proposed innovative ideas about the influence of relationships with others on the development of healthy or pathological functioning and on the use of the understanding and interpretation of transference in the treatment of psychopathology (Akhtar, 1989, 1994; Bacal, 1989; Brandshaft, 1989; Detrick, 1989; Donner, 1991; Gedo, 1989; Goldberg, 1988; Jackson, 1991; Levine, 1994; Lynch, 1991; Nicholson, 1991; Rowe & Isaac, 1989; Wolf, 1988, 1994).

The ideas of Kohut (1971, 1977, 1984) and his followers (Donner, 1991; Goldberg 1988; Jackson, 1991; Lynch 1991; Nicholson, 1991; Rowe & Isaac 1989; Wolf, 1988, 1994) will be presented in depth. The section entitled "basic concepts" includes (a) the importance of empathy as a tool for gathering data both for theory development and therapeutic interventions, (b) the concept of the self, and (c) the definitions and functions of selfobjects. The discussion of the nature and development of psychopathology will include what has been described as normal development and the events that lead to the development of psychopathology with descriptions of the various types of disorders of the self identified by self psychologists. In the section on the processes and goals of psychotherapy, there will be a description of the development and types of selfobject transferences and the interventions used by the therapist to treat psychopathology. Finally, a critique of the strengths and weaknesses of the theories of self psychology will be presented.

Basic Concepts

Empathy. Kohut (1984) stated that the "best definition of empathy . . . is that it is the capacity to think and feel oneself into the inner life of another person" (p. 82). Empathy also was defined as vicarious introspection, which was claimed to be one person's attempt to experience the inner life of another while retaining the stance of an objective observer. Empathy was regarded as having two important functions in the theory of self psychology. (1) Empathy is the primary method of observation used in psychoanalysis to gather data upon which theoretical formulations are based. (2) Empathy is a vital component of the therapeutic technique that allows the therapist to understand the patient's inner world and to explain it to the patient.

It was through the use of empathic immersion in the inner experience of patients with narcissistic personality disorders that Kohut (1971, 1977, 1984) made the observations that led to the development of the theoretical formulations of the psychology of the self. Through empathic immersion, Kohut found an understanding of the patient's internal mental life. When this understanding was communicated to the patient, the patient frequently reported an experience of relief and well-being. If, however, Kohut (1971, 1977, 1984) made observations or interpretations based upon a theoretical model removed from the patient's direct experience, the patient responded with disappointment, disillusionment, and/or rage. He found, however, that, if he communicated to the patient his empathic understanding that the patient had suffered the perception of being misunderstood, the patient often regained a sense of relief and well-being. Repeated observations of this sequence of responses in patients with narcissistic personality disorders led Kohut to develop his hypotheses about the development, psychopathology, and treatment of the self.

Kohut (1977) stated that "the essence of psychoanalysis lies in the scientific observer's protracted empathic immersion into the observed, for the purpose of data gathering and exploration" (p. 302). He further stated that psychoanalysis was unique among the sciences in that its theoretical propositions have been based consistently upon data drawn from introspection and empathy. Kohut (1984) emphasized that the empathy in self psychology used to understand the patient and to collect data was not different from the empathy used in any psychoanalytic treatment. However, he claimed that self psychology had supplied psychoanalytic thought with new theories that broadened and deepened the field of empathic perception to include intrapsychic experience not previously understood.

Goldberg (1988) pointed out that empathic observations made in the process of analysis either confirm the theory guiding the observation or suggest the need for changes in the theory. Patients with narcissistic personality disorder responded negatively to interpretations of their experience as manifestations of intrapsychic conflict. Kohut (1971, 1977, 1984) modified his explanation of their intrapsychic experience in an attempt to understand their negative response. He hypothesized that these patients had an internal experience of deficiency and emptiness rather than conflict. This hypothesis was confirmed by the patient when it was communicated to the patient. The confirmation of hypotheses about the internal state of deficiency led to further hypotheses about types and natures of deficiencies, compensatory defenses, the nature of the transference, and the mechanisms of change, all of which continue to be tested, confirmed, and changed on the basis of empathic observation and patient responses. The specific nature of these hypotheses will be discussed in later sections of this chapter.

The use of empathy as part of the therapeutic technique also will be explained in greater depth in the discussion of processes and goals of psychotherapy. In brief, through empathic immersion in what is communicated by the patient both verbally and nonverbally, the therapist is guided by both cognitive and affective understanding of what the patient is experiencing. The therapist communicates the understanding of the patient's experience to the patient. The patient responds to the communication of being understood with a sense of relief and well-being. The therapist's understanding serves as a calming and/or soothing function for the patient (Kohut, 1977, 1984; Lynch, 1991; Rowe & Isaac, 1989; Wolf, 1988). Temporary breaks or disruptions in the therapist's empathic understanding of the patient ideally provide an experience of "optimal frustration" (Kohut, 1984, p. 70), which promotes the patient's internalization of the psychological functions provided by the therapist. These internalizations gradually allow the patient to build and to improve the internal structure of the self. As the patient develops a cohesive, resilient structure of the self, the therapist uses empathy to explain to the patient the dynamics and etiology of the disturbances of the patient's self. These explanations provide an additional basis for the development of a cohesive self structure (Kohut, 1977, 1984; Lynch, 1991; Rowe & Isaac, 1989; Wolf, 1988).

Whether data are to be used to develop theory or in the process of treatment, the reliability of the data depends upon the training and experience of the observer. The observer using empathy must carefully monitor his or her assumed understanding of the patient for potential countertransferential distortions. Furthermore, empathy as used and defined in self psychology is a method of data collection with both cognitive and affective components. It should not be confused with sympathy, kindness, or support, and its use is not intended to gratify the needs of the patient (Kohut, 1977, 1984; Lynch, 1991; Rowe & Isaac 1989; Wolf, 1988).

The self. Kohut (1977) explained that the self is a concept that refers to the core of the personality, which evolves into a coherent and enduring configuration through the interaction of inherited and environmental factors. The self is the center of initiative, recipient of impressions, and repository of the individual's ambitions, ideals, talents, and skills. Ambitions, ideals, talents, and skills motivate the self and provide it with purpose and a sense of meaning in life.

Kohut (1971, 1977) initially described the self as bipolar in nature with a pole of ambitions, which contains the strivings of the self for recognition and appreciation, and a pole of ideals, that contains the guiding values and goals. Kohut further hypothesized what he called an “arc of tension” between the two poles of the self, which activates basic talents and skills and provides the basis for a plan of action and the subsequent activities that shape an individual's life. There will be a more extensive discussion of the nature and development of the bipolar self later in this chapter.

Kohut (1977) admitted that, despite his allocation of hundreds of pages of written material to the discussion of the psychology of the self, he "never explains how the essence of the self should be defined" (p. 310). He argued that the self is not knowable in its essence but only in its introspectively or empathically perceived manifestations. He stated that it may be possible to understand the intrapsychic experiences that lead to the development of a sense of self and the vicissitudes of these experiences. Some of the constituents of the self, such as the ambitions, ideals, talents, and skills, may be identified, and their genesis and functions may be explained. But, a complete understanding of the essence of the self is not possible.

Selfobjects. In his treatment of narcissistic personality disorders, Kohut (1971) came to recognize that the understanding provided by the therapist led to the development of transferential responses on the part of the patient in which the empathic understanding was used by the patient as a source of relief, soothing, calming, or regulation of self-esteem. Kohut defined this intrapsychic experience of the patient as the patient's subjective use of the therapeutic relation with the therapist as a selfobject. The term selfobject was defined by Kohut as an individual's internal experience of another person as a part of the self. Similarly, selfobject relations do not refer to the real interpersonal relations that exist between an individual and others in the environment. Selfobject relations are the intrapsychic experiences of an individual in which the individual uses interpersonal relations to build, to sustain, or to change the structure of the self. Selfobject relations are needed and used throughout life by all individuals to nourish the self (Kohut, 1971, 1977, 1984; Lynch, 1991; Rowe & Isaac, 1989; Wolf, 1988, 1994).

Kohut (1971, 1977) initially identified two primary types of selfobject relations that are manifested in the transference responses of the patients with disorders of the self. In the mirroring transference, the patient seemed to seek acceptance and confirmation of the goodness, wholeness, and specialness of the self. In the idealizing transference, the patient seemed to seek a merger with the calmness, wisdom, and power of the idealized selfobject.

The identification of the mirroring and idealizing transferences led Kohut (1971, 1977) to the hypothesis that these types of transferences were reflections or recapitulations of normal childhood experiences. Kohut (1971, 1977) proposed that the small child has intense needs to receive confirmation of his or her grandiose self. In other words, the small child seeks to be the center of attention and has yearnings to be special and important to the parents or other significant people in the environment. As the child matures, the sense of specialness is internalized through appropriate mirroring selfobject experiences, which are accompanied by optimal frustrations. In later childhood, the need for mirroring in selfobject relations becomes a need for confirmation of normal achievements and accomplishments. These experiences of appropriate mirroring, optimal frustration, and internalization of selfobject confirmations are the building blocks of the pole of ambitions in the structure of the self.

Similarly, Kohut (1971, 1977) proposed that the small child needs to idealize selfobjects in the environment. These idealized selfobjects provide a source of strength or calmness with which the child can merge at times when the child feels anxiety, fear, or other forms of upset. The merger with the idealized selfobject provides the child with a source of calmness, reassurance, and equilibrium. As with the mirroring selfobject experience, the idealized selfobject experiences become internalized with appropriate selfobject responses and optimal frustration. The idealized selfobject relations provide the basis of the development of the pole of ideals.

Kohut (1971, 1977) proposed that, as the child matures into adulthood with appropriate self-object relations, the two poles of the self also change and mature. The pole of ambitions that starts with a need for confirmation of grandiosity is transformed over time into healthy ambitions, the enjoyment of activities, and a zest for life. The pole of ideals shifts from a need for an omnipotent selfobject to a stable internalized configuration of values, ideals, and goals.

Kohut (1984) later identified a third selfobject relationship needed by individuals. The need for a twinship or an alter-ego selfobject relation also was identified through the analysis of transference responses of patients. The need for an alter-ego selfobject relation was described by Kohut as the need to experience an essential similarity with another individual. Kohut commented that the young child "obtains a vague but intense and pervasive sense of security as he feels himself to be a human among humans" (1984, p. 200). It should be noted, however, that in his writings about the alter-ego or twinship experience, Kohut neither discarded his theoretical formulations about the development of the bipolar self through selfobject experiences nor did he thoroughly integrate this newly identified need for twinship experiences with his previous formulations. He mentioned briefly that the need for twinship selfobject experiences may influence the development of the tension arc of talents and skills between the two poles of the self but did not explain the nature of the influence.

Wolf (1988, 1994) similarly introduced several "needed selfobject experiences" (1994, p. 72) with no integration of these selfobject experiences with the previously hypothesized development of the bipolar self nor with any explicit repudiation of the theory of the bipolar self. The problems with the idea of the bipolar self and its integration with later theoretical formulations will be considered further in the discussion of the strengths and weaknesses of self psychology.

Wolf (1994) stated that self psychology now recognizes at least seven types of selfobject experiences that are needed for the establishment and maintenance of a cohesive, energetic, and balanced self.

1. Mirroring selfobject experiences are the needs to feel recognized, affirmed, appreciated, and accepted.

2. Idealizing selfobject experiences are the need to be accepted by and to merge with a stable, calm, powerful, wise, protective other who possesses qualities lacking in the self.

3. Merged selfobject experiences are a primitive form of the mirroring need that finds confirmation for the self in the experience of being totally unified with the selfobject. For an infant, this is "a real experience of blissful well-being that forms the bedrock upon which healthy self-esteem is built" (p. 73).

4. Alter-ego selfobject experiences are the need to perceive an essential similarity or a likeness with the selfobject.

5. Adversarial selfobject experiences are the need to perceive the selfobject as a benevolently opposing other who continues to be supportive and responsive even when faced with active opposition. The adversarial selfobject experience confirms a sense of autonomy.

6. Efficacy selfobject experiences are the need to be aware of having initiated and caused a response from another.

7. Vitalizing selfobject experiences are the need of a child to perceive that the selfobject is affectively attuned to the changing moods and feelings of the child.

To summarize, selfobjects, when functioning appropriately, are hypothesized in self psychology to evoke and to maintain the self. The selfobject relationship is an intrapsychic one that is subjectively experienced. The actual interpersonal relations between persons may give rise to selfobject experiences but are not the same as selfobject experiences. Selfobject experiences are vitally needed throughout life to maintain the cohesion, vigor, and balance of a healthy self (Goldberg, 1988; Kohut, 1977, 1984; Lynch, 1991; Rowe & Isaac, 1989; Wolf, 1988, 1994).

The Nature and Development of Psychopathology

Normal Development. Kohut (1977) proposed that a precondition for the development of a healthy, cohesive self is that the child be born into an empathic, responsive human milieu of selfobjects just as the child must be born into an oxygen providing environment in order to survive physically. The selfobjects in the environment must be equipped with mature psychological organization that can be utilized to assess the needs of the child and to perceive and to respond emphatically to tensions and states of imbalance. The selfobjects include the child in his or her own psychological organization and take action to remedy tensions and imbalances that occur. When the child experiences anxiety, hunger, rage, or any other discomfort, the selfobject resonates empathically and comforts the child with gratification of needs or tactile or vocal comfort. The child experiences a merger with an omnipotent selfobject and participates in a rudimentary way with the highly developed psychic organization of the selfobject. The child experiences the feelings of the selfobject as communicated by touch and tone of voice as if they were the feelings of the child.

Kohut (1977) hypothesized that a repeated sequence of events in early childhood provided by the selfobject established the baseline for later internalization of self structure. The sequence described included the mounting anxiety of the child, followed by stabilized mild anxiety of a signal type, which is followed by gratification and/or calmness. This repeated sequence of psychological events accompanied by the merger with the empathic omnipotent selfobject provides the basis for the child's handling optimum, nontraumatic failures of the selfobject by using transmuting internalization to build the structure of the self. Optimal failures on the part of the selfobject are responses that are somewhat delayed, mildly deviant from beneficial norm of the child's usual selfobject experiences, or otherwise mildly disappointing. These optimal, nontraumatic frustrations stimulate the use of transmuting internalizations of the selfobject functions of the temporarily disappointing selfobject.

Kohut (1977) emphasized the importance for the child of the two steps of selfobject functions in this sequence of events when the child experiences anxiety and tension. The first step is the empathic merger with the selfobject's mature psychic organization and participation in the selfobject's containment of the anxiety to an affect signal before the anxiety increases to an overwhelming level. The second step is the need-satisfying actions performed by the selfobject. Kohut stated that the importance of this sequence cannot be overestimated: "if optimally experienced during childhood, it remains one of the pillars of mental health throughout life" (p. 87).

Small infants, thus, presumably experience themselves before self/object differentiation in a state of limitless merger with the selfobjects in the environment. As the infant experiences optimal frustrations, self/object differentiation begins, and there is an emergence of a transient sense of selfhood and self structure. During infancy, there is an easy, nonpathological oscillation between states of merger and transient states of self structure (Kohut, 1977; Wolf, 1988).

As the child continues to mature, the interactions between the child and the selfobjects continue to provide the basis for the development of the self. Through countless repetitions, the selfobjects empathically respond to certain potentialities of the child: the aspects of the grandiose self exhibited, aspects of idealized images admired, and different innate talents employed by the child. Other potentialities of the child are not empathically responded to by the parents. The "nuclear self" (Kohut, 1977, p. 100), thus, is formed by the deeply anchored, mostly unconscious responsiveness of the selfobjects in the environment, and their responsiveness is a function of their own nuclear selves (Kohut, 1977).

Kohut (1977) proposed that initiative and assertiveness are innate, healthy aspects of the potentialities of the child. If initiative and assertiveness are responded to with empathy and appropriate selfobject experiences, then normal adult assertiveness will evolve to be used in the pursuit of goals and the realization of ambitions. Kohut stated that destructive rage is triggered by a narcissistic injury inflicted by one of the selfobjects of childhood. Chronic narcissistic rage develops only when selfobject needs are chronically and traumatically frustrated.

Kohut (1977) proposed that, in the preoedipal period of development, the child has two opportunities to achieve consolidation of a cohesive, resilient self structure. The child may either establish a cohesive grandiose exhibitionistic self through the empathically responding "merging-mirroring-approving selfobject" (p.185), or the child may establish a cohesive idealized "parent image" (p. 185) through the relation with an empathically responding selfobject who permits and/or enjoys the child's idealization and merger with the selfobject. In either case, the child develops fully and internalizes one pole of the bipolar self and uses compensatory structures to compensate for the deficiencies in the other pole.

Kohut (1977) stated that this model of the bipolar self is a useful abstraction that can help explain the many differences that can be observed readily among different types of healthy, cohesive selves encountered in the analytic setting. With the aid of this theoretical assumption, the many varieties and types of nuclear selves, "ambitious or idealistic, charismatic or messianic, task-oriented or hedonistic" (p. 186), can be understood, and the impact and significance of environmental experiences with various selfobjects can be understood. The problems with this formulation of the development of the bipolar self will be discussed again at greater length in the discussion of the strengths and weaknesses of self psychology.

Given the development of a cohesive nuclear self during the preoedipal period of development, Kohut (1977) proposed that the principles of self psychology continue to exert an important influence in the oedipal period of development. Kohut proposed that the child with a cohesive self will experience the challenges of the oedipal phase of development with anticipation and joy despite the potential anxieties and conflicts of this stage of development.

Kohut (1977) proposed that, if the child enters the oedipal phase with a firm, cohesive, continuous sense of self, the child then will experience "assertive-possessive, affectionate-sexual desires for the heterogenital parent and assertive, self-confident, competitive feelings vis-a′-vis the parent of the same sex" (p. 230). However, the child's oedipal experiences cannot be evaluated in isolation. The child's experiences can be understood only within the context of the selfobject experiences in the environment.

Kohut (1977) suggested that the appropriate response of the selfobject parents in the oedipal phase has several components. First, both parents will react with joy and pride to the child's developmental achievement and to the child's vigor and assertiveness. Second, the empathic heterogenital parent will consciously or unconsciously grasp the fact of having become the target of the child's libidinal desires and will respond in an aim-inhibited fashion to the child's advances. Third, the homogenital parent will grasp the fact of having become the target of the child's rivalrous competition and will respond with aim-inhibited counter aggression to the child's hostility. The appropriate empathic perception of the child's intentions and the appropriate empathic responses of aim inhibition help the child to acquire intrapsychic structures that modulate the expression of aggressive and libidinal drives. Given the empathic, appropriate responses of selfobject parents with healthy cohesive selves, the normal child's oedipal experiences are filled with joy due to the child's inner awareness of a significant forward move in a psychological realm of new and exciting experiences and to the child's participation in the joy and pride of the parental selfobjects rejoicing in the growth and progress of their child.

Wolf (1994) proposed that, during the oedipal period, the child needs mirroring, idealizing, and alter-ego selfobject experiences in order to form adequate gender identity and to prevent the development of neuroses. Wolf stated that boys needed nonseductive confirmation of autonomy and maleness by the mother together with acceptance of his idealizing needs. Boys also need nonaggressive acceptance of their adversarial and alter-ego needs by their fathers. Girls need nonseductive confirmation of autonomy and femaleness by the father with acceptance of idealizing needs. Girls need their mothers to accept alter-ego and adversarial needs nonaggressively.

As the child enters the school-age years, there is a gradual expansion of selfobject experiences with teachers and peers becoming the providers of some selfobject needs. In adolescence, cognitive development leads to a recognition of parental defects which precipitate rapid de-idealization of the parents. The continuing need of the self for idealized selfobjects is met by the peer group, the adolescent subculture and its idols, and the heroes found in culture and history. The adolescent's increasing capacity for symbolic thought allows the development of the capacity to find idealized selfobjects in art, religion, philosophy, and history (Wolf, 1988, 1994).

In adulthood, a variety of selfobject needs are met through the marital relationship. Intimacy facilitates a controlled regression to primitive merger without loss of the autonomy of the self. Spouses usually fulfill several mirroring and idealizing selfobject needs. The adult also has the capacity to use symbolic selfobject experiences through art, music, or literature. The adult also may find selfobject sustenance in work or community activity (Wolf, 1988, 1994).

It should be noted that the need for different selfobject experiences at various developmental stages has been emphasized in the discussion of normal development in self psychology. The developmental stages were presented in rather vague, general terms with little specific commentary about the relation between stages in physiological or cognitive development and the psychological development of self structure. Most of the ideas about development in self psychology, particularly Kohut's (1971, 1977, 1984), are a hypothetical reconstruction of childhood experiences based upon transference responses. Wolf (1988, 1994) made a cursory effort to integrate Kohut's previously developed theories of normal childhood development with later infant research, particularly the work of Stern (1985), but very few findings in infant research were described in support of the developmental theories of self psychology.

The Nature and Development of Psychopathology

As may be inferred from the previous description of the theories of self psychology, psychopathology is thought to be the result of gross traumatic failures of empathic responsiveness in selfobject experiences (Kohut, 1971, 1977, 1984; Nicholson, 1991; Rowe & Isaac, 1989; Wolf, 1988, 1994). Kohut (1977) specifically stated that the great majority of the disturbances, fixations, and unsolvable inner conflicts of the disturbed adult personality are the result of "the specific pathogenic features of the atmosphere in which the child grows up" (p. 187). The traumatic events of childhood remembered in the treatment process are usually what Kohut calls "crystallization points" (p. 187) that demonstrate the general lack of empathy and responsiveness in the child's environment.

Kohut (1977, 1984) considered all forms of psychopathology to be disorders of the self. He proposed that there were "three genetic-structurally defined classes of disorders" (1984, p. 8): psychoses, narcissistic personality disturbances, and classical neuroses. Each group was differentiated by the nature and severity of the damage to the self and by the capacity of the individual suffering from the disorder to benefit from analysis.

