Object Relations in Psychoanalytic Theory
Mitchell, Stephen A. & Greenberg, Jay R (1983)
Chapter 7- D. W. Winnicott and Harry Guntrip

Klein and Fairbairn were system-builders. Each constructed a broad and novel vision of human experience and difficulties: Klein, in her slowly evolving, piece-by-piece redefinition and refocusing of Freudian theory; Fairbairn, in his dramatic refutation of Freud's work. Winnicott and Guntrip, by contrast, were concerned with single issues. Both declared allegiance to prior traditions: for Winnicott, his own personal blend of Freudian and Kleinian thought; for Guntrip, Fairbairn's recently fashioned object relations theory. Yet each felt that the tradition he emulated had omitted one crucial area of concern and attempted to correct that oversight.
     Winnicott and Guntrip present their own contributions as circumscribed and limited, mere emendations of earlier theoretical traditions. Winnicott's formulations concerning the emergence of the self, however, provide a foundation for developmental theory radically different from that of his Freudian and Kleinian predecessors. Guntrip's formulations concerning ego regression move Fairbairn's object relations theory in a direction which diverges from some of its most basic premises. 

D. W. Winnicott

Winnicott, an extremely innovative and influential contributor to the development of psychoanalytic theory and practice, has provided an intricate, subtle, and often powerfully poetic account of the development of the self out of its relational matrix. The form and manner of his work parallels some of his central thematic concerns in two striking ways. First, Winnicott's prose has an elusive quality. Almost all of his papers were originally presented as talks, and their style reflects an informality more suitable to the spoken than the written word. Each is short, with often scintillating clinical observations loosely strung together with pithy, almost epigrammatic theoretical formulations. The central themes are generally presented in the form of evocative paradoxes that entice the reader playfully. The arguments are more discursive than tightly reasoned; Winnicott follows his presentations where they take him. Because of this elusiveness, Masud Khan, his editor and foremost disciple, has aptly characterized his style as "cryptic."
     A second striking feature of Winnicott's presentation is his curious manner of locating himself with regard to psychoanalytic tradition. Winnicott claims great allegiance to his theoretical forebears, particularly Freud and, to a lesser extent, Klein. He presents his own contributions as a continuation of their work, which he describes in reverential terms. In fact, the central thrust of a rather scathing review by Winnicott and Khan (1953) of Fairbairn's work is a criticism of the latter's rejection of Freud's metapsychological formulations. However, Winnicott preserves tradition in a curious fashion, largely by distorting it. His interpretation of Freudian and Kleinian concepts is so idiosyncratic and so unrepresentative of their original formulation and intent as to make them at times unrecognizable. He recounts the history of psychoanalytic ideas not so much as it developed, but as he would like it to have been, rewriting Freud to make him a clearer and smoother predecessor of Winnicott's own vision. This tendency to absorb and rework the concepts of others is reflected in Khan's description of Winnicott's impatience with reading: "'It is no use, Masud, asking me to read anything! If it bores me I shall fall asleep in the middle of the first page, and if it interests me I will start re-writing it by the end of that page"' (1975, p. xvi). Harold Bloom (1973) has suggested that each major poet within the Western tradition distorts the vision of his most prominent predecessors to make room for his own personal vision. Winnicott's manner of positioning his own innovative and important contributions vis-a-vis the psychoanalytic tradition suggests such a process more than any of the other theorists considered in this volume. He could have been describing his own approach to psychoanalytic tradition when he says: "Mature adults bring vitality to that which is ancient, old and orthodox by re-creating it after destroying it" (1965b, p. 94).
     These formal characteristics of Winnicott's writing--his elusive mode of presentation and his absorption yet transformation of theoretical predecessors--parallel his central thematic interest: the delicate and intricate dialectic between contact and differentiation. Almost all his contributions center around what he depicts as the continually hazardous struggle of the self for an individuated existence which at the same time allows for intimate contact with others. Winnicott's depiction of the healthy self rests upon one of his many paradoxes--through separation, nothing is lost, but rather something is gained and preserved: "This is the place that I have set out to examine, the separation that is not a separation but a form of union" (1971, p. 115; italics in original). The achievement of such a state is by no means easy; the development of the self is fraught with dangers. How does the child discover himself within his mother's care without losing himself to her? How can the child differentiate himself yet retain maternal resources? How can one communicate without being depleted, be seen without being appropriated, be touched without being exploited? How can one preserve a personal core without becoming isolated? The formal and stylistic characteristics of Winnicott's presentation reflect these issues. He entices, baffles, and provokes his readers, valuing them highly but never confronting them directly. He reveres his theoretical forebears, prizing continuity with them; yet, he refashions and reshapes their work radically according to his own imagery and vision. Lack of contact with others as well as total accessibility to others pose, for Winnicott, grave dangers to the survival of the self.
     Winnicott was a prominent pediatrician prior to and throughout his career as a psychoanalyst, and his deep familiarity with babies and mothers pervades and informs his approach to psychoanalytic issues. He began his ten-year analysis with Strachey in 1923, three years before Melanie Klein's move to England; like Fairbairn, he was deeply influenced by her work. His second analyst, Joan Riviere, was among Klein's closest collaborators, and he was in supervision with Klein herself between 1936 and 1940. Winnicott felt that Klein's work overlapped some of his own early observations and helped him resolve issues with which he was struggling. He had worked with children who seemed never to have reached a stable and differentiated oedipal stage; in his early work on feeding disorders he had been struck by the predominance of greed in infants and the centrality of fantasies in young children concerning their own "insides" and the "insides" of the mother (1936, p. 34). Klein's depiction of early phantasies, anxieties, and primitive object relations spoke directly to Winnicott's earliest concerns.
     In 1945, following an incubation period involving considerable clinical work both with children and with psychotics, Winnicott began a series of papers which marked his departure from Freudian and Kleinian theory. Freud had illuminated neurosis; Klein had explored depression. His own work, Winnicott suggests, is an emendation, an application of prior psychoanalytic concepts to the relatively uncharted area of manifest psychosis. This diagnostic distinction was to lose meaning as the approach Winnicott developed broadened into a general theory of development and psychopathology markedly at variance with the formulations of Freud and Klein. The processes leading to the development or the inhibition of the self are depicted and understood solely in the context of the interaction between the child and the environmental provisions supplied by significant others. Thus, despite his protestations of continuity and allegiance, Winnicott's work constitutes an approach to human experience which rests solidly within the relational/structure model.

The Emergence of the Person
Winnicott's most important contributions to psychoanalysis begin with his observation that classical theory and the psychoanalytic treatment of neurosis take something very basic for granted: that the patient is a person. By this he means it is assumed that the patient has a unified and stable personality available for interactions with others. Freud, Winnicott suggests, presupposed the "separateness of the self and a structuring of the ego" (1960a, p. 41). Because of this presumption, two major problems have been overlooked: patients who are not "persons," either because of manifest psychosis or because they only appear to interact with others; and those features of the analytic situation which bear most directly on early developmental processes facilitating the emergence of personhood. These were precisely the areas Winnicott set out to explore. Almost all of his major contributions concerned the conditions making possible the child's awareness of himself as a being separate from other people, and he approaches this problem from different angles, through different formulations, and in different contexts.
     The mother provides experiences which enable the incipient self of the infant to emerge. The latter begins life in a state of "unintegration," with scattered and diffuse bits and pieces of experience. The infant's organization of his own experience is preceded by and draws upon the mother's organized perceptions of him. The mother provides a "holding environment" within which the infant is contained and experienced: "an infant who has had no one person to gather his bits together starts with a handicap in his own self-integrating task" (1945, p. 150). Winnicott termed the state of devotion that characterizes the mother, enabling her willingly to offer herself as an attentive medium for her baby's growth, the state of "primary maternal preoccupation." He regards the mother's absorption in fantasies of and experiences with her baby as a natural, biologically rooted, and adaptive feature of the last trimester of pregnancy and the first several months of the baby's life.
     In addition to "holding," the mother "brings the world to the child" and, in Winnicott's view, this function plays a crucial and intricate role in development. The infant when excited conjures up, or, more precisely, is on the verge of conjuring up, an object suitable to his needs. Ideally it is precisely at that moment that the devoted mother presents him with just such a suitable object--the breast, for example. This is the "moment of illusion." The infant believes he has created the object. Over and over the infant hallucinates, the mother presents, and the content of the conjuring approximates more and more closely to the real world.

