the William Alanson White Institute : In Memory

Contemporary Psychoanalysis, Vol. 20, No. 4 (1984)

Object Relations Theories and the Developmental Tilt

Stephen A. Mitchell, Ph.D.

It is the predicament of the neurotic that he translates everything into the terms of infantile sexuality; but if the doctor does so too, then where do we get?

Joseph Campbell, The Masks of God

THE DESIGNATION "OBJECT RELATIONS THEORY" has been used with reference to a wide range of very different kinds of formulations: from Klein's rich and complex depiction of unconscious fantasy, to Fairbairn's highly abstract, schematic structural model, to Bowlby's ethologically-based theory of attachment, to Winnicott's epigramatic paradoxes and pithy observations about children, to Mahler's powerfully evocative portrayal of the longing for symbiotic fusion, to Jacobson's causuistic emendations of Freud's drive theory.

To employ a common term for such a wide array of disparate points of view runs a risk—the fashionable popularity accruing to "object relations theories" in recent years has blurred important distinctions beneath a hazy aura connoting theory that is new, humanistic, often esoteric, and presumably pertaining to the deepest recesses of the mind and the earliest developmental phases. Is the employment of a common phrase to designate these different groups of theories useful or misleading? What do these heterogeneous theories have in common? What is essential and what is artifactual and political in their formulation? What has been their central role in the development of psychoanalytic ideas?

Since the multiplicity of theories of object relations has been parallelled by a proliferation of different histories and interpretations of what might be regarded as the "object relations movement, " there is no consensus concerning these questions. Perhaps

- 473 -

the must fundamental and interesting disagreement is characterized by the following divergence. Some observers (Kernberg, 1976) ; (Modell, 1968) (and, in a somewhat different vein), (Levenson, 1983) regard object relations theories as an extension of pre-existing theory—object relations theories add to drive theory and the structural model a consideration of the earliest relationships of the infant, which classical theory, in its focus on the Oedipus complex, does not fully illuminate. Other observers (Guntrip, 1971) regard object relations theories as a new, alternate paradigm, grounded in different assumptions concerning the nature of mind, and shifting psychoanalytic theory from a framework in which drives and their derivatives are understood to constitute the basic stuff of mental life, to one in which the primary ingredients are relational configurations, past and present, real and imaginary. If one conceives of traditional psychoanalytic theorizing as constituting a "mainstream, " the first view is that object relations theories have deepened the channel, whereas the second view is that object relations theories have rerouted the stream altogether. (See Lichtenberg, 1983b), (for an interesting development of the metaphor of the psychoanalytic "mainstream.")

Authors (such as Kernberg [1976] and Modell [1968]) who view object relations theory as an extension of traditional theory into earlier, pre-oedipal realms, regard those theorists who reject classical metapsychology (like Fairbairn [1952] and Bowlby [1969]) as extravagant in their claims, unnecessarily and wastefully disregarding the full richness of Freud's contributions. The abandonment of a conceptual framework as complex, elegant and serviceable as classical metapsychology is surely not a loss to be taken lightly. But what of the costs of the other path? What have been the implications and consequences of the attempt to absorb object relations theories into the mainstream? I will designate the most important device through which this absorption has been accomplished the "developmental tilt" and will demonstrate that the employment of this strategy has had pervasive and unfortunate consequences for the manner in which object relations concepts have been articulated and utilized in both psychoanalytic theory and technique.

The Strategy of Accommodation and the Developmental Tilt

Developing a psychoanalytic theory is a process not dissimilar to designing a house, the construction of spaces within which people

- 474 -

live and do things. Different kinds of spaces might be envisioned—the problem in design is to arrange those spaces in relation to each other so that they fit together, and so that the more microscopic and circumscribed clinical insights and emphases rest comfortably on the fundamental metapsychological premises of the theory.

In classical Freudian theory (pre-ego psychology), the conceptual foundation is provided by the concept of drive. All motivational, developmental and structural phenomena, both in life in general, and within the psychoanalytic situation, are understood in terms of drive derivatives and defenses against drive derivatives. Classical theory encompasses within it an account of relations with others, but these relations, like all other phenomena, are understood to consist of transformations of underlying drive pressures and defenses, serving either as vehicles for drive gratification, or as a bulwark in the ego's defenses against drive pressures. In this sense, classical drive theory is perfectly internally consistent, a well-designed and amply supported conceptual edifice.

In the more recent history of psychoanalytic ideas (since the late 1940s), increasingly greater emphasis has been placed on relations with others, past and present, real and imaginary. Psychoanalytic clinicians and theorists began to grant object relations a more central and more pervasive role than before. This created a crisis of design. What is the relationship between object relations and the underlying conceptual foundations of drive? The increasingly greater clinical and theoretical emphasis on object relations placed an enormous strain on the classical model, like a group of cantilevered beams which are called upon to bear more and more ornamentation until they threaten to collapse under the increased weight.

Greenberg and I (1983) have argued that the various strategies within the complex array of contemporary psychoanalytic theories can be grouped around two basic positions, which we have termed the strategy of radical alternative and the strategy of accommodation. Strategists of radical alternatives have abandoned the drive model completely, substituting an alternative conceptual framework to replace the weight-bearing function of the original foundation. Sullivan (1940), (1953), Fairbairn (1952) and Bowlby (1969) are the purest practitioners of this approach. Most other contemporary psychoanalytic authors maintain a loyalty to the classical model in some form, adapting it to enable it to encompass the more general shift in the direction of relational issues. These strategists

- 475 -

of accommodation have developed various and often ingenious devices for bracing and buttressing the drive model, stretching and altering it, to enable it to contain an increasingly greater emphasis on object relations. This more preservative approach characterizes most of the authors generally considered to be "object relations theorists."

How does one both preserve a theory and yet introduce into it new concepts which are at variance with its basic thrust and underlying assumptions? More specifically, how can one grant that Freud was correct in his characterization of psychopathology as entailing conflicts over drives and defenses centered on the Oedipus complex, yet also grant a primary and basic role to the development and patterning of relationships with others? One device has been to alter one or more component parts of the original model to encompass relational processes and issues. Thus, Hartmann (1939) transforms the concept "ego" from an agency whose sole purpose is the control and regulation of drives to an agency encompassing complex and primary relations with the environment (including the interpersonal environment), relations which are relatively independent of the drives. Other theorists have transformed the concept "id" so that the repository of the drives themselves is subject to the impact of early object relations (Jacobson), or actually comprised of relational configurations (Kernberg, 1976). Another device has been the strategic use of diagnosis (Kernberg, 1976) ; (Kohut, 1971) ; (Stolorow and Lachmann, 1980) —classical theory and the structural conflict it depicts is correct for neurosis; however, for more severe disorders (borderlines, narcissistic personality disorders, developmental arrests, etc.), a new model focused on object relations is required. (See Greenberg & Mitchell [1983] for an extended discussion of these various strategies.) One of the most important devices through which accommodation has been accomplished, leading to pervasive implications in the way object relations concepts have been shaped, has been the "developmental tilt"—i.e., Freud was correct in understanding the mind in terms of conflicts among drives; object relations are also important, but earlier.