Psychotic and borderline disorders were combined by Kohut (1977, 1984) as disorders in which no nuclear self was consolidated in early development. Borderline conditions were described as differing from psychotic conditions in that the borderline personality disorders have a well-developed peripheral layer of defensive structures. Paranoid and schizoid disorders were considered to be types of borderline disorders.

Psychoses were considered to be disorders of the self in which the damage to the self is relatively permanent and in which there is a lack of organized defenses. Psychoses may be either organically or experientially based. The experiential basis of psychotic conditions is the pervasive prolonged failure of the childhood selfobjects to meet the developing needs of the child (Kohut, 1977, 1984; Nicholson, 1991; Wolf 1988).

Wolf (1988) proposed that schizophrenia is the result of constitutional factors combined with deficient mirroring. Empty depression is based upon organic factors combined with a lack of joyful selfobject experiences. A lack of opportunity to experience merger with calm idealized selfobjects impedes the development of self-soothing and self-supportive structures, which predisposes the individual to mania and guilty depression.

In the narcissistic personality and behavior disorders, Kohut (1977, 1984) proposed that the outlines of a nuclear self had been established in early development but that the structure of the self had remained incomplete. The defects in the structure of the self make individuals with narcissistic disorders vulnerable to fragmentation, enfeeblement, or extreme disharmony when they experience a narcissistic injury. Fragmentation was described as a state of experience in which the aspects of one's self-experience no longer seem coordinated. The state of fragmentation is accompanied by feelings of apprehension, disorganization, moodiness, and malaise. Individuals with narcissistic disorders have difficulty maintaining self-esteem in times of stress.

Individuals with narcissistic behavior disorders attempt to bolster their shaky self-esteem through perverse, delinquent, or addictive behavior. Individuals with narcissistic personality disorders suffer symptoms of hypochondria, depression, hypersensitivity, lack of zest, irritability, insomnia, and inability to concentrate. Patients who suffer severe narcissistic disorders may display grandiose fantasies of self-importance, a sense of entitlement, and inability to empathize with others or to see others except as need-gratifying objects (Kohut, 1977, 1984; Nicholson, 1991; Rowe & Isaac, 1989; Wolf, 1988).

Narcissistic disorders are the result of a continuous lack of empathic responsiveness in early development. The small child experiences such gross empathic failures in selfobject relations that the child fails to internalize the self-confidence and inner security provided by appropriately empathic mirroring experiences and/or the strength and calmness provided by appropriate merger with the idealized parent. The extent and nature of the empathic failures of the parent and the concomitant disturbance of the individual with the narcissistic disorder is identified by the analysis of the transference that develops in treatment (Kohut, 1971, 1977, 1984; Kohut & Wolf, 1978; Nicholson, 1991; Rowe & Isaac, 1989; Wolf, 1988).

Kohut and Wolf (1978) proposed that individuals who are suffering from narcissistic disorders have intense needs and a conviction that they will not find empathic understanding. These feelings arouse deep shame, suppression of needs, depression, hopeless withdrawal, and sometimes bursts of rage. Grandiose fantasies and social isolation are defensive maneuvers to protect the individual from feeling inner emptiness and despair. Bursts of rage are a reaction to underlying feelings of helplessness and hopelessness.

Neuroses are the disorders of the self that occur after the formation of a cohesive, nuclear self. Kohut (1977, 1984) thought that neurotic disorders arose from failures in selfobject relations during the oedipal period. After appropriately responding to the child in the preoedipal period and providing selfobject experiences that help the child establish a cohesive nuclear self, the parents fail to respond to the oedipal child with empathic responsiveness. Usually, the inappropriate responses of the parents occur when they are alarmed by the child's emerging sexuality and aggression and are not responsive to the assertive affection and competition of the oedipal child.

The failure of the selfobject parent to be empathic with the whole self of the oedipal child has a disintegrating effect on the cohesive self, which begins to fragment. As a result of the fragmentation, isolated drive experiences and conflicts about them begin to occur. The child cannot assimilate the feelings of lust and hostility without the empathic response of selfobjects to the accompanying feelings of affectionate and competitive assertiveness (Kohut, 1977, 1984).

The individual with a neurotic disorder will have difficulty with the expression of healthy affection, vitality, and assertiveness. The neurotic individual may have difficulty in achieving goals and realizing ambitions. Neurotic disorders also may be characterized by conflict over the expression of healthy sexuality and aggression. Neurotic symptoms include anxiety, depression, phobia, obsessive thinking, compulsive behavior, or conversion disorders (Nicholson, 1991; Wolf, 1988).

Kohut and Wolf (1978) described five different character types with patterns of behavior that may be observed in any type of disorder of the self. As previously stated, all forms of psychopathology from psychoses to neuroses are considered to be, in the theory of self psychology, disorders of the self. Any individual with a self disorder may display one or more of the character types described.

People who have mirror-hungry personalities feel compelled to display themselves, demanding the attention of others in hopes that admiring responses will counteract their inner sense of worthlessness. Individuals with ideal-hungry personalities can only experience themselves as worthwhile when relating to others whom they admire and by whom they feel accepted. Individuals with alter-ego-hungry personalities seek confirmation by associating with another whose appearance, opinions, and values they share (Kohut & Wolf, 1978).

Individuals with contact-shunning personalities avoid social contact and any form of intimacy in an attempt to avoid rejection and as a defense against the merger experience they both yearn for and fear. Individuals with merger-hungry personalities seek the continuous presence of selfobjects, which will provide the self structure that they lack. They fail to distinguish themselves from their selfobjects and fail to maintain clear personal boundaries or a cohesive sense of self (Kohut & Wolf, 1978).

Kohut and Wolf (1978) also described four clinical syndromes that are manifestations of various types of disorders of the self. These clinical syndromes may be mixed in any given person at any given time.

(1) The understimulated self is based upon childhood experience in which selfobjects fail to provide stimulating responses to childhood needs. These individuals lack vitality and seek excitement. Symptoms of the understimulated self include addictions, perversions, compulsive masturbation, and social hyperactivity.

(2) The fragmenting self is based on childhood experience in which the selfobjects failed to assist the emerging self in the integration of early experience. The fragmenting self reacts to narcissistic disappointments with a loss of self cohesiveness. When self impairment is mild, the symptoms of fragmentation are a minimal amount of emotional and intellectual disorganization with some loss of physical coordination. In narcissistic personality disorders, the fragmentation may be severe with a marked deterioration of physical appearance, disorganization, and perhaps even disorientation. Fragmentation improves rapidly if the individual can establish an empathic relation with a selfobject. In treatment situations, unempathic responses may lead to sudden intense experiences of fragmentation. If the therapist can correctly interpret the connection between the unempathic response and the fragmentation, the patient recovers, and fragmenting symptoms may begin to disappear (Kohut & Wolf, 1978).

(3) The over-stimulated self is based on childhood experience in which the selfobjects consistently responded inappropriately and excessively to the grandiose and/or idealizing fantasies of the developing child. Thus, childhood grandiosity is not neutralized into developmentally phase-appropriate ambitions and goals. The overstimulated self cannot enjoy success and avoids attention. Creativity is often impaired. These individuals often experience painful tension and anxiety regarding their fantasies of greatness. With the overstimulation created by the needs of the parental selfobject, the individual's capacity for enthusiasm and vitality gets lost (Kohut & Wolf, 1978).

(4) The overburdened self has a childhood experience in which the need for merger with the calming and soothing functions of the idealized parents was frustrated by unempathic responsiveness. Calming and soothing functions, therefore, are not internalized as self functions. The environment is imagined to be hostile and dangerous. These individuals tend to be overly sensitive, irritable, and suspicious. They often suffer from intense anxiety. In treatment situations, the overburdened self may feel narcissistically injured by an unempathic response, even a simple question, and may react with paranoid suspiciousness. The selfobject transference may be restored by an empathic interpretation of what had occurred, and the patient may recover a calm, cohesive self (Kohut & Wolf, 1978).

In summary, it has been proposed that normal development is characterized by a meaningful sequence of changes in the nature of selfobject relations from infancy through adulthood. Kohut (1984) specifically stated that the idea that normal development is a progression from helpless dependence to autonomy and from self love to the love of others is erroneous and misleading. He proposed that both narcissism and object love (i.e., the capacity to love others) evolve from archaic or primitive forms to mature forms in two separate lines of development. In normal development with appropriately empathic responsiveness and optimal frustrations, there is "movement from archaic to mature object love" (p. 208).

Thus, in self psychology, psychopathology is understood to be based upon traumatic failures in empathic responsiveness to the normal needs of the developing child for merger, mirroring, idealizing, or alter-ego selfobject experiences. The severity of the psychopathology is related to the developmental level at which the traumatic failures occur. Psychoses and borderline disorders are caused by traumatic failures preceding or during the development of the nuclear self, and neurotic disorders are caused by failures in development during the oedipal period. Disorders of the self are also described as having characteristic behavior patterns reflective of the frustrated and undeveloped selfobject experiences and clinical syndromes that reflect the type of developmental failures that occurred (Kohut, 1984; Kohut & Wolf, 1978; Nicholson, 1991; Wolf, 1988).

The Processes and Goals of Psychotherapy

Self psychology theorists have emphasized the importance in therapy of the therapist's empathic understanding of patients' subjective experiences and their meaning. The empathic responsiveness of the therapist reactivates the frustrated selfobject needs of the patient, activating one or more selfobject transferences to the therapist. The selfobject transferences involve idealizing, mirroring, and alter-ego selfobject transferences (Donner, 1991; Kohut, 1984; Rowe & Isaac, 1989; Wolf, 1988).

The therapist uses empathy to understand the selfobject needs activated in the transference and communicates this understanding to the patient. The therapist also explains the dynamics and genetic origins of the frustrated selfobject needs when patients have sufficient self structure to integrate information about the dynamics and origins of defective self structure without experiencing such information as a narcissistic injury. Some individuals with disorders of the self may respond with relief to the experience of having their immediate feeling understood while reacting with intense shame and humiliation to the idea of an underlying deficit in their self structure. Other individuals benefit from dynamic and genetic interpretations from the beginning of treatment (Donner, 1991; Kohut, 1984; Rowe & Isaac, 1989; Wolf, 1989).

Inevitably in the course of treatment, the therapist will disappoint and fail to meet the patient’s expectations, usually through a failure in empathic responsiveness. The patient may react with rage, withdrawal, acting out, or other negative symptoms. It is the responsibility of the therapist to understand and to acknowledge the patient's subjective experience. The therapist's sensitivity to and understanding of the experience of the patient provides the patient with a sense of confirmation and efficacy. These optimal failures of the therapist promote the process of transmuting internalization, which is the acquisition of cohesive, flexible, enduring self-regulating structures based upon the interpretations made by the therapist. In the course of treatment, the patient gradually acquires empathy toward the self (Donner, 1991; Goldberg, 1988; Kohut, 1984; Wolf, 1988).

Kohut (1984) stated that the crucial emotional experience for human psychological survival and growth is the attention of a selfobject that attempts through empathy to understand and to participate in the individual's psychological life. The responsibility of the analyst is to focus attention on the inner life of the patient, and the successes and failures of this understanding activity "are the essential motor of the psychoanalytic process" (p. 38). The task of the therapeutic process was described as the exploration of the flaws in the structure of the self by the analysis of the selfobject transferences with a focus on understanding the dynamic and genetic dimensions of the flaws.

Kohut (1984) described the process of treatment as a three-step movement: (a) analysis of defenses, (b) the development of the transference, and (c) the establishment of empathic attunement between the self of the patient and the selfobject whom the therapist becomes in the transference. The third step was described as essential because the aim and result of the cure in self psychology is the establishment of empathic attunement between self and selfobject on a mature, adult level. The establishment of empathic attunement in the transference "permanently takes the place of the formerly repressed or split-off archaic narcissistic relationship; it supplants the bondage that had formerly tied the archaic self to the archaic selfobject" (p. 66).

Kohut (1984) explained that the therapist's consistent efforts to understand the patient leads to two results that are analogous to the outcome of normal childhood development: occasional failures of empathy, similar to the optimal frustrations of childhood, lead to the building of self structure, and the usually adequately maintained understanding inspires the patient's realization that, contrary to childhood experience, sustaining empathic attunement is available in the world. Kohut conceded that the empathic understanding provided by the therapist may thus provide a "corrective emotional experience" (p. 78), but he argued that this experience is only a single aspect of "the multifaceted body of the psychoanalytic cure" (p. 78). He also explained that the corrective emotional experience of empathic understanding during the interpretation and working through of a selfobject transference had been different from the corrective emotional experience in the analysis described by Alexander et al. (1946) in which the analyst plays a parental role to provide the patient with a different experience than had occurred with the parents in the patient's childhood. In the analysis of self psychology, the analyst does not play a role in order to provide the patient with an experience different from childhood experience. The analyst simply provides empathetic understanding of the patient's experience, which serendipitously happens to provide the patient with a different experience than had occurred in the patient's childhood.

Kohut (1984) commented that, after a successful analysis, most memories of the analyst fade away. Normal mental functioning, whether established in consequence of normal childhood development or in consequence of a successful analysis, depends upon smoothly interacting psychological structures. Neither the selfobjects of childhood nor the selfobject revived in transference should play a role in conscious healthy adult life. Functions formerly performed by selfobjects are internalized as parts of the structure of the self and largely function outside of awareness.

Kohut (1984) advised that confrontation should be used sparingly. He stated that it is not the task of the therapist to educate the patient by confrontation but is to cure defects in the self through the consistent interpretation of selfobject transference. He also cautioned that, in the three-step movement of the analysis, the stage of the analysis of the defenses may result in an improvement in the patient's behavior and mood, which is then followed by a seemingly ominous deterioration in the patient's condition. Kohut explained that this phenomenon "is nothing else but the transference clicking into place" (p. 178). When the patient fully enters into the transference, "the analytic situation has become the traumatic past and the analyst has become the traumatic selfobject of early life" (p. 178).

Kohut (1984) concluded that several goals will be achieved in a successful analysis.

1. The analyst will have explained that deficits in the structure of the self are the result of traumatic failures in selfobject experiences.

2. The analyst will have explained that the anxious clinging archaic selfobjects and their functions are an indication of persistent determination to complete the development of the self.

3. The analyst will have explained that the stalemated development of the self is the basis of the deeply felt needs and demands for an appropriately responsive selfobject experience that might facilitate new progress in the development of the self.

Finally, Kohut (1984) stated that analysis leads to a cure only through the employment of countless repetitions of understanding and explaining. Interpretation is "the analyst's only active function in the analytic process" (p. 209). Wolf (1988) and Goldberg (1988) similarly emphasized the primary analytic function as interpretation. Goldberg (1988) stated that compassion or sympathy in psychotherapy may be as natural a part of an interpersonal experience as simple politeness but has nothing to do with the therapy. Compassion or sympathy can even become a substitute for treatment. Goldberg emphasized that the curative process of therapy was the use of empathic understanding with cognitive meaning provided by appropriate interpretations that allow the patient to understand his or her internal experience.

Wolf (1988) stated that the therapist does not soothe or mirror. The therapist interprets the need for soothing or mirroring. The therapist does not approve of grandiose expectations but rather explains their role in psychic functions. It may be the case that the mere presence of the therapist or the understanding of the therapist may have a soothing or self-confirming effect on the patient, and such effects also are interpreted or explained.

Kohut (1984) stated that psychotic patients could not benefit from psychoanalysis because the lack of a cohesive self would preclude the activation of a stable transference. Kohut wrote that a nuclear self could not be created by therapy. However, he said that the psychotic patient could use the therapist as a selfobject to build and to reinforce defensive structures. Educational activities and empathic responses on the part of the therapist could help the psychotic patient manage defenses optimally.

Kohut (1984) proposed that the key to the successful analytic treatment of the borderline patient is largely the therapist’s sincere acceptance of the patient's reproaches as psychologically realistic to the patient. Prolonged and successful introspection is used to remove the inner barriers that impede empathic understanding of the patient. If the therapist can extend empathic understanding to the borderline patient, the borderline patient will "become [a patient with] an analyzable narcissistic personality disorder" (p. 182).

Kohut (1984) concluded that the analyzability of the borderline patient depends in many cases on two factors: (a) the therapist’s efforts to retain an attitude of empathy and concern despite the serious narcissistic injuries inflicted by the patient and (b) the therapist’s efforts to offer the patient empathic understanding of the patient's experience of the world. Empathic understanding encourages and allows the development of a selfobject transference which makes possible the gradual exploration of the dynamic and genetic causes of the underlying disorder. Understanding and exploration of the dynamics and etiology of the disorder enable the patient to reassemble the self.

Wolf (1988) stated that the primary difficulty in the treatment of borderline patients is the therapist's lack of understanding of the patient's internal experience. This lack of understanding interferes with the therapist's capacity for empathic attunement. Wolf hoped that, with continuing progress in the theory and treatment of self disorders, the category of borderline patients eventually would be dissolved into a small group who are psychotic and a much larger group who can be understood and treated as narcissistic behavior disorders.

Heuristic or Guide to Investigation

As was described previously, Kohut’s (1971, 1977, 1984) guide to investigation was empathic immersion in the patient. Hypotheses about the patients internal experiences were then offered to the patient in the form of interpretations. His theory about the internal structure and experience of narcissistic disorders was based upon data derived from empathic understanding of the patient, which was communicated to the patients and was confirmed by the patients.

Ontology

Essential Human Nature

Kohut (1977, 1984) proposed that the human individual is predisposed to develop a cohesive sense of self given appropriate environmental response to the needs of the developing individual. If the individual does not receive the appropriate, optimal environmental response in childhood, the individual in adulthood continues to seek the responses needed to develop a cohesive self. Psychopathology is based upon deficits in the experience of the individual, which may be corrected. The correction of deficits provides the basis for the resumption of normal development.

Kohut (1977) proposed that initiative and assertiveness are innate, healthy potentialities of the child. With appropriate, empathic selfobject experiences, the child's innate initiative will evolve into normal adult assertiveness used in the pursuit of goals. Destructive aggression and rage develop only when selfobject needs are chronically and traumatically frustrated.

The Nature and Development of Psychopathology

In self psychology, psychopathology is based upon traumatic failures in empathic responsiveness to the normal needs of the developing child for merger, mirroring, idealizing, or alter-ego selfobject experiences. The severity of the psychopathology is related to the developmental level at which traumatic failures occur. Psychoses and borderline disorders are caused by traumatic failures preceding or during the development of a nuclear self. Narcissistic disorders are based on traumatic failures in the preoedipal period after the establishment of a nuclear self but while self structure is still incomplete. Neurotic disorders are caused by traumatic failures during the oedipal period (Kohut, 1984; Kohut & Wolf, 1978; Nicholson, 1991; Wolf, 1988).

The theoretical model proposed in self psychology defines psychopathology as the result of gross traumatic failures of empathic responsiveness in selfobject experiences. All forms of psychopathology are considered to be self disorders that are distinguished by genetic and structural differences.

The Role of the Therapist

Kohut (1984) recommended analysis as the treatment of choice for narcissistic and neurotic disorders. In analytic treatment, the primary role of the therapist is the provision of empathic understanding of patients' subjective experiences and their meanings. The empathic responsiveness of the therapist reactivates the patient's frustrated selfobject needs, and one or more selfobject transferences develop, the idealizing, mirroring, and/or alterego transferences (Donner, 1991; Kohut, 1984; Rowe & Isaac, 1989; Wolf, 1988).

The therapist uses empathy to understand and to interpret selfobject needs activated in the transference. The patient experiences relief. Occasional failures of empathy, similar to the optimal frustrations of childhood, promote the patient's use of transmuting internalization to build flexible, enduring, self-regulating structures. As the patient develops self structure, the therapist explains the dynamics and genetic origins of the frustrated selfobject needs (Donner, 1991; Kohut, 1984; Rowe & Isaac, 1989; Wolf, 1988).

Kohut (1984) stated that analysis leads to a cure only through countless repetitions of understanding and explaining. Interpretation is the analyst's most important activity in the treatment process. Kohut conceded that the empathic understanding provided by the therapist may provide a type of corrective emotional experience but that this type of experience is only a single aspect of the multifaceted psychoanalytic cure. The effects of the empathic attunement are also interpreted and explained in the exploration of the dynamics and etiology of the deficits in the self.

The Mechanism of Change

The optimal failures of the therapist in the context of a general atmosphere of empathic understanding stimulate the patient to internalize selfobject functions provided by the therapist. Internalization of selfobject functions are used by the patient to build and to improve the structure of the self. As the patient develops a cohesive, resilient self structure, the therapist interprets the dynamics and etiology of the disturbances in the patient's self. These interpretations provide an additional foundation for the development of cohesive self structure. Interpretation is the primary and essential mechanism of change, but it should be noted that interpretations can only be understood and effectively used by the patient in the context of the empathic attunement of the therapist and the internalization of selfobject functions provided by the therapist (Goldberg, 1988; Kohut, 1984; Wolf, 1988).

The Goals of Treatment

Kohut (1984) specified three goals to be achieved in a successful analysis: (a) the analyst will have explained how deficits in self structure were the result of traumatic failures in selfobject experiences; (b) the analyst will have explained that the anxious archaic intrapsychic selfobject relations were a reflection of the individual's persistent determination to complete self development; and (c) the analyst will have explained how the stalemated development of the self was the source of the deeply felt need for a responsive selfobject experience that might facilitate new progress in self development. Kohut also stated that the aim and result of a cure in self psychology was the patient’s development of a capacity for empathic attunement on a mature adult level. Finally, several self psychologists have mentioned that a goal of treatment is the development in the patient of empathy for the self (Donner, 1991; Goldberg, 1988; Kohut, 1984; Wolf, 1988).

