Object Relations in Psychoanalytic Theory
Mitchell, Stephen A. & Greenberg, Jay R (1983)
Chapter 1- Object Relations & Psychoanalytic Models

The daily work of the psychoanalyst is intimately bound up with his patients' relations with other people. Like everybody else, patients spend a good deal of their time talking about people. Even when their associations run toward concerns somewhat divorced from the mainstream of social intercourse--to dreams, fantasies, symptoms, and so on -- the presence of others can always be inferred. Moreover, the patient in analysis is talking to someone; his communication is shaped by his understanding of and relation with the person he is talking to. All theories of psychoanalysis recognize this. In Freud's earliest drive theory (1905a), the object of a drive (in broad terms, the person toward whom the drive is directed) was, along with its source and aim, postulated as one of its essential characteristics. Although the object was seen as the most variable element of drive, one not inherently or originally connected with it, there could be no expression of drive demand without at least an implicit object. Drive, insofar as it was psychologically rather than physiologically comprehensible, became known through its derivatives, through its direction toward some object. In this sense, all psychoanalytic knowledge must begin with the individual's relations with others.
    Psychoanalytic approaches to object relations are, however, infinitely complicated by the realization that the "people" about whom the patient is talking do not necessarily behave in a way that another observer of those same people would confirm. This fact appeared dramatically in the first psychoanalytic treatment ever attempted, that of Anna 0 (Breuer and Freud, 1895). Anna O's pseudocyesis and her belief that Breuer was acting like her lover rather than like her physician, initially led not to increased understanding of the dynamics of human interaction but to the premature termination of her therapy. It was left to Freud, with his courage in the face of conventionally unacceptable phenomena, to interpret Anna O's reaction in a way that deepened our comprehension of people's interaction with each other. The resulting theory of transference made it impossible ever again to assume that the "objects" about whom patients talk necessarily correspond in a one-toone fashion with the "real people" of the external world. The concept of transference suggests that the "object" of the patient's experience (be it analyst, friend, lover, even parent) is at best an amended version of the actual other person involved. People react to and interact with not only an actual other but also an internal other, a psychic representation of a person which in itself has the power to influence both the individual's affective states and his overt behavioral reactions.
     Examples of this are commonplace in clinical practice outside of the transference in the more technical and limited sense of the term. For instance, a patient, a man of middle age who lives alone and who in treatment has been talking about feelings of shame at having wasted many years of his life pursuing chimerical goals, reports in session that his niece and her boyfriend will be visiting him for the holidays. In preparation for this visit he has been polishing his lamps, wiping fingerprints off the walls, and, in general, preparing for the onslaught of intensely critical, parental intruders. His mood as he describes his preparation matches the story; he is apprehensive, timid, and embarrassed by the living conditions his guests will observe. In the next session, following the holiday, he says that the visit went surprisingly well, except that the "kids" who stayed with him were "bums" and "slobs," wanting only to lie around in bed all day, reveling in the freedom to do this away from their watchful parents. His affect once again matches the situation as he describes it. He is haughty and contemptuous, condescending and judgmental.
     What is striking about this very ordinary example is the tangential relationship between the patient's description of his responses to his guests and any "actual" characteristics of the niece and her boyfriend. The only consistency between the two sessions lies in the relationship described, that of an angry, critical, and self-righteous parent and a misbehaving, shameful child. In the first session the guests are assigned the role of the parental figures; in the second the role is assumed by the patient. We may speculate that it was the visitors' failure to conform to the patient's expectations -- their failure to act parental -- that instigated the dramatic shift in his perception of them (and of himself). For the purpose of the present discussion, however, what matters is that this man's account of an experience with other people is decisively shaped by a pattern of relationship that includes a template of the other and, in everyday language, is "carried around in his head." The relative impact of the characteristics of actual people and of these internal images varies widely among different individuals, but their presence and activity is to some extent demonstrable in everyone.
     The existence of these mental representations of others, sharing as they do some of the characteristics of "real" people as well as some of their capacity to trigger behavioral response, yet being demonstrably "different," raises critical conceptual problems for any dynamic theory of the mind. Such images go under various names in the psychoanalytic literature. In different theoretical systems they are called variously "internal objects," "illusory others," "introjects," "personifications," and the constituents of a "representational world." Their functions within the psychic economy are likewise a matter of debate. They may be understood as serving as a kind of loose anticipatory image of what is to be expected from people in the real world; as becoming closely entwined with the individual's experience of who he is; as persecutors, fulfilling the function of a kind of critical internal fifth column; or as a source of internal security and resource, invoked in times of stress and isolation.