For many strategists of accommodation the pillar of classical metapsychology, the structural model, is understood to provide an adequate framework for an account of human experience, both normal and pathological, and that account depicts the conflict

- 476 -

among various drive derivatives, and between drive derivatives and the defensive functions of the ego and the superego. When a theorist following this strategy wants to introduce various relational needs and processes as primary in their own right, as irreducible, as neither merely gratifiers nor defenders against drives, they are often introduced as operative before the tripartite structures of id, ego and superego have become separated and articulated. Theorists concerned with linear continuity necessarily preserve the classical theory of neurosis as centered around sexual and aggressive conflicts at the oedipal phase. They set object relations formulations into pre-existing theory by arguing that they pertain to a developmental epoch prior to the differentiation of psychic structures, in the earliest relationship of the mother and infant. The traditional model is jacked up, and new relational concepts are slid in underneath. To return to our architectural metaphor, it is as if a new, complex and roomy foundation level has been set beneath an older edifice; the upper stories remain just as they were, but the center of gravity has shifted downward. The original structure is intact, but unoccupied; the scene of the action has moved downward to the lower levels.

The Developmental Tilt and Its Distortions

Melanie Klein evolved an elaborate account of human experience as a passionate struggle between murderousness, malevolence and envy towards significant others, and a deep sense of love, gratitude and a wish to save and restore them. Michael Balint depicts human relations as a search for a perfect "unconditional" love, offering the possibility of a passive surrender to a trusted and caring nurturance. D. W. Winnicott came to see psychopathology as centering on a struggle between an authentic and spontaneous expression of impulses and wishes and a need to shape oneself around the way others see one, according to the image others provide and seem to require. Margaret Mahler locates the experience of self in a pervasive dialectic between a need for autonomy and self-definition and a desperate longing to surrender to and fuse with another. Heinz Kohut characterizes the self as a bi-polar structure generated from the tension between a need for a warm and embracing recognition, and a need to identify with admired others.

Each of these contributions (presented here in obviously greatly

- 477 -

collapsed, over simplified and schematic form) constitutes an object relations theory generally applicable to human experience at all points within the life cycle. Each offers an account of life's central passions, an account which is at considerable variance from that provided by classical metapsychology, in which human experience is portrayed as a struggle to negotiate between the claims of body-based, asocial psychic tensions and the demands of social reality. In each object relations account, the human organism is seen as inherently social, embedded in a matrix of relationships, seeking relatedness with others in a primary and fundamental fashion. In each account, the passions depicted characterize human longings and fears at all ages. The struggle between destructiveness and hopeful benevolence, the search for all-embracing love, the tension between self-expression and pandering, between autonomy and a longing to fuse, the need for supportive recognition and admired heroes—these are fundamental dimensions of human relations, from infancy through senescence. These various theories all draw on what Greenberg and I have termed the "relational model, " whose basic premises are at variance with the classical drive model. The most essential and salient feature of object relations theories, we argue, is precisely this broad and pervasive departure in fundamental paradigm.

Yet, each of these theorists—Klein, Balint, Winnicott, Mahler, and Kohut—maintains a loyalty, in one form or another, to classical drive theory. One (Mahler) maintains the earlier model in its essentials; another (Klein) preserves its language while changing its meanings; another (Winnicott) proclaims his loyalty although the original model no longer figures meaningfully in his formulations. Despite this diversity in degrees of fealty, each author requires accomodation to make room for his or her own contribution, and therefore many of these innovations have been introduced into psychoanalytic theory via the developmental tilt; consequently, the dynamic issues they depict tend to get characterized as infantile, pre-oedipal, immature, and their persistance in later life is often regarded as a residue of infantilism, rather than as an expression of human relational needs extending throughout the life cycle. We find this tendency even in object relations theorists like Guntrip and Bowlby, who have disgarded drive theory completely, yet whose thinking has been greatly influenced by those major innovators like Melanie Klein and Winnicott who used the developmental tilt to preserve allegience to the classical system.

- 478 -

Authors who preserve some form of allegiance to drive theory yet introduce relational dynamics as earlier, often end up with a bifurcated view of the life cycle. To regard relational issues as prior to drive issues separates human development into two kinds of concerns—young infants have relational needs; older children and adults (those who are healthy or suffer only from neurotic difficulties) struggle with conflicts between instinctual impulses and defenses. Thus, Winnicott distinguishes between early "needs" and later instinctual "wishes"; Stolorow and Lachmann distinguish between "developmental arrests" and later structural conflicts; Kohut distinguishes between disorders of the self and later structural neuroses; Mahler distinguishes between disorders involving the separation-individuation process and later oedipal conflicts, etc. (See Feinsilver, 1983), (for an incisive critique of such dichotimizations as pre-oedipal/post-oedipal, conflict/deficiency, interpretation/repair of deficits, insight/corrective emotional experience.) Is it accurate or feasible to limit relational issues to the earliest developmental phases? Do relational issues emerge sequentially over the course of early infancy, becoming progressively resolved, allowing the child to move on? The latest thinking of some of the more prominent infant researchers suggests that they do not.

Stern (1983), for example, challenges the notion that the "separation-individuation" issue, as depicted by Mahler (1967), is accurately assigned to an early phase of development. Developmental theorists like Mahler and Spitz, have tended to regard one early phase of life as bringing to a head and essentially resolving a particular major life-cycle issue, relational in nature, such as the establishment of basic trust, autonomy, separation-individuation, etc. Stern argues, by contrast, that these issues are most accurately viewed as life-long struggles. The dialectic between union/fusion and differentiation/autonomy experiences, for example, is a perpetual facet of human existence, manifesting itself in the young infant in visual gaze behaviors (Stern, 1977), in the toddler through motility, and in the older child and adult in various symbolic processes. (See Lichtenberg, 1983a), (for a discussion of the recent evidence suggesting that very young infants can differentiate self and object images, which challenges the notion of a specific symbiotic phase.) These are differences not in meaning, or dynamic issues, but in the equipment, motoric and cognitive, through which the child is able to experience the same issue. Thus, collapsing life-long relational issues to early, circumscribed phases via the developmental

1Hartmann warned psychoanalytic theorists against what he called the "genetic fallacy"—the equation of a behavior with its origins, or the assumption that a behavior originating out of conflict is inevitably forever linked to and fueled by conflictual difficulties (1960, p. 93). The distortions Hartmann was addressing pertain to all psychoanalytic theory employing the framework of developmental phases. The potential misuse of the "developmental tilt" might be considered a subcategory of the genetic fallacy, wherein particular kinds of life cycle relational issues are collapsed into their earliest manifestations so as to preserve later developmental epochs as the province of drive-related issues.

