| home | directory | reading room |  about therapy | online counseling | mental health information | about us  

It is inevitable that psychoanalysts interested in developmental theory would seek correspondences between the work of Margaret Mahler (1979; Mahler and Furer, M., 1968; Mahler, Pine, & Bergman, A., 1975) and that Heinz Kohut (1971, 1977, 1984), the two authors many recognized as having most significantly influenced American psychoanalysis in the current, post classical era of object relations and self-development. Their methodologies, of course, are significantly different. Mahler based her theory on systematic observation and careful documentation of the psychologically relevant behavior of infants and toddlers in interactions with their parents, providing a view of the development of the "objective self," a concept we put forward in an effort at integrating Kohut and Mahler (Shane and Shane, 1980). The term objective self in our formulation was based on the vantage point from which these observations on the development of the self were made. Kohut, unlike Mahler, derived his theory exclusively from the psychoanalytic situation, creating a view of the development of the "reconstructed self," as we termed it in that same publication, based again on the vantage point from which his observations were made.

But despite this difference in their vantage points, because both Kohut and Mahler began their respective inquiries with the same general frame of reference, that is, psychoanalytic, and because both stress the significance of the infant-caretaker unit in the earliest period of life, congruencies between their investigative findings as well as their conceptual innovations were to be expected. And, in fact, it seems that both Mahler and Kohut had, at the beginning at least, that same expectation. Kohut (1980), in a discussion of our 1980 paper, described an interchange he had had with Mahler in the early '70s. Mahler had written a letter at that time asking him what he saw as the differences between her theory and his. Kohut answered Mahler's query in such a way that conceptual distinctions between them were played down, saying that he viewed himself and Mahler as " digging from different directions into the same area of the mountain" (1980). Thus, in the 1970s, he seemed to infer that the approaches of the observer of children and of the reconstructor of children might generate findings that could meet, or be integrated, the same position we had taken. The fact is, however, that Kohut, in discussing our 1980 paper, indicated that since the exchange of letters between himself and Mahler some years before, he had changed his mind about the possibility of integration between them, not because of the differences in his and Mahler's methods of data gathering, but because of differences that later evolved in their respective schools of thought.

In this chapter we will begin with a rather full review of the developmental sequence put forward by Mahler, using as much as possible her own language and theoretical underpinnings. Unless otherwise indicated the formulations are contained in Mahler, et al. (1975). We will then examine selected aspects of this developmental scheme from the viewpoint of Kohut's conceptualizations, noting where the views overlap and where they part company. At the same time we will assess Mahler's and Kohut's developmental postulates from the vantage point of current infant observation.

Mahler's chief contribution to psychoanalytic developmental theory is her study of what she terms the separation-individuation phase, or process, which culminates in object and self-representations and stable identity formation. To this end, Mahler conceived a research design using analytically trained observers who applied classical, analytic, genetic-dynamic insights to infant-mother interactions. She discovered what she considered to be a normal, ubiquitous process occurring in children who successfully maneuver the first three years of life. As Harley and Weil (1979) stress, "Mahler's intent was not to add any new theory; to the contrary, our ultimate aim was to integrate into our existing body of developmental knowledge detailed in systematic findings on the beginning enfoldment of object relations" (pp. xiv-xv).

This separation-individuation phase, Mahler asserts, establishes a sense of separateness from and relation to reality, particularly in regard to the relationship between one's own body and the primary love object. This phase is never finished but occurs primarily from four months to 36 months of age. Separation, defined by Mahler as an intrapsychic achievement not to be confused with physical separation, refers to an emergence from symbiotic fusion with the caregiver; individuation refers to the child's assumption of his or her own individual characteristics. While these processes are intertwined, there may be either lag or precosity in one or the other. For example, a mother may interfere with separation strivings in her infant but may encourage progressive development of cognitive, perceptual, and affective function.

The Forerunners of the Separation-Individuation Process

Mahler identified two phases that precede separation-individuation, the normal autistic phase and the normal symbiotic phase, which phases were less systematically studied by Mahler through direct observation and therefore are less behaviorally detailed than the finely noted aspects of the subsequent developmental sequence, which is, after all, her main focus. The concepts of autism and symbiosis were derived for the most part through Freud's (1900, 1905, 1915, 1920, 1923, 1926) theories of early development and Mahler's own psychoanalytic reconstructions of the study of psychotic and borderline children and adults.

Mahler's normal autistic phase describes the initial state of the infant, who is viewed as existing in a monadic system (as contrasted to a dyadic system), self-sufficient in hallucinatory wish fulfillment. The principal task of this phase is the achievement of homeostatic equilibrium through somatopsychic mechanisms. Only gradually does the infant become aware of need satisfaction stemming from outside the self.

This dim awareness of the need-satisfying object marks the beginning of the normal symbiotic phase. Mahler is denoting a social symbiosis, a matrix of physiological and sociobiological dependency on the mother for which structural differentiation takes place, leading to a functional ego. The physiological need gradually becomes a psychological wish, and the aspects become object bound. Concomitantly the body image is delineated. The infant's inner sensations from the core of the self remain the central crystallization point of the feeling of self around which the sense of identity will be established.

Normal autism and normal symbiosis are prerequisite to normal separation-individuation. The normal autistic phase promotes homeostasis; the normal symbiotic phase creates the ability of the human being to cathect the other within a vague, dual entity, creating the primal soil of which all human relationships form. The subsequent separation-individuation subphases are characterized by a steady increase in awareness of the self separate from others, which coincides with the origins of the sense of self and of true object relationships, as well as an awareness of the reality in the outside world.