The infant comes to the breast when excited, and ready to hallucinate something fit to be attacked. At that moment the actual nipple appears and he is able to feel it was that nipple that he hallucinated. So his ideas are enriched by actual details of sight, feel, smell, and next time this material is used in the hallucination. In this way he starts to build up a capacity to conjure up what is actually available. The mother has to go on giving the infant this type of experience. (1945, pp. 152-153)
In the "moment of illusion" the infant's hallucination and the object presented by the mother are taken to be identical. The infant experiences himself as omnipotent, the source of all creation; this omnipotence, Winnicott suggests, becomes the basis for the healthy development and solidity of the self. (Kohut was later also to argue that the basis for a healthy self lies in the opportunity for a prolonged experience of infantile omnipotence.) The necessity for maternal devotion in this process is apparent. The mother's empathic anticipations of the baby's needs and her precise timing are crucial. To make illusion possible, "a human being has to be taking the trouble all the time to bring the world to the baby in understandable form" (p. 154). The simultaneity of infantile hallucination and maternal presentation provide the repetitive experiential basis for the child's sense of contact with and power over external reality.
     Healthy development requires a perfect environment, but only briefly. By perfect, Winnicott means a mother whose maternal preoccupation makes possible a very close and accurate sensitivity to her infant's needs and gestures. As Winnicott describes it in his later writing, the mother functions as a mirror, providing the infant with a precise reflection of his own experience and gestures, despite their fragmented and formless qualities. "When I look I am seen, so I exist" (1971, p. 134). Imperfections in the reflected rendition mar and inhibit the child's capacity for self-experience and integration and interfere with the process of "personalization." When the mother is able to resonate with the baby's wants and needs, the latter becomes attuned to his own bodily functions and impulses, which become the basis for his slowly evolving sense of self. The failure of the mother to actualize the child's gestures and needs undercuts the child's sense of hallucinatory omnipotence, constricting his belief in his own creativity and powers and driving a wedge between the evolution of the psyche and its somatic underpinings. "The mind has a root, perhaps its most important root, in the need of the individual, at the core of the self, for a perfect environment" (1949a, p. 246).
     Another approach Winnicott takes to the same issues is his consideration of the conditions necessary for the development of the capacity to be alone. He suggests that it is extremely important for the mother not only to shape the world to the infant's demands but also to provide a non-demanding presence when the infant is not making demands or experiencing needs. This makes it possible for the infant to experience needlessness and complete unintegration, a state of "going-on-being" out of which needs and spontaneous gestures emerge. The mother's non-demanding presence makes this experience of formlessness and comfortable solitude possible, and this capacity becomes a central feature in the development of a stable and personal self. "It is only when alone (that is to say, in the presence of someone) that the infant can discover his own personal life" (1958b, p. 34).
     Fortunately for everyone involved, the mother's exquisite responsiveness is not necessary for very long. Once hallucinatory omnipotence is firmly established, it is necessary for the child to learn the reality of the world outside his control and to experience the limits of his powers. What makes this learning possible is the mother's failure, little by little, to shape the world according to the infant's demands. As the mother recovers from her maternal preoccupation and becomes interested once again in other areas of her life, the child is forced to come to terms with what he cannot do, cannot create, cannot make happen. These harsh realities are assuaged by a push within the child toward separateness. Thus, the mother's ego coverage and responsiveness decrease in fine syncrony with an increase in the exercise of active ego functions on the part of the infant. As the infant matures, the mother does not actualize his wishes so much as receive and respond to his gestures. An increasingly greater differentiation and interaction characterizes their relationship. The early mother who materializes the infant's passive hallucinatory wish gradually gives way to the mother who responds to needs which are now actually expressed through gestures and signals. The mother's "graduated failure of adaptation" (1949a, p. 246) is essential to the development of separation, differentiation, and realization.
     Winnicott suggests that deficiencies in maternal care, more specifically the failure to provide a perfect environment and its graduated withdrawal, have a debilitating impact on the emotional development of the child. Maternal failures are of two kinds: inability to actualize the hallucinatory creations and needs of the infant when he is in excited states; and interference with the infant's formlessness and unintegration when he is in quiescent states. Both kinds of maternal deficiencies are experienced by the child as a terrifying interference with the continuity of his own personal existence, and both result in the experience of the "annihilation of the infant's self" (1956a, p. 304). The infant's personal existence is rooted both in his formless states and in his omnipotent creative gestures. Ideally, the mother is the medium for formlessness and the instrument of omnipotence. Any interference with these functions is experienced by the infant as an "impingement." Something from the outside is making claims on him, demanding a response. He is wrenched from his quiescent state and forced to respond, or he is compelled to abandon his own wishes, to accept prematurely the feeble and unrealistic nature of his own demands, and to mold himself to what is provided for him.
     The major consequence of prolonged impingement is fragmentation of the infant's experience. Out of necessity he becomes prematurely and compulsively attuned to the claims and requests of others. He cannot allow himself the experience of formless quiescence, since he must be prepared to respond to what is asked of and provided for him. He loses touch with his own spontaneous needs and gestures, as these bear no relation to the way his mother experiences him and what she offers him. Winnicott characterizes the resulting fragmentation as a split between a "true self," which becomes detached and atrophied, and a "false self on a compliant basis." The "true self," the source of spontaneous needs, images, and gestures, goes into hiding, avoiding at all costs the possibility of expression without being seen or responded to, the equivalence of complete psychic annihilation. The "false self" provides an illusion of personal existence whose content is fashioned out of maternal expectations and claims. The child becomes the mother's image of him. The "false self" comes to take over in some sense the caretaking functions which the environment has failed to provide. The "false self" covertly protects the integrity of the "true self"; it functions "to hide the true self, which it does by compliance with environmental demands" (1960b, p. 147). The false self draws on cognitive functions in its anticipations of and reactions to environmental impingements, resulting in an overactivity of mind and a separation of cognitive processes from any affective or somatic grounding (1949b, pp. 191-192).
     Winnicott regards the formation of "transitional objects" as another aspect of this larger process entailing the development of the person. The most important dimension of transitional phenomena is not the objects themselves, but the nature of the relationship to the objects, representing a developmental way station between hallucinatory omnipotence and the recognition of objective reality. The emergence of the person entails a movement from a state of illusory omnipotence, in which the infant, through the mother's facilitation, feels he creates and controls all features of the world he lives in, to a state of objective perception, in which the infant accepts the limits of his powers and becomes aware of the independent existence of others. The move between these states is not a one-way, linear progression; both children and adults continually vacillate between them. Winnicott contrasts these two different states starkly with each other: solipsistic subjectivity with objective perception; the inner world with the world of outer reality; the world of "subjective objects" over which one has total control with the world of separate and independent others. Relations with transitional objects constitute a third, intermediary, and transitional realm between these two worlds.
     How is it possible for an object to be neither under illusory, omnipotent control nor part of objective reality? Herein lies the paradox which is the essence of transitional experiencing. Winnicott suggests that what is necessary for the establishment of a transitional object (such as a blanket or teddy bear) is a tacit agreement between the adults and the baby not to question the origin and nature of that object. The parent proceeds as if the baby had created the object and maintains control over it, yet also acknowledges its objective existence in the world of other people. Thus, the parent who understands this paradox allocates the object to neither of the two realms, and the agreement not to challenge the baby's special rights and privileges over his object creates the transitional realm. The transitional object is neither under magical control (like hallucinations and fantasies) nor outside control (like the real mother). Transitional experience lies somewhere between "primary creativity and objective perception based on reality-testing" (1951, p. 239). Because of this ambiguous and paradoxical status, transitional objects help the baby negotiate the gradual shift from the experience of himself as the center of a totally subjective world to the sense of himself as a person among other persons. Transitional experiencing is not merely a developmental interlude, but remains a cherished and highly valuable realm within healthy adult experience. It is here we can let our thoughts wander, concerned neither with their logic and validity in the real world nor with the threat that our musings will lead us into a totally subjective, solipsistic realm, causing us to lose the real world altogether. Transitional experience is rooted in the capacity of the child to play; in adult form it is expressed as a capacity to play with one's fantasies, ideas, and the world's possibilities in a way that continually allows for the surprising, the original, and the new. In transitional experience, we maintain access to the most private wellspring of our thoughts and imagery, without being held accountable for them in the clear and harsh light of objective reality.
     In later writings Winnicott depicts yet another feature of the emergence of the person, based on the distinction between "objectrelating" and "object usage." These formulations highlight his understanding of the function of aggression and destruction in the process of separation. "Object-relating" is defined as subjective, projective experiencing in which the other is under the infant's illusory control. "Object-usage" is the perception of and interaction with the other as independent and real, outside the infant's omnipotent control. Once again, Winnicott tries to focus our attention on the precise mechanisms which make this transition possible, and once again they revolve around a paradox. The child "destroys" the object because he has begun to experience it as separate and outside his subjective control; the child "places" the object outside his omnipotent control because he is aware of having destroyed it. Thus, the child "uses" and "destroys" the object because it has become real, and the object becomes real because it has been "used" and "destroyed." The survival of the object is crucial. The mother's non-retaliatory durability allows the infant the experience of unconcerned "usage," which in turn aids him in establishing a belief in resilient others outside his omnipotent control.
     For Winnicott the emergence of a healthy, creative self is contingent upon the specific environmental provisions he has grouped under the term "good-enough-mothering." These provisions make it possible for the infant to start "by existing and not by reacting" (1960b, p. 148). They make possible the affective shift from infantile dependence to independence and the cognitive shift from omnipotent conception to realistic perception. They determine the structure, coherence, and vitality of the person's sense of self: "individuals live creatively and feel that life is worth living or else ... they cannot live creatively and are doubtful about the value of living. This variable in human beings is directly related to the quality and quantity of environmental provision at the beginning or in the early phases of each baby's living experience" (1971, p. 83).
     Ideally, the true self, nurtured in a non-impinging environment, represents "the inherited potential which is experiencing a continuity of being, and acquiring in its own way and at its own speed a personal psychic reality and a personal body-scheme" (1965b, p. 46). Ideally, human experience entails the generation of spontaneous impulses and expressions, while the true self "does no more than collect together the details of the experience of aliveness" (1960b, p. 148). Yet, even under the best of circumstances, Winnicott suggests, personhood is a fragile and tenuous phenomenon, and there is always tension between subjective experience and objective reality. We all begin life completely dependent on our caretaker's recognition and facilitation of our wishes and gestures to provide even the chance for us to know and become ourselves. This total dependence necessitates a total vulnerability to non-responsiveness and intrusions, which are experienced as annihilations of personal continuity. The inevitable residue of this vulnerability is a private citadel of subjective reality held forever inaccessible to public, objective light. "At the centre of each person is an incommunicado element, and this is sacred and most worthy of preservation" (1963, p. 187). No matter how firmly anchored in objective reality the person is, no matter how fluidly and resiliently one negotiates the gap between subjective creativity and objective externality, the fear of the exploitation of the true self persists as the deepest dread and therefore there remains a "non-communicating self, or the personal core of the self that is truly isolate." "The question is: how to be isolated without having to be insulted" (1963, pp. 182, 187). Winnicott's answer to this question is reflected in the substance as well as the stylistic qualities of his contributions to psychoanalytic ideas, in the tension in his own work between openness and elusiveness, directness and cryptic ambiguity, loyalty to tradition and the destruction and rearrangement of that tradition.

Winnicott and the Models
Winnicott's innovative contributions to psychoanalytic thought operate within the relational/structure model. There is no such thing as a baby, he insists, only a nursing couple. The concept of the mother-infant unit, brought along from his experience in pediatrics, led him to establish his frame of reference not in processes taking place solely within the child, but in the relational field between the child and the caretakers: "The centre of gravity of the being does not start off in the individual. It is in the total set-up" (1952, p. 99). With an emphasis greatly reminiscent of Sullivan, Winnicott declares the uselessness "in describing babies in the earliest stages except in relation to the mother's functioning" (1962a, p. 57) and the impossibility of understanding psychopathology by viewing the individual as an "isolate" (1971, pp. 83-84). Although physical holding and ministrations are of tremendous importance in the holding environment, in Winnicott's view the relationship between mother and infant consists of complex and mutual emotional needs and is not essentially physical. In fact, he explicitly rejects Mahler's term "symbiosis" as "too well rooted in biology to be acceptable" (1971, p. 152), stressing instead the interactional, emotional nature of the exchange between mother and child. In his system various aspects of the early relationship between the infant and the mother serve as the foundation for the differentiation and structuralization of the self.
     Winnicott's strategy for positioning himself vis-a-vis the drive/structure tradition, given his relational /structure framework, might be characterized as one of benign neglect. He did not, like Fairbairn, abandon the use of the drive theory altogether. Nor did he, like Jacobson, Kernberg, and other figures within American ego psychology, attempt to blend relational concepts with the older drive/structure framework. Rather, he establishes object relations on a footing that is autonomous and separate from instinctual processes. In classical drive theory, object relations are derivatives of the vehicles for drive gratification and defense. In Winnicott's theory, the earliest object relations consist of interactions between developmental needs within the child and maternal provisions offered by the mother, entirely separate from drive gratification. He does not challenge the drive concept directly, but he crowds it out, relegating it to a peripheral and secondary status.
     According to Winnicott, the child needs relatedness with the mother. This need for contact consists of a built-in orientation and anticipation rather than a set of specific a priori images of the kind Klein had suggested; there is a readiness and expectancy rather than an object itself. Play "enables the baby to find the mother" (1948a, p. 165), and, despite his reluctance to align himself with Fairbairn's work, Winnicott speaks of a "drive that could be called object-seeking" (1956b, p. 314). The infant needs the maternal provisions which define good-enough mothering including: an initial perfectly responsive facilitation of his needs and gestures; a non-intrusive "holding" and mirroring environment throughout quiescent states; the collusive agreement to respect transitional objects; survival, despite the intensity of the infant's needs; and the failure to retaliate against the destructive features of object-usage. Winnicott differentiates the need for these maternal provisions from instinctual wishes: "a need is either met or not met, and the effect is not the same as that of satisfaction and frustration of an id impulse" (1956a, p. 301). These relational needs are a developmental imperative; if they are not met, no further meaningful growth can take place.
     Winnicott emphasizes the separation between key relational processes and the drives. "There is a relationship between the baby and the mother ... it is not a derivation of instinctual experience, nor of object relationship arising out of instinctual experience. It antedates instinctual experience, as well as running concurrently with it, and getting mixed up with it" (1952a, p. 98). In classical drive theory the capacity to enjoy life is rooted in the possibility for drive gratification and sublimation. Winnicott emphasizes the priority of relational processes leading to the emergence of the self.