- 479 -

tilt distorts the very nature of those issues and the ways they manifest themselves at different points throughout the life cycle.1

The developmental tilt has generated what at times seems to be an infinite regress in claims to developmental priority. A psychodynamic account which the author regards as more basic, more primary than structural conflict, is presented as earlier, leading to the attribution of extra-ordinarily complex affective and cognitive capacities to the newborn (Klein), great weight granted to prenatal and birth experiences (Winnicott, 1949), and even speculations on the effects on the embryo, in its first days, of parental attitudes at the point of conception (Laing, 1976). Deeper is transformed into earlier, rather than more fundamental, as if dynamics attributable to the first months of life or even to prenatal existence still occupy the most basic layers of experience, underlying and governing psychic events and processes of later chronological origin. Thus, theorists attempting to accomodate the drive model to object relations issues attempt to keep instinctual and relational issues temporally separable. By pushing relational issues into an earlier developmental era, they preserve the oedipus complex as still fundamentally instinctual. This mode of introducing theoretical innovation strains credulity; it also skews these innovations in a peculiar way, by collapsing relational issues into the interaction between the mother and infant during the earliest months of life.

Let us consider as a representative example an excerpt from the work of Balint (1968), who introduced rich and clinically useful object relations concepts while remaining loyal, in basic respects, to drive theory. Balint developed the concepts of "primary love" and the "basic fault" in an innovative and clinically useful effort to account for transference/countertransference impasses with certain

- 480 -

kinds of difficult patients. The principle of abstinence central to classical technique, Balint points out, was developed in the context of drive theory. The patient's impulses and wishes must not be gratified, lest they become further entrenched rather than transformed into memory and renounced. However, certain patients, Balint argues, become stuck in analysis, demanding a responsiveness from the analyst, without which they seem unable to progress. Balint characterizes these longings and the patient's efforts to gratify them as a need for "primary love."

In my view, all these processes happen within a very primitive and peculiar object-relationship, fundamentally different from those commonly observed between adults. It is definitely a two-person relationship in which, however, only one of the partners matters; his wishes and needs are the only ones that count and must be attended to; the other partner, though felt to be immensely powerful, matters only in so far as he is willing to gratify the first partner's needs and desires or decides to frustrate them; beyond this his personal interests, needs, desires, wishes, etc., simply do not exist. (1968, p. 23)

Balint has provided an account of the analytic encounter which is based on relational concepts and is alternative to that generated by the drive model. It is not gratification of specific impulses that the patient is seeking, Balint argues, but the need to establish a certain kind of relationship—a state of unconditional love. What is puzzling about Balint's description is his restriction of such longings to the earliest and most "primitive" object relationships. It appears that Balint's depiction of the longing for primary love has wide applicability. Surely, we might define "mature" love as a relationship characterized by mutuality—"When the satisfaction or the security of another person becomes as significant to one as is one's own satisfaction or security, then the state of love exists" (Sullivan, 1940, pp. 4243). Such mutuality, however, seems clearly an ideal, not a normative practice. No matter how mature and healthy, all love relationships are characterized by periodic retreats from mutuality to self-absorption and demands for unconditional sensitivity and acceptance. Many patients (not at all as "regressed" as those Balint sees as suffering from a "basic fault") take many years before their relationships are weighted more in the direction of mutuality than self-absorption. Sullivan argued that most of us are chronically juvenile, integrating relationships on the basis of our own ego-centric concerns, lacking the capacity

- 481 -

for intimacy, for seeing things from the other's perspective as well. Further, it seems particularly odd to depict the emergence in the analytic situation of a preoccupation by the patient with his or her own needs and an experience of the analyst as existing only in relation to those needs as "primitive." Although such longings and demands are organized and expressed differently in different developmental eras and in different types of patients, they are almost inevitable. One might argue that the analytic situation is defined precisely in this way. Free-association, for example, might be considered to embody precisely the freedom to ignore any concern about the analyst's needs; the "fundamental rule" thus is designed to encourage this kind of transient narcissism (Greenberg, personal communication). Most patients have the experience, or struggle to resist the experience, of the analyst as existing only vis-a-vis them; in fact, the absence of such feelings is often understood to reflect a "resistance to the transference." Thus, Balint's concept of primary love provides an illuminating account of relational longings and conflicts throughout the life cycle, but, as with many object relations formulations, these accounts have been collapsed into earliest infancy, "a very primitive and peculiar object-relationship."

It might be argued that the impact upon object relations concepts produced by the developmental tilt is insignificant. The basic concepts are there anyway, such a position would claim. What difference does it make whether relational issues are understood as operating essentially prior to the differentiation of psychic structure and the inception of instinctual conflict?

Such a view would necessarily minimize considerations of the aesthetics and economics of theory-construction, since theories employing the developmental tilt tend to be exceedingly and often unnecessarily complex and contrived. Relational issues are granted temporal priority, but the theory must move inexorably towards the establishment of instinctual conflict at the core of "classical" neurosis. Bridging this conceptual gap is not easy, and often requires the kind of ingenuity for which Rube Goldberg was famous. Kohut's (1971) postulation of two separate libidinal energies and developmental lines (narcissistic-libido and object-libido) and his "principle of complementarity, " and Kernberg's (1976) use of "general systems theory, " are the clearest examples of strained arguments and shifting terminology which serve as bridging concepts, allowing the theorist to start with relational assumptions and

- 482 -

arrive at the traditional version of the Oedipus Complex (Mitchell, 1979). The resultant theories have an oddly unsettling, implausible quality, reminiscent of the architecturally notorious residential college built at Yale whose exterior facade, facing earlier buildings, is done in traditional Gothic style, while the interior facade opens onto a fashionable (for the time of construction) colonial courtyard. One enters the building (or theory) in one century and exits in another! External continuity is preserved at the price of internal contradiction and tension. However, the most important impact on object relations concepts of the developmental tilt is in terms of their clinical applications, and it is to these I now turn.