     What is generally agreed upon about these internal images is that they constitute a residue within the mind of relationships with important people in the individual's life. In some way crucial exchanges with others leave their mark; they are "internalized" and so come to shape subsequent attitudes, reactions, perceptions, and so on. This observation presents the psychoanalytic theorist with a range of difficult questions to which a great deal of contemporary theorizing is directed. 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?
     The term "object relations theory," in its broadest sense, refers to attempts within psychoanalysis to answer these questions, that is, to confront the potentially confounding observation that people live simultaneously in an external and an internal world, and that the relationship between the two ranges from the most fluid intermingling to the most rigid separation. The term thus designates theories, or aspects of theories, concerned with exploring the relationship between real, external people and internal images and residues of relations with them, and the significance of these residues for psychic functioning. Approaches to these problems constitute the major focus of psychoanalytic theorizing over the past several decades.
     Discussion of theories of object relations is complicated by the fact that the term has been used in many different contexts and with any number of different connotations and denotations, resulting in considerable ambiguity and confusion. Some authors take exception to the broad usage just described, arguing that it threatens to deprive the term of all specificity of meaning and to blur significant areas of theoretical disagreement (Lichtenberg, 1979). They prefer to restrict the designation "object relations theory" to the works of a particular theorist. But in this usage a further problem develops, because the term is applied to more than one, often incompatible, theoretical stance. It is often used exclusively to describe the approach developed by Melanie Klein, and equally often with respect only to the theory of W. R. D. Fairbairn, despite the fact that the nature, origin, and content of "objects" varies dramatically in these two approaches. In more recent years, Otto Kernberg (1976) has applied the term to his own particular blend of the ego psychology of Jacobson and Mahler, influenced by terms and emphases derived from the writings of Melanie Klein.
     A further difficulty with the term is that its use by many authors has tendentious and polemical overtones. Guntrip (1969) opposes the "object relations theory" of Fairbairn and Winnicott to what he considers the "mechanistic" psychology of Hartmann and the American ego psychologists, a distinction that allows him to conclude that the former is better because it is "more human." Theorists operating within the more orthodox psychoanalytic tradition use the term as one of opprobrium. It is a way of accusing another author of concentrating on the psychological superficialities (behavior with other people) at the expense of the mental depths. "Object relations theory" in the view of these authors implies a concession to behaviorism; the phenomena to which it refers are more adequately described by the concept of "drive derivative," referring to the manifestations of drive as they appear in the experience of the individual (Brenner, 1978). Another group of theorists, particularly the followers of Harry Stack Sullivan, view `object relations" as a weak term of compromise (Witenberg, 1979). Object relations, they argue, are interpersonal relations, but the term allows its user to proclaim continued allegiance to drive theory.
     It might be argued that the term object relations theory is overused, too variable and hopelessly entangled with theoretical dispute to warrant retention. With so unclear a referent, this argument might run, to attempt a theoretical approach to the treatment of object relations is a project based on shifting sands. However, because the term is so widely used in psychoanalytic literature, the substitution of a set of novel, more narrowly defined terms would necessitate unfeasible translation of massive segments of the existing psychoanalytic literature into a new language. Simply dropping the term would result in more rather than less confusion, particularly in a project that centers around an exposition and comparison of different theoretical traditions. Accordingly, we will retain the term object relations in a general sense; it is important, however, that we specify our definition at the outset, and that we keep in focus the term's specific referents for the different theories we examine.
     The concept of object relations originated as an inherent part of Freud's drive theory. The "object" in Freud's language is the libidinal object (in the later theory also the object of the aggressive drive). In this sense the meaning of the word "object" parallels its dual usage in everyday English, in which it refers both to a thing and to a goal or target. Freud's object is a thing, but it is not any thing; it is the thing which is the target of a drive. The "object" of psychoanalysis is thus not the "object" of academic psychology, that is, simply an entity existing in time and space (see Piaget, 1937). In its original usage, the concept of object was intertwined with and contingent upon the concept of drive. Despite this connection, some psychoanalytic theorists have retained the terms "object" and "object relations" although they eliminated entirely the concept of drive in the classical Freudian sense (see Fairbairn, 1952; Guntrip, 1969). Other theories which stress the role of the object and which hold that they deal with the problems of object relations are drive theories (see Jacobson, 1964; Kernberg, 1976). Thus, the term, although seemingly theory-bound, does not adequately discriminate between those approaches which accept Freudian libido theory and those which do not.