We now see that it is not instinctual satisfaction that makes a baby begin to be, to feel that life is real, to find life worth living. In fact, instinctual gratifications start off as part-functions and they become seductions unless based on a well-established capacity in the individual person for total experience, and for experience in the area of transitional phenomena. It is the self that must precede the self's use of instinct; the rider must ride the horse, not be run away with. (1971, p. 116)
     How can drive gratification provide a seductive distraction from more basic developmental needs? Here the distance Winnicott has come from the drive/structure model is apparent. In the latter, drive gratification constitutes the underlying foundation for and the latent, essential nature of object relations. Even within Klein's work, gratification is essential to the development of object relations. The mother becomes "good" through a good feed; the infant loves the mother by swallowing and internalizing her. The mother becomes "bad" by frustrating the infant. Winnicott has separated these two realms. The self emerges and becomes structuralized through relational experiences with specific maternal provisions. What is crucial in these provisions is the position of the object, the mother's function in "holding" the infant, actualizing his gestures, surviving his attacks, and so on. According to Winnicott, gratification by itself does little to affect the position of the object; maternal provisions are independent of the mother's function in satisfying instinctual needs. "A baby can be fed without love, but lovelessness or impersonal management cannot succeed in producing a new autonomous human child" (1971, p. 127). The satisfaction of instinctual needs can, in fact, be offered as a substitutive distraction. The infant can be "`fobbed off' by a satisfactory feed" (1963, p. 181).
It must be understood that when reference is made to the mother's adaptive capacity this has only a little to do with her ability to satisfy the infant's oral drives, as by giving a satisfactory feed. What is being discussed here runs parallel with such a consideration as this. It is indeed possible to gratify an oral drive and by so doing to violate the infant's ego-function, or that which will later on be jealously guarded as the self, the core of the personality. A feeding satisfaction can be a seduction and can be traumatic if it comes to a baby without coverage by ego-functioning. (1962a, p. 57)
Thus, although Winnicott preserves the concept of instincts, they are relegated to a secondary and peripheral status in development. His greatest concern vis-a-vis physically based instinctual wishes is that they can become a means for interference with more basic developmental needs (1952b, p. 225).
     Winnicott's approach to psychopathology and treatment reflects relational/structure premises. Mental health in his view is constituted by the relative integrity and spontaneity of the self. Psychopathology (apart from a politically aimed diagnostic slight of hand to be considered shortly) entails corruption and constriction in the movement and expression of the self. The necessary and sufficient factor responsible for mental health is appropriate parental provisions--good-enough mothering. Winnicott defines psychosis specifically as an "environmental deficiency disease"; yet all psychopathology within his system involves impairment in the functioning of the self and is thus, by definition, a product of parental deficiency. The parents' personalities, Winnicott demonstrates over and over throughout his clinical illustrations, have an enormous impact on the development of the child, and parental pathology, when it interferes with the provision of nurturance and proper infant care, reverberates clearly in the psychopathology of the child: "the child lives within the circle of the parent's personality and . . . this circle has pathological features" (1948b, p. 93).
     Winnicott's relational/structure understanding of the nature of psychopathology is reflected in his treatment of the phenomenon of regression. Regression, he suggests, is not a return to points of libidinal fixation or specific erotogenic zones. Regression represents a return to the point at which the environment has failed the child. Appropriate parental provisions are the sine qua non of emotional growth; where they are missing, development stops, and the absent developmental "needs" dominate subsequent living. Developmental needs are very different from "wishes" derived from drives. Needs are a developmental necessity; nothing else can happen until they are filled. Within the drive/structure model, regression is pathological and dangerous in that it provides a surfeit of infantile gratification. In Winnicott's version of the relational/ structure model, regression is a search for missing relational experiences. "The tendency to regression in a patient is now seen as part of the capacity of the individual to bring about self-cure" (1959, p. 128).
     Winnicott sees the curative factor in psychoanalysis, not in its interpretive function, but in the manner in which the analytic setting provides missing parental provisions and fills early developmental needs. The function of psychoanalysis is to compensate for parental failures in adaptation, and "to provide a certain type of environment" (1948a, p. 168). The person of the analyst and the analytic setting "hold" the patient; in the reliability, attentiveness, responsiveness, memory, and durability of the analyst, the aborted self of the patient becomes unstuck and continues to grow. Winnicott also sees the psychoanalytic process in terms of mutual play between patient and analyst; when the patient is constricted in this capacity, the analyst functions to rekindle it (1971, p. 38). Whereas Freud's major emphasis in discussing the liberating value of psychoanalysis is on the freedom from illusion, Winnicott emphasizes the increasing freedom to create illusion, and this is intimately tied up with the capacity to play (R. Bank, personal communication).
     Winnicott defended himself against orthodox critics who charged that his approach to treatment is too regressive and too gratifying of infantile wishes, by arguing that gratification in regression is the result not of libidinal satisfaction, but of the fact that the "self is reached" (1954, p. 290). This view of treatment is an outgrowth of, and consistent with, his relational/structure assumptions concerning maturation, development, and psychopathology. Specific relations with a maternal caretaker are essential to the development of the person. When provided, they set the child free to grow and function freely as a person in the world; when they are missing, the incipient self is ensnared and imprisoned, wrapped in a protective cocoon, hidden from the world of others experienced as unsafe for authentic and spontaneous living. Only if the appropriate facilitating environment is provided can the true self be reached and allowed to continue its growth.