Clinical Consequences of the Developmental Tilt

The relationship between patient and analyst has, from the very beginnings of psychoanalysis, occupied a central place in all theorizing about the analytic situation and its therapeutic action. The manner in which that relationship has been conceived, however, has undergone many intricate variations and transformations. Although any generalization about this complex conceptual history runs the risk of oversimplification, it is not at all misleading to note that in recent decades, the analytic relationship has been understood as more and more of a real and new relationship than previously. For Freud, the relationship with the analyst was a re-creation of past relationships, a new version struck from the original "stereotype plate" (Freud, 1912). The here-and-now relationship was crucial, but as a replication, as a vehicle for the recovery of memories, the filling in of amnesias, which cured the patient. Contemporary views of the analytic relationship tend to put more emphasis on what is new in the analytic relationship. The past is still important, but as a vehicle for understanding the meaning of the present relationship with the analyst, and it is in the working through of that relationship that cure resides. (See Racker, 1968), (and Gill, 1983, for an extended treatment of this contrast.)

Object relations theorists have played an important part in this redefinition of the nature of the analytic relationship. Not surprisingly, the analyst is seen not just as a projection screen for and interpreter of old object relations, but as offering an opportunity for the development of a new relationship. What does this new relationship consist of? There is a wide range of different accounts. Fairbairn (1952) puts it this way—in order for the patient

- 483 -

to relinquish his tie to bad objects, the tie which is at the core of all psychopathology, he must experience the analyst as a "good object." Objectlessness is impossible; one can't relinquish old attachments unless new ones seem possible and compelling. The analyst must become a good object—this is a formula with which object relations theorists of all persuasions would agree. But what does it mean to say the analyst becomes a "good object"? "Good" in what sense? The analyst provides possibilities for relatedness hithertofor unavailable to or unutilizable by the patient. But what sort of opportunities for relatedness does the analyst provide? It is here that the developmental tilt becomes crucial, since the developmental tilt collapses relational needs in general into the kinds of interactions which characterize the relationship of the small infant and the mother. For many of these authors, the analyst is seen as providing various dimensions of relatedness which appear to characterize encounter and intimacy throughout the life cycle: a containment (Bion, 1967) or holding (Winnicott, 1949) of the other, merger experiences (Mahler, 1967), admiration and occasions for idealization (Kohut, 1971), a generally caring impact (Klein, 1957), etc. Yet instead of conceptualizing these dimensions of the analytic relationship as providing the patient with a richer, more complex, more adult kind of intimacy that his previous psychopathology allowed him to experience, the developmental tilt leads to a view of these dimensions essentially as developmental remediations. Rather than being enriched in the present, the patient is seen as having past omissions corrected, developmental gaps plugged up. This lends a regressive cast to the whole analytic enterprise and seriously distorts the nature of these experiences. Let us consider several examples.

The following is an excerpt from a case discussed by Melanie Klein in Envy and Gratitude (1957). The patient is a woman described as aggrieved about every aspect of her life.

She had been breast-fed, but circumstances had otherwise not been favourable and she was convinced that her babyhood and feeding had been wholly unsatisfactory. Her grievance about the past linked with hopelessness about the present and future É The patient telephoned and said that she could not come for treatment because of a pain in her shoulder. On the next day she rang me to say that she was still not well but expected to see me on the following day. When, on the third day, she actually came, she was full of complaints. She had been looked after by her maid, but nobody else had taken an interest in her. She described to me that at one

- 484 -

moment her pain had suddenly increased, together with a sense of extreme coldness. She had felt an impetuous need for somebody to come at once and cover up her shoulder, so that it should get warm, and to go away again as soon as that was done. At that instant it occured to her that this must be how she had felt as a baby when she wanted to be looked after and nobody came.

It was characteristic of the patient's attitude to people, and threw light on her earliest relation to the breast, that she desired to be looked after but at the same time repelled the very object which was to gratify her. The suspicion of the gift received, together with her impetuous need to be cared for, which ultimately meant a desire to be fed, expressed her ambivalent attitude towards the breast. (p. 204)

Here Klein depicts a woman whose view of her own life and relations with others is characterized by a sense of deprivation, hopelessness, cynicism, and a methodical refusal to allow herself to be given to by anyone. Klein's formulations concerning envy (a deliberate spoiling of the "good") provide a rich metaphorical context for illuminating the patient's dynamics. However, Klein reduces this lifelong refusal to allow anyone to give her anything, to allow anyone to become important to her, to her relationship as an infant with the breast. Klein is clear on this point. The breast is not a metaphor for nurturance and hope. Neither is Klein suggesting that the feelings toward the breast are the first in a series of relationships with others in which the patient deals with hopelessness and anxiety through envious spoiling. "Her impetuous need to be cared for É ultimately meant a desire to be fed." Various expressions of the need to be cared for, surely a fundamental relational need throughout the life cycle, are collapsed by Klein into symbolizations and transformations of the earliest longings vis-à-vis the breast.

Balint's writings reveal a similar tilt in his understanding of important interpersonal events within the analytic process. Balint (1968) tells of his work with an "attractive, vivacious, and rather flirtatious girl in her late 20s, " who entered treatment complaining of "an inability to achieve anything." She had been academically successful, but unable to complete her final exams, and socially popular, but unable to really become involved with a man.

Gradually, it emerged that her inability to respond was linked with a crippling fear of uncertainty whenever she had to take any risk, that is, take a decision. She had a very close tie to her forceful, rather obsessional, but most reliable father; they understood and appreciated each other; while

- 485 -

her relationship to her somewhat intimidated mother, whom she felt to be unreliable, was openly ambivalent.

It took us about two years before these connections made sense to her. At about this time, she was given the interpretation that apparently the most important thing for her was to keep her head safely up, with both feet firmly planted on the ground. In response, she mentioned that ever since her earliest childhood she could never do a somersault; although at various periods she tried desperately to do one. I then said: "What about it now?"—whereupon she got up from the couch and, to her great amazement, did a perfect somersault without any difficulty. (pp. 12829)

This interaction proved to be an important breakthrough in the treatment; "many changes followed in her emotional, social, and professional life, all towards greater freedom and elasticity."

How does Balint understand the somersault, the "crucial event" in this case? He characterizes it as a "regression, " which he carefully defines as the "emergence of a primitive childish form of behavior after more mature, more adult, forms have firmly established themselves" (p. 129). This is a peculiar and unpersuasive characterization. Why is turning a somersault "childish" and "primitive"? Against what faded and anemic vision of adulthood is this being measured? Here is a young woman who lives an adulthood of great caution, constriction and uninvolvement. Given the interpretive context Balint and the patient had developed, and given the patient's subsequent progress, the somersault seems clearly a metaphorical enactment of her new willingness to take risks, to plunge herself into things without knowing exactly how they will turn out, to act in ways other than a cautious placing her feet slowly one in front of the other. Why "childish" and "primitive" then? The meaning of the act is clearly a progression, not a regression, an expansion of the patient's maturity and potentials, not a diminution of them. Is the behavior itself so "childish" and "primitive"? Adults are not supposed to make spontaneous physical gestures, to play in this way?