     Because of this ambiguity we reject narrow definitions of object relations. Dispute as to whose theory constitutes a "true" object relations approach is a barren enterprise that has caused endless confusion to students of psychoanalysis. In this book the term refers to individuals' interactions with external and internal (real and imagined) other people, and to the relationship between their internal and external object worlds. We believe that the term retains utility only in this broad usage. All psychoanalytic theories contain theories of object relations; they must if they are to maintain contact with the day-to-day experience of the individual. Various approaches are differentiated by their use of observations concerning relations with others and by the extent to which these observations are integrated with classical drive theory. It follows from this that our usage explicitly dissolves any assumption of a tie between the terms "object" and "object relations" and the concept of underlying drives. Despite its origin in drive theory, we believe that the term "object," divorced from that origin, retains its theoretical utility:
     1. Within the history of psychoanalysis the term has been used to describe both real people in the external world and the images of them that are established internally. This dual connotation is useful in describing the interchange between "inside" and "outside" that occurs in every analytic treatment (see Modell, 1968; Stierlin, 1970).
     2. The word "object" is vague enough in ordinary usage to connote a wide range of characteristics. This accords well with the experience of patients, whose world can be populated by "objects" which are active or static, benign or malignant, alive or dead, and so on. The very generality of the term indicates the variability of one's experience with other people.
     3. Although the term itself is general, the concept object suggests tangibility. Again this accords well with the experience of patients, who see exchanges with their objects as having all the experiential reality of transactions in the external world. Although in the phenomenology of the patient's experience "internal objects" are felt actually to exist, our use of the term does not imply the physical reality of such objects. These are certainly not entities, or homunculi within the mind.
     4. Returning to the ordinary usage of the term, an object, despite its durability, can be manipulated and modified. It can be reshaped, repainted, cut in two, repaired, even destroyed. This connotation lends itself well to the psychoanalytic concept of intrapsychic operations that can be performed upon objects, and the experiences that many patients report corresponding to these operations.

Conceptual Models in Psychoanalytic Theory
Psychoanalysis is, by its very nature, an interpretive discipline. Psychoanalytic theorizing operates within a continual dialectic of crossfertilization with clinical data. Freud developed his theory from the clinical material provided for him by his patients; the theory in turn continually shapes and illuminates the newly emerging clinical data. Although the practicing psychoanalyst attempts to suspend his formal theoretical preconceptions as he listens to his patients, to stay as close as possible to the phenomenology of the patient's experience, theory must enter at some point. The very nature of psychoanalytic practice as a collaborative inquiry into the patient's life presupposes that something is missing in the patient's experience of himself. Whether this is conceptualized as being the result of repression (Freud), inattention (Sullivan), disclaimed action (Schafer), self-deception (Fromm), or bad faith (Sartre), the assumption is that some salient aspects of the patient's reality, some crucial dimension of meaning, is absent in his account of his own experience, whether or not the patient is aware of it (Ricoeur, 1970).
     Psychoanalytic theories provide interpretive possibilities aimed at supplying missing dimensions in the patient's account of himself. Each theory selects from the complexity of life certain aspects or dimensions which are understood to lie at the center of human concerns, coloring much of the seemingly diffuse and variegated aspects of the patient's experience. This dimension provides content for interpretations, a reservoir of meanings within which the clinical material can be understood. The basic concepts within each psychoanalytic theory become the warp and woof out of which the complex tapestry of human experience is woven. In the psychoanalytic process itself, if the work is proceeding in a rich and vital way, broad underlying theoretical principles are almost invisible. The analyst, once past his earliest apprenticeship, does not retain his theory in mind for active, focused use as a lexicon of meanings. He does not continuously shift back and forth between his patient's communications and his theory, deciphering piece by piece as he goes along. He listens to the patient's account of his experience, and his attention is drawn now this way, now that. Certain themes stand out, certain pieces do not fit together. Gradually his thoughts about the patient crystallize. Some underlying processes in the patient's life and his communication of his experience emerge. The analyst begins to have a sense of what has been omitted, and this awareness shapes his thoughts concerning possible interpretive interventions. How can he communicate something of his experience and understanding of the salient omissions in the patient's account in a way which will enrich the patient's understanding? Throughout this process, theoretical concepts as such may be missing from the analyst's thoughts. Nevertheless, they provide the invisible backdrop, the unseen framework, within which the analyst hears the patient's story. Thus, basic concepts within psychoanalytic theory provide interpretive possibilities for orienting the clinician
toward crucial and hidden dimensions of meaning by informing his sensibilities as a listener.