Winnicott and the Tradition
Winnicott is very careful throughout his writings to place himself within the earlier tradition of psychoanalytic ideas. The two figures he was most concerned with, both explicitly and implicitly, were Klein and Freud. Winnicott took considerable pains to present his contributions as a continuation of, not a departure from, their systems, and criticized Fairbairn for directly challenging Freud's drive theory. Yet his own formulations operate wholly within the relational/structure model--and this posed serious political problems. Despite the relational nature of his concepts,
Winnicott aligns himself with Klein and Freud through a combination of assimilation, distortion, and strategic avoidance.
     Winnicott's use of Klein's theory reflects a marked ambivalence. On the one hand, several of her concepts and emphases provided key intellectual tools in the development of his own thought. The notions of an inner world, internal objects, primitive greed, the importance of phantasy--all these concepts occupy a central place within Winnicott's system. He openly acknowledged this debt and as late as 1948 defended Kleinian theory against its critics, most notably Glover. On the other hand, Winnicott began directly challenging Kleinian theory as early as 1941. That year he took objection to the notion of a priori knowledge and imagery of the father's penis (1941, p. 63). In 1949 he argued that birth is not experienced in terms of the projection of aggression--"the stage has not yet been reached at which this means anything" (1949b, p. 185). In 1959 he suggests that the concept of the death instinct is "unnecessary" (1959 p. 127). And in a retrospective review of Klein's contributions he suggests that her attempt to date complex cognitive processes earlier and earlier in infancy "spoilt" her later work (1962b, p. 177).
     Winnicott's broadest critique of Klein's system concerned her emphasis on internal processes at the expense of relations with real others; his major theoretical departure from her vision lies in his stress on the interpersonal environment. In Winnicott's theory of mind, object relations are rooted in and constituted by both the mother's performance of caretaking functions and her character. His major criticisms of Kleinian concepts address her attempt to derive object relations from inherent, constitutional sources such as a priori object images and innate aggression. What Klein derives from constitution, Winnicott derives from environmental provisions and failures.
     "Melanie Klein represents the most vigorous attempt to study the earliest processes of the developing human infant apart from the study of child-care. She has always admitted that child-care is important but has not made special study of it" (1959, p. 126; italics in original). As a pediatrician and director of a child psychiatric clinic, Winnicott was much more aware of battering and neglecting mothers than was Klein, who had a fashionable West End practice. This difference undoubtedly bears on the difference in the weight they give to the reality of parental behavior and character (James Grotstein, personal communication).
     In his challenge of the central pillars of Klein's theory, Winnicott became disaffiliated with the Kleinian group; in later years, he seems to have regarded this with a touch of bitterness and regret. "I never had analysis by her, or by any of her analysands, so that I did not qualify to be one of her group of chosen Kleinians" (1962b, p. 173)--a puzzling statement, given his analysis with Joan Riviere. Winnicott used his position as a member of the "c-group" within the British Psychoanalytic Society (devotees of neither Klein nor Anna Freud), to attempt to heal the split within the society and reconcile Kleinian formulations with main-line Freudian theory.
     Despite Winnicott's open departure on many issues, the treatment of Klein in his writing reflects a considerable effort to demonstrate his continuity with her views. The major device employed to preserve continuity is his tendency to reinterpret Klein's formulations into a more fully relational/structure framework. At times Winnicott's alterations are openly acknowledged. For example, he maintains Klein's emphasis on unconscious phantasy as the pervasive underpinning of mental life, but he explicitly separates phantasy from presuppositions concerning a priori knowledge. For Klein, with closer ties to the drive/structure model, phantasy is primarily an internal phenomenon, generated by the drives and related to the world of real others only secondarily. In Winnicott's system the primacy of phantasy is preserved but the content is altered. "Fantasy is more primary than reality, and the enrichment of fantasy with the world's riches depends on the experience of illusion" (1945, p. 153). For him phantasy is oriented toward personalized reality from the start, manifesting itself in a readiness to develop illusions of control over what the real world actually provides. Through phantasy the infant is poised, at the "moment of illusion," for relational interchange with the outside world.
     At other points Winnicott's alteration of Klein's formulations is much more covert. For example, he considered Klein's development of the concept of the depressive position to be her greatest contribution to the history of the psychoanalytic ideas; it "ranks with Freud's concept of the Oedipus complex" (1962b, p. 176). By introducing personal, as opposed to social, sources of the sense of guilt, Winnicott felt that Klein had opened up a whole new realm of psychoanalytic investigation concerned with the "idea of an individual's value," in addition to questions of "health" (1958a, p. 25). Nevertheless, in presenting Klein's formulations, he changes them. The development of the capacity for concern (Winnicott prefers the term "concern" to Klein's "guilt") is presented as a feature of the transition from infantile omnipotence to objective perception and relationship. In this transition the synthesis is effected between the two different "mothers" within the infant's experience: the caretaker, environment mother, who provides the holding function in quiescent states, and the "object" mother, who is the victim of the infant's "ruthless" fantasies and attacks in excited states.
      In his greedy excitement, Winnicott suggests, the infant uses the mother with no regard for her feelings or even survival. He is aware only of his own wishes. The depressive crisis is precipitated by the realization that the mother who is the object of these excited states is also the mother who provides the holding environment between excitements, the mother the infant depends on and loves. This synthesis and realization arouses a deep concern for the mother. Two aspects of maternal functioning are crucial, according to Winnicott, for the infant to sustain and integrate the capacity for concern. First, the mother must survive the excited states and "hold" the situation in time, so that the child can come to trust her durability and perceive the less-than-omnipotent destructiveness of his own needs and phantasies. The mother's survival demonstrates the resistance and resiliency of the real world in the face of the infant's wants and demands. Second, the mother must provide the baby with the "opportunity to contribute," to make up to the mother, to console her. Only if reparation is possible can the guilt the child feels over his destructive impact be tolerated and the capacity for concern emerge.
     In Winnicott's reworking of Klein's formulations concerning the depressive position, several basic changes are apparent. He views depressive anxiety and guilt as much more directly concerned with the person of the real mother than did Klein. The child cannot simply "repair" the mother in fantasy and play, but needs to be given the "opportunity to contribute," to console the mother in actuality. This poses difficult problems for children of inconsolable, depressed mothers. "Their task is first to deal with the mother's mood ... creating an atmosphere in which they can start on their own lives" (1948b, p. 93). The depressive position, therefore, is more fully grounded in the child's actual interpersonal world. Further, the very issue at the heart of the depressive position is different in Winnicott's formulations. Klein had depicted depressive anxiety as arising from the integration of the good breast (the repository of the child's projected love plus gratifying experiences with the mother) with the bad breast (the repository of the child's projected hate plus frustrating experiences with the mother). Without acknowledging that he is changing anything, Winnicott depicts depressive anxiety as arising from the integration of the environment mother (who "holds" the infant in quiescent states) with the object mother (who is the victim of the infant's greedy appropriations in excited states). These are not simply parallel constructions.
     Klein's formulation reflects the residues of the drive model notion that the central task of early psychic development is the regulation and integration of internally arising drive energies; Winnicott's formulation reflects the more fully relational model notion that the central task of early psychic development is the integration of various caretaking functions provided by the mother. To employ Klein's concepts without openly modifying them, it is necessary for Winnicott to misread them. This becomes most clear in his discussion of the function of aggression. "I have used the expression primitive love impulse, but in Klein's writings the reference is to the aggression that is associated with the frustrations that inevitably disturb instinctual satisfactions as the child begins to be affected by the demands of reality" (1958a, p. 22). To modify Klein's work by deriving aggression from actual experiences of frustration is not only possible, but is a compelling alteration. To present such modifications as Klein's own view obscures the fundamental differences between Klein and Winnicott with respect to their basic, underlying presuppositions.
     Guntrip reports that Winnicott urged him to "have your own relation to Freud and not Fairbairn's" (1975, p. 151). In another context Guntrip implies that Winnicott himself actually had two relationships to Freud, one public and one private. Privately, he suggests, Winnicott acknowledged his departure from Freud's drive-based approach to psychopathology in the direction of a more relational view. "We disagree with Freud," Guntrip reports Winnicott as saying. "He was for `curing symptoms.' We are concerned with whole living and loving persons" (Mendez and Fine, p. 361). Why was this open divergence not reflected in Winnicott's writings? Winnicott, Guntrip suggests, was "clinically revolutionary and not really interested enough in pure theory to bother to think it out." This assertion is puzzling if one examines references to Freud in the work of Winnicott, who takes great pains, at times involving elaborate and intricate argumentation, to proclaim himself at one with Freud in all respects. His relationship to Freud cannot be, as Guntrip asserts, the product of laziness or disinterest; rather, it appears to be the result of a systematic strategy by Winnicott to present his contributions as a direct continuation of, rather than a marked departure from, Freud's work. The major devices used in these efforts were a systematic misreading of Freud's formulations, and a use of diagnostic distinctions which give the appearance of preserving Freud's theory of neurosis intact.
     Let us consider several of the more striking of Winnicott's misreadings of Freud. Freud's concept of primary narcissism is a stumbling block for any relational/structure model, since it explicitly presupposes that the infant is at first not oriented toward others, thereby making object relations secondary and derivative phenomena. Both Klein and Fairbairn directly challenged the concept of primary narcissism, the former by arguing the presence of internal object relations inherent in narcissism, the latter by arguing that libido is directed toward reality and others from the start. Winnicott takes a different tack: he recognizes the difficulties Freud's concept poses, then says he prefers to think that Freud did not really mean what he said. Winnicott quotes Freud's reference to the infant as a "completely narcissistic creature . . . totally unaware of her [the mother's] existence as an object." He goes on to remark, "I like to think that Freud was feeling round this subject without coming to a final conclusion because of the fact that he lacked certain data which were essential to the understanding of the subject" (1949b, p. 175). Despite the total absence of any such tentativeness in Freud's writing, Winnicott proceeds to use the concept of "primary narcissism" as if it did not imply an early objectless state, but was in fact equivalent to his own antithetical view of the early dependence of the infant on the mother.
     The presupposition of an innate aggressive drive is one of the twin pillars of Freud's "dual-instinct theory." Klein preserved and extended this concept; Fairbairn explicitly rejected it. In Winnicott's work there is no aggressive drive of the sort Freud had formulated, yet he continues to use the term extensively. Winnicott achieves the appearance of continuity with respect to the concept of innate aggression by simply adopting the term and redefining it. He emphasizes the importance of aggression throughout his work, employing Freud's term as if it carried the same meaning. Yet, at several points he notes that aggression and destruction do not entail anger or hate. "Aggression" for him refers not to a specific instinctual drive, but to a general vitality and motility. He equates it with a life force and argues that "at origin, aggressiveness is almost synonymous with activity" (1950, p. 204). He suggests that it consists of a need for something to bump up against, something outside the self to be encouraged and struggled with: "it is the aggressive component that . . . drives the individual to a need for a Not-Me or an object that is felt to be external" (1950, p. 215). The "destruction' in Winnicott's late work on object-usage is thus an innocent, nonbelligerent desire for engagement: "this destructive activity is the patient's attempt to place the analyst outside the area of omnipotent control, that is, out in the world." The aggressive drive "creates the quality of externality" (1971, pp. 107, 110).
     Winnicott's treatment of Freud's theory of oedipal guilt provides another example of forced continuity. In Freud's theory the Oedipus complex is constituted by the clash between instinctual forces driving the child toward incestuous and murderous impulses, on the one hand, and the fear of retaliation by the real parent and later by the superego,
on the other. It is a product of the tension between the press of drive and the fear of retribution from social reality. Winnicott describes Freud's Oedipus complex quite differently:

In the simplest possible terms of the Oedipus complex, a boy in health achieved a relationship with his mother in which instinct was involved and in which the dream contained an in-love relationship with her. This led to the dream of the death of the father, which in turn led to the fear of the father and the fear that the father would destroy the child's instinctual potential. This is referred to as the castration-complex. At the same time there was the boy's love of the father and his respect for him. The boy's conflict between that side of his nature which made him hate and want to harm his father, and the other side by which he loved him, involved the boy in a sense of guilt. (1953a, p. 17)
This description is an account of the Oedipus complex as modified by Klein's formulations concerning the depressive position. Freud's conflict between drive (both libidinal and aggressive) and social reality has been replaced by Klein's conflict between love and hate. (Freud sometimes also speaks of ambivalence in connection with oedipal conflicts, but an ambivalence derived from constitutional bisexuality, body-based and drive-derivative, rather than the more fully emotional ambivalence Klein and Winnicott depict.) Klein explicitly addressed the differences between her account of the oedipal crisis and Freud's. Winnicott does not; he rewrites Freudian theory through a Kleinian perspective, preserving an illusion of consensus and unbroken tradition.
     A final example of his systematic misreading of Freud is provided by Winnicott's argument that the most novel and innovative of his contributions were actually implicit in Freud's work all along:
It would appear to me that the idea of a False Self ... can be discerned in the early formulations of Freud. In particular I link what I divide into a True and False Self with Freud's division of the self into a part that is central and powered by the instincts (or by what Freud called sexuality, pregenital and genital), and a part that is turned outwards and is related to the world. (1960b, p. 140)

This is an extremely misleading parallel. Winnicott's distinction between the true and false self contrasts authentic and spontaneous living with compliant, overly adaptational living. Freud's distinction between the id and ego contrasts primitive, asocial, undirected impulses with a necessary knowledge of and facility with the outside world. Freud's distinction does not address itself to the issue of inauthenticity, which is at the center of Winnicott's concern. A comparison between the two concepts is interesting and revealing. Freud's concern, consistent with the drive/structure model, is with the division between drives and regulatory functions, between energy and its organization and use. Winnicott's concern, consistent with the relational/structure model, is with different forms of relation between self and others. Placing Winnicott's formulations beside Freud's, one can see the distance he has moved from the drive/structure model. Winnicott is concerned with minimizing that distance.
     The second device used by Winnicott to position himself in continuity with Freud's work is found in his approach to diagnosis. In one of the earliest papers in which he put forth his original views (1945) Winnicott makes a tripartite distinction among categories of mental disorder: preself disorders (psychotics, schizoids, borderline cases, and false selves)--a dysfunction within the earliest, most primitive object relations; depressive disorders--difficulties with inner-world issues involving conflicts between love and hate as characterized by Melanie Klein; whole person disorders (neurosis)--oedipal conflicts as characterized by Freud. This classificatory system reflects Winnicott's relation to tradition: Freud was right with respect to neurosis; Klein was right with respect to depressives; Winnicott takes as his own province the relatively unexplored area of psychotic and borderline-psychotic phenomena.
     By 1954 he has placed the "majority of so-called normal people" in the middle group, as Kleinian depressives (1954b, pp. 276-277). The third group, Freudian neurotics, now consists only of "quite healthy people" who have managed to establish a stable and vital enough self to be confronted with the oedipal problems Freud had described. By 1956 the middle classification of Kleinian depressives has dropped out, and the majority of sufferers from mental dysfunction are understood to be grouped in the first category, those who, as a result of deficiencies in parental provisions, lack an integrated, vital self. Winnicott has come to use the false self concept as a single diagnostic principle, representing a continuum of psychopathology from psychotic states, in which the false self has collapsed, to nearly healthy states, in which the false self mediates selectively and sparingly between the true self and the outside world (1960b, p. 150). The other category of human beings, that realm within which Freudian theory still applies, is no longer regarded as a form of psychopathology at all. In neurosis, adequate parenting has produced a healthy self. "True neurosis is not necessarily an illness ... we should think of it as a tribute to the fact that life is difficult" (1956c, pp. 318-319).
     What is Winnicott accomplishing with these shifting diagnostic distinctions? Initially he portrays his work as an application of Freud's concepts to a realm of pathology not considered by Freud. As his work developed, however, it became apparent that Winnicott was proposing not an extension, but an alternative to Freud's approach. He is offering a framework for understanding psychopathology which, firmly rooted in the relational model, is at odds with classical formulations based on drive and defense. Thus, the diagnostic group Winnicott takes as his own gradually swells; that left to Freud gradually shrinks. It is necessary for Winnicott's political positioning of himself in continuity with Freud, however, for him to designate neurosis as a phenomenon adequately understood only within the framework of classical drive theory. However, because Winnicott's own system is so comprehensive and so much at odds with drive/structure principles, it is not easy for him to allow Freud's theory a meaningful place. In fact, the manner in which he depicts Freud's view of neurosis is itself badly distorted. He defines neurosis as "the illness that belongs to intolerable conflict which is inherent in life and in living as whole persons" (1959, p. 136). Neurosis, he suggests, is the fate of individuals who have had adequate parenting and hence possess a stable and vital self. Their struggles concern universal instinctual conflicts, and their own constitutional excesses and deficiencies, balances and imbalances. Neurosis is the province of the "individual," the "personal factor," in contrast to false-self disorders, which are the product of environmental deficiencies.
     Freud, despite his emphasis on constitutional factors, never separated neurosis from environmental factors. In fact, the interaction between constitutional and environmental factors formed one of Freud's "complemental series" and is at the heart of his understanding of the development of psychopathology. Further, Freud did not view neurotics as existential heroes, as Winnicott suggests, but distinguished very clearly between neurotic suffering and the "common unhappiness" of everyday life. In separating out "neurosis" as a preserve in which Freud's theory remains unchallenged and merely amended, Winnicott perpetuates not Freud's original vision, but a distorted icon.