Critiques of Self Psychology

A major criticism of Kohut (1971, 1977, 1984) and his followers (Donner, 1991; Goldberg, 1988; Lynch, 1991; Jackson, 1991; Nicholson, 1991; Rowe & Isaac, 1989; Wolf, 1988, 1994) has been "their exaggerated claims to uniqueness and originality" (Greenberg & Mitchell, 1983, p. 366). The theoretical model developed in self psychology has striking similarities to models proposed by previous theorists who presented their ideas from an interpersonal or object relations perspective. These similarities were not discussed by Kohut and, in some cases, were dismissed overtly on grounds that seem specious and based upon inadequate appreciation of what was actually proposed by these previous theorists.

Greenberg and Mitchell (1983) pointed out several similarities between the ideas of Kohut (1971, 1977) and Sullivan (1953, 1956, 1964, 1972). Kohut's ideas about the selfobject relations are similar to Sullivan's ideas about the interpersonal field. Kohut's discussion of the sensitivity of the child to the feelings of the selfobject is similar to Sullivan's ideas about empathic linkage between parent and child. Kohut's description of the analytic process as dependent on the empathic understanding of the therapist are very similar to Sullivan's description of the therapist as a participant observer.

There are also similarities between Kohut's (1971, 1977, 1984) ideas and the theories of several object relations theorists. Fairbairn (1952) proposed that pathological drive manifestations were based upon failures in the relational experiences of the developing child just as Kohut suggested that drive manifestation and conflict occur only when selfobjects are not appropriately empathically responsive. Kohut's proposition that selfobject experiences are needed throughout life differs little from Fairbairn's conception that the final stage of mature adult development is mature dependence.

The similarities between the ideas of Kohut (1971, 1977, 1984) and Winnicott (1958, 1965) are numerous. Both defined the intrapsychic development of the infant as dependent upon experiences with the significant others in the environment. Both defined the crucial role of infantile grandiosity and omnipotence as a basic foundation for the development of the self. Winnicott emphasized the importance of empathic mirroring activity of the mother in the development of the structure of the self. Both Winnicott and Kohut emphasized the importance of slow, progressive failures in optimal responsiveness as the building blocks of the development of intrapsychic structures internalized when experiencing optimal frustration. Finally, Winnicott proposed the importance of the internal experience of transitional objects in all creative endeavors just as Kohut proposed the need for continuing selfobject experiences for the achievement of ambitions and goals (Greenberg & Mitchell, 1983).

Kohut (1977, 1984) made two arguments to distinguish his theories from previous theories proposed by interpersonal and object relations theorists. First, Kohut argued that theories and ideas of self psychology were based solely on the intrapsychic experience of selfobject relations rather than on the real interpersonal relations that form the basis of interpersonal and object relations theories. Both interpersonal and object relations theorists have probably been quite taken aback at this line of reasoning. Most of Jacobson's (1964, 1967, 1971) ideas have been based on her understanding of intrapsychic experience as have been Kernberg's (1976, 1980) formulations. Both Jacobson and Kernberg have proposed that the internalization of object relations is the building block of intrapsychic structure. One can only assume that Kohut did not completely understand that their theoretical formulations of the development of intrapsychic structure were based upon relations between self and object just as he formulated that the development of self structure was based upon selfobject relations.

Second, Kohut (1977, 1984) argued that the basic flaw in the theoretical perspective of the object relations theorists was their ideas about separation-individuation and the achievement of autonomy. Kohut emphasized the need for selfobjects throughout the lifespan of the individual and seemed to regard the individuation and autonomy proposed by object relations theorists as a "desolate and schizoid state of total self-sufficiency" (Greenberg & Mitchell, 1983, p. 370). Kohut, himself, wrote that the individual should develop from archaic needs for mirroring, idealizing, and twinship selfobject experiences to a mature reliance on empathic attunement with significant others in the adult environment. The object relations theorists certainly would endorse the idea that healthy human functioning requires continuing relations with others throughout life (Fairbairn, 1952; Jacobson, 1964, 1967, 1971; Kernberg, 1975, 1976, 1980; Mahler et al, 1975).

Greenberg and Mitchell (1983) also criticized Kohut's (1977) idea that, given appropriate selfobject experiences in either normal development or through analysis, the individual possesses the possibility of an essentially conflict-free existence in which the healthy self structure can manifest a spontaneous, joyful, and creative self with internal resources of a sense of humor and wisdom. Greenberg and Mitchell (1983) stated unequivocally that,

Neither of us has known anyone who is entirely free of dependency strivings, greed, envy, separation conflicts, divided loyalty, and so on, and we question Kohut's vision of healthy development. Are these problematic phenomena deterioration products or are they an inevitable part of life? It seems to us more economical conceptually to acknowledge that life, both for the child and for adults, is fraught with struggles and conflicts, no matter how ideal the parenting. (p. 371)

Finally, Greenberg and Mitchell (1983) agreed with Gedo (1989) that there is more to the development of the self and intrapsychic structure than can be explained in the theories of self psychology. The proponents of self psychology seem to claim that most human development is dependent on the formation of self structure and that the child’s development of self is based on the internalization of the empathic mirroring of the parent and/or the opportunity to idealize the parent. They commented that "Kohut has stretched the concept of narcissism past the point of usefulness, setting it apart from other features of interpersonal relations" (p. 371).

Kohut (1977, 1984) proposed that the development of healthy narcissism and the development of mature object love were two separate lines of development. However, his formulations about psychological development were devoted exclusively to the development of a healthy structure of the self and healthy narcissism. He did not describe or explain the transformation of archaic, infantile object love to its mature manifestations of healthy, mature love of significant others in adult life.

It generally is agreed that the major contributions made by Kohut (1971, 1977, 1984) were the propositions about the nature and development of normal narcissism and the conditions that lead to the development of narcissistic personality disorders and his emphasis on the importance of understanding the subjective experience of the self (Gedo, 1989; Greenberg & Mitchell, 1983). Wolf (1988) stated that, just as the ideas of classical psychoanalysis have promoted a more rational attitude toward sexuality, the formulations of self psychology have promoted a more rational attitude about understanding the need for self esteem. Wolf proposed that, in contemporary culture, narcissism has been viewed pejoratively as associated with selfishness and egotism. Wolf stated that, with the contributions of Kohut (1971, 1977, 1984) about the necessity for self esteem for healthy psychological functioning, it is to be hoped that "reasonable people will consider it legitimate to talk about and study narcissism without any pejorative connotations" (p. 28).

Greenberg and Mitchell (1983) stated that Kohut's (1971, 1977) emphasis on the cohesiveness, continuity, and integrity of the subjective experience of the self was theoretically original and has been clinically useful. Of particular importance has been Kohut’s interest in the influence of significant relationships on the development of self-experience. Greenberg and Mitchell also acknowledged the significance of Kohut’s ideas about analytic technique. Kohut’s extensive exploration of the influence of empathic understanding on success in treatment has contributed a great deal to clinical understanding of the importance of the manner, delivery, and timing of interpretation. Furthermore, his formulations about the emergence of infantile grandiosity and idealization in transference as developmentally normal transference phenomena to be understood and eventually explained has provided a clinically valuable alternative to the idea that infantile grandiosity and idealization should be defined as regressive and defensive operations, which should be interpreted as such and thereby resolved.

Another major contribution of Kohut (1971, 1977, 1984) was his willingness to depart from conventional psychoanalytic theory in order to understand and to explain the dynamics and etiology of narcissistic personality disorders. It is important to note that, prior to Kohut's theoretical formulations, patients with narcissistic personality disorders were considered untreatable within an analytic framework. Kohut's ideas about the preoedipal experiences in the development of a cohesive self structure provided a framework for the understanding and treatment of narcissistic personality disorders (Gedo, 1989).

Kohut's ideas about the development of a cohesive self structure and its importance in the regulation of self esteem offer a clinically useful understanding of what might be considered variations on what is commonly understood to be a narcissistic personality disorder with its diagnostic criteria of grandiose flawlessness, a sense of entitlement, and a lack of empathy with others. Whether or not one agrees with the idea that all forms of psychopathology are disorders of the self, individuals in many different diagnostic categories suffer from difficulty in the maintenance and regulation of self esteem. It might be clinically useful in the treatment of any individual with such difficulty to monitor the transference for the emergence of selfobject needs for mirroring, idealizing, or twinship as identified by Kohut.

Wolberg (1988,1995) criticized Kohut (1971, 1977) for his attempts to base a theory of normal infantile development on the observations of the transference reactions of adult narcissistic patients. He also criticized Kohut for using language and concepts that were complex and confusing. He stated that some of Kohut's clinical data are interesting and useful. Wolberg observed that the active use of empathy might be very helpful in the treatment of severely disturbed patients with borderline and narcissistic personality disorders. However, Wolberg commented that the modification of the conventional passive psychoanalytic technique in order to treat severely disturbed patients did not require a complete modification of classical psychoanalytic developmental theory.

In the same vein of thought, a major criticism of self psychology, in general, and Kohut, in particular (1977, 1984), has been the claim that all psychopathology can be understood and treated within the theoretical structure of self psychology. Gedo (1989) particularly criticized the idea that any form of psychopathology could be regarded as simply a developmental arrest that can be cured by the provision of selfobject experiences that repair the deficits in the structure of the self. Gedo argued that Kohut ignored the possibility of the formulation of pathogenic internal structures that must be understood and undone before attempting the restoration of healthy functioning. Kohut indeed did emphasize the importance of explaining or interpreting the dynamics of faulty self structure, but he did not identify or explain these dynamics in any specific or systematic way.

A general weakness of Kohut's (1971, 1977, 1984) theories is the lack of a clear, specific, and systematic explanation of the development of the structure of the self. Kohut proposed that the self structure was built in normal development by transmuting internalization in response to optimal frustration. He further proposed that self structure was composed of a grandiose pole, which formed the core of ambitions, and an idealizing pole, which formed the core of goals or values. He also proposed a tension arc between the two poles comprised of skills and abilities, which forms the program of action the self may take to realize its potential. In his later writing, Kohut (1984) suggested that the selfobject experience of twinship was needed for the realization of the program of action of the tension arc. However, the development and structural formation of the bipolar self with its tension arc or what later self psychologists refer to as a tripolar self (Donner, 1991; Nicholson, 1991; Rowe & Isaac, 1989) remains vague, unclear, and seems to be an arbitrary, artificial, metaphysical construct that confuses more than it clarifies.

All of Kohut's (1971, 1977, 1984) ideas about the development of the self structure and the selfobject experiences needed to form the structure of the self were based upon the analysis of transference phenomena in adult patients. Whether one can clearly understand the developmental needs, experiences, and internal structure of preverbal and prerational children based upon the emotions, needs, and demands that arise in the treatment of an adult patient must be considered. Perhaps more specific and systematic ideas about the development of intrapsychic structure in very young children could be gathered from the observation of infants and small children as was done by Mahler et al. (1975) or by Stern (1985).

One also wonders what effect Wolf's (1988, 1994) identification of seven selfobject experiences has upon the ideas in self psychology about the development of the structure of the self. How do these needed selfobject experiences contribute to the bipolar or tripolar development of self structure? Do self psychology theorists now propose a heptapolar (seven-pole) development of self structure? Wolf, perhaps wisely, did not write about the issue, but the questions remain.

The idea of transmuting internalization also has been rather inadequately explained in self psychology. How exactly does a transmuting internalization differ from other forms of internalization used by the developing child to assimilate the functions, strengths, weaknesses, and pathologies of significant others in the environment?

Furthermore, what is the effect on the developing child of experiences with significant others that do not fit neatly into what Kohut (1971, 1977, 1984) or Wolf (1988) described as selfobject experiences? Gedo (1989) criticized Kohut for attempting to describe and to explain all mental life within the parameters of the vicissitudes of the building of self structure "in various circumstances involving idealization/disillusionment and affirmation/lack of empathy" (p. 423). Gedo suggested that Kohut failed to take into account the richness and complexity of mental life based upon the richness and complexity of human relations in his efforts to fit all intrapsychic experience into a model circumscribed by what he described as the selfobject experiences needed to build a cohesive structure of the self.

Gedo (1989) stated that Kohut (1971) made a significant contribution in his departure from the restrictive paradigm of the transference neuroses. Kohut's ideas about narcissistic personality disorders gave clinical substance to the revision of psychoanalytic theory to take into account data from cases in which the clinical material involves issues different from those in the neurotic cases most familiar to psychoanalysts. Gedo commented that Kohut convincingly demonstrated that the problems of narcissistic personality disorders could be treated in traditional psychoanalysis if the analyst could accept and understand the narcissistic transferences he described.

However, Gedo (1989) rejected the idea that Kohut's (1971, 1977, 1984) data about the transference reactions of narcissistic personality disorders could be used to formulate a new clinical theory with universal applicability. Gedo said that Kohut's efforts to establish a theory with universal applicability resulted in another reductionistic theory that fails to consider the legacy and complexity of all developmental stages. Instead, Kohut seemed to remain fixated on the developmental stage that produced the psychopathology of the particular cases that led to the formulation of his theory. Gedo commented that psychoanalytic clinical theory must integrate the observations of children and adults with all types of personality organization. Both pathological and normal derivatives of each developmental stage must be integrated in a hierarchical manner, which would contain a place for the findings of Kohut. Gedo's (1988) final remark was that "there are more things in heaven and on earth than are dealt with by self psychology" (p. 426).

Chapter 8: Examination and Evaluation of the Theories

As was described in Chapter 1, the purpose in this study has been to identify and to clarify the underlying assumptions and postulates of four major theoretical models of psychotherapy: (a) the psychoanalytic interpersonal school, (b) object relations theory, (c) self psychology, and (d) the Atlanta school of experiential psychotherapy. In previous chapters, each model was discussed. The heuristic, or the method used for gathering data, and the ontology, or the basic assumptions that define the model as a unique theoretical entity were identified and described. The assumptions in each model that were considered to be most important in defining the theory were the assumptions about (a) essential human nature, (b) the nature and development of psychopathology, (c) the role of the therapist, (d) the mechanism of change in psychotherapy, and (e) the goals of treatment.

The goal in this chapter is to identify commonalities and differences in the basic concepts of these theories and to distinguish semantic differences from substantive differences. When substantive differences are identified, an effort will be made to discover whether support for differing theoretical assumptions can be obtained from either empirical data about psychotherapy processes and outcome or criteria in the philosophy of science.

Thus, in this chapter, there will be a brief review and summary of the criteria for the evaluation of scientific theory as specified in the philosophy of science. There will then be a summary of empirical findings from psychotherapy process and outcome research. The theoretical models will be discussed with a preliminary evaluation intended to place the models within an historical and clinical context followed by a thorough and detailed comparison of the theories. The heuristic and ontological assumptions of each theory will be considered separately.

As the heuristic and the ontology of each theory are compared, it will be assumed that some seeming differences in assumptions are based upon semantics or the use of different terminology for similar concepts. Regardless of the specific terminology used in each theoretical model, an attempt will be made to find common underlying meanings.

In some cases, it is assumed that substantive differences will be found among concepts in the different theoretical models. These substantive differences will be examined in the context of research on psychotherapy to establish whether empirical support exists for any given theoretical position. The four theoretical models under consideration then will be evaluated and assessed based on criteria proposed in various philosophies of science. There will be a particular emphasis on assessment of conceptual strengths and weaknesses based on Laudan’s philosophy (1977, 1984).

There then will be a summary of the findings and conclusions drawn from the comparison of the assumptions of the four theoretical models. Finally, recommendations will be made for further research.

Evaluation of Theories: Criteria from the Philosophy of Science

There are currently several philosophers of science who state different criteria for the evaluation of scientific theories. As described by Laudan (1990), the primary philosophical positions concerning theory evaluation are: (a) epistemic relativism in which it is stated that any scientific theory is as accurate and relevant as any other because all scientific theories are subjective; (b) empirical positivism in which it is proposed that scientific theory is accurate to the degree that it is supported by data collected by agreed upon, correct, scientific methods; (c) the realist ontology of science in which it is proposed that theories function successfully when the model approximates the structure of the object under consideration; and (d) the pragmatist's position in which it is argued that the value of scientific theories lies in their observable consequences and their success in organizing experience. Each of these theoretical positions has been discussed by various psychologists and psychoanalysts as appropriate to the evaluation of theories of psychoanalysis and psychotherapy (Adams, 1984; Leary, 1984; Manicas & Secord, 1983; Mulaik, 1984).

Another philosophical position used to evaluate theories in psychotherapy is existential phenomenology. Several psychological theorists have argued that a phenomenological approach is the most appropriate approach to the study of the intrapsychic experiences of the human mind. Phenomenological investigations seek to identify patterns of experience based upon descriptions of experience provided by the subjects of investigation. The goal of phenomenological investigation is to understand rather than to predict or to explain. It is a research method deliberately and explicitly designed to investigate human experience (Dreyfus & Wakefield, 1988; Fisher & Fisher, 1983; Sass, 1988; Woolfolk, et al., 1988).

The criteria for evaluation of the robust, explanatory nature of scientific theory in the investigation of clinical phenomena was described by various authors as (a) the identification and understanding of patterns of human experience (Fischer & Fischer, 1983); (b) the extent to which the theoretical model explains the phenomena under investigation (Manicas & Secord, 1983); (c) the extent to which theoretical formulations describe what works (Mulaik, 1984); (d) the extent to which theoretical formulations are clinically significant (Adams, 1984); (e) the extent to which new theories accommodate the data and successes of previous theories and explain new data not integrated into previous theories (Lakatos, 1978); (f) the simplicity of the theory; (g) the rigorous, well-defined terminology of the theory; and (h) the extent to which the theory stimulates growth, progress, and competition in theoretical development (Laudan, 1977, 1984).

A Summary of Empirical Data about Psychotherapy Process and Outcome

Orlinsky and Howard (1986) found that therapeutic conditions, therapeutic interventions, and other factors are correlated significantly with success in treatment across a collection of controlled outcome studies. The significant therapeutic conditions were (a) role preparation or clear communication to the patient of the expectations about the patient's contributions to the treatment process, (b) verbal activity of the patient in sessions, and (c) the establishment of a collaborative therapeutic relationship. A collaborative therapeutic relationship was described as the therapist's encouragement of the patient's initiative and the patient's assumption of an active role in the resolution of problems rather than therapeutic relationships in which the therapist is directive and the patient compliant and/or dependent. Therapeutic interventions that are correlated with success in treatment were (a) confrontation, which was defined as any activity of the therapist that raises the self awareness of the patient; (b) focus on the patient's affect; (c) focus on the transference; and (d) interpretation. Other factors that are correlated with success in treatment were (a) the patient's expression of negative affect, particularly in early sessions; (b) the therapist's engagement and confidence; (c) the empathy, warmth, and acceptance of the therapist; and (d) the length of time in treatment.

Other interesting data from psychotherapy research have been gathered by Strupp (1980a, 1980b, 1980c, 1980d), who found that patients who did not succeed in therapy did not relate well to the therapist, kept interactions on a superficial level, and/or had more hostile interactions with the therapists than did successful patients. The therapists observed in these treatment situations did not modify their approach in working with the unsuccessful patient but instead provided essentially the same responses and interventions as had been provided to the successful patient. Strupp et al. (1988) proposed that work with these "more difficult" patients might be more successful if the therapist would modify his or her treatment approach.

A Preliminary Integration and Evaluation of the Theories of Psychotherapy

Before attempting to identify the major areas of agreement and disagreement among the theories, it may be helpful to place the theories and their originators into historical and clinical context. The ideas discussed in this study represent 70 years of theoretical development in a rapidly changing cultural and theoretical milieu. In the early part of this century, classical Freudian concepts dominated the field of psychiatric treatment (Greenberg & Mitchell, 1983). Many of these earlier theories were attempts to explain clinical phenomena that were not accounted for in classical analytical thought. Most of these theorists were concerned with understanding dynamics in development and treatment that seemed interpersonal in nature rather than instinctual (Fairbairn, 1954; Klein, 1964, 1975; Sullivan, 1953, 1954, 1956, 1971; Whitaker & Malone, 1953/1981; Winnicott, 1958, 1964, 1965, 1971, 1989).

Sullivan's (1953, 1954, 1956, 1971) writings were collected from papers originally published after 1923. Much of his early clinical work was with schizophrenic patients whom he thought had a severe disturbance in their capacity to relate to others. Sullivan's efforts to understand schizophrenia led to the development of his ideas about the importance of interpersonal relations both in development and treatment (Chapman, 1976; Greenberg & Mitchell, 1983).

Klein (1964, 1975) first published articles in 1919. Many of her theoretical formulations were based on attempts to understand the factors involved in the development and treatment of preoedipal children and severely disturbed borderline and psychotic patients. Fairbairn (1954) began to develop his ideas in the 1930s as did Winnicott (1958, 1964, 1989). Both Fairbairn and Winnicott were concerned with interpersonal dynamics in the development and treatment of severely disturbed patients. Winnicott, a pediatrician, was interested in integrating his observations of mother-infant interaction with analytic theories of development. Both Fairbairn and Winnicott were influenced by Klein's ideas about the development of object relations in the very young child.

Whitaker and Malone (1953/1981) started developing their ideas in the 1940s. They stated that their motivation in developing their theories was a desire to identify and to explain interpersonal processes that are common to all forms of psychotherapy. Sullivan (1947) was listed in their bibliography, and thus, it may be assumed that his ideas had some influence on the development of their ideas.

These theorists were practicing in widely different locations and treated different types of patients. Accordingly, there are some interesting similarities in the ideas of Sullivan (1953, 1954, 1971), Fairbairn (1954), Winnicott (1958, 1964, 1965, 1971, 1989), and Whitaker and Malone (1953/1981), particularly in their ideas about the importance of interpersonal processes in development and treatment. They agreed that essential human nature is fundamentally invested in the need and the search for satisfying, healthy, interpersonal relationships. They agreed that normal development is crucially dependent upon optimal gratification and responsiveness of the parents to the child. They also agreed that psychopathology develops as a result of deficits in the parents’ responsiveness to the needs of the child. Finally, they all agreed that the therapeutic relationship is, at least, part of the curative factor and mechanism of change in psychotherapy. It is of interest to note that these theorists had similar reactions to classical analytic theories, which contain a more negative view of the nature of humans than did their theories, an emphasis on the role of instinctual drives in development, and a focus on interpretation and insight as the mechanism of change in treatment.