     Does the characterization of psychoanalysis as an interpretive discipline challenge its credibility as a scientific discipline? Not at all. Recent approaches to the philosophy and history of science have highlighted the presumptive and interpretive features of all scientific disciplines and shed considerable light on the ways in which psychoanalytic theories function.
     Until the last several decades, Western philosophy of science has been dominated by an understanding of science and its theories grounded in a thoroughgoing empiricism and, in this century, elaborated by the philosophical doctrines of logical positivism. Within this understanding, termed by recent philosophers of science "The Received View," there is an assumed one-to-one correspondence between good theory and actual events and processes in the real world: facts are established irrefutably through objective observation; theories offer different explanations of these facts on the basis of which testable predictions can be made; experimentation determines the correctness or error of the theory; science proceeds through a gradual accumulation of neutral observations and confirmed hypotheses; scientific understanding changes through the absorption of earlier, more limited theories into increasingly broader and more inclusive theories.
     In the 1950s and 1960s this view of the philosophy and history of science came under increasing criticism, and a very different understanding was developed by a series of authors all employing what has been termed "weltanschauung" analyses (Suppe, 1977). In this view, science is not an unbroken line of ideas representing a closer and closer approximation to the "truth." Steeped in the current scientific climate heavily influenced by relativity theory, Heisenberg's Uncertainty Principle, and the seemingly infinite regress of particle physics, contemporary philosophers of science have reconsidered the relation between science and "objective" reality. There are no purely objective facts and observations which lie outside of theory, according to this new view. One's theory, one's understanding, one's way of thinking, determine what are likely to be taken as facts, determine how and what one observes. Observation itself is understood to be "theory-laden." Science, in this view, necessarily takes place within a community, employing a common language, taking for granted certain basic and ultimately untestable premises. Science thus operates within a conceptual perspective, a weltanschauung. Among the different theories developed within this approach, the contributions of Thomas Kuhn (1962) have been most influential and most widely employed. Truth, he argues, is unknowable; the concern of science is with problem solving, and the history of scientific ideas consists of a series of models, "ways of seeing the world" (p. 4) which are more or less useful in solving problems. The more comprehensive and influential of these models, or paradigms, do not necessarily presuppose or incorporate the preceding paradigms. They represent a series of alternative solutions, pictures of the universe, which are crystallized out of pieces of data and concepts. Kuhn suggests that those contributions which are to be understood as paradigms have two defining characteristics: "their achievement was sufficiently unprecedented to attract an enduring group of adherents away from competing modes of scientific activity. Simultaneously, it was sufficiently open-ended to leave all sorts of problems for the redefined group of practitioners to resolve" (p. 10). Each major paradigm forms the framework for observation and theorizing for a period of time and is eventually replaced by a new crystallization, a new picture, casting a different light on things, useful in solving different problems. The reigning paradigm at any particular time has an enormous influence on scientific activity. It determines what data are taken to be meaningful and real, what methods are to be considered valid, and where the scientist is to be positioned in relation to his object of study.
     One of the most intriguing aspects of Kuhn's work is his focus on the transition between paradigms. If science consists of a series of discontinuous models, how does one move from one paradigm to the next? What are the features of such a transition? His depiction of this process highlights what might be considered to be the more "political" features of the history of scientific ideas. Paradigms, because they are models of reality taken for the "truth," inspire loyalties. During the peak period of influence of a paradigm, nearly all workers within the particular field are under its sway. There is shared agreement on the epistemological assumptions, the methodological approaches, and the observational perimeter which it provides. As that peak period is passed, new data, new ideas begin to emerge outside the boundaries legitimized by the paradigm. At that point an array of different strategic options presents itself. Some remain loyal to the old paradigm and discount the validity of new, discordant data and concepts altogether. Another strategy, which might be termed accommodation, entails attempts to stretch the concepts and boundaries of the old paradigm to encompass what is new. This may work for some time, depending on the novelty of the new data, the resiliency of the old paradigm, and the interpreter's skill at casuistic explanations. However, at some point, Kuhn suggests, "retooling" is unavoidable -a shift in paradigms is "economically" necessary and essential to further progress. Then the old paradigm fades and a new one emerges.