Harry Guntrip

Guntrip has been the foremost historian, synthesizer, and popularizer of the study of object relations within the writings of Klein, Fairbairn, and Winnicott. His particular vantage point as an analysand of both Fairbairn and Winnicott, the larger historical context he provides, and the fluidity and lucidity of his prose (in comparison with that of the other major British theorists) all contribute to the effectiveness of his overview and synthesis. Guntrip's enthusiastic expositions of Fairbairn's work, in particular, were largely responsible for drawing attention to the latter's contributions.
     Guntrip not only surveyed and synthesized the work of his predecessors, however; he also moved the theory in a very specific direction, according to his own unique vision of human experience and suffering. In 1960 he introduced what he characterized as a modification and extension of Fairbairn's theory of ego-splitting and object relations. This "extension" radically alters the thrust of Fairbairn's approach and generates clinical hypotheses antithetical to those derived from his system.

The Emergent Synthesis and the "Regressed Ego"
Guntrip's history of psychoanalytic ideas is first and foremost a moral history. From the ministry and pastoral counseling he turned to psychoanalysis in search of a deeper, psychologically more sophisticated approach to human experience. But the scientism and the antipathy toward religion he discovered in Freud's work was disturbing to him. Guntrip came to understand Freud as torn by an internal tension between his clinical observations, which were concerned with persons and relationships, and the depersonalized, biologistic theoretical edifice he had constructed in accordance with the mechanistic, Helmholtzian intellectual milieu surrounding him. Guntrip sees the drive theory, the formulations concerning the "psychic apparatus," and the functional analysis of the ego as products of the dehumanizing currents within Freud's intellectual environment, and therefore as unacceptable and dangerous.

While Freud hovered between a psychology of the organism and a psychology of the person, a theory of instincts and a theory of object relations, his theory in toto remained fundamentally oriented to biology. Thus, he makes character dependent on the organic maturing of the sexual instincts rather than dealing with sexual functioning as controlled by the extent to which character has matured in human relationships. Moreover, his radical subordination of objects to the role of mere means to the gratification of instincts is unsatisfactory from a sociological and from a human point of view, since it treats of personal relationships on a subpersonal level. (1961, p. 29)
Guntrip's moral critique of Freud is deep and far-reaching and extends to modern science and technology as a whole. "Science has to discover whether and how it can deal with the `person,' the `unique individual,' we will dare to say the `spiritual self' with all the motives, values, hopes, fears and purposes that constitute the real life of man, and make a purely 'organic' approach to man inadequate." The stakes are high; what hangs in the balance is the "final fate of mankind. If nuclear physics threatens us with the possibility of universal destruction, genuine psychodynamic understanding, if only it be given time to work quietly, gives at least a realistic hope of new life" (1961, pp. 15-16).
     Guntrip's own innovation in theory and practice centers on his development of the concept of the "regressed ego." Fairbairn had described fragmentation of the ego as resulting from the compensatory establishment of internal objects as substitutes for relations with real others. According to him, pieces of the ego remain attached to these internal objects, siphoning off libido from the increasingly emptied central ego, which remains directed toward real others in the external world. He understands the schizoid sense of emptiness and weakness as reflecting the withdrawal of energy from the real world into the world of internal object relations.
     Around the time that Fairbairn's terminal illness forced the premature ending of Guntrip's analysis with him, Guntrip began to develop the view that "ego-weakness" is reflective not just of the withdrawal of libido from external objects, but of the withdrawal of libido from objects altogether, whether external or internal. He argues that Fairbairn's "libidinal ego," which acts as the repository of all disappointed longings and hopes for contact and nurturance, undergoes a "final split." Part of it remains attached to the "exciting object," as Fairbairn had described, perpetually seeking relatedness. Another part becomes split off from the exciting object and becomes even more withdrawn, renouncing object-seeking altogether. This "regressed ego" is constituted by a profound sense of helplessness and hopelessness. The depriving experiences with real others have produced a fear of and antipathy toward life so intense and pervasive that this central portion of the ego has renounced all others, external and internal, real and imaginary; it has withdrawn into an isolated, objectless state. In this flight from life, Guntrip suggests, the regressed ego seeks to return to the prenatal security of the womb, to await a rebirth into a more hospitable human environment. Thus, regression entails a flight and a longing for renewal. When the flight aspect is more prominent, the regression is experienced as a longing for death-- relief from conflictual relations with external and internal objects. When the hope aspect is more prominent, the regression is experienced in connection with a return to the protection of the womb.
     After some initial unclarity about the relationship of his own formulations to Winnicott's work (1969, p. 74), Guntrip concluded that his concept of the "regressed ego' includes both the split-off portion of the libidinal ego in flight from all objects as well as all "unevoked potentials" (Winnicott's "true self"), which because of maternal deprivation have never been fully experienced or expressed in relation to others in the first place. Guntrip further pieces together aspects of Fairbairn's and Winnicott's concepts with his own formulation as follows: when maternal deprivation is experienced as "tantalizing refusal" (Fairbairn's "exciting object"), the result is an active oral libidinal ego bound to internal objects; when maternal deprivation is experienced as simple neglect or impingement (Winnicott), there is a more pronounced retreat into a passive, withdrawn state where the ego is in hiding and potentials never develop (1969, p. 70).
     In Guntrip's view the regressed ego exerts a powerful pull out of life, both from the world of real others and from the world of internal object relations. He attributes to this regressive pull an increasingly more central role in the dynamics of all psychopathology. The lure of the regressed ego threatens to deplete the total personality, plunging the patient into isolation and dysfunction. Guntrip suggests that the early traumas generated by inadequate mothering are essentially frozen in time: the helpless and terrified infantile ego, overwhelmed by unrequited longings and dread of abandonment, remains alive within the regressed ego, in the heart of the personality. "Ego-weakness" is both the experiential product of regressive longings which perpetually tug at whatever ego is left attached to objects as well as an actual structural impairment, reflecting the helpless terror of the infant, encased in the heart of the personality, generating a perpetual sense of inner dread and vulnerability. In the face of the constant threat of total depersonalization and disorganization, reasoned Guntrip, the ego continually struggles to remain attached to life. All mental life and involvements with others, real and imaginary, operate most basically as a defense against regressive longing. Thus, the concept of the "regressed ego" becomes a conceptual black hole, swallowing up everything else. Conflictual relations with others and masochistic attachments to bad internal objects serve as the ego's protection against regression. Oral, anal, genital fantasies reflect "a struggle . . . to `stay born' and function in the world of differentiated object-relations as a separate ego," as defenses against the central part of the personality that has "gone back inside." The dangers threatened by the regressed ego constitute the "tap-root" of all psychopathology; all forms of pathology are defenses against the "schizoid problem." Psychoanalytic treatment is a controlled "exhaustion illness" (1969, pp. 79, 53, 215, 78). All defensive attachments to objects, real and imagined, are relinquished; the weak and helpless infantile ego emerges; and the largely maternal relationship with the analyst, through "replacement therapy," allows the ego to reintegrate and enter the world on a positive basis.

Guntrip's Divergence from Fairbairn
In championing Fairbairn as an alternative to Freud, Guntrip blurs fundamental differences between his own approach and Fairbairn's, and this obscures important conceptual issues and clinical choices. One broad difference is their overall approach to the history of psychoanalytic ideas. Fairbairn and Guntrip both begin with a critical evaluation of classical drive theory, but their critical basis is quite different. Fairbairn's objections rest on conceptual and pragmatic grounds: the presuppositions of pure pleasure-seeking and the separation of energy from structure are anachronistic and misleading; his own theory of endopsychic structure is closer to the clinical data, more economical, and provides more interpretive possibilities. Guntrip's objections to the classical tradition are less conceptual than moral and aesthetic. Freud's "theories of instinctive sex and instinctive aggression have done as much harm to our general cultural orientation in this century, especially in the atmospheres engendered by two world wars, as his opening up of the field of psychotherapy in depth has done good" (1971, p. 137).
     Guntrip objects to Freud's motivational theory (man as governed by impersonal drives), to machine metaphors, and to the presupposition of innate aggression. His fundamental concern is with the ethical implications of Freud's language and system.