The theories of Jacobson (1964), Mahler (1972, Mahler et al., 1975), Kernberg (1976, 1980, 1984), and Kohut (1971, 1977, 1984) were developed from the 1950s through the 1980s and were based upon increasingly more complex and detailed data about normal development and about patients in treatment. These theories included a consideration of both innate and environmental factors in the essential nature and development of the human individual and a recognition of organic factors in the development of psychopathology, particularly the psychoses. The curative factors in psychotherapy were thought to be based upon both interpersonal dynamics in the therapeutic relationship and interpretation leading to insight.

Both Kernberg (1977, 1980, 1984) and Kohut (1971, 1977, 1984) developed extensive theoretical models to integrate data about severely disturbed borderline and narcissistic patients into psychoanalytic theories of development and treatment. The models presented by Kernberg and Kohut are the most comprehensive of all the theories that have been considered in this study. In accordance with the ideas of Lakatos (1978) about research programs that new theories must accommodate the data and successes of previous theories and must explain new data, the theories of both Kernberg and Kohut are progressive within the research program of analytic theory and thought. In an historical context, their theories have been the most recently developed, which allows the integration of the most recent data.

It also should be noted that although the origins of interpersonal psychoanalysis lie in the work of Sullivan (1953, 1954, 1956, 1971), both clinical and theoretical efforts from the interpersonal perspective have been produced up to the present time. The theories of interpersonal psychoanalysis are less comprehensive in their explanation of intrapsychic structure and development among different diagnostic categories than are the theories of Kernberg (1976, 1980, 1984) and of Kohut (1971, 1977, 1984). However, according to the interpersonal theorists, there is no need for elaborate, complex theorizing about hypothetical, metaphysical intrapsychic constructs or their development or for the identification of diagnostic categories. The interpersonal theorist and therapist are concerned with understanding the actual interpersonal processes that occur between patient and therapist and in helping the patient understand and change interpersonal processes (Bromberg, 1989; Levenson, 1983; Stern, 1987).

A similar argument may be made for the Atlanta school of experiential psychotherapy. The ideas developed by Whitaker and Malone (1953/1981) have been used clinically and recently have been reexamined theoretically by Gantt (1984) and Felder and Weiss (1991). The focus in experiential psychotherapy is on the therapeutic experience rather than on diagnosis or hypothetical intrapsychic structure. The primary concern in the experiential model is an understanding of the actual processes that occur in psychotherapy and of the interpersonal factors considered to be the mechanism of change.

Heuristics

In most of these theories, the primary method of data collection was the observation of patients and/or descriptions of the subjective experience of the therapist. The interpersonal therapists based most of their ideas about patients on their observations of the interpersonal patterns that develop in the therapeutic relationship (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982). Fairbairn's (1954) and Klein's (1964, 1975) ideas were based on observations of patients. Whitaker and Malone (1953/1981) and Felder and Weiss (1991) derived their formulations from experiences with patients in treatment.

Kohut (1971, 1977, 1984) derived his ideas from his empathic immersion in the experience of patients with narcissistic personality disorder. He also argued that the use of empathy and introspection for observations and data collection was the essence of psychoanalytic investigation. Goldberg (1988) stated that Kohut's hypotheses about the dynamics and appropriate treatment of narcissistic patients have been confirmed by their improvement in treatment.

Winnicott (1958, 1964, 1965, 1971, 1989) developed his ideas through attempts to integrate his observation of patients, his observations of normal mother-child interaction, and psychoanalytic theories about development and treatment. As has been stated earlier, Winnicott did not develop an integrated theoretical model encompassing normal development, the origins of psychopathology, and methods of treatment. However, his essays about mother-infant interaction and psychoanalytic experiences often beautifully capture moments of human experience and illuminate them with enriched depth and meaning.

Jacobson's (1964, 1971) theoretical formulations were an attempt to integrate theories about the development of object relations with drive/structure theory. Some of her ideas, particularly her ideas about the management of transference with severely disturbed patients, were drawn from her clinical experience. However, many of her formulations were based on a unique synthesis of previous theories of intrapsychic development.

Mahler's (1972, Mahler et al., 1975) theoretical model was derived from and supported by a longitudinal study in which mother-child interaction was observed. Kernberg (1976, 1980, 1984; Kernberg et al., 1972) developed his theories on the basis of information from several sources including (a) the psychotherapy research project of the Menninger Foundation, which was a systematic, detailed study of the treatment of borderline patients over several years; (b) the theories of Mahler (1972, Mahler et al., 1975) about the normal stages of development in early childhood, which, in turn, were derived from her longitudinal study; (c) some of the ideas of Klein (1964, 1975) and Fairbairn (1954) about object relations and primitive defenses, and (d) many of the ideas of Jacobson (1964, 1971) concerning the interacting effects of internalized object relations, drives, and the development of ego and superego. In the exposition of his theories, Kernberg also integrated, referred to, and gave credit to the ideas of many other analytic theorists far too numerous to cite.

Evaluation: The heuristics of these various theoretical positions differ in the extent to which observation of normal mother-child interaction is integrated with the data taken from controlled studies and the extent to which either is integrated, if at all, with patient observation. In several theories, assumptions were based primarily upon the observations of patients and on the subjective experience of the therapist. The theories also differ in the degree to which theorists integrated previous theories of psychological development and treatment with their own theory.

According to the criteria for theory evaluation taken from the philosophy of science, the following propositions should be noted. (1) The heuristics of all of the models discussed made some contributions to understanding patterns of human experience in development and/or in the process at psychotherapy. Fairbairn (1954) and Klein (1964, 1975) contributed alternative theories of normal and abnormal human development that are alternatives to classical Freudian psychoanalysis, particularly in the development processes that lead to severe psychopathology. Sullivan (1953, 1954, 1956, 1971) and Winnicott (1958, 1964, 1965, 1981) offered original and insightful ideas about patterns of experience both in development and in the process of psychotherapy. Whitaker and Malone's (1953/1981) primary original contributions were descriptions of patterns of human experience in the process of psychotherapy. Thus, the heuristic of patient observation used by these theorists met the phenomenological criteria for theory evaluation (the identification and understanding of human experience) described by Fischer and Fischer (1983). The specific strengths and weaknesses of the ideas formulated will be discussed more fully in the sections concerned with the specific concepts described by each theorist.

(2) Kohut (1971, 1977, 1984) used the heuristic of empathic immersion in the experience of patients with narcissistic personality disorders, which allowed him to identify patterns of human experience in development and in psychotherapy, thus satisfying Fischer and Fischer's (1983) criteria for theory evaluation. He also explained transference phenomena in narcissistic patients and the successful clinical management of such transference phenomena, thus also satisfying the theoretical criteria proposed by Manicas and Secord (1983) of explaining the phenomenon under investigation, by Mulaik (1984) of describing treatment modifications that work, and by Adams (1984) of offering clinically significant formulations. On the other hand, Kohut's heuristic is based almost exclusively on emphatic understanding of patients and does not include a thorough consideration or integration of the successes of previous theories, for which omission he has been criticized several times (Akhtar, 1989; Gedo, 1989; Summers, 1994). Thus, Kohut's heuristic satisfied Lakatos's (1978) criteria that new data be explained without meeting the further stipulation that new data be integrated with previous successful theories.

(3) Of all the heuristics discussed, Kernberg's (1976, 1980, 1984) heuristic was the most comprehensive in scope and involved a combination of theoretical integration of research studies, observation of patients, and integration of previous theoretical ideas. His theoretical formulations about the dynamics and treatment of borderline patients and other severely disturbed patients within the context of his ideas about normal and abnormal development meet most of the criteria for theory evaluation previously described. He described patterns of human experience, which satisfied criteria suggested by Fischer and Fischer (1983), explained the phenomena of borderline personality organization, which satisfied criteria suggested by Manicas and Secord (1983), and proposed clinical approaches to the effective treatment of borderline conditions, which satisfied criteria suggested by Mulaik (1984) and Adams (1984). Kernberg also made extensive efforts to accommodate the data and successes of previous theories while explaining new data, which satisfied criteria suggested by Lakatos (1978).

Ontology

Essential human nature

In these theories, there are two different conceptualizations of the essential nature of the human individual. The first is that the individual’s essential nature is predisposed to healthy functioning, and given appropriate environmental support, a healthy, well-functioning individual will develop. The second is that different individuals are predisposed to the manifestation of aggressive drives and/or libidinal drives that are actualized and given substance by environmental circumstances, which in combination determine adaptation and healthy or pathological functioning.

The most extreme position was taken by Klein (1964, 1975). She stated that the infant has powerful innate drives. Aggressive drives emanate from an innate death instinct. The new born infant is threatened by paranoid anxiety aroused by the prevalence of hostile, destructive, and aggressive instincts. Experiences of gratification arouse the innate life instinct, but libidinal internalization of the good, gratifying object can be disturbed by a predominance of innate, aggressive instincts. However, even with Klein's emphasis on the influence of innate aggressive drive, she also made it clear that both libidinal and aggressive drives are greatly influenced by environmental factors. Innate aggressiveness is reinforced by frustrating experience, while innate libido is reinforced by gratifying experience. Instinct and environment interact to develop a predominance of either good or bad object relations.

The most striking similarities in theoretical assumptions are in the ideas of Fairbairn (1954) and Sullivan (1953, 1954, 1956, 1971). Both agreed that the human individual has an innate tendency to seek, to maintain, and to secure satisfying interpersonal relationships. Sullivan proposed an inherent tendency toward health dependent upon satisfactory interpersonal relations. Throughout life, the individual seeks interpersonal relations that will rectify unpleasant experiences in previous relationships thereby removing obstacles to continuing progress toward health. Fairbairn reformulated the concept of libido to mean object seeking rather than pleasure seeking. Aggressive impulses only emerge in response to frustration. Thus, the infant is solely motivated, from birth, toward achieving healthy, satisfying relationships.

The idea that essential human nature is predisposed to health through participation in satisfying relationships is a basic assumption of several other theorists. Winnicott (1958, 1964, 1965, 1971, 1989) repudiated the concept of an innate death instinct and proposed that destructive aggression only arises in response to frustration. Normal aggression was described by Winnicott as a healthy need to grow, to explore, and to move. The proponents of the Atlanta school of experiential psychotherapy also said that the human individual is innately disposed toward growth and healthy relationships. Healthy aggression was described as the normal efforts to gratify needs, and destructive aggression was described as a response to frustration (Felder & Weiss, 1991; Whitaker & Malone, 1953/1981).

Kohut (1971, 1977, 1984) stated that the human child has an innate need to develop a cohesive structure of the self. Because the self can only be developed through the internalization of selfobject functions, it must be presumed that, in agreement with Fairbairn (1954) and Sullivan (1953, 1954), Kohut claimed that the human individual has a primary, innate need for satisfactory relationships. Kohut also proposed that destructive aggression was only a response to the frustration of normal, healthy, narcissistic needs.

Jacobson (1964, 1975), Mahler (1972, Mahler et al., 1975), and Kernberg (1976, 1980, 1984) integrated the concept of innate aggressive and libidinal drives with the development of object relations. Pleasant and unpleasant interpersonal experiences actuate impulses derived from libidinal and aggressive drives. In turn, good and bad fused selfobject representations become invested with libido and aggression and are correspondingly idealized and devalued. As the child forms whole self and object representations, parental love promotes the investment of libido in self and object. Appropriate demands and frustrations encourage the child to direct aggression toward the frustrating objects and to invest libido in the self, fostering the development of healthy narcissism. Healthy narcissism supports the development of autonomous ego functions with concomitant pride in realistic achievements. Realistic achievements, in turn, support the continuing healthy narcissistic endowment of the ego.

As has been described previously, the development of object relations, ego and superego autonomy, and manifestations of libidinal and aggressive drives mutually influence one another. The maturation of intrapsychic structure into an autonomous ego and superego based on complex self and object representations releases libidinal and aggressive energy that can be used for rapid progress in social, intellectual, cultural, and physical activities.

Kernberg (1976, 1980, 1984) proposed that the development of healthy, normal object relations that provide the foundation of higher-level intrapsychic structure is dependent on an interaction between the genetic endowment of the individual and environmental influences. Pathology may result from innate excessive or restrictive affects or from a lack of optimal responsiveness in the environment or from an interaction between innate and environmental factors.

Thus, there is a substantive disagreement among these theories. The majority of the theorists agree that human individuals are primarily motivated to seek and to maintain satisfactory relationships. Many of the theorists proposed that destructive aggression is a response to frustration (Fairbairn, 1954; Kohut, 1977, 1984; Sullivan, 1953, 1954, 1956, 1975; Whitaker & Malone, 1953/1981; Felder & Weiss, 1991). Winnicott (1958, 1964, 1989) and the experiential therapists described a healthy type of innate aggression that is manifested in movement, exploration, and the efforts to gratify needs. On the other hand, several theorists proposed that humans are primarily and innately motivated by affect states that are linked to pleasurable and unpleasurable experiences, which activate libidinal and aggressive drive derivatives that, in turn, become intricately invested in and influenced by the development of object relations and later intrapsychic structures. In the psychoanalytic literature, the various impulses that arise from aggressive and libidinal drives are usually referred to as drive derivatives (Jacobson, 1964, 1975; Kernberg, 1976, 1980, 1984; Mahler, 1972; Mahler et al., 1975).

Evaluation: There are two substantive disagreements among the theorists about essential human nature. The first disagreement concerns the basic motivation of the human individual. The second disagreement concerns the definition and description of an innate aggressive drive. Each disagreement will be considered separately.

1. It has been hypothesized that an essential, primary motivation of human behavior is a drive to establish healthy, satisfactory interpersonal relations with others. Destructive aggression is a response to traumatic frustration (Fairbairn, 1954; Felder & Weiss, 1991; Kohut, 1971, 1977, 1984; Sullivan, 1953, 1954, 1956, 1971).

2. It has been hypothesized that the motivations for human behavior are complex interactions of innate libidinal and aggressive drives that are often, but not always, activated in response to interpersonal experience (Jacobson, 1964, 1971; Kernberg, 1976, 1980, 1984; Mahler et al., 1975).

From the realist theoretical position described by Manicas and Secord (1983), the success of a theoretical formulation is the extent to which it explains the phenomena under investigation. From this realist perspective, it seems doubtful that the assumption that the human individual is primarily motivated to have successful and satisfactory interpersonal relations can adequately explain the complexity and diversity of human behaviors. Without even considering destructive human behavior, intellectual curiosity, creativity, or even the pleasure small children have in their physical development can not be explained as being based on a primary drive to establish interpersonal relationships. Children climb trees, jump on beds, and solve puzzles, and adults climb mountains, jump out of airplanes and do scientific research probably, at least in part, in order to explore the environment and to gratify feelings of mastery and personal achievement. Philosophically, it would seem that there is more to human motivation than a desire to achieve and to sustain interpersonal relations.

However, the question is not resolved. Innate human motivations also may be based upon more complex interacting needs than can be explained by manifestations of libidinal and aggressive/assertive strivings. Observations of infants suggest an inherent motivation toward interpersonal relations (Mahler et al., 1975; Stern, 1985). Observations of infants also indicate that they seek stimulation from the beginning weeks of life and will interrupt feedings and social interaction to look, to listen, and to orient toward interesting visual or auditory stimuli (Stern, 1985). Thus, these observations of infants support the hypothesis that human children are powerfully motivated to form and to sustain interpersonal relations and also support the notion that human motivation is more complexly determined then can be fully explained by the need for interpersonal relations. Thus, even if libidinal drives were redefined as the drive to seek satisfying interpersonal relations and other human motivations emanating from the aggressive drive, how is the aggressive drive defined and explained? This leads to the second substantive disagreement among the theorists.

1. The aggressive drive is an innate instinctual drive to express anger, hostility, frustration, and/or destructive impulses (Jacobson, 1964; Kernberg, 1976, 1980, 1984; Klein, 1964, 1975; Mahler et al., 1975). Kernberg (1984) specifically described the aggressive drive as a manifestation of an innate capacity to hate as opposed to the libidinal drive described as a manifestation of an innate capacity to love.

2. Winnicott (1958, 1964, 1965, 1971, 1989) proposed that normal healthy aggression can be observed in the child's inherent needs to move and to explore, which is similar to the ideas of Whitaker and Malone (1953/1981) that healthy aggression is the normal striving to gratify needs. Aggression is defined as an innate need for movement, achievement, mastery, and/or the successful striving to gratify needs. Destructive aggression is a response to traumatic frustration.

Given the extent of human violence both in current times and historically, it would be difficult to dismiss conclusively the idea that humans have an innate tendency to express anger and/or destructive hostility. The realist philosophers of science (Manicas & Secord, 1983) stated that the success of a theoretical formulation is the extent to which it explains the phenomena under investigation. The hypothesis that humans have an innate aggressive drive, which may or may not be brought under various levels of control by the socialization processes of childhood, may explain the varieties of manifestation of aggression.

On the other hand, the idea that destructive aggression exists only as a response to traumatic frustration also can explain adequately the varieties of expressions of human aggression. However, if one accepts this idea, it would seem that one has assumed an innate human capacity for aggression, which only is exhibited in the conditions of traumatic frustration.

These assumptions can then be restated.

(1) There is an innate human aggressive drive that differs from individual to individual at birth.

(2) There is an innate human capacity for aggression that differs from individual to individual dependent upon the degree and extent of the experience of environmental trauma and frustration.

(3) There are perhaps differing innate capacities for aggression that develop and emerge in response to differing amounts of environmental trauma and frustration.

With one exception, the empirical data from psychotherapy research do not provide useful findings to resolve this conflict in these three assumptions. The exception is the research finding that the establishment of a collaborative therapeutic relationship has a significant effect on outcome in treatment. A collaborative therapeutic relationship was described as one in which the therapist encourages the patient’s initiative and the patient takes an active role in his or her treatment. This particular finding provides some support for the assumption of an innate need for assertive, aggressive strivings from the individual.

The criteria for theory evaluation provide some basis for evaluating the strength of each of these three positions. Observations of infant behavior indicate that infants differ from birth in some behaviors that are characteristic of temperament, although none of these differing characteristics has been identified as being indicative of an innate aggressiveness (Stern, 1985). However, once again from the realist position in the philosophy of science as described by Manicas and Secord, (1983), one might propose that the third assumption most adequately explains the phenomena of human aggression. If one assumes differing amounts of innate aggression interacting with the influence of differing degrees of traumatic frustration, one can explain the fact that individuals respond differently to similar traumas.

The possibility that humans may have innate and differing capacities for anger and/or destructive aggression does not preclude the possibility that humans also may have the type of assertive or aggressive drive described by Winnicott (1958, 1964, 1971, 1989) and Whitaker and Malone (1953, 1981). Once again from the realist position in the philosophy of science as discussed by Manicas and Secord, (1983), the assumption that there is a human drive for achievement, accomplishment, mastery, success, and/or gratification of needs contributes to understanding the complexity of human motivation.

In conclusion, in accordance with the realist position that a theory may be evaluated as successful or unsuccessful dependent upon the extent to which the theoretical model explains the phenomena under investigation, human motivation might best be understood through an integration of the theoretical positions considered. Humans, thus, may be assumed to be motivated by an innate need to seek and to maintain interpersonal relationships and by varying degrees of innate aggressive potential, which may be displayed both in anger and/or destructive behavior in response to frustration or trauma and also may emerge as strivings for achievement, success, and mastery.

Normal Development

With the exception of the experiential theorists (Felder & Weiss, 1991; Whitaker & Malone 1953/1981(who did not provide a theory of normal development)), all of these theorists proposed normal stages of development with different developmental tasks at each stage. Interpersonal psychoanalytic theory, self psychology, and object relations theory provide theories of normal development from infancy to adulthood that have substantial agreements as well as substantive disagreements (Felder & Weiss, 1991; Kernberg, 1976, 1980, 1984; Kohut, 1971, 1977, 1984; Jacobson, 1964, 1971; Mahler et al., 1975; Sullivan, 1953, 1954, 1956, 1971; Whitaker & Malone, 1953/1981).

There seem to be areas of disagreement in the developmental theories of the object relations theorists, but many of these disagreements that concern timing of developmental stages have been resolved by Mahler's (Mahler et al., 1975) observations of the stages of development. Klein's (1964, 1971) and Fairbairn's (1954) ideas that some sort of organized ego is present from birth has not been supported by observation of normal infants (Mahler et al., 1975; Stern, 1985).

In general, the object relations theorists have described preoedipal development as a time when internalization of significant objects contributes to and/or provides the basis for intrapsychic structure. All of these theorists hypothesize early gross introjections of the objects with splitting used to separate and to protect good fused selfobject representations from bad fused selfobject representations. Further development leads to increasingly selective identifications, separation of self and object, and integration of good and bad self and object representations. In general, as data were collected, object relations theories of development became increasingly complex and sophisticated in order to accommodate previous theoretical formulations and to integrate new data and observations culminating in Kernberg's (1976, 1980, 1984) comprehensive theory (Fairbairn, 1954; Klein, 1964, 1971; Jacobson, 1964, 1971; Mahler et al., 1975; Winnicott, 1958, 1964, 1965, 1971, 1989).