The most striking feature of Balint's account of this intriguing clinical moment, however, is what is omitted in his characterization of it as an "emergence." According to Balint's account, this act didn't simply emerge—it was invited! It was Balint, the adult analyst, who suggested that the patient try a somersault; what was new for the patient was her ability to respond to this invitation. The patient was closely tied to her obsessional but reliable father. Her analyst of several years, doubtlessly also obsessional and reliable, acts in a very different fashion from the cautious father—he

- 486 -

invited her to play, to take a risk, and in so doing takes a risk himself. He seduces her, in a fashion; or, perhaps, allows himself to respond to her hobbled seductiveness. Here is a man, despite his respectability, who is not bound by convention, willing to try something very different, whose outcome is unknown and unknowable. Should we characterize the analyst's invitation as regressive? This seems an extraordinarily misleading way to depict a brilliant and creative piece of clinical work. The patient and analyst have recreated in the transference a powerful attachment mediated through reliability and cautiousness, in which the decorum and professionalism of the analytic situation are symbolic equivalents of the parents' timidity and deep fear of life and spontaneity. Perhaps the crucial event wasn't the patient's somersault at all, but the analyst's invitation, through which he stepped out of the transferential integration in which he was participating and thereby transformed the relationship. Thus, Balint's clinical data suggest that the patient's psychopathology is strongly bound up with her attachment to her parents and their character pathology. The clarification of that attachment, and the mutual development of new forms of relation with respect to the analyst, are ameliorative. These new forms of relation reflect a playfulness, spontaneity, a willingness to take risks. The bias generated by the developmental tilt leads to a characterization of these events as the emergence, even if benignly, of a childishness, which, it seems to me, strikingly distorts its likely meaning.

The developmental tilt is evident not just in the writings of authors from the British School, but also in the work of theorists in the tradition of American Ego Psychology. Here structural conflict over sexual and aggressive impulses is seen as dominating later childhood and subsequent development. When relational issues are added to the theory, most importantly in the contributions of Mahler, Jacobson and Kernberg, they are introduced as pertaining to the earliest developmental phase; their evidence later in life is regarded as a regressive residue of very early disturbance. Consider this clinical excerpt from Blanck and Blanck (1974), who have synthesized various ego psychological contributions and applied them to clinical practice.

Mrs. Fletcher:

I always feel unwanted. My husband only wanted me for sex, but he never held me just because he liked me.

Therapist:

Everyone needs to be held at times, but when do we need it most?

- 487 -

Mrs. Fletcher:

You mean when we were babies? You seem to be telling me that when I think of a woman, even if sexually, that it really reflects the way I yearned to be held, cuddled, and loved by my mother.

Therapist:

Do you see now why you asked me whether I am a "butch"?

Mrs. Fletcher:

Oh, it upsets me. I want a woman.

Therapist:

But do you understand why?

Mrs. Fletcher:

I need mothering.

Thus the patient arrives at the realization that her homosexual wishes contain the intense yearning for mothering that was unfulfilled in the age-appropriate symbiotic phase. (p. 306)

Consider the therapist's first intervention. The patient has expressed the view that her husband uses her for sex, without feeling any tenderness or liking for her. The therapist pays lip service to the need for tenderness throughout life, and then immediately collapses such a need into the infant's need for tenderness from the mother, "when we need it most." Relational needs which might reasonably be regarded as aspects of all adult relationships, a longing to be held and cherished, are here depicted as regressive, symbiotic yearnings, unresolved residues from earliest childhood. The introduction of Mahler's concept of symbiosis as pre-structural, rather than as a depiction of the tension between autonomy and surrender throughout the life cycle, necessitates the collapsing of the need for tenderness and the longing for fusion into the earliest relationship with the mother. Such yearnings vis-a-vis the mother are not depicted as the first in a series of similar longings in later relationships, but as the only developmental forum in which such needs make sense. (Bergman [1971]) (provides a moving Mahlerian account of adult love as inevitably drawing on symbiotic yearnings, yet these are still, by definition, regressive, even if regression in the service of romance.)

Whether or not relational issues are tilted toward infancy has important implications in the handling of a clinical material, as the following example illustrates.

A young professor of English literature (who knows much of the psychoanalytic literature) has been struggling with phobic anxiety about presenting his work to his peers. He comes from a tight-knit extended family, very closed in on itself, regarding the external world and particularly people who move successfully through it, with great suspicion. The patient felt great conflict between his intellectual endeavors and upwardly mobile

- 488 -

ambitions, and his deep loyalty to the anti-intellectual and paranoid traditions of his family. His mother was a long-suffering daughter/wife/mother who induced great guilt and expected her children to stay with and protect her; his father was a brittle, narcissistic and grandiose man who was disdainful and deeply fearful of life outside the narrow confines of his interests. The patient never felt support or admired for his accomplishments, which he kept essentially hidden and devalued, convinced that they would destroy both parents and his connections with them, which he both dreaded and longed for. After working on many facets of his phobic anxiety, the patient began a session apologetically reporting a recent success. A long-feared meeting at which he was to present his work had gone very well; in fact, he felt exuberant in his display of his powers; he felt that he should be able to go on to other matters, but he still seemed to "need" to tell the analyst all about it, hoping to elicit approval and pride in him. He regarded this need for "mirroring" (he'd been reading Kohut) as childish and very embarassing, a sign of how deeply he'd been damaged in his ability to sustain a sense of self-worth.

What is the nature of this patient's hesitantly expressed, wished-for interaction? He wants to revel in his success, to crow, to elicit the analyst's admiration, pride, perhaps envy. He regards this wish as childish, and is embarassed by it. This attitude toward his wish is consistent with the approach taken toward many relational needs generated by object relations theories introduced through the developmental tilt; it is the position taken by Klein toward her patient's wish for nurturance, by Balint toward the somersault, by Blanck and Blanck toward the patient's wish to be held and cherished. The analyst in this case did not experience the patient's wish to share his success as resembling an infant seeking self-recognition in mother's eyes, or a little boy showing off, but rather as a man who was fearfully prideful of his success and newly discovered powers. However, the patient's apologetic display pulls for reassurance from the analyst (either explicitly or implicitly), a request for permission to show his powers, which preserves both the characterological defense of the patient and a subtle protectiveness for the other, who, it is assumed, cannot bear to fully witness the patient's struggles and triumphs. The resultant interaction is a blend of expansive vitality, solicitous protectiveness, deferential obsequiousness and ultimate, secret triumph. Is the prideful man unrelated to the boastful boy or the yearning baby? Probably not. These might be usefully regarded as expressions, at different developmental levels and through different cognitive and symbolic modalities, of the same fundamental relational need. To collapse

- 489 -

the various transformations of that need into its earliest manifestation, however, is to seriously distort its meaning and to infantalize the patient as well.