Where does the difficulty about Freud's position lie? It is not that his description of the actual sexual situation of civilized man in our time is inaccurate ... His picture of the state of sexual frustration inside marriage, and its wider repercussions ... is both true and challenging. Large numbers of human beings experience a strong and persistent pressure of sexual need either conscious or repressed, and the upsurge of sexual impulses, in a way that finds no gratification within the limits of monogamous marriage and civilized sexual morality ... The question concerns the interpretation to be put upon these strong sexual impulses. If they are indeed solely manifestations of an innate, constitutionally powerful instinct, then we have little option but to tolerate rebels or to endure the spread of neurosis. (1969, p. 71)
Guntrip believes the conclusions Freud's theory produces regarding human possibilities too grim. He considers drive theory degrading to mankind and, on that basis, unacceptable. He sees the shift toward the study of the ego and the emphasis on its resources in Freud's later writings as moving psychoanalytic theory in a more acceptable direction. "Here, truly, is a reprieve for and a reinstatement of the ego of ordinary consciousness to a position of dignity from which Freud's earlier, biologically oriented, instinct-theory threatened to dethrone it" (1969, p. 100). This teleological approach to theory is a bit like criticizing the "big bang" theory of the origin of the universe because of its implications for the eventual fate of the cosmos. Schafer points out that in Guntrip's writings, "It sounds as if kindly theory is better theory, or a kindly theoretician a better theoretician" (1976, p. 118). This seems an apt characterization, and constitutes a distinct and significant departure from Fairbairn's concerns.
     Guntrip's most important departure from Fairbairn's system centers on his claim that the regressed ego constitutes the "core" of all psychopathology. Although he suggests that this concept is simply an extension of Fairbairn's formulations, a closer look reveals it to be a radical shift in a direction leading to antithetical clinical hypotheses and interpretations. Guntrip's premise is that the dominant dynamic pull within human experience is a total retreat from others, real and imagined, in a deep longing for a "return to the womb." This is a puzzling formulation, even on its own terms. What are the ingredients of such a longing? Since Guntrip does not explicitly suggest the existence of prenatal memories (as Winnicott does), fantasies of a withdrawal to a completely nurturative, supportive environment can only derive from elaborations of pieces of actual experience with caretaking figures. Unless it is a memory, a longing for a return to the womb can only be a metaphor for or fantasy of perfect mothering. Guntrip states emphatically that the regressive flight seeks objectlessness, and he distinguishes regressive fantasies concerning a return to the womb from fantasies concerning the breast precisely on that point. However, it is not clear why the womb, any more than the breast, represents an objectless state rather than a particular form and representation of "exciting object." It is not clear why the regressed ego is seen as withdrawn from all objects rather than seeking a particular fantasied object of an infinitely perfect and totally supportive nature. Another reading of Guntrip's theory of regression would be to regard the "womb" as a biological metaphor standing for a psychological and developmental state of undifferentiation, not so much objectless as pre-object. This reading brings it very close to formulations concerning an early undifferentiated matrix in the work of Jacobson, Loewald, and others within the ego psychological tradition. Again, such an approach is a marked departure from Fairbairn, who emphasizes the child's objectrelatedness and reality-orientation from the start.
     Guntrip emphasizes regression and withdrawal exclusively--which seems both unconvincing and at odds with the basic thrust of Fairbairn's work. He points out that flight is a ubiquitous reaction to conflict and deprivation and amasses various clinical examples revolving around a wish to retreat and escape. He then argues that this regressive pull toward flight is not simply reactive and transitory, but so pervasive and powerful that it becomes the dominant motivational thrust within the personality. All other motivations, all relations with external and internal objects, become defensive bastions against this regressive pull. This conclusion is novel and startling. It is one thing to say that flight is a reaction to difficulty; clearly withdrawal from others in the face of deprivation, anxiety, and conflict is a common response, and Guntrip's depiction of schizoid withdrawal is a clinically useful account of the phenomenology of such reactions. However, it is something else altogether to argue that flight is the predominant motivation in human experience. Surely this does not coincide with Fairbairn's view. Fairbairn had argued that attachments to objects are so necessary and adhesive that bad objects are relinquished only when good object relations are experienced as a real possibility. Compulsive attachments to bad objects are maintained, since objectlessness is impossible, both conceptually and experientially. For Guntrip, objectlessness is not only possible; the threat of depersonalization and devitalization created by the lure of regression to an objectless state is the deepest and most pervasive anxiety within the psyche.
     Guntrip has stood Fairbairn on his head by subtly reversing priorities. In Fairbairn's system, object-seeking, the need for contact and relation, is primary; for Guntrip, withdrawal is primary, and object-seeking is a secondary defensive reaction against the terror of regressive longing. For Guntrip, the regressed ego abandons objects. For Fairbairn, the ego can never rid itself of objects; it is by its very nature entangled with them. Fairbairn sees even the most regressed and seemingly withdrawn behavior as deriving from powerful ties to internal objects. (For example, he [1954] suggests that in "auto-eroticism" the genitals symbolize objects; libido is always object-related.) He considers the greatest resistance in psychoanalysis to be the libidinal attachment to bad objects. Guntrip considers the greatest resistance in psychoanalysis to be the terror of regressive ego weakness and depersonalization. In line with this reversal, he uses the designation "ego psychology" in his later writings to characterize his point of view. "Object relations theory" has become a more superficial layer of analysis: "`depression has to be understood . . . from the point of view of object-relations,' i.e., the need and struggle to retain object-relations (guilt being an object-relation), but ... the deeper problem of regression which it masks, has to be understood `in terms of ego psychology"' (1969, p. 144). Guntrip retains the complexities of prior theory, but he reinterprets the fundamental function of all dynamic processes in terms of defending against regression.
     To highlight the clinical implications of the difference between Fairbairn and Guntrip in understanding psychopathology and resistance, let us consider a piece of autobiographical material provided by Guntrip himself. In a posthumously published article (1975) he presents a moving and candid retrospective consideration of the significance in his life of his personal analyses with Fairbairn and Winnicott, which centers around his own internal struggle with regressive longings. Guntrip differed with Fairbairn's interpretations of this material, and his dissatisfaction with the latter's approach led him to formulate his own innovative concept of the "regressed ego." (Kernberg, 1980, argues that Guntrip's unresolved transference toward Fairbairn was responsible for his tendency to distort the latter's ideas.)
     Guntrip portrays his mother as a woman who, deprived of maternal care and compelled to take care of younger siblings, approached the experience of motherhood with a sense of duty and deep resentment. She nursed Harry, her firstborn, in the hope that nursing would forestall a second pregnancy. She refused to nurse her second son, who subsequently died, at which time she renounced all sexual relations and devoted herself to a business career. Guntrip, according to his mother's account, walked into her room at the age of three and a half to discover his dead brother on her lap. He subsequently developed a severe, mysterious illness. He was sent away from the mother to an aunt, where he recovered, but he remained for several years sickly and demanding. Guntrip suffered a complete amnesia for the experiences surrounding his brother's death. Yet one major indication of their importance in his later life was a recurrent "illness" involving total exhaustion that would develop following the departures of close fraternal figures. Between "illnesses" he was compulsively active and preoccupied with work.
     Themes and images about death, tombs, buried men, and so on recurred in dreams throughout his life. Although Guntrip had sought analysis to aid in the recovery of these early memories and the cure of the psychogenic illnesses which he felt derived from them, the memories had remained inaccessible during his two analyses. They finally emerged in a series of dreams in his later life, when both Fairbairn and Winnicott were dead. The dreams were precipitated by a retirement forced upon him for reasons of health, which he experienced as a surrender to his mother's wishes to destroy him or to let him die. They consist of a series of images of the mother as immobile, unreachable, frozen, and, finally, a figure with no face, arms, or breasts, holding on her lap the dead baby brother. Guntrip understood these images as the reemergence from repression of the memories of his mother as paralyzing, aloof, and totally schizoid. The dreams were followed by a mood state consisting of a dull, mechanical, lifeless apathy--a final, diminishing echo of the earlier illnesses of collapse.
     What is the nature of the images of Guntrip's mother as frozen and mutilated? What is the dynamic significance of his exhaustion illness? From Guntrip's point of view, the images and illness represent the final retreat of the schizoid "heart of the self" from life, because of gross and terrifying maternal deficiencies. This interpretation derives from his revision of Fairbairn's theory, proposing a final schizoid split in which the center of the self detaches itself completely from objects and renounces life. The exhaustion illness represents a desperate escape from the horrifying visage of the mother and the dead brother. Throughout his analysis he felt that Fairbairn shed little light on these problems. He reports that Fairbairn founded his interpretations on "oedipal" dynamics, on what he felt were Guntrip's attempts to get his mother to mother him. Guntrip does not explain or elaborate Fairbairn's thinking on this point; nevertheless, it is possible to construct a Fairbairnian approach from the latter's basic theoretical principles. In Fairbairn's system, Guntrip's images and feeling states would be viewed not as an escape from his mother, but as a return to her, a longing for the reestablishment of his early connection to her, in her depression and aloofness, her morbidity and desolation--an unconscious yet tenacious holding on to her. This holding on is reflected in a dream Guntrip recounts, illustrating in the most concrete and literal terms Fairbairn's notion of object tie: "I was working downstairs at my desk and suddenly an invisible band of ectoplasm tying me to a dying invalid upstairs, was pulling me steadily out of the room. I knew I would be absorbed into her. I fought and suddenly the band snapped and I knew I was free" (1975, p. 150). In Fairbairn's system the "invisible band" is the object tie which Guntrip actively, although unconsciously, perpetuates, despite his counter-depressive defenses and disclaimers.
     The implications of these alternative interpretations are quite different. From Guntrip's point of view, his mother was unable to support life, and his perception of this was horrifying and traumatic. Although he tried to win her love later in childhood through provocation, his deepest fears were based on his massive flight from his mother and all other objects, a flight which manifests itself in episodic collapses. Within Fairbairn's system, the central issue is not flight from, but devotion and allegiance to, the depressed and desolate mother of Guntrip's early years. The collapses, so dreadful to him, represent a longing for a reunion with the dead and lifeless core of the mother, with whom the dead brother remains in envied union. As he expresses it in the dream, "I knew I would be absorbed into her." The faceless mother is the "exciting object" which Guntrip seeks continually in his descents into exhaustion and lifelessness. Within Fairbairn's system, Guntrip has turned his profound longing for his schizoid mother (manifested in his exhaustion illnesses) into an illusion of escape from her.
     Guntrip's reworking of Fairbairn poses several serious problems in its clinical application. The patient for him is an innocent and passive victim of parental deficiency. At the core of all forms of psychopathology is a helpless, frightened child who has gone into hiding. Guntrip takes considerable pains to absolve the patient of accountability for perpetuating his condition--the problem is "pre-moral" (1969, p. 10), deriving solely from primitive terror, never from conflicts or ambivalence. In this respect Guntrip's vision closely resembles Freud's earliest theory of neurosis based on infantile seduction and Rank's theory of the birth trauma. Guntrip objects to Rank's emphasis on the physical aspects of the birth experience rather than on the personal, emotional atmosphere provided by the mother, as well as to his call for a quick cure, a direct and immediate assault on the affective residues of the birth trauma rather than a gradual analytic working through of defenses. However, he echoes Rank's view that underlying all neurosis is an encapsulated traumatic (hysterical) neurosis which provoked and preserves a flight from life.
     Guntrip's account of psychopathology and the analytic process resembles the child's fable "Sleeping Beauty." A terrible trauma has occurred in early childhood, inflicted on the passive and innocent child from the outside (a disgruntled fairy). The terror and helplessness of this traumatic event remain imbedded in the heart of the personality, awaiting a call back to life by a more hospitable environment. The self of the patient is ultimately passive; lack of good mothering produces a retreat to a lifeless withdrawal, until the analyst (the prince) awakens it from its slumber. We have noted Fairbairn's tendency to portray the patient as an innocent victim. This tendency is somewhat balanced in his system by his view of neurosis as an active perpetuation of bad object ties. Guntrip removes this balance; object ties are merely a defense against more basic fearful flight. In his system terror has replaced active allegiance as the bottom line of neurosis and the source of the greatest resistance within the psychoanalytic process.
     Corollary to Guntrip's presentation of the patient as victim is his utopian premise that complete happiness, free from anxiety and conflict, is possible within human experience.
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 interests and abilities, within an overall good ego-development, in good object-relationship. (1969, 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 extra-familial 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. (1969, p. 357)
      Guntrip attributes the universal failure of mankind (to live up to this myth of mental health) to inadequate mothering--depriving, insensitive, or malevolent. His absolution of the patient is paralleled by a tendency to villify the actual parents, in contrast to the ideal parenting and love he calls upon the therapist to provide "in loco parentis" (1969, p. 350). This polarity between the good parenting of the therapist and the good life to which it presumably leads and the bad parenting of the actual caretakers perpetuates both a splitting of object relations and the transference, and a disclaiming of accountability for the active perpetuation of the neurosis. What is at stake in therapy, in Guntrip's vision, is the "saving of the ego" (1969, p. 213). Just as humane psychodynamic theory constitutes the hope for mankind, the therapist, in his view, becomes an heroic nurturative figure, rescuing the helpless patient from the terror and paralysis generated by disastrous mothering.