Kernberg (1976, 1980, 1984) described a complex, intricate interaction between self and object representations, libidinal and aggressive drives, and development of ego and superego autonomy in normal development. Emotional maturity is based on the capacity to discriminate subtle aspects of both self and others and on selectivity in accepting and/or internalizing the qualities of others. Intrapsychic object representations provide love, confirmation, support, and guidance to the self-enhancing autonomous ego functions that can be put to optimal use in social, intellectual, cultural, and physical activity. The intrapsychic and interpersonal worlds of the individual mutually support, influence, reinforce, and modify each other throughout life. Kernberg (1976, 1980, 1984) based his theory of development on the ideas of most of the other major object relations theorists (Fairbairn, 1954; Klein, 1964, 1975, Jacobson, 1964, 1971; Mahler et al., 1975) but did not include or discuss explicitly Winnicott's (1958, 1964, 1965, 1971, 1989) ideas of development in infancy.

Winnicott's (1958, 1964, 1965, 1971, 1989) hypotheses were based primarily on his observations of mother and infant dyads in his pediatric practice. Winnicott described in detail processes that he called (a) absolute dependence (equivalent to the symbiotic stage) and (b) relative dependence (equivalent to the stage of separation/individuation). Winnicott also identified and described several psychological conditions and processes that occur in mothers, infants, and the mother/infant dyad. According to him, primary maternal preoccupation promotes the development of the appropriate empathic response to the infant of the good-enough mother. He described the sense of hallucinatory omnipotence of the infant, which provides a basis for a healthy sense of self. He suggested the existence of what he called the “holding environment,” which contains and manages the infant’s experience while protecting it from impingement, allowing the infant to experience the state of "going on being" from which the spontaneous feelings and needs of the true self can emerge. He described the mirroring function of the parents, which supports the development of self acceptance and appreciation. He explained the mother's graduated failure of adaption, which facilitates separation and differentiation and the development of ego functions. He described the development of the capacity for concern in the infant, which develops into a healthy concern for others if the mother demonstrates her capacity to survive the infant's aggression and allows the child opportunities to make reparation. He proposed the concept of the “transitional object” as a symbolic replacement of the omnipotently controlled mother of early infancy. Transitional objects provide comfort in the subjective world of internal experience and facilitate the growing awareness and acceptance of external reality.

According to interpersonal theory, normal development proceeds from infancy to adulthood with a continuing need for satisfying interpersonal relationships. There is splitting of the self and other in infancy followed by later integration, the nature of which is determined by the predominant nature of the child's interpersonal experience. The individual's expectations about, and experiences of, interpersonal relations mutually influence, modify, and shape one another. In psychoanalytic interpersonal theory, no assumptions are made about the development or nature of intrapsychic structure. Personality is determined by interpersonal experience, and satisfying interpersonal relations can at any time in life correct problems developed through previous unsatisfying interpersonal relations (Bromberg, 1989; Cashdan, 1982; Kiesler, 1982; Sullivan, 1953, 1954, 1956, 1971).

Kohut (1971, 1977, 1984) also argued for a need for interpersonal relationships throughout life to sustain and to nourish the self. His statements about the normal development of intrapsychic structure are primarily concerned with the development of the bipolar self through the transmuting internalization of mirroring and idealizing selfobject functions. He argued that the development of a cohesive self structure is a separate line of development from the maturation of object relations.

Thus, two substantive differences in theories about normal development can be identified. (a) From the interpersonal perspective, the hypothetical intrapsychic structures and processes described by Kernberg (1976, 1980, 1984) and Kohut (1971, 1977, 1984) are not necessary to the understanding of human personality and behavior. (b) Although both Kernberg and Kohut agree that intrapsychic processes and structure are useful in understanding personality and behavior, there are substantial disagreements in their descriptions of the intrapsychic processes involved. Kohut proposed that self structure and object love evolve from primitive to mature forms in two separate lines of development. Kernberg argued that self structure invested with healthy narcissism develops concurrently and is interwoven with the maturing of object relations.

Evaluation: The conflict in opinion about the usefulness of hypothesizing intrapsychic structure and processes is clearly a substantive difference in the theoretical positions that have been discussed. Proponents of the interpersonal perspective have criticized other theories of psychotherapy for inventing hypothetical, metaphysical constructs such as the ego and superego, the bipolar self, projective identification, transmuting internalization, et cetera (Bromberg, 1989; Stern, 1987; Sullivan, 1953).

Sullivan (1953) proposed a theory of development that is focused on interpersonal processes rather then on intrapsychic ones. He proposed that, from infancy onward, the developing personality and expectations about relationships are based upon interpersonal interactions. As the mothering one approaches the infant with varying degrees of tenderness or anxiety, the child develops a prevailing personification of the mother as either predominantly good or bad and a corresponding approach to future interpersonal relationships. The child also develops a self-system in response to interpersonal experience composed of a good-me, bad-me, and not-me. Throughout life, interpersonal relationships continue to influence both the self system and expectations about relationships.

It should be noted that Sullivan (1953) proposed several hypothetical intrapsychic structures: the personification of the good or bad mother, the set of expectations about future relationships, and the self-system composed of the good-me, bad-me, and not-me. Thus, even within the interpersonal perspective, there is an implicit acknowledgment that it is difficult to explain human behavior without making an assumption that experience creates or molds internal processes and structure that influence later behavior, expectations, and experiences.

In terms of theory evaluation, Sullivan (1953) identified and described some patterns of human experiences in his theory of development that satisfy criteria suggested by Fischer and Fischer (1983), but it is questionable that his developmental theory completely explains the phenomena under consideration, and thus, his theory does not satisfy criteria suggested by Manicas and Secord (1983). As was mentioned previously, human behavior is too complex and diverse to be completely described in terms of the influence of interpersonal relationships.

However, at the time Sullivan (1953) was developing his ideas, his theory was an attempt to explain his observations, that is, the importance of interpersonal interaction and relationships on psychological functioning, which had not been explained in previous classical analytic theory. Evaluating Sullivan’s developmental theory by Laudan’s criteria one finds the following. (a) There is a simplicity to Sullivan's theory; it is not overly complex. (b) There are several terms, such as personification, self system, and prototaxic and parataxic styles of thinking, that are poorly defined and described. (c) Sullivan's (1953) theories of development do not seem to have stimulated subsequent progressive theorizing about human development. None of the interpersonal theorists reviewed has elaborated or even much commented upon his theory of development (Laudan, 1977, 1984).

Kohut's (1971, 1977, 1984) theories of development are focused upon the development of the self. Most of his ideas concerning the development of the self in the growing baby and child are based upon assumptions that the needs and feelings aroused in transference reactions of analytic patients reflect the needs and feelings of the small child. Kohut made little attempt to integrate his ideas with previous theories and stated that the development of object love and object relations was a separate line of development from the development of a cohesive self structure. He stated that his primary contribution to analytic theory was his identification and explanation of the development of the self and the treatment of self-disorders. The evaluation of Kohut's (1971, 1977, 1984) theory of development will be reserved for consideration along with the theories of development taken from object relations theories (Kernberg, 1976, 1980, 1984; Winnicott, 1958, 1964, 1965, 1971 1989).

Kernberg's (1976, 1980, 1984) theory of development, as has been previously mentioned, integrated some of the ideas of Fairbairn (1954) and Klein (1964, 1975) with the theoretical models of Jacobson (1964, 1971) and Mahler (1972; Mahler et al., 1975). The theoretical model of Mahler was based upon a longitudinal study of small children. Kernberg's theory of development, thus, extensively accommodated the data and successes of previous theories although he did not thoroughly consider or integrate Winnicott's (1958, 1964, 1965, 1971, 1989) ideas and formulations. He largely satisfied the criteria for theory evaluation proposed by Lakatos (1978).

In Kernberg’s (1976, 1980, 1984) theory of development, he also thoroughly examined and explained most of the phenomena under investigation while describing intricate, interlocking patterns of human experience. As has been previously described, Kernberg, building upon the formulations of Jacobson (1964) and Mahler (1972), described a complex interaction between normal stages of development, interpersonal experience, the activation of libidinal and aggressive drives, the maturation of primitive merged selfobject experiences into differentiated whole self and whole object representations, and the development of an autonomous ego and superego. Parental love promotes the investment of libido in self and object which contributes to the development of object love and healthy narcissism. Healthy narcissism promotes the development of autonomous ego functions with resulting realistic achievements, which, in turn, promote further the healthy narcissistic endowment of the ego. The establishment of mature and autonomous ego and superego based on complex self and object representations releases libidinal and aggressive energy, which then can be used for social, intellectual, cultural, creative, or physical accomplishments.

Kernberg's (1976, 1980, 1984) emphasis in his theory of development was on the growth and maturation of intrapsychic structure and processes. Pleasant and unpleasant interpersonal experience was described by him as having a profound influence on the development of intrapsychic structure, but Kernberg did not describe in detail the nature and influence of specific interpersonal interactions. Kernberg's theories of development, thus, may be complemented by some of the ideas presented by Winnicott (1958, 1964, 1965, 1971, 1989) and Kohut (1971, 1977, 1984), which are most concerned with specific interpersonal interactions between parent and child and the effect of those experiences on intrapsychic structure. Winnicott's ideas about primary maternal preoccupation, hallucinatory omnipotence, the holding environment, "going on being,” and the mirroring function of the parents can easily be incorporated with Kernberg's ideas about the internal processes at the symbiotic stage of development. Winnicott's ideas about the mother's graduated failure of adaptation, the development of the infant's capacity for concern, and transitional object similarly do not contradict any of Kernberg's hypotheses about the processes occurring in the separation/individuation stage of development. Winnicott's ideas about development may be considered to enrich and to expand Kernberg's hypotheses.

There are obvious similarities between Winnicott's (1958, 1964, 1965, 1971, 1989) formulations and many of Kohut's (1971, 1977, 1984). Winnicott's ideas about primary maternal preoccupation, which promotes the development of maternal empathy, hallucinatory omnipotence, and the holding environment are very similar to Kohut's emphasis on the importance of an empathic parental response and the small infant's need for merger type experiences. Winnicott's description of the parent’s mirroring function as providing the basis of self acceptance and appreciation does not seem to differ drastically from Kohut's idea that the parent mirrors the child's need for admiration and approval. Winnicott's idea that the mother's graduated failure of adaptation leads to the development of ego functions and the separation and differentiation of the self is very similar to Kohut's idea that optimal, nontraumatic frustration stimulates the internalization of selfobject functions and the development of self structure. Finally, Winnicott's ideas about the development and emergence of the true self from an appropriate holding environment are similar to Kohut's ideas about the development of cohesive self structure through the experience of an appropriately responsive interpersonal environment. Kohut's theories about the development of a cohesive self were much more complex than Winnicott's description of the development of the true self.

Thus, the similar parental behaviors and their effects described by both Winnicott (1958, 1964, 1965, 1971, 1989) and Kohut (1971, 1977, 1984) are not in conflict with Kernberg's (1976, 1980, 1984) theory of development and could conceivably enhance his hypotheses about intrapsychic processes by providing a description of interpersonal processes that affect intrapsychic structure. Similarly, Kohut's ideas about idealizing self object experience could provide a description of interpersonal experience that could influence the development of ideal self representations, ideal object representations, and the ego ideal.

On the other hand, many of Kohut's (1971) formulations about development are vaguely defined, overly complex, and/or confusing. “Selfobject experience” is a term used repeatedly to describe real interpersonal interactions, but selfobjects are defined as internalized representations rather than as real people in the environment. The entire structure of the “bipolar self with its tension arc connecting the poles” is poorly defined and needlessly complicated. Furthermore, Kohut never explicitly discarded the concepts of id, ego, and superego, but neither did he integrate these concepts with his theories about the development of the self.

Thus, from the criteria suggested in the philosophy of science, one may evaluate these various theories of development as follows.

(1) The identification and understanding of patterns of human experience were accomplished by Kernberg (1976, 1980, 1984), Kohut (1971, 1977), and Winnicott (1958, 1964, 1965, 1971, 1989). Kernberg's theory is the most comprehensive and complex, but his theory also could be enhanced and enriched by incorporating some of the ideas of the other two theorists.

(2) Kernberg's theory also contains the most extensive explanation of the phenomena under investigation which satisfied criteria suggested by Manicas and Secord (1983). However, an integration of Winnicott’s and some of Kohut's ideas would make Kernberg's theory even more comprehensive and explanatory.

(3) Kernberg extensively accommodated the data and successes of previous theories and explained data not previously integrated into theory, which satisfies criteria suggested by Lakatos (1978). Winnicott also integrated previous theories with his own ideas and with new data derived from his observations of mothers and infants. Kohut formulated his theory based upon the data collected from patients in analysis without concern for integrating or accommodating the success of previous theories.

(4) Neither Kernberg’s nor Winnicott's theories seems overly complex, which satisfies criteria suggested by Laudan (1977). Kernberg's theories are intricate and complex but explain intricate and complex processes. His explanations of the interactions of various intrapsychic processes are clear and logical. Winnicott's ideas are often almost deceptively simple, even when describing complex human interactions and internal processes. Kohut's ideas, on the other hand, are all too often confusing and convoluted.

(5) Most of the terminology used by Kernberg is well defined, which satisfies criteria suggested by Laudan (1977). Winnicott had a tendency to use common every day language often to create images or metaphors (such as the infant's state of "going on being"), which illuminate his meaning. However, some of his ideas are vague and poorly defined. Kohut has been frequently criticized for using poorly defined terminology or for using terms that are given one definition and then used in a way that necessitates a different definition.

In summary, when evaluated as a scientific theory, Kernberg's (1976, 1980, 1984) theory of development is the most comprehensive, well-defined, and comprehensible. However, almost all of Winnicott's (1958, 1964, 1965, 1971, 1989) and many of Kohut's (1971, 1977, 1984) ideas about interpersonal interactions and their effects do not contradict Kernberg's theory. An integration of the ideas of the three theorists could provide a more comprehensive understanding of early development than any of the three provide on his own.

The Nature and Development of Psychopathology

There is general agreement among almost all of the theorists that the development of psychopathology is the result of traumatic disappointments or frustrations in childhood experience. There is general understanding that traumatic disappointments experienced in very early childhood produce more serious and severe psychopathology than traumatic disappointments later in childhood (Fairbairn, 1954; Felder & Weiss, 1991; Kernberg, 1976, 1980, 1984; Klein, 1964, 1975; Kohut, 1971, 1977, 1984; Jacobson, 1964, 1971; Mahler et al., 1975; Sullivan, 1953, 1954, 1956, 1964, 1971; Whitaker & Malone, 1953/1981; Winnicott, 1958, 1964, 1965, 1971, 1989).

Several substantive theoretical disagreements can be identified. (a) The interpersonal and experiential theorists have little interest in placing psychopathology into diagnostic categories and identifying the dynamics of each category (Bromberg, 1989; Stern, 1987; Sullivan, 1953, 1954, 1956, 1971). (b) Kohut (1971, 1977, 1984) thought that all psychopathology was based upon deficits in the structure of the self. (c) Kernberg (1976, 1977, 1984) explained all psychopathology as emanating from faulty intrapsychic structure.

Evaluation: From the data collected in psychotherapy research, the interpersonal and experiential schools of psychotherapy do not accommodate the findings of Strupp et al. (1988) that different types of patients have different responses to therapeutic interventions that led Strupp et. al. (1988) to recommend that treatment be modified to accommodate the needs of "more difficult" patients. In these theories, there is no consideration of the dynamics and behavior of differing patients with differing diagnoses that require different treatment approaches.

On the other hand, several of the object relations theorists (Jacobson, 1964, 1971, Kernberg, 1976, 1980, 1984; Winnicott, 1958, 1964, 1965, 1971, 1989) and the self psychology theorists (Kohut, 1971, 1977, 1984; Wolfe, 1978) described the possible underlying dynamics of patients with borderline personality organization and narcissistic personality disorders and made recommendations for modifications in the treatment of these patients. Accordingly, these theorists were following the recommendations of Strupp et al. (1988).

Kohut's (1971, 1977, 1984) theories about the dynamics of patients with narcissistic personality disorders originally were focused upon understanding the treatment reactions of these patients and on modifying the treatment to make it more appropriate to their needs. (a) Kohut’s theories about psychopathology provided a unique and original identification and understanding of some patterns of human experience, which satisfied criteria suggested by Fischer and Fischer (1983). (b) Kohut’s model explained the phenomena of narcissistically disturbed patients, which satisfies criteria suggested by Manicas and Secord (1983). (c) Kohut proposed a treatment approach to narcissistic patients that worked, which satisfies criteria suggested by Mulaik (1984) and was, thus, clinically significant, which satisfies criteria suggested by Adams (1984).

All of these points apply to Kernberg's (1976, 1980, 1984) theories about the dynamics and treatment of patients with borderline personality organization. The differences between Kernberg's and Kohut's (1971, 1977, 1984) contributions to the understanding of psychopathology lie in their differing approaches to the consideration of other types of disorders from the ones with which each of them were originally concerned.

Kohut (1971, 1977, 1984), particularly in his later writing, attempted to explain all psychopathology as based upon disorders of the self. Kernberg (1976, 1980, 1984) classified psychopathology into psychotic, borderline, and neurotic categories, maintaining previously understood and agreed upon conceptions about each type of disorder while using his model of object relations based development to explain borderline dynamics.

Evaluation of each of these theories yields the following conclusions.

(1) Kernberg's (1976, 1980, 1984) model extensively accommodated the data and successes of previously stated theories. Kohut (1971, 1977, 1984), on the other hand, explained data not previously understood but did not integrate these data with previous theories. Therefore, according to the criteria proposed by Lakatos (1978), Kernberg's model is more successful.

(2) In attempting to expand his model to explain all psychopathology, Kohut's (1971, 1977, 1984) theory became overly complex and full of vague and poorly defined terms. Kernberg's (1976, 1980, 1984) model, although complex, has been presented clearly and logically, and the terminology has been rigorously defined. Therefore, the criteria proposed by Laudan (1977, 1984) support the idea that Kernberg's model is more robust than Kohut’s.

(3) Laudan (1977, 1984) also proposed that the strength of a theory can be evaluated by the extent to which it stimulates growth, progress, and competition in theoretical development. There is little doubt that the theories of both Kohut (1971, 1977, 1984) and Kernberg (1976, 1980, 1984) continue to stimulate progress and competition.

The Role of the Therapist and the Mechanism of Change

All these theorists agree about two elements that are essential to successful treatment. First, the interpersonal relationship with the therapist has a major effect. Secondly, interpretation of the meaning of behavior of which the patient is not consciously aware can promote insight and understanding, which also has an important role in the success of a treatment. The major source of disagreement is the extent to which the therapist directly gratifies infantile needs with or without interpretation of the gratification. The ideas in each theory of the role of the therapist and the mechanism of change are necessarily dependent upon the extent to which the interpersonal aspect of the therapy contributes to change (Fairbairn, 1954; Felder & Weiss, 1991; Kernberg, 1976, 1980, 1984; Klein, 1964, 1975; Kohut, 1971, 1977, 1984; Jacobson, 1964, 1971; Mahler et al., 1975; Sullivan, 1953, 1954, 1956, 1964, 1971; Whitaker & Malone, 1953/1981; Winnicott, 1958, 1964, 1965, 1971, 1989).

Evaluation: The most important substantive disagreement to be discussed is whether the therapeutic relationship alone, interpretations alone, or the combination of the two provide the mechanism of change in treatment. If the therapeutic relationship is a part of the mechanism of change, then to what extent does it and/or should it provide the gratification of infantile needs? Part of the confusion about this issue might be eliminated with some discussion about what is meant by the gratification of needs within the therapeutic relationship. There then will be a discussion of the support that can be found in psychotherapy research for the mechanism of change in each theory.

Substantive areas of disagreement include the following ideas.

1. The interpersonal perspective focuses on actual patterns of maladaptive interpersonal functions without hypothesizing internal dynamics as part of the determining forces. The therapist is described as a participant observer in the interpersonal behavior patterns demonstrated in the therapeutic relationship. The therapist both participates in the relationship and comments upon it. Stern (1987) remarked that these comments on chronic behavior patterns create a new experience for the patient that helps bring about change. The question in which the therapist hopes to engage the patient's interest is "What's going on around here?" The relationship between patient and therapist, thus, was described as part of the mechanism of change because it provides an opportunity for the manifestation of interpersonal patterns of behavior. Both the relationship and interpretation are part of the mechanism of change (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982).

2. Kernberg (1976, 1980, 1984) stated that it was important to reduce the use of defenses through interpretation thereby allowing the emergence of repressed material into consciousness, which resolves intrapsychic conflict within the neurotic patient. With the borderline patient, interpretation and clarification of primitive defenses, such as splitting, strengthen the ego and promote integration of part self and object representations into whole self and object representations. Interpretation was the mechanism of change, and the importance of the therapeutic relationship was minimized.

3. Whitaker and Malone (1953/1981) and Felder and Weiss (1991) recommended that the therapist make extensive use of his or her authentic self and provide the gratification of deep infantile needs. They also agreed with the interpersonal idea that it is important to break repetitive patterns of maladaptive interpersonal behavior by not playing the appropriate role called for in the pattern. They did not agree explicitly with the interpersonal emphasis on interpreting or commenting on the maladaptive patterns. As infantile needs are gratified and maladaptive patterns are interrupted, the patient's anxiety diminishes, and there is a resumption of normal, healthy growth.

4. Kohut (1977, 1984) emphasized the need for the therapist to empathize with the selfobject needs activated in the transference. Otherwise, the role of the therapist and process of treatment are fairly similar to Kernberg's (1984). The therapist interprets and analyzes the defenses and the transference. Self structure develops in response to empathic attunement, optimal failures, and accurate interpretations. In Kohut's theory, both the therapeutic relationship and interpretation provide the mechanisms of change.

It can be argued that there is always gratification of infantile needs in the therapeutic relationship. Few parents would disagree with the proposition that children need and demand attention, and it is difficult to imagine an effective therapeutic relationship in which attention is not provided to the patient. Kernberg (1984) admitted that the empathy, warmth, and concern of the therapist may far exceed the emotional gratifications available in the family of origin. He argued, however, that these gratifications would only provide a foundation for the therapeutic alliance within which the patient can achieve insight, which is the real vehicle for therapeutic change. It should be noted, however, that it has been suggested in the analytic literature that there is no such thing as a therapeutic alliance with a borderline patient. If the therapist succeeds in involving a borderline patient in a therapeutic alliance, then the borderline pathology has been cured (Adler, 1979).