The skewing of relational issues created by the developmental tilt is sometimes accompanied by two additional clinical emphases—a tendency to minimize the importance of conflict, and a tendency to portray the patient as essentially passive. These two qualities characterize, in particular, the clinical approaches developed by Winnicott, Guntrip, Balint and Kohut.

Drive theory is conflict theory—asocial impulses clash with socially-inspired defenses against impulses, and it is from this clash that all mental life is generated. Theorists introducing relational issues through the developmental tilt have often tended to present these issues not only as earlier, but also as non-conflictual, pre-conflictual. Relational needs are not asocial, leading inevitably to conflict with the social environment. Relational needs are social by definition; what is sought is some form of relatedness; if the interpersonal environment provides the opportunity for that relatedness, there is no conflict. If the interpersonal environment does not provide such opportunities, what results is not conflict but deprivation. Winnicott (1954) expresses this point of view most clearly, in distinguishing between "needs" and "wishes." Wishes derive from instinctual impulses and eventually clash with social reality; if they are not gratified, they can be repressed, sublimated, transformed into aim-inhibited gratifications, etc. Needs are developmental necessities; the child requires certain kinds of parenting behaviors to provide necessary experiences. If the parent provides them, the child continues to develop; if the parent does not provide them, the child stops developing, becomes frozen. Similarly, if the analyst does not provide these object relational opportunities in some fashion, nothing else can happen. It is not gratification of impulses; it is a question of reaching the self by providing necessary experiences. Serious psychopathology, in Winnicott's view, is always a result of inadequate provision of needs, always an "environmental deficiency disease." In Winnicott's model, the simple provision of maternal functions produces in the child non-conflictual experience and the simple unfolding of the self.

Guntrip (1969) similarly operates on the premise that a seamless, conflict-free existence is humanly possible, and certainly desirable.

- 490 -

If we imagine a perfectly mature person, he would have no endopsychic structure in the sense of permanently opposed drives and controls. He would be a whole unified person whose internal psychic differentiation and organization would simply represent his diversified interest and abilities, within an overall good ego-development, in good object-relationship. (p. 425)

Proper parenting results in a perpetual internal harmony and equilibrium.

Then the grown-up child is free without anxiety or guilt to enter an erotic relationship with an extrafamilial partner, and to form other important personal relationships in which there is a genuine meeting of kindred spirits without the erotic element, and further to exercise an active and spontaneous personality free from inhibiting fears. This kind of parental love, which the Greeks called agape as distinct from eros, is the kind of love the psychotherapist must give his patient because he did not get it from his parents in an adequate way. (p. 357)

In developing his "self psychology in the broad sense, " Kohut (1977) takes a very similar position—if parenting is adequate in providing appropriate self-object functions, life proceeds rather simply and easily. Even the peak of the oedipal stage, the climax of instinctual sturm und drang in classical theory, is experienced as a joyful exercise of functions.

The dramatic conflict-ridden Oedipus of classical analysis, with its percepts of a child whose aspirations are crumbling under the impact of castration fear, is not a primary maturational necessity, but only the frequent result of frequently occuring failures from the side of narcissistically disturbed parents É (p. 247)

Similarly, suggests Kohut, if the analyst does not subject the patient to "empathic failures, " the analysis proceeds smoothly and non-conflictually.

To regard conflict as the exclusive property of drive theory and to present relational concepts as fundamentally non-conflictual in nature is to seriously limit the clinical utility of object relations contributions. It misses the importance of conflicts between and among different relationships and identifications, where ties and loyalties to one parent, for example, are experienced as (and in reality may very well be) a threat to ties and loyalties to the other. Also missed is the clinical importance of conflict within a single relationship. Intimacy is never a primrose path, but a process which includes risks, choices and anxiety. Particularly for patients

- 491 -

whose past efforts at relatedness have been severely dashed, warmth, nurturance, connection, can be a frightening prospect. Otto Will (1959) notes that for some patients, paradoxically, "closeness to another implies anxiety, separation and death" (p. 213).

Surely, a patient's retreat, fragmentation and withdrawal may be caused by a missed connection on the part of the analyst, but not necessarily so. To assume that it is, unnecessarily limits clinical options. It is often not the experience of "empathic failure, " but the experience of empathic success that precipitates withdrawal, devaluation and fragmentation. For someone who has experienced repeated failure of meaningful connection, whose essential attachments are to constricted and painful relationships, either in actuality or fantasy, hope is a very dangerous feeling. It may be precisely the sense of meaningful connection that precipitates the patient's withdrawal, because the possibility of such connection calls into question the basic premises of the patient's painfully constricted subjective world. Sullivan's (1953) formulation of the "malevolent transformation, " Klein's (1951) concept of envious spoiling, and Bion's (1967) depiction of "attacks on linking" all point to the dangers of hope and the conflictual nature of relational needs. The minimization of the importance of conflict, which sometimes accompanies object relations concepts introduced through the developmental tilt, leads to a view of relational processes which is simplistic and overlooks their essential ambivalence in the psychoanalytic situation.

A closely related clinical emphasis sometimes generated by the developmental tilt is the tendency to portray the patient as passive, detached and victimized. Psychopathology is a direct product of deprivation, "environmental failure." Certain kinds of interpersonal experiences are necessary for the growth of the self; when these are lacking, central features of the child remain buried, unevoked, frozen. The patient as he presents himself for treatment is an empty shell vacated by this missing core, which can only be brought to life through the analyst's creation of a more receptive environment; the passive, "true self" of the patient awaits this call. Guntrip (1971) states most clearly the premises of this approach to treatment, which I have characterized as the "Sleeping Beauty" model (Greenberg & Mitchell, 1983). Psychotherapy is

the provision of the possibility of a genuine, reliable, understanding, and respecting, caring personal relationship in which a human being whose

- 492 -

true self has been crushed by the manipulative technique of those who wanted to make him "not be a nuisance" to them, can begin at last to feel his own true feelings, and think his own spontaneous thoughts, and find himself to be real. (p. 182)

Guntrip sees the neurotic as a "neglected physically grown-up child" having been deprived of the "elementary right to the primary supportive relationship that can alone enable him to live" (1971, p. 156). Thus, the analyst brings to the frightened child in the patient missed possibilities for life. "At the deepest level, psychotherapy is replacement therapy, providing for the patient what the mother failed to provide at the beginning of life" (1971, p. 191).