The Relational Model in Perspective
It would be convenient, perhaps, if the history of psychoanalytic ideas consisted of an unbroken line of progressive advances, with each new theorist standing squarely on the shoulders of his predecessor, using prior theory as a solid foundation for further, incremental exploration. Instead, the major figures within the history of psychoanalytic ideas have a complex and discontinuous relationship. Psychoanalytic theory is not simply additive; it consists of a collection of uniquely fashioned crystallizations of ideas and data, often overlapping, but with different centers and organizational principles. The authors we have considered in "The British School" do not constitute a "school" by virtue of subscribing to a set of shared beliefs, but, like a school of painters, by virtue of a shared set of problems and sensibilities.
     The most fundamental common problem addressed by these theorists, as well as by theorists within the American interpersonal school, is the transformation of psychoanalytic metapsychology from a theoretical framework based on drives to a framework which makes relations with others, real and imagined, the conceptual and interpretive hub. The various versions of the relational/structure model share a common set of assumptions which set them apart from earlier drive/structure theory: the unit of study of psychoanalysis is not the individual, but the relational matrix constituted by the individual in interaction with significant others. The stuff of personality and the patterns that characterize psychopathological functioning are formed from that relational field. While physiological needs, bodily events, temperament, and other biological factors significantly affect human experience and behavior, they operate within the context of an interactive matrix and are subsumed by the preeminant motivational thrust toward the establishment and maintenance of relations with others. Each major theorist of the British school made important contributions to the movement of psychoanalytic theory from the drive/structure model to the relational/structure model. The nature and the style of presentation of these contributions, however, vary considerably; they begin with a common starting point and arrive at a common destination, but each covered the intermediate conceptual expanse differently.
     Klein began her researches into the mental life of children with a total dedication to classical drive theory. However, the data yielded by her efforts were rife with powerful, primitive phantasies involving loving and hateful relations with others--both with real, external figures as well as with characters in an internal drama of passionate and tragic intensity. Her interpretive focus shifted from an early emphasis on psychosexuality, to an almost exclusive emphasis on aggression, to a more balanced view of mental life centered around a profound struggle between love and hate, restoration and destruction. Although she retained the language of drive theory throughout and presented her work as an extension of classical theory, Klein's formulations alter, in a subtle yet pervasive way, the nature and function of the drives. Drives, in her system, are bound inextricably to objects, built-in with considerable specificity to the very experience of desire itself. Mental life in health and in pathology is a complex fabric woven out of relational strands--each component is an informed and personal engagement between the self and an other, real or imagined, external or internal. Gone are the basic components of the drive/structure model--directionless impulses seeking tension reduction. Gone is the classical closed energy system in which psychic energy is finite and distributed through alternate channels. Underneath Klein's drive theory language, a new vision of the mind had begun to emerge. The pursuit of pleasure and the avoidance of pain recede into the background in this motivational framework; the struggle between hateful destruction and fragmentation, between loving restoration and integration, occupies center stage.
     Klein's work made possible the contributions of Fairbairn, Winnicott, and Guntrip. Fairbairn applied her depiction of internal object relations to a thorough reconsideration of classical motivational and structural theory. He also retains some of Freud's language, though, unlike Klein, he explicitly redefines the terms. Libido for Fairbairn is not pleasure-seeking but object-seeking; psychic energy is not directionless and structureless but organized and oriented toward the reality of other people. Development, in his view, consists of a maturational unfolding of different modes of connection to other people, and all psychopathology is constituted by a failure in that development and a subsequent attachment and allegiance to infantile object ties, established as internal residues.
     Fairbairn's system, although often highly sparse and schematic, represents, along with the interpersonal theory of Sullivan, the purest and most consistent formulation of the relational/structure model. Fairbairn and Sullivan view the development of personality and psychopathology from a similar perspective: the child is enmeshed in relations with others and discovers himself in interaction. This embeddedness of the child with others is the overriding feature of early development, and the need for attachment, connection, integration with others is the preeminent motivational thrust of the human organism throughout life. While Sullivan delineates this common relational vision largely by studying patterns of behavior and interaction, Fairbairn's emphasis is on intrapsychic residues of relational experience--the internal fragmentation and patterning of the personality.
     Guntrip's emendation of Fairbairn's theory with the introduction of the concept of the "regressed ego" raises the retreat from object relations to a superordinate motivational principle, displacing Fairbairn's emphasis on attachment to objects. Guntrip's point of view, however, with its analysis of human experience and difficulties as derivative of vicissitudes of relations with others, remains as an alternative theory within the relational /structure model.
     Winnicott negotiated the expanse between the drive/structure model and the relational/structure model by sidestepping rather than directly challenging classical theory. He addresses himself to the development of an integrated and experientially real self, which he suggests is a problem not explored by Freud and Klein, and antedates or underlies the kinds of issues they concerned themselves with. Winnicott delineates a set of unfolding relational needs for specific maternal provisions which have a peremptory nature; the structure of the self and the organization of psychopathology derive from the fate of these early relational needs. Vicissitudes of and conflicts concerning the drives become important, he suggests, only in the context of these more basic processes. By this device, he introduces a theory of motivation, development, structure, and psychopathology that is based on relations between the self and others and operates squarely within the relational/structure model.
     In the work of Klein, Fairbairn, Winnicott, and Guntrip, object relations occupy the central focus and are understood to constitute the basic stuff of human experience. However, the nature of the object, its origins and qualities, is quite different in the formulations of these different theorists. Each sees particular aspects of objects as most crucial for development and psychic structure. For Klein, objects tend to have universal features. In many of her theoretical statements she stresses the a priori origins of object images as: part of a phylogenetic inheritance, built onto the experience of desire itself, construed from early sensations, or derived from the drives through projections. Although different in terms of frequency and severity, the content of these objects is the same for everyone--good and bad breasts, good and bad penises, babies, united parental couples. Klein stresses the importance of real people in the child's life; however, here, too, the universal features of these real objects are most important--their anatomical characteristics as representatives of the human species, their durability in the face of phantisied attacks against them, their inevitable mixture of gratifying and depriving features. The dramatis personae within the external and internal object worlds is standard.
     For Fairbairn, on the other hand, objects are highly specific and personalized. Internal objects are fashioned exclusively out of the particular features of the child's actual experience with the parents. The exciting object entices in precisely the manner in which the parent seemed to offer contact; the rejecting object attacks and withholds in precisely the manner in which the parents failed to provide contact; the ideal object provides contact precisely through the parents' actual pleasures and values. Fairbairn's structural theory parallels Sullivan's account of the formation of the self-system in this crucial respect. The personality of the child is patterned and organized in direct complementarity to the character of the parents, their actual behavior, appearances, subtle differences in responsiveness to the child, and so on. Even in light of this specificity, however, Fairbairn's categories are uniform and narrow. He sees the superordinate need of the child as the longing for emotional nurturance. "Good" objects are those features of the parents that provide for infantile dependence; "bad" objects are the features of the parents that fail to provide fully for infantile dependence. Other areas of the parents' personalities, unrelated to the issue of dependency, fade into the background. Sullivan, in contrast, considers many different features of parent-child interactions in addition to needs for early nurturance, and he extends his relational analysis beyond early infancy to later childhood and adolescence. Guntrip further narrows Fairbairn's singleness of focus in this regard; maternal nurturance for the very young infant assumes an exclusive focus and all subsequent development and psychopathology serve as reactions against failures in the earliest maternal contact.
     For Winnicott the nature of the object is also constituted by early maternal provisions, but these are somewhat more varied. The child has built-in needs for: a holding environment; mirroring; the actualization of his omnipotence; the opportunity for object-usage; the toleration of the ambiguities of his transitional experience; the opportunity to console. The child's early objects are prepatterned according to the templates provided by the child's own developmental needs. However, Winnicott, like Fairbairn, also brings the particularities of the actual parents into prominence. The specific characterological features of the parents which stand in the way of their fulfilling their caretaking functions become crucial for the child; the consolation of the actual parent in terms of his or her own difficulties in living becomes a concern and a prerequisite to the child's further development.
     Partially because of the difference in conceptualizing the origins and nature of objects, there is a marked difference in sensibility regarding private, personal fantasy between Klein and Winnicott, on the one hand, and Fairbairn and Guntrip, on the other. Klein and Winnicott regard the inner world of idiosyncratic fantasy as the most basic level of experiential reality and the external world of real others as a secondary, although important, realm. For Klein, phantasy, generated out of the drives themselves, constitutes the bedrock of experience; primitive
phantasies dominate early development and phantasies concerning one's internal object world underlie the sense of self and the nature of reality. The internal world, in her system, provides life's greatest resources and deepest torments. Winnicott likewise stresses the depth, beauty and primacy of the most private experiences. Subjective reality is the basis for all creativity prior to the transition to the objective external world; one's deepest being is removed from contact with others, remaining forever isolate.
     Fairbairn and Guntrip, by contrast, regard the inner world of internal object relations as secondary and compensatory. In their view, the child is oriented from the beginning toward contact with the parents as real people. The turn inward, the establishment of internal object relations (Fairbairn), and the regression of the ego (Guntrip) are substitutive replacements for what is missing in actual relations with the parents. Thus, although internal object relations are crucial to the understanding of psychopathology in the systems developed by Fairbairn and Guntrip, internal object relations are considered to be essentially masochistic and defensive rather than the underlying foundation and resource Klein and Winnicott take them to be.
     In devising a relational model theory, the central concern of each author has been to provide a compelling and clinically useful account of the origins and development of relations with others. However, abandoning the drive/structure model as a theoretical base creates other problems which each relational theorist must address. The classical drive theory provides a way of conceptualizing the source of psychic energy, the principles upon which the mind becomes patterned or structuralized, and the phylogenetic endowment which the child brings to his experience of the world. In abandoning drive theory, relational model theorists must provide alternative solutions to these issues, and the manner in which they have approached them varies considerably.
     In the drive/structure model the energy which fuels mental phenomena is drawn from transformations of drive tensions. To eliminate the drive concept is to eliminate the energy source, and a new source, or a new way of conceptualizing the problem, is necessary. Klein deals with this question by retaining the language of drives as energic forces, while changing their meanings into relational configurations. Libido and aggression become constellations of loving and hateful affects, images, and relationships respectively. Drives still fuel mental phenomena, but drives are impulses toward different relational patterns. Winnicott, although he does not address himself at length to this issue, tends to adapt a similar strategy--for example, by redefining aggression as motility or a kind of elan vital. Thus, aggressive energy becomes a relational thrust which is either facilitated by the mother or squelched, turning it into destructiveness. Both Sullivan and Fairbairn present more radical challenges to the classical theory of psychic energy. For them, the separation between mind as a set of structures and energy as the fuel which drives them is fundamentally misconceived. Mind is energy. For Sullivan the self is not a quasi-entity, but a dynamism, a pattern of energy transformation; for Fairbairn ego structures are energic, manifesting themselves in relational impulses.