Kohut (1971, 1977, 1984) proposed that vulnerable and fragile patients may need to internalize selfobject functions in order to build psychic structure before they can utilize dynamic and genetic interpretations that, in turn, build more inner structure. Kernberg (1984) proposed that one of the mechanisms of change in supportive psychotherapy is partial identification with the therapist's benign concern and empathy.

With the patients described by Kohut (1971, 1977, 1984), one gets the impression that they are drawn into the treatment situation by the availability of the mirroring and idealizing selfobject experiences in the transference. The unavoidable empathic failures of the therapist then promote internalization of the selfobject transference phenomena which stimulates the development of self structure, which, in turn, enables the patient to make use of deeper interpretations.

When Winnicott (1971) proposed that one of the functions of the analyst is to foster regression by directly gratifying infantile needs, his description of the behavior of the analyst is well within the normal parameters of an analyst’s behavior described in any other theory. Winnicott wrote that the analyst's quiet, accepting, reflecting, and clarifying presence provides maternal functions of being there, holding, and surviving. The analyst's undemanding presence provides conditions similar to those described as maternal protection from impingement in the infant's quiescent state of going on being. Thus, the analytic conditions permit and encourage the reemergence of the true self. Perhaps the most important part of this idea is that the analyst's normal, attentive, quiet, empathic, warm concern is similar to some parental functions and gratifies some infantile needs.

In considering the extent to which the therapist gratifies the needs of the patient, one must consider the responsibility of the therapist to protect the symbolic transferential nature of the therapeutic relationship. Balint (1968) differentiated between benign and malignant regression. To tell the therapist about the experience of a deep devastating need such as the need for a mother's love and involvement can be a healing, benign experience. However, if the therapist and patient believe that the therapist can directly meet intense regressive needs, the regression may become addictive and malignant. The patient may come to feel that the relationship with the analyst is the only source of gratification.

Stern (1987) stated that it is vital that the therapist not become just another "real" object, one of those people the patient has contact with on a regular basis who provides regular emotional support. The position of the therapist is most powerful when he or she is neither real (responding as if involved in a normal interaction) nor unreal (uninvolved, detached, and unavailable) but rather in what Winnicott (1971) called the “transitional realm.” If the therapist becomes a real and important source of gratification of deep emotional needs, the patient may become excessively dependent and/or the therapist may be offering much more than can be consistently delivered.

Whitaker and Malone’s (1953/1981) proposal that the therapeutic relationship progresses to a core stage of treatment in which both therapist and patient imagine the patient to be a small child and the therapist to be a symbolic primordial parent may be a recommendation for an unrealistic and perhaps dangerous therapeutic enterprise. It is probably either impossible or inadvisable for a therapist to have the kind of intense feelings for a patient that a parent has for a child. It is also probably inadvisable to encourage a patient to believe that a therapist can be a "second-chance" parent as described by Felder and Weiss (1991).

However, if the experiential theorists are referring to the type of gratifications and provisions of maternal functions described by Winnicott (1971), there is little doubt that gratification of infantile needs for attention, concern, warmth, nonintrusiveness, and reflection are a part of every therapeutic relationship. Jacobson (1964, 1971), Kernberg (1976, 1980, 1984), and Kohut (1977, 1984) may have emphasized repeatedly the crucial importance of interpretation as the only vehicle for therapeutic change and growth. However, in the clinical examples and case studies used to illustrate the concepts discussed by these theorists, there are frequent examples of the use of the self and the relationship to facilitate the therapeutic process.

None of the above discussion of the importance of the therapeutic relationship in keeping the patient in treatment and providing some of the foundation for change is intended to dismiss interpretation and insight as powerful agents for therapeutic change. Indeed, all of the theorists discussed the necessity of interpretation and insight as part of the therapeutic process.

Despite the seeming differences in their theoretical approaches, perhaps the actual behavior of the therapist in the analytic setting is very similar regardless of the theoretical orientation of the therapists. Interpretations usually are not complicated explanations of theoretical intrapsychic processes but are rather respectful linkings of feelings with behavior, identification of patterns as manifested in the transference, or other formulations made in everyday language that can be integrated into the experience of the patient.

Psychotherapy outcome research provides considerable support for the idea that both interpretation and the therapeutic relationship are vital parts of the mechanism of change. Two therapeutic interventions that are correlated with success in treatment are any activity of the therapist that raises the self awareness of the patient and interpretation (Orlinsky & Howard, 1986), both of which support the proposals of the interpersonal theorists (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982), of Kernberg (1976, 1980, 1984), and of Kohut (1971, 1977, 1984). These findings also directly contradict the idea proposed by Whitaker and Malone (1953/1981) and Felder and Weiss (1991) that the experience of the therapeutic relationship is a more significant mechanism of change than are understanding and self awareness arrived at by confrontation, clarification, and interpretation.

The following conditions that are correlated with success in treatment support the therapeutic relationship as being a part of the mechanism of change: (a) therapeutic interventions that focus on the transference, (b) the therapist's engagement and confidence, (c) the empathy, warmth, and acceptance of the therapist, and (d) the establishment of a collaborative therapeutic relationship in which the therapist encourages the patient's initiative and active role in the resolution of problems rather than a therapeutic relationship in which the patient is compliant and/or dependent (Orlinsky & Howard, 1986).

Other factors that are correlated with successful outcome include (a) verbal activity of the patient, (b) therapeutic interventions that focus on the patient's affect, and (c) the patient's expression of negative affect particularly in early sessions. All of these findings taken together seem to support what Winnicott (1958, 1964, 1965, 1971, 1989) has proposed as a model of treatment. The therapist should provide an attentive, quiet, warm, empathic presence and also should provide interpretations that allow the patient to feel understood. Such a stance would encourage the verbal activity of the patient, give implicit permission and support for the expression of negative affect, and encourage the patient to take an active role in treatment.

These research findings may be used to evaluate the different theoretical positions as follows.

(1) The model for psychotherapy proposed by the interpersonal theorists is supported by the findings.

(2) Kernberg's (1976, 1980, 1984) theories about psychotherapy seem to be deficient in discussion of the use of the therapeutic relationship as part of the mechanism of change.

(3) The theories of experiential psychotherapy certainly seem to be deficient in discussion of the importance of interpretation, clarification, and confrontation. The experience of the therapeutic relationship does not seem to be sufficient to bring about all the changes in treatment. Insight, understanding, and increased self awareness are also important mechanisms of change.

(4) Kohut's (1971, 1977, 1984) theories about the process of psychotherapy seem to be supported. In particular, his advocacy of the precedence of the use of empathic understanding as a therapeutic intervention in the beginning of treatment is supported by several findings. His approach probably would encourage the patient's verbal activity, support the patient’s taking an active role, focus therapeutic interventions on affect and transference, probably would increase the self awareness of the patient, and would encourage the expression of negative affect. However, Kohut also seemed to advocate using interpretation primarily to increase the patient’s awareness of deficits in his or her self structure. As has been previously discussed, human behavior, motivation, and psychopathology are too complex to be sufficiently understood or explained solely by an examination of the functioning of self structure.

The research findings of Strupp et al. (1988) give further grounds for evaluation of the theories. Of the theories considered, only a few of the object relations theorists (Jacobson, 1964, 1971; Kernberg, 1976, 1980, 1984; Winnicott, 1958, 1964, 1965, 1971, 1989) and Kohut (1971, 1977, 1984) made specific concrete suggestions about modifications in treatment in order to meet the treatment needs of more difficult patients.

In summary, the research findings about psychotherapy process and outcome seem to support best the style of treatment recommended by Winnicott (1958, 1964, 1965, 1971, 1989). An attentive, quiet, emphatic stance on the part of the therapist with interpretations that provide understanding and enhance the patient's self-awareness do not necessarily rule out the usefulness of the treatment approaches recommended by the interpersonal theorists (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982) and support Kohut's recommended specific strategies for working with patients with narcissistic disorders. Nor would one want to dispense with Kernberg's (1976, 1980, 1984) valuable insights into the dynamics and treatment of the patient with borderline personality disorder, but the empirical findings indicate that Kernberg's treatment approach inappropriately minimized the importance of the therapeutic relationship as part of the mechanism of change.

The Goal of Treatment

For the interpersonal theorists, the goals of treatment were identified as increased awareness, the correction of interpersonal deficits, and an improved capacity for interpersonal relations with others (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982).

For Kernberg (1976, 1980, 1984), the goals of treatment depended on the type of treatment used as has been described in the summary of his ideas. In general, the goals of treatment are to strengthen the ego, to resolve intrapsychic conflict, to foster the development of a modulated, realistic superego, to encourage the development of integrated self and object representations thereby creating mature object relations, and to improve interpersonal functioning.

Whitaker and Malone (1953/1981) stated that goals of treatment are for the patient to accept fantasies while maintaining the capacity to function in reality and for the patient to learn to seek the gratification of physical and emotional needs. Felder and Weiss (1991) stated that the goals of treatment should be set by the patient. The only specific goal acknowledged by Felder and Weiss is that the patient realizes he or she is growing and that growth will continue.

Kohut (1977, 1984) stated that the goals of treatment include the patient’s understanding through therapeutic interpretations of deficits in self structure as the result of traumatic failures and the patient’s insight into how anxious, immature selfobject relations reflect his or her need to find selfobject experience to facilitate self development. He also stated that one goal of treatment is for the patient to develop the capacity for mature empathic attunement.

There are obviously differences of opinion in these theories, but all of the theorists seem to agree that the goal of the treatment is the improvement of the patient, particularly in interpersonal relations. The theoretical disagreements reflected in the differences in these therapeutic goals are congruent with the theoretical differences that have already been discussed and evaluated.

Summary of Findings

The substantive theoretical differences in the different models can be considered best by reviewing the major theoretical differences among the psychoanalytic interpersonal theory, the experiential theory, the object relations theories, and the theory of self psychology. There will be a review of the evaluations made of the theoretical differences in the heuristic and ontology of these theories. Conclusions will be formulated first by comparing the theories as independent entities and then by considering the possibility of integrating ideas from the different theories. Finally, recommendations for research will be made (Fairbairn, 1954; Felder & Weiss, 1991; Kernberg, 1976, 1980, 1984; Klein, 1964, 1975; Kohut, 1971, 1977, 1984; Jacobson, 1964, 1971; Mahler et al., 1975; Sullivan, 1953, 1954, 1956, 1964, 1971; Whitaker & Malone, 1953/1981; Winnicott, 1958, 1964, 1965, 1971, 1989).

1. It has been hypothesized by the interpersonal theorists that there is no need to identify or describe complex, hypothetical intrapsychic structure. Patterns of human interpersonal interaction provide all the information needed to identify or to describe personality and behavior. There is no need to classify psychopathology based on the identification of underlying intrapsychic dynamics. Pathological interpersonal behavior patterns are unique for each individual and are based upon personal history. These patterns can be identified by the therapist who functions as a participant observer in the treatment process. The pattern is broken when the therapist comments upon it and provides a new interpersonal experience for the patient. The patient is drawn into the therapeutic investigation of "what's going on around here?" New experiences and understanding of pathological behavior patterns correct deficits in interpersonal relations, increase awareness of unconscious behavior patterns, and improve the capacity for interpersonal relating (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982).

2. The Atlanta school of experiential psychotherapy also seems to be unconcerned with the identification or understanding of either complex intrapsychic structure or the classification of psychopathology because neither was discussed in their theories. These theorists stated that psychopathology could be successfully treated by the gratification of deep infantile needs and the disruption of maladaptive behavior patterns. This can be accomplished by not assuming the roles demanded by the patterns but not necessarily by commenting upon them. The mutative forces in therapy are based upon new experience rather than on insight, interpretation, or understanding. The goal of therapy is the resumption of growth in interpersonal relating with an acceptance of fantasy (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone, 1953/1981).

3. Kohut (1971, 1977, 1984) stated that there is important explanatory value in understanding the development of the intrapsychic structure of the bipolar self and the internalization of selfobject functions. Kohut proposed that the development of self structure is dependent on the appropriate responsiveness and optimal frustrations of significant others in the environment who were internalized as selfobjects. Optimal frustration promotes the development of self structure as the individual uses transmuting internalizations to internalize the selfobject functions of mirroring, idealizing, and alter-ego selfobject experiences. The development of the self structure was conceived as a separate line of development from the development of object relations. Both lines of development progress from primitive, archaic narcissism and object use to mature self esteem and object love.

Kohut (1971, 1977, 1984) stated that all psychopathology is based upon deficits in the development of self structure. Psychotic and borderline (including schizoid and paranoid pathology) pathology develop when traumatic failures of selfobjects interfere with the development of a nuclear self. Narcissistic disorders develop after the formation of a nuclear self but before the nuclear self has been invested with cohesive self structure and reliable internalizations of the mirroring, idealizing, and alter-ego functions of selfobjects. Neurotic disorders develop when there are traumatic disappointments in the selfobject's appreciation and appropriate aim-inhibition in the oedipal period.

Classification and understanding of the differing developmental disappointments and intrapsychic structure of different types of psychopathology assist the therapist in determining appropriate treatment strategies. Empathic attunement with the selfobject needs activated in the transference promotes the internalization of selfobject functions of the therapist when the patient is faced with transitory empathic failures. Development of self structure allows the patient to appreciate and use dynamic and genetic interpretations to build the structure of the self. One goal of treatment is to help the patient understand that the deficits in the structure of the self manifested in disturbed interpersonal relations are the result of traumatic failures of the parents to provide optimal selfobject mirroring, idealizing, and alter-ego experiences. Another goal is to help the patient understand that maladaptive interpersonal relations are based on the needs of the self to find appropriate selfobject experience that will facilitate construction of the self and growth to healthier interpersonal relations, which can then be used to support and maintain self structure.

4. Kernberg (1976, 1980, 1984) stated that understanding the underlying intrapsychic dynamics of the patient is crucial to the success of psychotherapy. Treatment of patients with borderline personality organization was particularly contingent upon the therapist's understanding of the manifestation of part self and object relations in the transference and the mechanism of splitting to explain the intense, contradictory emergence of affects in the transference.

Kernberg (1976, 1980, 1984) said that the organization of the self was not a separate line of development from the organization of object relations. Self and object representations form concurrently with libidinal investments providing part of the base of both healthy narcissism and object love. Normal parental demands and prohibitions promote the development of aggressive drives directed at object representations and the investment of libidinal energy in the developing self. Healthy narcissism promotes the development of autonomous ego functions and a well-modulated autonomous superego.

All psychopathology was described by Kernberg (1976, 1980, 1984) as based on faulty intrapsychic structure. The psychotic patient is fixated or regressed at the symbiotic stage of development and does not have adequate boundaries between self and object representations. The patient with borderline personality organization is fixated or regressed to the point at which good, libidinally invested representations of the self and object have not been integrated into whole self and object representations. The individual without identity and object constancy suffers from borderline personality organization. Narcissistic personality organization was described as a pathological condensation of intrapsychic representations of the real self, the ideal self, and the ideal object accompanied by repression of unvalued self representations, devaluation of object representations, and blurring of normal ego-superego boundaries.

Neurotic psychopathology was described as being caused by the development of an overly punitive superego, producing intrapsychic conflict between the normal strivings of the ego for gratification and the overly restrictive and prohibitive responses of the punitive superego. Thus, all psychopathology was understood by Kernberg (1980, 1984) as the product of faulty, malfunctioning intrapsychic structure. The development of self representation culminating in a sense of personal identity was not considered to be a separate line of development from the integration of good and bad object representations into object constancy. On the contrary, integration of good and bad self representations complemented and enhanced the integration of good and bad object representations. Both integrations provided the foundations for ego and superego structure and autonomy.

Kernberg (1984) stated that the interpretation of defenses reduced the use of defenses, strengthened the ego, promoted the emergence of intrapsychic conflict, and modulated the functions of the superego to more realistic, less punitive functions for guidance and support. Interpretation of primitive defenses in patients with borderline personality organization also strengthened the ego and promoted the development of integrated good and bad self and object representations.

The evaluations that have been made of theoretical differences in the heuristics and ontology of these theories can be summarized as follows.

1. Heuristics: Kernberg's (1976, 1980, 1984) heuristic was judged to be the most comprehensive in scope, meeting all of the philosophical criteria for a successful theory.

2. Ontology: Essential human nature. It was concluded that essential human nature might be understood best through an integration of these theoretical models. Humans, thus, may be assumed to be motivated by an innate drive to seek and to maintain interpersonal relationships and by varying degrees of innate aggression that may be manifested destructively or as strivings for achievement, success, and mastery.

3. Ontology: Normal Development. Kernberg's (1976, 1980, 1984) theory was found to be the most comprehensive, well-defined, and successful as judged by the criteria of theory evaluation taken from the philosophy of sciences. However, the emphasis in his theory is on intrapsychic development, and his theory of development could be enhanced by an integration of interpersonal processes described by Winnicott (1958, 1964, 1965, 1971, 1989) and Kohut (1971, 1977, 1984).

4. Ontology: Nature and Development of Psychopathology. Kohut (1971, 1977, 1984) has provided a clinically useful and theoretically successful model for understanding patients with narcissistic personality disorders. Kernberg (1976, 1983, 1984) has developed a clinically useful and theoretically successful model for understanding patients with borderline personality organization. Kernberg's theories about the nature and development of all types of psychopathology are more robust than are Kohut's hypotheses. However, it was suggested that Kernberg's model would be even more clinically useful if it were expanded to include a category of narcissistic personality organization, which could be placed developmentally between the categories of borderline personality organization and the neurotic disorders.

5. Ontology: The Role of the Therapist and the Mechanism of Change. The empirical findings from psychotherapy research seemed to support best the model of treatment proposed by Winnicott (1958, 1964, 1965, 1971, 1989). However, there is also empirical support for the treatment approach recommended by the interpersonal theorists (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982), the treatment approach recommended by Kohut (1971, 1977, 1984), particularly in treating patients with narcissistic disorders, and for the modifications in treatment recommended by Kernberg (1976, 1980, 1984) for working with severely disturbed patients.

Essentially, it seems that the style of treatment used by the therapist must vary in response to the level of psychopathology presented by the patient. Both the provision of a responsive, attentive, therapeutic relationship and the use of clarification, confrontation, and interpretation as techniques that increase self awareness serve as mechanisms of change. The importance of the therapeutic relationship in providing a mechanism of change seems significantly dependent on the level of psychopathology. The more serious the psychopathology, the more the patient seems to need the quiet, accepting empathy and understanding provided by the therapist. Well-functioning neurotic patients, on the other hand, may derive more therapeutic benefit from interpretation, clarification, and confrontation than from the therapeutic relationship from the beginning of treatment.

These hypotheses do not necessarily contradict the recommendations for the treatment of borderline patient given by Kernberg (1976, 1980, 1984). Confrontation and interpretation of primitive defenses do not preclude the offering of such interpretations in the spirit of empathic understanding. The borderline patient who is living in a state of emotional chaos is quite likely to feel profound relief if a therapist accepts and explains how splitting, for example, might be causing the extremes of love and hate felt with alternating intensity toward a significant other in the patient's life.

Whitaker and Malone (1953/1981) may have been true pioneers in understanding the importance of the therapeutic relationship as a mechanism of change particularly in their emphasis on the empathic understanding of powerful unconscious fantasies acted out in transference. Whitaker and Malone (1953/1981) should be commended for their early and innovative awareness of the power of the therapeutic relationship as a mechanism for change, particularly in their understanding of the reawakening of powerful, childlike unconscious feelings and fantasies aroused in the transference and the necessity for the therapist to respond with empathy, thereby gratifying some deep infantile needs for attention, acceptance, and concern.

However, the research and further developments in theory indicate that Whitaker and Malone (1953/1981) overemphasized the importance of pure experience and unconscious processes over the whole course of psychotherapy. It cannot be assumed that interpersonal experience without interpretation or analysis is superior to the increased awareness provided by the commentary and/or interpretation as described in the theories of the psychoanalytic interpersonal theorists, the theories of self psychology as described by Kohut (1971, 1977, 1984), or the theories of object relations as integrated by Kernberg (1976, 1980, 1984). Empirical support was provided for interpretation and insight as a curative factor in making the patient aware of pathological processes. Experience does not take precedence over understanding. Thus, the theories of the experiential psychotherapist (Felder & Weiss, 1991; Gantt, 1984; Whitaker and Malone, 1953/1981) have been found clearly to be inadequate in explaining the complex process of psychotherapy.

Conclusions

At this point, there are two approaches that may be taken to evaluate the theories under consideration. First, it could be postulated that the theories under consideration are essentially incompatible and that each theory should be evaluated on the basis of the data about psychotherapy process and outcome and by the criteria for theory evaluation described in the philosophy of science. Second, there could be the possibility of integrating the different models. Each will be considered separately.

Evaluation of the Theories as Incompatible Theories

From the philosophies of science concerned with the evaluation of theory, the following points are clear.

(1) Phenomenology (Fischer & Fischer, 1983): Interpersonal, self psychology, and object relations theories are all equally successful in identifying patterns of human behavior and explaining them. The interpersonal theorists simply identify and comment upon patterns. The approach of self psychology and the object relations theorists has been to identify patterns of different categories of psychopathology and proposes a reasonable method of treatment.

(2) The realist approach (Manicas & Secord, 1983): The interpersonal approach exhibits significant deficiencies in its explanation of the phenomena under investigation. Both self psychology and object relations theory offer a more detailed, explanatory description of normal development, the development of psychopathology, and the processes and goals of psychotherapy. Accordingly, from the realist approach, the theories of self psychology and object relations explain more than the theories of the psychoanalytic interpersonal school.