     Kuhn and other proponents of weltanschauung analyses have come under criticism of various sorts. Kuhn's term "paradigm" has been attacked as so vague and elastic to be meaningless (Masterman [1971] catalogues twenty-two different meanings). His view of science as proceeding through successive "revolutions" has been judged applicable to certain historical periods and scientific disciplines but not to others. Proponents of weltanschauung analyses in general have been accused of giving too little weight to the place of rationality in science and to the extent to which objective, reasoned argument plays an important role in choices among different theories. Thus, while "The Received View" was based on an unquestioned belief in the objectivity of science, and weltanschauung analyses emphasized nonobjective, presumptive features, contemporary philosophers of science are struggling toward a middle course. Kuhn, in response to these criticisms and suggestions, has revised his approach. In his later work he specifies the kind of objective criteria which do play a part in theory choice, and has himself rejected the term "paradigm" as so broad and vague as to invite oversimplification and misuse.
     Scientific communities hold in common many complex kinds of beliefs and "cognitive commitments," Kuhn now suggests, and this vast array of beliefs constitutes a "disciplinary matrix." Among the various kinds of conceptual tools employed within a disciplinary matrix for explaining phenomena are many different kinds of models. Some serve as simple analogies or heuristic devices-"A behaves or can be regarded like B." Other models have a much deeper and more pervasive role within a scientific community, in which the model provides a basic framework of orientation and belief, serving as "objects of metaphysical commitment: the heat of a body is the kinetic energy of its constituent particles, or, more obviously metaphysical, all perceptible phenomena are due to the motion and interaction of qualitatively neutral atoms in the void" (1977, p. 298). It is precisely this sort of model which psychoanalytic theorizing organizes itself around; thus, in speaking of models within psychoanalytic theory, we are employing the term to refer to the kind of metaphysical commitment that Kuhn describes. In employing his approach, we are not necessarily implying its applicability as a general account of all sciences, nor are we entering the complex philosophical tangle concerned with questions of objectivity, subjectivity, and verification of theories. We are suggesting that Kuhn's approach to the development of scientific ideas and his definition of models as metaphysical commitments are highly applicable to the history of psychoanalytic thought and constitute a useful way to approach the different strategies of theory-construction.
     Psychoanalytic theories operate as models reflecting metaphysical commitments because they are based upon untestable premises concerning four fundamental issues. The first issue, the approach to which shapes many other aspects of the theory, concerns the basic unit of analysis. What is primary and what is derivative? What are the basic constitutive building blocks of experience: drives, wishes, values, goals, relations with others, identifications, choices, action, and so on? Of what are the "structures," the patterns, the "stuff" of the personality composed? Once this broad foundation has been delineated, three other more specific and overlapping dimensions of theoretical concern must be addressed. First, motivation: What do people want? What are the prevailing and underlying goals of human activity? Second, development: in the transformation from the relatively unformed infant to the relatively patterned adult, what are the crucial events? Third, structure: What gives an individual his or her distinctive shape, governing the regularity of behavior, events, and relationships within an individual life? What mediates between past events and present experience and behavior? Concepts in each of these areas touch on, interact with, and are contingent upon concepts in other areas. Taken together, the approaches to issues within each area form a broad framework for generating clinical hypotheses and interpretive possibilities. When different theorists reflect fundamental similarities in their approaches we consider them as working within the same model. Psychoanalytic theories which deal with these issues in a fashion which is fundamentally different from preceding solutions are considered to be operating out of an alternative model.
     The most significant tension in the history of psychoanalytic ideas has been the dialectic between the original Freudian model, which takes as its starting point the instinctual drives, and an alternative comprehensive model initiated in the work of Fairbairn and Sullivan, which evolves structure solely from the individual's relations with other people. Accordingly, we designate the original model the drive/structure model and the alternative perspective the relational/structure model. We have chosen these terms as a means of highlighting the differences between the models in their metaphysical commitments concerning the underlying content of mind. All psychoanalytic theories presuppose enduring, characteristic patterns and functions that typify the individual personality. Such patterns and functions organize experience and mediate between experience and subsequent behavioral response. Most theorists term them "psychic structure," to emphasize their consistency and temporal continuity. (Sullivan [1953], wary of the danger of reification inherent in the use of concepts like "structure," substituted "dynamism.") We retain the term structure, mindful that such usage implies a metaphor of "spatialization" that is not to be taken literally (see Schafer, 1972). We are using it in the sense suggested by Rapaport, who designates "structures" within psychoanalysis as configurations "having a slow rate of change" (1967, p. 701), and Lawrence Friedman, who characterizes structures in terms of a "reactive stability" (1978b, p. 180).