     In the drive model, the mind is forged out of the necessity for drive gratification and regulation. To eliminate the drive concept is to eliminate the basis for the patterning of mental phenomena, and a new basis for that patterning is provided by each relational theorist. For Klein psychic structure is derivative of phantasy. The world of internal objects develops out of early relational needs, and its organization, set by the predominant, recurring phantasies, underlies the individual's experience of himself and his interpersonal world. Fairbairn, Sullivan, and Winnicott all see the personality as patterned around the necessity for maintaining the best possible connection with the parents. For Fairbairn disturbances in object relations necessitate the establishment of compensatory internal objects, and these become the kernels of different components of the personality. For Sullivan the self becomes organized first for the purpose of avoiding anxious interactions with caretakers, and subsequently for maintaining levels of anxiety at a minimum. Winnicott portrays the structuralization of the self into true and false dimensions as a consequence of the child's dual need to engage the parents through the caretaking which is provided, yet to protect the self from being overwhelmed or exploited. Hence, each relational theorist derives psychic structure not from the need to regulate drive tensions but from central relational needs.
     In the drive/structure model the infant is portrayed as equipped with a complex phylogenetic inheritance, both drives and innate ego capacities. These inborn, physiologically based characteristics set a prior cast to the individual's life, contributing both to character formation and to the direction in which psychopathology develops. The world of reality and actual others becomes relevant and meaningful to the extent to which they approximate (either through fulfillment or frustration) the various needs derived from the drives. In abandoning the drive concept, each relational theorist also necessarily disassociates himself from this particular manner of assigning to the body and to the physiological endowment of the infant a central role in the development of personality and psychopathology. In fact, adherents of classical drive theory often criticize relational model theories on the grounds that they constitute an extreme and naive environmentalism, viewing mental life as a simple registering of external events. In this view, relational model theories, by abandoning the drive concept, omit the central role of the body in human development and the importance of innate factors in general.
     Physical sensations are the basis for all experience. The infant's life is dominated by physiological needs; bodily images and preoccupations pervade much of later psychopathology. The difference between the approach to the body in relational model and drive model theories is not in terms of whether the body is important, but in what way. Bodily needs are not viewed, as they are by drive structure theorists, as the originator of important psychological intentions and meanings. Bodily events and processes are regarded instead as providing a language for experience, a vehicle for the expression of intentions and meanings, relational in nature. Thus, in Klein's theory bodily tensions do not create motives which demand actions to relieve tensions; the child loves or hates, and uses bodily processes to express these motives. Fairbairn characterizes erogenous zones as "channels" to the object; Sullivan stresses the manner in which "zones of interaction" color the different experiences of the infant with the caretaker. Sullivan grants bodily tensions (as the major source of the infant's "needs for satisfaction") a prominent place in his theory. They are the major force drawing the infant into interpersonal configurations with caretakers. However, what is important psychologically is not whether such needs are gratified or frustrated but the quality of the interaction, the anxious or nonanxious nature of the relationship. Similarly, Winnicott emphasizes the importance of physical handling in early experience, not in providing specific forms of gratification but in expressing and mediating caring and responsiveness. Relational model theorists tend to regard the body not as generating independent psychological motives which shape experience and behavior but as the major medium of exchange between the infant and his or her caretakers. This is not to deny that the body manifests independent physical needs for food, oxygen, and so on, interference with which has grave consequences. For relational/structure theorists, these independent physical needs do not in themselves play a significant etiological role in shaping personality and psychopathology, whose major determinants are the vicissitudes of more purely relational considerations. "Average expectable" physical care is presumed. The human infant is viewed as an organism whose experience is fully mediated through the body and whose means of expression are limited to bodily events and processes, yet whose psychological nature is dominated by a search for connection, attachment, engagement with other human beings. It is this search that subsumes and imparts meaning to all other dimensions of human life.
     Does the abandonment of the drive concept and the view of the body as the vehicle for, rather than the cause of, psychological motives suggest that in the relational/structure model all innate factors have been eliminated? Do the purer and more explicit relational theories like Sullivan's and Fairbairn's regard personality and psychopathology as a direct, unmediated product of input from others, registered on an organism that brings nothing significant to experience, a psychological blank slate?
     Within learning theory, the earlier account (Skinner, for example) of all experience as a product of learning, with no contributions from innate factors, has been superseded by the concept of "prepared" or "directed" learning, in which experience is understood to register on an organism oriented with a priori sets of expectations, "wired" in a particular fashion (see Konner, 1982, pp. 26-29). Similarly, each relational theory posits, either explicitly or implicitly, a motivational thrust toward interpersonal engagement, a "drive" toward object-relatedness. Experience with significant others becomes important not simply because environmental influences are powerful and the baby is responsive but because the baby is "looking" for certain kinds of experiences, primed for specifically human engagement.
     In Sullivan's principle that "needs for satisfaction" operate as integrating tendencies, in Fairbairn's assertion that "libido is object-seeking," in Bowlby's claim that the infant is preprogrammed for attachment is the common premise that the infant brings to his experience an elaborate phylogenetic endowment. However, that inheritance is understood to consist not of an array of loosely organized body-based tensions but of a complex, coherent set of interests, sensitivities, and expectations which draw the infant into human relationship. Spitz's studies (1965) demonstrated that the infant prefers the human face to all other visual stimuli. A long line of subsequent research findings have catalogued the exquisite synchrony between the infant's inborn visual, auditory, tactile preferences and rhythms and physical attributes of human caretakers as well as their intuitive responsiveness to the baby. This discovery of innate factors in the baby's experience does not pose a problem for relational model theories; rather it supports its central premise.
     The child is primed for certain kinds of experience. What happens after that experience is encountered? Do relational model theorists suggest that experience, interchanges with others, are recorded in a direct, unmediated fashion? Do they presume that the account which the adult maintains of parental figures, for example, is wholly veridical, an undistorted replica of the actuality of those parents? Here too, the answer is "no." Both Sullivan and Fairbairn assume that the infant brings to his or her experience not only expectations but particular a priori principles for organizing that experience. Sullivan delineates a developmental sequence of "prototaxic," "parataxic," and "syntaxic" modes (bearing some resemblance to Piaget's stages of cognitive development) in which all experience is registered. In his theory, psychopathology is not a simple reflection of actual events; it is a complex transformation of events processed and refashioned through different perceptual and cognitive organizational patterns. Similarly, Fairbairn's account of the formation of endopsychic structures does not presume a direct transfer from actual experience to internal residues. Experience with caretakers undergoes a complex sequence of splitting and recombining operations. The content of internal objects is derived from actual experience, but experience that has been transformed according to an innately given set of organizational processes.
     We have considered innate expectations and organizational principles which are fairly standard from one baby to the next. What of those innate features which vary from baby to baby? A growing body of research on temperamental differences in young infants suggests that constitutional differences are stable over time and important influences on personality development (Thomas and Chess, 1980). The drive/structure model accounts for these considerations in terms of different distributions of drive energies (and in later ego psychology, through different constitutional strengths of autonomous ego functions). Is there room in the relational/structure model for innate temperamental differences? Observations of such differences are not inconsistent with relational model premises (although they are often neglected by relational model theorists). The relational model highlights the caretakers' success or failure in meeting the infant's relational needs, in providing nurturance, bonding, a "facilitating environment." What causes failure in these efforts? Early infant researchers tend to characterize such discordances in terms of a lack of "fit" between the particular mother and the particular infant (Stern, 1977). Each baby brings to encounters with caretakers his own particular rhythm of engagement, level of activity, distinct affective and behavioral displays. Each caretaker brings to his encounter with the baby his own style and intensity of responsiveness, attention span, level of interest, anxieties, and so on. The fullest delineation of infant/ caretaker interactions, the most complete relational model account, would take into consideration all these factors, viewing the relational residues of the child as a complex blend of temperamental and experiential influences continually interpenetrating each other. Discordances in these interactions would be understood to represent a lack of fit between the coparticipants.
     Thus, a consideration of innate constitutional factors is not outside the interpretive realm of the relational/structure model, not inconsistent with its basic premises. However, these factors are the least developed area of relational structure theories. Other than Bowlby's important efforts to ground his theory of attachment in biological principles, the inherent characteristics, expectations, and rhythms of relational needs tend to be only vaguely and globally described by relational model theorists. The inherent organizational principles through which experience is coded and retained have been delineated only incompletely. The greatest area of relative omission has been in the consideration of temperamental differences among babies. Sullivan and Fairbairn both write as if the only crucial variable is the caretaker, anxious or not, emotionally available or not. They do not emphasize the extent to which the caretaker's responsiveness or lack of responsiveness is keyed to the particular baby's style and rhythm. For example, Sullivan traces the caretaker's anxious responses to the baby's various behaviors back to the caretaker's own difficulties in living. He does not consider that different babies might display more or less of particular behaviors which are likely to set off the caretaker's anxiety. Although they are not compelled to by the premises of their model, relational /structure theorists do tend to deemphasize constitutional factors and treat the character and emotional presence of the caretaker as the major determinant of personality development.
     What are the reasons for this imbalance? First, the two purest relational model theories, those of Sullivan and Fairbairn, are explicitly theories of psychopathology, not general theories of personality development. Sullivan provides no account of the organization of the self apart from the need to avoid anxiety; Fairbairn provides no account of the structuralization of the self apart from the establishment of compensatory internal objects. Both blame psychopathology on parental failure. This does not imply that innate features of the child are unimportant for a general theory of human development, only that they are not a crucial determinant in creating psychopathology per se. Second, the history of psychoanalytic ideas, like that of most intellectual disciplines, tends toward broad dialectical swings. Freud's early seduction theory placed full responsibility for neurosis on molesting caretakers. His realization of the apocryphal nature of his patients' seduction "memories" led to the development of the drive/structure model, in which psychopathology unfolds from the innate recesses of the child's mind. Klein's formulations of a priori object images and relationships represent the furthest swing of the pendulum in this direction. The contributions of the relational model theorists, both of the interpersonal and British traditions, constitute a reaction to this feature of Freudian and Kleinian thought. Although the principles of the relational/structure model itself do not demand it, they tend to portray the infant as fairly uniform and innocent, with the blame for psychopathology set back once again on the parents. We expect that this slant will disappear as the influence of early infant research on relational model theory increases, with its stress on the interactive nature of temperamental and experiential variables.