(3) Pragmatism (Adams, 1984; Mulaik, 1984): The assessment of clinical theory should be based upon what works or what leads to therapeutic success. The empirical findings from psychotherapy research provide support for the models of treatment proposed by Winnicott (1958, 1964, 1971, 1989) but also provide support for the theories of the interpersonal theorists (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982), of Kohut (1971, 1977, 1984), and of Kernberg (1976, 1980, 1984). No one of these models has been demonstrated to be clearly superior to the others.

(4) Lakatos (1978) proposed that a theory is progressive if it accommodates the success of previous theories and explains new data. With this definition, interpersonal psychotherapy is not progressive. It continues to explain new data in terms of its existing theories. In contrast, self psychology explains new clinical data about narcissistic patients with new theoretical formulations. However, self psychology tends to ignore rather than accommodate the success of previous theories. Object relations theories as integrated by Kernberg (1984) allow the most comprehensive accommodation of previous theoretical successes with the explanation of new data.

(5) Laudan (1977, 1984) proposed that theory evaluation could be based on the complexity of the theory, with overly complex theories representing degenerating research traditions, the extent to which terms are adequately defined, and the extent to which theories generate lively competition and continued development. Interpersonal psychotherapy has an elegant simplicity. Self psychology might be viewed as overly complex. For example, selfobject relations are defined as being internalized and intrapsychic. The development of self structure is based upon the internalization of selfobject relations, which are already defined as being internalized. Object relations theory, on the other hand, is complex and intricate and might be viewed as either overly or sufficiently complex.

(6) Furthermore, there are few poorly defined terms in either interpersonal or object relations theories. Self psychology, however, is fraught with poorly defined and confusing jargon. Therefore, interpersonal and object relations theories are theoretically more progressive than theories of self psychology.

(7) Finally, Laudan (1977, 1984) proposed that the strength of a theory was based upon its generating lively competition and continued development. The psychoanalytic interpersonal school, Kohut's self psychology, and Kernberg's theories of object relations continue to stimulate lively competition and theory development.

(8) Thus, although criteria for theory evaluation indicate weaknesses in the psychoanalytic interpersonal school and the theories of self psychology, one cannot definitely establish that Kernberg's theories of object relations provide the most comprehensive understanding of the nature of humans, the processes of normal development, the nature of psychopathology, or the processes and goals of psychotherapy. A confirmation of the most progressive theoretical system awaits the production and analysis of data from future research.

Integration of the Theories

The following attempt to integrate these theories will be divided into sections for the theoretical components previously discussed as the heuristic and the ontology. It is to be expected that even within an integration of the theories some deficiencies will be identified.

Heuristics. As was previously discussed, Kernberg’s (1976, 1980, 1984) method of investigation is the most comprehensive in all the theories. His theoretical formulations were based on data taken from research studies, observation of patients, and integration of previous theoretical ideas. However, it was hypothesized that Kernberg's theory of development would be even more comprehensive and explanatory if it integrated and incorporated many of the ideas of Winnicott (1958, 1964, 1965, 1971, 1989) and Kohut (1971, 1977, 1984) about interpersonal processes and their effect on intrapsychic processes into his theory, which is predominately concerned with intrapsychic development. Winnicott’s (1958, 1964, 1971, 1989) observations of mother-infant dyads provided invaluable insights into some of the processes that take place both in normal development and in the development of psychopathology. Winnicott’s observations of the behavior of patients in therapy and the similarities between the mother-infant relationship and the therapist-patient relationship provided useful insights into the role of the therapist and the mechanism of change in psychotherapy.

Kohut’s (1971, 1977, 1984) method of empathic immersion in the patient’s experience certainly seems to have provided important contributions to understanding the role of the therapists and the mechanism of change in psychotherapy. His heuristic has been particularly helpful in the treatment of patients with narcissistic disorders but also can be applied to the treatment of most patients.

The emphasis of the interpersonal theorists (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982) on observing the specific interpersonal patterns that develop in each unique therapeutic relationship also provide a useful guide to understanding the role of the therapists and the mechanism of change in psychotherapy. Whatever the diagnostic category of any particular patient, there is no standard set of procedures that can be followed in treatment. The interpretation of defenses recommended by Kernberg (1976, 1980, 1984) is often very helpful in relieving symptoms but only when the patient is able to hear and to understand the interpretation. During any particular therapy session, it might be assumed that any experienced therapist from any theoretical orientation constantly will consider the question proposed by the interpersonal theorists: “What’s going on around here?”

Thus, an integrated heuristic for the development of a comprehensive theory of psychotherapy would include several elements. First, an integrated heuristic would include the integration of useful concepts from previous theories as was done extensively by Kernberg (1976, 1980, 1984). Second, an integrated heuristic would include the integration of data taken from controlled studies of normal childhood development as Kernberg used Mahler’s (1972, Mahler et al., 1975) research as part of the foundation of his theory. Third, an integrated heuristic would include the integration of data taken from psychotherapy research such as was done by Kernberg (1976, 190, 1984). Fourth, an integrated heuristic would include the integration of observations of mother-child interactions and the behavior of children in naturalistic settings as was done by Winnicott (1958, 1964, 1971, 1989). Fifth, an integrated heuristic would include the integration of observations made about similarities in the behavior of patients in treatment and the behavior of children as was done by Winnicott (1958, 1964, 1971, 1989) and Mahler (1972, Mahler et al., 1975). Sixth, an integrated heuristic would include the integration of data derived from empathic immersion in the experience of the patient as was done by Kohut (1971, 1977, 1984). Finally, an integrated heuristic would include integration of sensitive and acute observations of the patient’s behavior in therapy sessions and particularly the patient’s responses to therapeutic interventions such as was recommended by every theorist who has been discussed and particularly by the interpersonal theorists (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982). Only through the use of such a complex and determined method of investigation using previously successful theoretical paradigms, research findings, and observations would it be possible to develop a fully robust and comprehensive theory.

Ontology: Essential Human Nature. The most reasonable and theoretically progressive hypotheses supported by research on the observation of infants about essential human nature is an integration of all the theoretical approaches. Humans may have an innate need to establish and to maintain interpersonal relationships, but this does not adequately explain the complexity of human behavior. It also is reasonable to assume an innate potential for aggression. This innate potential for aggression may include an innate predisposition for anger and/or destructiveness, but it is likely that the manifestation of destructive aggression is shaped by experience with trauma and frustration. There also may be innate needs for achievement, mastery, and success, which also may be shaped by environmental factors. Thus, an individual innately endowed with the potential for destructive aggression may become a successful professional football player given the circumstances of environmental understanding and support, while another may become a criminal given a not extraordinary amount of frustration and trauma in the environment. Similarly, an innate propensity for mastery and success in an optimally responsive environment may produce a pioneering research physicist, but the same degree of innate need for mastery and success in a frustrating, traumatic environment may result in a successful burglar. The issue is complex and awaits much further research to arrive at a complete understanding of human nature.

Ontology: Normal Development. A preliminary integration of the theoretical models was discussed earlier in the chapter. As was described then, Kernberg’s (1976, 1980, 1984) theory of normal development was primarily concerned with the development of intrapyschic structure. His ideas were about the development of complex intrapsychic self and object representation, ego, superego autonomy, and the healthy narcissistic investment in the ego. He proposed that these developments occur within the context of appropriate interpersonal experience including gratification of needs and realistic parental demands, prohibitions, and expectations. However, Kernberg did not describe the exact nature of parental behaviors that provide the basis for the optimal development of intrapsychic structure.

Winnicott (1958, 1964, 1971, 1989) and Kohut (1971, 1977, 1984) both described parental behaviors that might be considered to provide the basis for the development of the intrapsychic structure described by Kernberg (1976, 1980, 1984). Winnicott’s concept of primary maternal preoccupation in the earliest stages of infancy would certainly describe a mother who would be prepared to provide optimal gratification during the symbiotic stage. Similarly, Winnicott’s concepts about the maternal provision of a holding environment, her protection of the infant who is in the state of “going on being,” and the mirroring function of the parents seem to describe parental behaviors that would provide the optimal gratification of the infant’s needs both during the symbiotic stage of development and during the stages of separation and individuation. Winnicott stated that these types of parental activities allowed an individual’s true self to emerge. It is thus possible that, intrapsychically, an emergence of the true self is the equivalent of the development of an autonomous ego invested with healthy narcissism.

The similarities between the developmental theories of Kohut (1971, 1977, 1984) and Winnicott (1958, 1964, 1971, 1989) were described extensively earlier in this chapter. It should be noted that the mirroring function described by both of them could promote the investment of healthy narcissism in the autonomous ego described by Kernberg (1976, 1980, 1984). Furthermore, Winnicott (1958, 1964, 1971, 1989) proposed the concept of the mother’s graduated failure of adaptation which stimulates the development of ego functions. Kohut (1971, 1977, 1984) proposed that the optimal, nontraumatic frustration stimulates the internalization of self object functions and the development of self structure. Both ideas complement Kernberg’s (1976, 1980, 1984) idea that appropriate parental demands, expectations, and prohibitions help form the basis for the development of the autonomous ego and superego. Thus, it can be seen that Kernberg’s developmental theory may be enhanced through the incorporation of most of Winnicott’s and some of Kohut’s descriptions of appropriate and/or optimal parental behavior.

Finally, it should be noted that Kernberg (1976, 1980, 1984) explicitly stated that relationships with others continue to influence intrapsychic structure throughout life with increasingly subtle identifications that are internalized by the individual. Both Sullivan (1953) and Jacobson (1964) described the way in which human relationships continue to influence the individual in post-oedipal childhood and adolescence. Jacobson (1964) described development into adulthood. Some of their ideas may also be integrated with Kernberg’s concept of increasingly complex and subtle identifications that modify constantly maturing intrapsychic structure.

Both Sullivan (1953) and Jacobson (1964) described post-oedipal childhood as the time when the child has the opportunity to have new interpersonal experiences with peers and authority figures outside the family. Jacobson stated that these experiences provide the basis for new identifications and reinforce a sense of personal identity. Sullivan also stated that such associations can promote the child’s acceptance of his or her own individuality.

Sullivan (1953) described adolescence as a time when a need for intimacy and the successful attainment of an intimate relationship can repair internal damage from earlier relationships. Jacobson (1964) described adolescence as a time when a sense of identity is consolidated and when there is a stabilization of superego function. Ego and superego autonomy form the basis for satisfying emotional, social, intimate, and vocational growth in adult life.

Thus, an integration of Kernberg’s (1976, 1980, 1984) comprehensive theory of the normal development of complex intrapsychic structures can be enhanced by an integration of Winnicott’s (1958, 1964, 1971, 1989) and Kohut’s (1971, 1977, 1984) descriptions of optimal parental activities. Furthermore, both Sullivan (1953) and Jacobson (1964) described some of the environmental intrapsychic influences that continue to influence individual development throughout life.

Ontology: The Nature and Development of Pyschopathology. Kernberg(1976, 1980, 1984) has provided the most comprehensive theory of the nature and development of psychopathology. He provided extensive explanations of the differences in the functioning of psychotic, borderline, and neurotic patients. He also provided an innovative and clinically useful guide to the understanding and treatment of borderline patients. However, Kohut’s (1971, 1977, 1984) theories about patients whom he identified as narcissistic provides insight into their dynamics and behavior and the methods for treating these patients. Therefore, Kohut's model provides a clinically useful guide to the understanding and treatment of narcissistic patients.

The most comprehensive theory of psychopathology might be one in which Kohut’s (1971, 1977, 1984) formulation about the narcissistic patient is integrated with Kernberg’s theory. Perhaps the narcissistic patient could be placed in a category between a borderline level of functioning and a neurotic level of functioning. It might also be clinically useful to integrate Winnicott’s (1958, 1964, 1971, 1989) ideas about the development of a false self into Kernberg’s (1976, 1980, 1984) theory of psychopathology. Winnicott essentially proposed that the false self develops in response to environmental demands that impinge upon the true self. It might be assumed that, when these demands come very early in the child’s life, development of the false self is at the borderline level of functioning, with the accompanying splitting of the self into all good and all bad internal experiences of the self. If the excessive environmental demands described by Winnicott (1958, 1964, 1971, 1989) occur later in the child’s life, perhaps narcissistic disorders develop with the false self being the grandiose, idealized self described by Kohut (1971, 1977, 1984). Finally, if the excessive environmental demands occur in the pre-oedipal or oedipal stage of development, the false self might develop as part of the neurotic character structure. When intrapsychic conflict occurs due to the harsh demands of the overly primitive and demanding superego described by Kernberg (1976, 1980, 1984), the false self may emerge displaying neurotic denial of normal human needs and impulses. It might be thought that the overly demanding superego develops as an internalization of the excessive environmental demands as described by Winnicott.

Thus, the best understanding of the nature and development of psychopathology might be through an integration of the theories of Kernberg (1976, 1980, 1984), Kohut (1971, 1977, 1984), and Winnicott (1958, 1964, 1971, 1989). Each theory provides useful concepts for understanding patients with different types of psychopathology and different paradigms from which to select interpretations that would best fit the level at which any individual patient is capable of understanding the interpretation being made.

Ontology: The Role of the Therapist, the Mechanism of Change, and the Goals of Treatment. The empirical findings from psychotherapy research have been identified and were summarized earlier in this chapter. An explanation of the treatment recommendations in each of the theories being evaluated reveals that each of them contains more than one suggestion for the role of the therapist and/or a mechanism of change that meets the criteria established for interventions and/or conditions that are significantly correlated with treatment success. Thus, an integration of the treatment recommendations from each theory may provide the most valuable theoretical model for therapeutic success in treatment. An integration of the these theories may provide a constructive and useful guide in deciding how treatment may be adapted to accommodate the different treatment conditions and interventions that would be most effective with different types of patients.

The recommendations for the role of the therapist, the mechanism of change, and the goals of treatment proposed by psychoanalytic interpersonal theory are congruent with the variables that are associated with positive outcome. These theorists emphasized the importance of establishing a warm, empathic therapeutic relationship. They recommended interpretation of maladaptive interpersonal behavior patterns. They suggested increasing the patient’s awareness through experiential confrontation by emphasizing the importance of both patient’s and therapist’s active involvement in answering the question “What is going on around here?” Finally, these therapists proposed a therapeutic approach in which the patient is encouraged to take an active role in the treatment process while the therapist assumes the role of participant-observer (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Stern, 1987; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982).

Kernberg (1976, 1980, 1984) may have overemphasized the importance of interpretation as the mechanism of change in treatment. However, he made invaluable contributions in his recommendations for different types of treatment for different types of patients. His theories about the intrapsychic dynamics and typical transference reactions of the borderline patient have been found to be clinically useful to many practicing therapists and certainly meet the criteria suggested by Strupp (1980a, 1980b, 1980c, 1980d) that treatment approaches be modified to work with different types of patients.

Similarly, Kohut (1971, 1977, 1984) provided a theoretical model for the treatment of patients with narcissistic disorders thereby also satisfying the criteria suggested by Strupp (1980a, 1980b, 1980c, 1980d). Kohut’s proposed model of treatment also emphasizes the communication of empathic understanding as a primary intervention from the beginning of treatment. As has been stated previously, this approach would encourage the patient's verbal activity, encourage the patient to take an active role in his or her treatment, focus therapeutic interventions on affect and transference, probably would increase the self awareness of the patient, and would encourage the expression of negative affect. However, as has been stated previously, it is probably unnecessarily restrictive and potentially counter productive to limit interpretation to comments about the patient’s self structure.

In the comparison of the theories as incompatible paradigms for successful therapy, the Atlanta experiential theorists were described as having developed the least successful theory in terms of what psychotherapy research has established as effective therapeutic conditions. However, if these theorists’ proposals about treatment are included in an integrated theory, they could augment the success of treatment. Their recommendation that the therapist attend closely to the transference responses of the patient satisfies the empirical finding that a focus on transference is correlated with success in treatment. Their recommendation that the therapist should provide verbal commentary and interpretation of the patient’s nonverbal behavior satisfies the empirical finding that experiential confrontation is correlated significantly with success in treatment. Their recommendation that the therapist use silent listening as a therapeutic technique and their insistence that the patient is responsible for setting treatment goals meets the criteria for encouraging the patient to take an active role in the treatment process. And, finally, their recommendation that the therapist provide a warm, empathic, and accepting relationship for the patient satisfies empirical findings. Perhaps their most important recommendation to be integrated with the other theories is their emphasis on attention paid to nonverbal behavior and increasing the patient’s awareness of nonverbal behavior (Felder & Weiss, 1991; Gantt, 1984; Whitaker & Malone 1953/1981).

Thus, an evaluation of the hypotheses in the different theories under consideration once again indicates that an integration of theoretical model provides the most comprehensive set of hypotheses. Both Winnicott (1958, 1964, 1965, 1971, 1989) and Kohut (1971, 1977, 1984) offer models most broadly supported by the research, but the models proposed by Kernberg (1976, 1980, 1984) and the interpersonalists (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982) also offer reasonable proposals that may be used effectively. The interpersonalists do not specifically recommend treatment modifications that depend upon the level of psychopathology, but their entire system is based upon a sensitive understanding of the interpersonal style of the patient with the goal being a mutual exploration by patient and therapist of the processes that have been aroused in transference. The entire basis of their treatment approach is an empathic awareness of interpersonal processes regardless of the level of pathology. Similarly, Kernberg's recommendations for different types of interventions based on different categories of psychopathology do not preclude empathic understanding of the defenses and dynamics of the patient. In fact, Kernberg has stated specifically that the empathy, concern, and understanding of the therapist are absolutely necessary conditions for any successful treatment.

The empirical findings of psychotherapy research support the hypotheses about the process of psychotherapy presented by Kohut (1971, 1977, 1984) and Winnicott (1958, 1964, 1965, 1971, 1989) concerning the mechanism of change being based upon both the provision of an emphatic therapeutic relationship and the use of interpretation. However these hypotheses about the process of psychotherapy and the best use of therapeutic interventions would be enhanced and made more thorough with an integration of the ideas of the theorists of the interpersonal school of psychoanalysis (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982) and the object relations theorists as represented best by Kernberg (1976, 1980, 1984). I have described and explained in my evaluation how each theory complements the other theories in the consideration of the separate ontological issues. Some value can be found in each theoretical position concerning essential human nature, the processes of normal development, the nature and development of psychopathology, the role of the therapist, and the mechanism of change in psychotherapy. All the theories provide insight into essential human nature. The process of normal development and the nature and development of psychopathology were most comprehensively described by Kernberg (1976, 1980, 1984) but can be enhanced and complemented by the ideas of Winnicott (1958, 1964, 1965, 1971, 1989) and Kohut (1971, 1977, 1985). The role of the therapist and the mechanism of change were most accurately described by Kohut and Winnicott but can be expanded and enhanced by the hypotheses of Kernberg (1976, 1980, 1984) and the interpersonalists (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982). In summary, all of these theorists provided valuable contributions to the issues that have been evaluated.

Recommendations for Research

The investigation and results of this study suggest that future research might be directed to the systematic and controlled study of the outcomes with psychotherapy patients who are treated in therapy by analytic therapists who are identified as interpersonal, self psychology, or object relations in their theoretical orientation or practice treatment from an integration of these differing theories. It would be particularly useful if the patients were divided diagnostically into groups that were categorized as neurotic, narcissistically disturbed, and borderline by independent observers and if the treatment outcomes were assessed by independent observers.

In conclusion, among these theoretical models, the psychoanalytic interpersonal school (Bromberg, 1989; Cashdan, 1982; Chrzanowski, 1982; Kiesler, 1982; Levenson, 1983; Sullivan, 1953, 1954, 1956, 1971; Wachtel, 1982), Kohut's (1971, 1977, 1984) self psychology, and Kernberg's (1976, 1980, 1984) integration of object relations theory, all offer unique and interesting insights into essential human nature, the processes of normal development, the nature and development of psychopathology, and the processes and goals of treatment. Kernberg's theories have slightly more support based on theory evaluation in the philosophy of science, but the true value of each theoretical approach awaits further research. Finally, all these theoretical approaches emphasize the importance of increasing the awareness of unconscious processes and dynamics in patients in treatment and focusing on interpersonal relations in the understanding of normal development, the nature of psychopathology, and the curative factors in psychotherapy. The importance of the interpersonal relationship and increasing awareness of unconscious processes have been supported clearly by empirical data from psychotherapy research.

References

Adams, H. E. (1984). The pernicious effects of theoretical orientations in clinical psychology. Clinical Psychologist, 37, 90-94.

Adler, G. (1979). The myth of the alliance with borderline patients. American Journal of Psychiatry, 136, 642-645.

Akhtar, S. (1989). Kernberg and Kohut: A critical comparison. In D. W. Detrick & S. P. Detrick (Eds.), Self psychology: Comparisons and contrasts. Hillsdale, NJ: Analytic Press.

Anchin, J. C. (1982). Interpersonal approaches to psychotherapy: Summary and conclusions. In J. C. Anchin & D. J. Kiesler (Eds.), Handbook of interpersonal psychotherapy. New York: Pergamon Press.

Alexander, F., French, T. M., & Bacon, C. L. (1946). Psychoanalytic therapy: Principle and applications. New York: Ronald Press.

Arnkoff, D. B. (1983). Common and specific factors in cognitive therapy. In M. J. Lambert (Ed.), Psychotherapy and patient relationships. Homewood, IL: Dorsey Press.

Balint, M. (1968). The basic fault. London: Tavistock.

Balint, M. (1979). The basic fault: Therapeutic aspects of regression. New York: Bruner/Mazel.

Beitman, B. D., Goldfried, M. R., & Norcross, J. C. (1989). The movement toward integrating the psychotherapies: An overview. American Journal of Psychiatry, 146, 138-147.

Bellock, L., & Faithorn, P. (1981). Crises and special problems in psychoanalysis and psychotherapy. New York: Brunner/Mazel.

Bowlby, J. (1984). Psychoanalysis as a natural science. Psychoanalytic Psychology, 1, 7-21.