This view of the patient as an abandoned, deprived, detached infant overlooks the extent to which psychopathology often entails an active clinging to, often an insistence on, symptomatic behaviors and painful experience. Fairbairn's notion that underneath all forms of psychopathology one finds an attachment to "bad objects" points to this active dimension which Guntrip's later formulations (when he was under the influence of Winnicott's work) lose. Psychopathology is not simply an absence or fearful avoidance of good relatedness. We often observe not just an avoidance of the positive, but a fascination with the negative. Patients with repetitive disturbances in relations with others are drawn, like the moth to the flame, to specific negative types of relations—sadistic, skittish, withdrawn, or debilitated. This compulsive repetition of painful early experience seems to reflect not just a detachment from some forms of relationship, but an attachment to others. The masochistic character seeks abuse partially because the violence imparts a fantasy of connection and caring from others who are experienced as inaccessible in other ways. The depressed character seeks deprivation often because it makes possible a deep and often fantasied sense of connection with a schizoid or depressed parent, so unavailable in other ways. What the patient is attached to is often not actual attributes of the parents, but fantasied attributes, not satisfying features of their relationship, but precisely what is missing. It is the deprivation, the pain, the depression which serves as a vehicle for attachment. Embedded in much psychopathological experience and behavior are personifications of others, to whom the analysand feels tied through the pathology. The patient does not simply miss or exclude from consciousness signals which would

- 493 -

lead to nurturance and attachment—he looks for different cues, which draw him into attachments not based on caring and support but on pain, misery and so on. The danger of the new dimension of the analytic relationship is that it challenges these allegiances; the patient must choose between attachments to fantasied images and presences which impart an often subtle sense of safety and connection, and the possibility of attachment to real others, with all the attendent risks. Thus, analysands often speak of the dread of a profound isolation in giving up their neurosis; psychopathology is not merely a state of aborted, frozen development, but a cocoon actively woven out of fantasied ties to significant others. Consider the dream of an analysand in the termination phase of treatment.

The patient's parents suffered from deep depressions and considerable misfortune, that peaked when he was six, leading them to withdraw into their own isolation and depression in a very global sort of way. He developed into an extremely competent and resourceful man, who suffered from a very low sense of self-esteem, bouts of depression, and a tendency to form symbiotic relations with lovers who were greatly disadvantaged in some fashion. The dream followed a period of work in which he had begun to experience himself and his relationships with other people in a more positive, even joyful way. This movement made him anxious; he feared that it was his depression and sensitivity to depression in others that made him a desirable person. Here is the dream:

I am on a small island off the mainland with my parents and sister. I take a boat to the mainland, where there is a sort of carnival going on. I walk around, watching the people, participating, having a great time. Then I remember that I must return to the island. I get in the boat and try to go back, but insects come and sting me. If I move back and stop rowing, they stop. I start to move toward the island and they sting again. I stop; they stop. I am very conflicted about what to do. After a long time of trying and stopping, I give up with a sense of relief, and rejoin the activities on the mainland.

The dream seemed to fit his experience at that time. He felt a sense of the rich possibilities which life and other people offer. Yet he also felt bound by his loyalties to his family and their ways. The connection to them was maintained through a stinging pain. As long as he suffered like they had, remaining isolated from others, he was bound up with them. To live more fully is to abandon them

- 494 -

and the comfort which the tie to them provides. Paradoxically, the deadness of his early experience with his parents, recreated in his current life, provided him a largely fantasied connection with others, which he felt kept him alive and encased in an illusory safety. Beneath a seemingly passive "detachment" is often a secret attachment, largely unconscious, but experienced as necessary and life-sustaining. The relational issues depicted in the contributions of object relations authors greatly illuminate patients' struggles, both past and present, yet the tendency to collapse these issues into early infancy and to portray the patient as nonconflictually and passively awaiting a reawakening distorts their nature and the processes through which they are perpetuated.

Object Relations Theory: Divergent Clinical Applications

All relational model theories rest, either explicitly or implicitly, on a broad developmental perspective. Human relations are understood to constitute the basic stuff of experience, and the pursuit and maintenance of relatedness is seen as the essential motivational thrust both in normality and in psychopathology. Relations take different forms across the life cycle—early relationships between the infant and caretakers are precursors and, in some sense, prerequisites to later, more complex relationships. A commonly held tenet of all versions of relational model theories is the premise that disturbances in the earliest relationship with caretakers significantly interferes with subsequent relateness, and is a predisposing factor in the generation of later psychopathology.

With respect to clinical applications, however, object relations theories often diverge not around the question of what the patient's problem was (i.e. what went wrong in his or her early relationships), but around the question of what the problem now is, and what is best done about it. The patient's development and capacity for relatedness has been warped by early difficulties in relations with significant others. Agreed. The question is, what is the present nature of these difficulties, and what is the most effective point of remediation. Authors whose vision passes through a lens of theory skewed by the developmental tilt tend to view the patient as an infantile self in an adult body, fixed in developmental time and awaiting interpersonal conditions which will make further development possible. In this view, what was missed is still missing,

- 495 -

and needs to be provided for essentially in the form in which it was missed the first time around. The analyst must enter at the point of the "environmental failure, " providing relational experiences as "replacements" for those which the infant never encountered. It is this view of psychopathology as the encapsulation of past infantile needs that Levenson points to in characterizing object relations theory as viewing the patient as an adult "stuck with an incorporated infant, like a fishbone in the craw of his maturity" (1983, p. 142). However, this is not the only possible clinical application of object relations concepts. In fact, it can be argued that it violates the basic meaning of Fairbairn's claim that the analyst must become a "good object." Let us return to the commonly held developmental perspective to trace out an alternative position.

The pursuit and maintenance of human relatedness are understood to constitute the basic maturational thrust in human experience. Disturbances in early relationships with caretakers seriously distort subsequent relatedness, not by freezing or fixing infantile needs in place, but by setting in motion a complex process through which the child builds an interpersonal world (or world of object relations) out of what is available. The child simply cannot do without relationships, without ties to others, both in terms of real interactions and in terms of a sense of interconnection, belonging. To be human means to be in relation to others, to be embedded in a relational matrix. Thus, Fairbairn (1952) chronicles the processes through which the child's early experience is fragmented and internalized: gratifying contact with parents is preserved through real relatedness, while ungratifying contact is fractured and preserved in fantasies of "objects" to which portions of the child's "ego" become attached, fantasies of attachments to internal objects being necessary to fill the child's need for a fuller human connection. Fairbairn depicts the patient's deepest longing (libidinal ego) as focused on aspects of the parents which seemed to offer something (the "exciting object")—an appearance of vitality, or warmth, or sexuality they never had access to, which they could not reach. As the child grows to adulthood, subsequent relationships are filtered through this constellation of infantile fantasies, and, to a greater or lesser extent, experienced in their terms.