Chapman, A. H. (1976). Harry Stack Sullivan: The man and his work. New York: Putnam.

Chrzanowski, G. (1980). Problem patients or troublemakers? Dynamic and therapeutic considerations. American Journal of Psychotherapy, 34, 26-38.

Chrzanowski, G. (1982). Interpersonal formulations of psychotherapy: A contemporary model. In J. C. Anchin & D. J. Kiesler (Eds.), Handbook of interpersonal psychotherapy. New York: Pergamon Press.

Colby, K. M., & Stoller, R. J. (1988). Cognitive science and psychoanalysis. Hillsdale, NJ: Erlbaum.

Colson, D., Lewis, L., & Horwitz, L. (1985). Negative outcome in psychotherapy and psychoanalysis. In D. T. Mays & C. M. Franks (Eds.), Negative outcome in psychotherapy and what to do about it. New York: Springer.

Detrick, D. W., & Detrick, S. P. (1989). Self psychology: Comparisons and contrasts. Hillsdale, NJ: Analytic Press.

Dreyfus, H. L., & Wakefield, J. (1988). From depth psychology to breadth psychology: A phenomenological approach to psychopathology. In S. B. Messer, L. A. Sass, & R. L. Woolfolk (Eds.), Hermeneutics and psychological theory: Interpretive perspectives on personality, psychotherapy, and psychopathology. New Brunswick, NJ: Rutgers University Press.

Eagle, M. N. (1983). The epistemological status of recent developments in psychoanalytic theory. In R. S. Cohen & L. Laudan (Eds.), Physics, philosophy and psychoanalysis: Essays in honor of Adolf Grünbaum. Dordrecht, Holland: D. Reidel.

Elkin, I., Shea, T., Watkins, J. T., Imber, S. D., Sotsky, S. M., Collins, J. F., Glass, D. R., Pilkonis, P. A., Leber, W. R., Docherty, J. P., Fiester, S. J. & Parloff, M. B. (1989). National Institute of Mental Health treatment of depression collaborative research program: General effectiveness of treatment. Archives of General Psychiatry, 46, 971-982.

Eysenck, H. J. (1952). The effects of psychotherapy: An evaluation. Journal of Consulting Psychology, 16, 319-324.

Fairbairn, W. R. D. (1952). An object-relations theory of personality. New York: Basic Books.

Feiner, A. H. (1982). Comments on the difficult patient. Contemporary Psychoanalysis, 18, 397-411.

Felder, R. E. (1967). The use of self in psychotherapy. In D. S. Arbuckle (Ed.), Counseling and psychotherapy: An overview (pp. 100-109). New York: McGraw-Hill.

Felder, R. E., & Weiss, A. G. (1991). Experiential psychotherapy: A symphony of selves. New York: University Press of America.

Fischer, H. J. (1991). The science of psychotherapy. New York: Hemisphere.

Fischer, C. T., & Fischer, W. F. (1983). Phenomenological-existential psychotherapy. In M. Herson, A. E. Kazdin, & A. S. Bellock (Eds.), The clinical psychology handbook. Elmsford, NY: Pergamon Press.

Friedman, L. (1978). Trends in the psychoanalytic theory of treatment. Psychoanalytic Quarterly, 47, 524-567.

Frieswyk, S. H., Allen, J. G., Colson, D. B., Coyne, L., Gabbard, G. O., Horwitz, L., & Newsome, G. (1986). Therapeutic alliance: Its place as a process and outcome variable in dynamic psychotherapy research. Journal of Consulting and Clinical Psychology, 54, 32-38.

Gantt, S. P. (1984). A view of psychotherapy: An integration of experiential psychotherapy and existential-phenomenological philosophy. Dissertation Abstracts International, 45 (08), 2686B. (University Microfilms No. AAI84-24595)

Garfield, S. L. (1992). Major issues in psychotherapy research. In D. K. Freeshheim (Ed.), History of psychotherapy: A century of change (pp. 335 – 359). Washington DC: American Psychological Association.

Garfield, S. L., & Kurtz, R. (1976). Clinical psychologists in the 1970s. American Psychologist, 31, 1-9.

Gedo, J. E. (1989). Self psychology: A post-Kohutian view. In D. W. Detrick & S. P. Detrick (Eds.), Self psychology: Comparison and contrasts. Hillsdale, NJ: Analytic Press.

Gedo, J. E., & Goldberg, A. (1973). Models of the mind. Chicago: The University of Chicago Press.

Gholson, B., & Barker, P. (1985). Kuhn, Lakatos and Laudan: Applications in the history of physics and psychology. American Psychologist, 40, 755-769.

Giovacchini, P. L. (1982). Structural progression and vicissitudes in the treatment of severely disturbed patients. In P. L. Giovacchini & L. B. Boyer (Eds.), Technical factors in the treatment of the severely disturbed patient. New York: Jason Aronson.

Giovacchini, P. L. (1984). The psychoanalytic paradox: The self as a transitional object. Psychoanalytic Review, 71, 81-104.

Goldberg, A. (1988). A fresh look at psychoanalysis. Hillsdale, NJ: Analytic Press.

Goldfried, M. R. (1985) In vivo intervention or transference. In D. W. London (Ed.), Therapist's dilemmas. London: Harper & Row.

Greenberg, J. R., & Mitchell, S. A. (1983). Object relations in psychoanalytic theory. Cambridge, MA: Harvard University Press.

Grünbaum, A. (1984). The foundations of psychoanalysis: A philosophical critique. Berkeley, CA: University of California Press.

Howard, K. I., Kopta, S. M., Krause, M. S., & Orlinsky, D. E. (1986). The dose-effect relationship in psychotherapy. American Psychologist, 41, 159-164.

Hughes, J. M. (1989). Reshaping the psycho-analytic domain: The work of Melanie Klein, W. R. D. Fairbairn, & D. W. Winnicott. Berkeley, CA: University of California Press.

Jacobson, E. (1964). The self and object world. New York: International Universities Press.

Jacobson, E. (1971). Depression: Comparative studies of normal, neurotic and psychotic conditions. New York: International Universities Press.

Johnson, M. E., & Brems, C. (1991). Comparing theoretical orientations of counseling and clinical psychologists: An objective approach. Professional Psychology: Research and Practice, 22, 133-137.

Karasu, T. B. (1986). The specificity vs nonspecificity dilemma: Toward identifying therapeutic change agents. The American Journal of Psychiatry, 143, 687-695.

Kazdin, A. E., & Bass, D. (1989). Power to detect differences between alternative treatments in comparative psychotherapy outcome research. Journal of Consulting and Clinical Psychology, 57, 138-147.

Kernberg, O. (1975). Borderline conditions and pathological narcissism. New York: Aronson.

Kernberg, O. (1976). Object relations theory and clinical psychoanalysis. New York: Aronson.

Kernberg, O. (1980). Internal world and external reality. Northvale, NJ: Aronson.

Kernberg, O. (1984). Severe personality disorders: Psychotherapeutic strategies. New Haven, CT: Yale University Press.

Kernberg, O., Bernstein, E., Coyne, L., Appelbaum, A., Horwitz, L., & Voth, H. (1972). Psychoanalysis and psychotherapy: Final report of the Menninger Foundation Psychotherapy Research Project. Bulletin of the Menninger Clinic, 36, 1-275.

Kiesler, D. J. (1973). The process of psychotherapy: Empirical foundations and systems of analysis. Chicago: Aldine Publishing.

Kiesler, D. J. (1982). Interpersonal theory for personality and psychotherapy. In J. C. Anchin & D. J. Kiesler (Eds.), Handbook of interpersonal psychotherapy. New York: Pergamon Press.

Klein, M. (1964). Contributions to psychoanalysis, 1921-1945. New York: McGraw-Hill.

Klein, M. (1975). Envy and gratitude and other works, 1946-1963. New York: Delacorte Press.

Koch, S. (1981). The nature and limits of psychological knowledge: Lessons of a century of quasi "science." American Psychologist, 36, 257-269.

Kohut, H. (1971). The analysis of the self. New York: International Universities Press.

Kohut, H. (1977). The restoration of the self. New York: International Universities Press.

Kohut, H. (1984). How does analysis cure? Chicago: University of Chicago Press.

Kohut, H., & Wolf, E. S. (1978). The disorders of the self and their treatment: An outline. International Journal of Psycho-Analysis, 59, 413-425.

Lakatos, I. (1978). The methodology of scientific research programmes: Philosophical papers – Volume 1. [Edited by J. Worrall & G. Currie]. Cambridge: Cambridge University Press.

Lakatos, I. (1978). Mathematics, science and epistemology: Philosophical papers – Volume 2. [Edited by J. Worrall & G. Currie]. Cambridge: Cambridge University Press.

Lambert, M. J., & Hill, C. E. (1994) Assessing psychotherapy outcomes and processes. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change. (pp 72-113). New York: Wiley.

Lambert, M. J., & Bergin, A. E. (1992). Achievements and limitations of psychotherapy research. In D. K. Freedheim (Ed.), History of psychotherapy: A century of change (pp 360-390). Washington, DC: American Psychological Association.

Lambert, M. J., Shapiro, D. A., & Bergin, A. E. (1986) The effectiveness of psychotherapy. In S. L. Garfield & A. E. Bergin (Eds.), Handbook of psychotherapy and behavior change. New York: Wiley.

Laudan, L. (1977). Progress and its problems. Berkeley, CA: University of California Press.

Laudan, L. (1981). Science and hypothesis. Boston: Reidel.

Laudan, L. (1983). The demise of the demarcation problem. In R. S. Cohen & L. Laudan (Eds.), Physics, philosophy and psychoanalysis: Essays in honor of Adolf Grünbaum. Dordrecht, Holland: D. Reidel.

Laudan, L. (1984). Science and values. Berkeley, CA: University of California Press.

Laudan, L. (1990). Science and relativism. Chicago, IL: The University of Chicago Press.

Leary, D. E. (1984). Philosophy, psychology, and reality. American Psychologist, 39, 917-919.

Levenson, E. (1972). The fallacy of understanding: An inquiry into the changing structure of psychoanalysis. New York: Basic Books.

Levenson, E. (1983). The ambiguity of change: An inquiry into the nature of psychoanalytic reality. New York: Basic Books.

London, P. (1988). Eclectic psychotherapy gets "uppity." Contemporary Psychology, 33, 697-698.

Luborsky, L., Singer, B., & Luborsky, L. (1975). Comparative studies of psychotherapies: Is it true that "Everyone has one and all must have prizes?" Archives of General Psychiatry, 32, 995-1008.

Mahler, M. (1968). On human symbiosis and the vicissitudes of individuation. New York: International Universities Press.

Mahler, M. (1972). A study of the separation-individuation process. The Psychoanalytic Quarterly, 41, 487-506.

Mahler, M., Pine, F., & Bergman, A. (1975). The psychological birth of the human infant. New York: Basic Books.

Manicas, P. T., & Secord, P. F. (1983). Implications for psychology of the new philosophy of science. American Psychologist, 38, 399-413.

Manicas, P. T., & Secord, P. F. (1984). Implications for psychology: Reply to comments. American Psychologist, 39, 922-926.

Meichenbaum, D. (1988). What happens when the "brute data" of psychological inquiry are meanings: Nurturing a dialogue between hermeneutics and empiricism. In S. B. Messer, L. A. Sass, & R. L. Woolfolk (Eds.), Hermeneutics and psychological theory: Interpretive perspectives on personality, psychotherapy, and psychopathology. New Brunswick, NJ: Rutgers University Press.

Meichenbaum, D., & Gilmore, J .B. (1984). The nature of unconscious processes: A cognitive-behavioral perspective. In K. S. Bowers & D. Meichenbaum (Eds.), The unconscious reconsidered New York: Wiley.

Mendelsohn, E., & Silverman, L. (1987). The empirical study of controversial clinical issues in psychoanalysis: Investigating different proposals for treating narcissistic pathology. In R. Stern (Ed), Theories of the unconscious and theories of the self. Hillsdale, NJ: The Analytic Press.

Merbaum, M., & Butcher, J. N. (1982). Therapists' liking of their psychotherapy patients: Some issues related to severity of disorder and treatability. Psychotherapy: Theory, Research and Practice, 19, 69-77.

Michels, R. (1977). Treatment of the difficult patient in psychotherapy. Canadian Psychiatric Association Journal, 22, 117-121.

Michels, R. (1985). Introduction to panel: Perspectives on the nature of psychic reality. Journal of the American Psychoanalytic Association, 33, 515-519.

Michels, R. (1986, October). Repetition: Recurrent themes (knots) in life and therapy. Paper presented at the annual institute and conference of the American Academy of Psychotherapists, Atlanta, GA.

Modell, A. (1968). Object love and reality. New York: International Universities Press.

Mulaik, S. A. (1984). Realism, pragmatism, and the implications of the new philosophy of science for psychology. American Psychologist, 39, 919-920.

Neill, J. R. (1979). The difficult patient: Identification and response. Journal of Clinical Psychiatry, 40, 209-212.

Norcross, J. C., & Prochaska, J. O. (1982). A national survey of clinical psychologists: Affiliations and orientations. The Clinical Psychologist, 35, 1-6.

Norcross, J. C., & Prochaska, J. O. (1983). Clinicians' theoretical orientations: Selection, utilization, and efficacy. Professional Psychology: Research and Practice, 14, 197-208.

Norcross, J. C., & Prochaska, J. O. (1988). A study of eclectic (and integrative) views revisited Professional Psychology: Research and Practice, 19, 170-174.

Norcross, J. C., & Saltzman, N. (1990). The clinical exchange: Toward integrating the psychotherapies. In N. Saltzman & J. C. Norcross (Eds.), Therapy wars: Contention and convergence in differing clinical approaches. San Francisco: Jossey-Bass.

Norcross, J. C., Saltzman, N., & Giunta, L. C. (1990) Contention and convergence in the psychotherapies in N. Saltzman & J. C. Norcross (Eds.), Therapy Wars: Contention and convergence in differing clinical approaches. (pp 242 – 260). San Francisco, CA: Jossey-Bass.

Orlinsky, D. E., Growe, K., & Parts, B. K. (1994). Process and outcome in psychotherapy – Noch einmal. In A. E. Bergin & S .L. Garfield (Eds.), Handbook of psychotherapy and behavior change. (pp 270 – 376). New York: Wiley.

Orlinsky, D. E., & Howard, K. I. (1986) Process and outcome in psychotherapy. In S. L. Garfield & A. E. Bergin (Eds.), Handbook of psychotherapy and behavior change. New York: Wiley.

Pine, F. (1990). Drive, ego, object and self: A synthesis for clinical work. New York: Basic Books.

Popper, K. R. (1962). Conjectures and refutations. New York: Basic Books.

Prochaska, J. O. (1979). Systems of psychotherapy: A transtheoretical approach. Homewood, IL: Dorsey Press.

Prochaska, J. O., & Norcross, J. O. (1982). The future of psychotherapy: A Delphi poll. Professional Psychology: Research and Practice, 13, 620-627.

Robinson, D. N. (1984). The new philosophy of science: A reply to Manicas and Secord. American Psychologist, 39, 920-921.

Russell, R. L. (1986). The inadvisability of admixing psychoanalysis with other forms of psychotherapy. Journal Of Contemporary Psychotherapy, 16, 76-86.

Sass, L. A. (1988). Humanism, hermeneutics, and the concept of the human subject. In S. B. Messer, L. A. Sass, & R. L. Woolfolk (Eds.), Hermeneutics and psychological theory: Interpretive perspectives on personality, psychotherapy, and psychopathology. New Brunswick, NJ: Rutgers University Press.

Searles, H. F. (1986). My work with borderline patients. Northvale, NJ: Aronson.

Singer, B. A., & Luborsky, L. B. (1977). Countertransference: The status of clinical versus quantitative research. In A. S. Gurman & A. M. Razin (Eds.), Effective psychotherapy: A handbook of research. New York: Pergamon.

Skodal, A., Buckley, P., & Charles, E. (1983). Is there a characteristic pattern to the treatment history of clinical outpatients with borderline personality. Journal of Nervous and Mental Disease, 171, 405-410.

Smith, D. (1982). Trends in counseling and psychotherapy. American Psychologist, 37, 802-809.

Smith, M. L., Glass, G. U., & Miller, T. I. (1980). The benefits of psychotherapy. Baltimore: Johns Hopkins University Press.

Smith, R. J., & Steindler, E. M. (1983). The impact of difficult patients upon treaters: Consequences and remedies. Bulletin of the Menninger Clinic, 47, 107-116.

Spett, M. C. (1983). All psychologists are not members of divisions 12 and 17. American Psychologist, 38, 498.

Spyer, T. C. (1972). Ethical problems and some guidelines. In G. D. Goldman and G. Stricker (Eds.), Practical problems of private psychotherapy practice. Springfield, IL: Charles C. Thomas.

Stern, D. B. (1987a). Unformulated experience and transference. Contemporary Psychoanalysis, 23, 484-491.

Stern, D. B. (1987b). An interpersonal understanding of an object relations view. Contemporary Psychotherapy Review, 4, 67-79.

Stern, D. N. (1985) The interpersonal world of the infant: A view from psychoanalysis and developmental psychology. New York: Basic Books.

Stern, R. (1987). Introduction. In R. Stern (Ed.), Theories of the unconscious and theories of the self. Hillsdale, NJ: Analytic Press.

Stolorow, R. D., & Atwood, G. E. (1987). From the subjectivity of science to a science of subjectivity. In R. Stern (Ed.), Theories of the unconscious and theories of the self. Hillsdale, NJ: Analytic Press.

Stolorow, R. D., Brandshaft, B., & Atwood, G. E. (1983). Intersubjectivity in psychoanalytic treatment. Bulletin of the Menninger Clinic, 47, 117-128.

Stone, M. H. (1985). Negative outcome in borderline states. In D. T. Mays & C. M. Franks (Eds.), Negative outcome in psychotherapy and what to do about it. New York: Springer.

Strupp, H. H. (1980a). Success and failure in time limited psychotherapy: A systematic comparison of two cases. Comparison 1. Archives of General Psychiatry, 37, 595-603.

Strupp, H. H. (1980b). Success and failure in time limited psychotherapy: A systematic comparison of two cases. Comparison 2. Archives of General Psychiatry, 37, 708-716.

Strupp, H. H. (1980c). Success and failure in time limited psychotherapy: Further evidence. Archives of General Psychiatry, 37, 947-954.

Strupp, H. H. (1987). Psychotherapy research: Current trends. Psychotherapy in Private Practice, 5 (4), 1-14.

Strupp, H. H. (1989). Psychotherapy: Can the practitioner learn from the researcher? American Psychologist, 44, 717-724.

Strupp, H. H., Butler, S. F., & Rosser, C. L. (1988). Training in psychodynamic therapy. Journal of Consulting and Clinical Psychology, 56, 689-695.

Strupp, H. H., & Hadley, S. W. (1979). Specific vs nonspecific factors in psychotherapy. Archives of General Psychiatry, 36, 1125-1136.

Strupp, H. H., & Hadley, S. W. (1985). Negative effects and their determinants. In D. T. Mays & C. M. Franks (Eds.), Negative outcome in psychotherapy and what to do about it. New York: Springer.

Strupp, H. H., & Howard, K. I. (1992). A brief history of psychotherapy research. In D. K. Freedheim (Ed.), History of psychotherapy: A century of change. Washington, DC: American Psychological Association.

Sullivan, H. S. (1947). Conceptions in modern psychology. Washington, DC: William Alanson White Foundation.

Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton.

Sullivan, H. S. (1956). Clinical studies in psychiatry. New York: Norton.

Sullivan, H. S. (1972). Personal psychopathology. New York: Norton.

Summers, F. (1994). Object relations theories and psychopathology: A comprehensive text. Hillsdale, NJ: The Analytic Press.

Waldinger, R. J., & Gunderson, J. G. (1985). Completed psychotherapies with borderline patients. American Journal of Psychotherapy, 38, 190-202.

Wallerstein, R. S. (1986). Forty-two lives in treatment: A study of psychoanalysis and psychotherapy. New York: Guilford Press.

Weiner, M. F. (1982). The psychotherapeutic impasse. New York: The Free Press.

Weiner, M. F. (1978). Therapist disclosure: The use of self in psychotherapy. Boston: Butterworths.

Whitaker, C. A., & Malone, T. P. (1980). The roots of psychotherapy. New York: Brunner/Mazel.

Williston di Bella, G. A. (1979). Educating staff to manage threatening paranoid patients. American Journal of Psychiatry, 136, 333-335.

Winnicott, D. W. (1958). Through paediatrics to psychoanalysis. New York: Basic Books

Winnicott, D. W. (1965). The maturational process and the facilitating environment. New York: International Universities Press.

Winnicott, D. W. (1971). Playing and reality. London: Routledge.

Winnicott, D. W. (1989). Psycho-analytic explorations. Cambridge, MA: Harvard University Press.

Winnicott, D. W. (1988). Human nature. New York: Schocken Books.

Wolberg, L. R. (1995) The technique of psychotherapy. (4th ed.) Northvale, NJ: Aronson.

Wolfe, B. E., & Goldfried, M. R. (1988). Research on psychotherapy integration: Recommendations and conclusions from an NIMH workshop. Journal of Consulting and Clinical Psychology, 56, 448-451.

Wolf, E. S. (1988). Treating the self: Elements of clinical self psychology. New York: Guilford Press.

Woolfolk, R. L., Sass, L. A., & Messer, S. B. (1988). Introduction to hermeneutics. In S. B. Messer, L. A. Sass, & R. L. Woolfolk (Eds.), Hermeneutics and psychological theory: Interpretive perspectives on personality, psychotherapy, and psychopathology. New Brunswick, NJ: Rutgers University Press.

Zook, A., & Walton, J. M. (1989). Theoretical orientations and work settings of clinical and counseling psychologists: A current perspective. Professional Psychology: Research and Practice, 20, 23-31.