How does the patient break out of this closed system? The analyst offers him or herself as a "good" object. However, the "good" object must not be confused with any of the patient's internal objects

- 496 -

or fantasies. The patient has never known a "good" object, which is why the fragmentation which underlies psychopathology occurs. Surely the "good" object is not equivalent to the "exciting object, " the patient's image of an impossible, unreachable nurturance which fantasy sustained him or her in the absence of real relationships. No, Fairbairn's "good object" operates outside the closed system of the patient's internalized object relations (as does Racker's [1968] portrayal of the analyst as interpreter); the "good object" must offer something real, something authentic, which makes possible the leap out of the closed world of the patient's fantasied object ties.

Thus, the theory that analytic cure lies in the provision of a replacement for missed infantile experience is actually coterminous with the patient's own infantile fantasy of a magical cure—the analyst attempts to become the "exciting object, " the "magic helper" (Fromm, 1947, p. 70), to make good on the patient's "happy thought" (Sullivan, 1956, p. 203). Some analytic work done under the aegis of object relations theory via the developmental tilt is thus marred by a collusion between the patient's fantasy and the analyst's theory; the patient is jointly viewed as an exquisitely delicate and brittle infant to be handled in just the right fashion by a uniquely sensitive caretaker, leading to a splitting of the transference and a removal of the analysis from the world of real people, from which it never returns. Other analytic work done under the aegis of object relations theory via the developmental tilt, such as Balint's invitation to the somersault, seems to be excellent analysis explained in a curious fashion. The analyst interacts with the patient in a warm, spontaneous, concerned, or possibly risk-taking fashion. Dimensions of relatedness are expressed which, in another context, would be regarded as an important component of intimacy throughout the life cycle, including intimacy between adults. Yet the interaction is collapsed into mother-infant terms, translated into the romance of the nursery.

We began by noting the multiplicity and heterogeneity of theories considered to be "object relations" theories and the divergent strategies of theory-construction which they reflect. Most broadly, object relations concepts serve as a new model for viewing all of development, offering an alternative metapsychology, supplying a hermeneutics fundamentally different from drive theory. However,

- 497 -

this model has been positioned via classical theory in different ways. In theorizing, as in living, no choice is without its price. Object relations theories like Fairbairn, which abandon the drive model, paid a price in the loss of continuity and the finely-honed elegance of classical theory. Object relations theorists who have preserved a continuity between new relational concepts and the drive model framework have also paid a price, in the distortions generated by the developmental tilt. In using object relations contributions and evaluating their place in the history of psychoanalytic ideas, it is crucial to separate out the conceptual substance from the packaging, the vision of human experience from the positioning of that vision vis-à-vis prior traditions.

REFERENCES

Balint, M. 1968 The Basic Fault London: Tavistock.

Bergmann, M. 1971 Psychoanalytic observations on the capacity to love In Separation-individuation: Essays in Honor of Margaret S. Mahler ed. J. McDevitt and C. Settlage. New York: International Universities Press.

Blanck, G., & R. Blanck 1974 Ego Psychology: Theory and Practice New York: Columbia University Press.

Bowlby, J. 1969 Attachment New York: Basic Books, Inc.

Bion, W. 1967 Second Thoughts New York: Jason Aronson, Inc.

Fairbairn, W. R. D. 1952 An Object Relations Theory of the Personality New York: Basic Books.

Feinsilver, D. 1983 Reality, transitional relatedness and containment in the borderline Contemporary Psychoanalysis 19 537560

Freud, S. 1912 The dynamics of transference Standard Edition XII pp. 99108

Fromm, E. 1947 Man For Himself New York: Fawcett.

Gill, M. 1982 Analysis of Transference Volume I New York: International Universities Press.

Greenberg, J., & Mitchell S. 1983 Object Relations in Psychoanalytic Theory Cambridge, Mass.: Harvard University Press.

Guntrip, H. 1969 Schizoid Phenomena, Object Relations and the Self New York: Basic Books.

Guntrip, H. 1971 Psychoanalytic Theory, Therapy and the Self New York: Basic Books.

Hartmann, H. 1939 Ego Psychology and the Problem of Adaptation New York: International Universities Press.

Hartmann, H. 1960 Psychoanalysis and Moral Values New York: International Universities Press.

Jacobson, E. 1964 The Self and the Object World New York: International Universities Press.

Kernberg, O. 1976 Object Relations Theory and Clinical Psychoanalysis New York: Jason Aronson.

- 498 -

Kernberg, O. 1980 Internal World and External Reality New York: Jason Aronson.

Klein, M. 1957 Envy and gratitude In Envy and Gratitude and Other Works, 19461963 New York: Delacorte Press, 1975

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.

Laing, R. D. 1976 The Facts of Life: An Essay in Feeling, Facts and Fantasy New York: Pantheon.

Levenson, E. 1983 The Ambiguity of Change New York: Basic Books.

Lichtenberg, J. 1983a Psychoanalysis and Infant Research Hillsdale, N.J.: The Analytic Press.

Lichtenberg, J. 1983b Is there a weltanschauung to be developed from psychoanalysis? In:The Future of Psychoanalysis ed. Arnold Goldberg. New York: International Universities Press.

Mahler, M. 1967 On Human Symbiosis and the Vissitudes of Individuation Volume 1 Infantile Psychosis New York: International Universities Press.

Mitchell, S. 1979 Twilight of the idols: Change and preservation in the work of Heinz Kohut Contemporary Psychoanalysis 15 170189

Modell, A. 1968 Object Love and Reality New York: International Universities Press.

Racker, H. 1968 Transference and Countertransference New York: International Universities Press.

Stern, D. 1977 The First Relationship: Mother and Infant Cambridge, Mass.: Harvard University.

Stern, D. 1983 Affect attunement. Address given at the William Alanson White Institute December 7, 1983

Stolorow, R., & F. Lachmann 1980 Psychoanalysis of Developmental Arrests New York: International Universities Press.

Sullivan, H. S. 1940 Conceptions of Modern Psychiatry New York: Norton.

Sullivan, H. S. 1953 The Interpersonal Theory of Psychiatry New York: Norton.

Sullivan, H. S. 1956 Clinical Studies in Psychiatry New York: Norton.

Will, O. 1959 Human relatedness of the schizophrenic reaction Psychiatry 22 pp. 205223

Winnicott, D. 1949 Birth memories, birth trauma, and anxiety In Through Paediatrics to Psycho-analysis London: Hogarth Press, 1958

Winnicott, D. 1954 Metapsychological and clinical aspects of regression within the psychoanalytical setup In Through Paediatrics to Psychoanalysis London: Hogarth Press, 1958

- 49