Images of the love object, as well as images of the bodily, and later a psychic, self, emerge from the increasing memory traces of pleasurable and unpleasurable experience. But even the most primitive differentiation from the mother depends upon the matching of discharge patterns of mother and infant, and later interactional patterns discernible in mutual cueing. The mother's availability is thus essential to progression into the next phase.

At the age of 4 to 5 months, the peak of symbiosis, differentiation is observed. By this time the achievement of the social smile indicates the establishment of the specific bond. A central core of dim body awareness is hypothesized, based on behaviors that demarcate self from other. For example, the infant is observed molding to his mother, distancing from her, feeling his own and is mother's bodies, and handling transitional objects.

Six to seven months is importantly marked by hatching, defined by Mahler has a gradual ontogenetic evolution of the sensorium, characterized by a new look of alertness, persistence, and goal directedness. The child is observed pulling at his mother's hair and ears, feeding her, straining back for a better look at her, all in contrast to the earlier molding into mother when held.

The child's creation at this time of transitional objects is a monument to his need for contact with the mother's body. In addition, the mother's preferred soothing or stimulating pattern is taken over, assimilated as a transitional pattern.

At seven to eight months, the infant develops a checking-back-to-mother pattern. Mahler calls this the most important normal pattern of cognitive and emotional development. The baby begins with comparative scanning, comparing mother with the other. He becomes more and more familiar with how she feels, tastes, smells, and looks.

When symbiosis is delayed or premature, differentiation is similarly delayed or premature. Such disturbed symbiosis can be caused by the mother's indifference, ambivalence, intrusiveness, or unpredictability.

Mahler notes that as important as inborn potential is for eventual harmonious personality development, a favorable mother-child interaction also affects subphase adequacy. The mother, too, has to adapt, particularly at crucial periods.

Mahler and her colleagues (1975) conclude

It is the specific unconscious need of the mother that activates out of the child's infinite potentialities those in particular that create for each mother "the child" who reflects her own unique and individual needs. This process takes place of course within the range of the child's innate endowment [p. 60].

The Second Subphase: Practicing (10-17 Months)

As the child enters the practicing subphase, he and his mother are less able to take undisturbed pleasure in their close physical contact with one another, but they can enjoy one another at a greater distance. The mother who has difficulty relating to her child tends to want him to grow up quickly as soon as he seeks to distance himself. Such children, in turn, find it difficult to give up a demanding closeness. On the other hand, children who have had the best relationships with their mothers can now venture the farthest from her; but when they become concerned about separation, they are able to turn to her once again. How the child perceives the widening world depends on mother, for while the thrust for individuation is an innate given, it can be experienced as painful if mother is not available to alleviate fears, falls, and hurts.

The central feature of the practicing subphase is the late investment in the exercise of autonomous functions, especially motility, to near exclusion of interest in the mother. Optimal distance allows for free exploration at the same time that it permits refueling, a "perking up" for further exploration. Mahler concludes that during this early practicing subphase, all children have some separation anxiety and require the use of the distance modalities of hearing and seeing mother in order to enable them to move physically from her.

The practicing subphase proper, from 10 or 12 to 16 or 18 months, is characterized by Greenacre's (1957) concept of a "love affair with the world". Children attain a new visual level in upright locomotion, which offers them continuously new and interesting sites. This period constitutes the height of narcissism, when the child invests his own functions and his own body, as well as objects and objectives of his expanding reality. Impervious to falls and other frustrations, he is more easily able to substitute other adults for his mother to get reassurance.

Mahler emphasizes the importance to emotional development of the capacity to walk, which allows the child to test reality and his own control and mastery of the world. Individuation and identity formation are augmented. The mother both renounces and continues to possess the child's body. Delay in the child's locomotion results in concomitant delay in the experience of a love affair with the world. Walking thus has great symbolic meaning for both mother and child, indicating to the mother that the child will make it and adding to the child's developing self-esteem.

The child's mood can be described as low-key in mother's absence. Gestures and performance are slowed down, interest in surroundings diminished, and the child appears preoccupied with inwardly concentrated attention, with "imaging." When a person other than the mother actively attempts to comfort the child, his emotional balance is lost, and tears result. The low-keyed state ends with mother's return. This state, and the inferred attempt to image the mother, is seen by Mahler as a miniature anaclitic depression and an attempt to hold on to the ideal state of the self.

A disturbed practicing subphase may include greater than average separation anxiety, more than average shadowing of the mother or impulsive darting away from her, and excessive sleep disturbance.

The Third Subphase: Rapprochement (17-24 Months)

In the early rapprochement subphase, mother is no longer just home base, but also a person with whom the child wishes to share his discoveries, wanting her interest and participation. His elation with locomotion wanes, and he shifts to a wish for social interaction with his mother and with other children. For both boys and girls, the discovery of anatomical differences produces a sense of their own bodies. For girls, the penis symbolizes what other children have that they cannot get, and for both boys and girls a claim to gender identity ensues. The child shows a characteristic negativity toward mother, which results both from a sense of expanded autonomy and more intense recognition of the father.

The child reacts more strongly to mother's absence, not with low-keyedness, but with increased activity and restlessness. Both responses indicate sadness, but the activity is seen as a defense instituted against that painful emotion. The child can now cope better and can relate to substitute adults and engage in symbolic play to master the separation experience. The period of early rapprochement ends at about 18 months and appears to be a temporary consolidation and acceptance of separation; however, harbingers exist of impending crisis, including almost ubiquitous temper tantrums, vulnerability, rage, helplessness, recurrence of stranger reactions, and beginning loyalty conflicts between mother and others.

As awareness of separateness grows, there is an increased need for the object's love. This constitutes the basis for rapprochement. During this period there is a strong need for optimal emotional availability of the mother. It is the mother's love of the toddler and her acceptance of his ambivalence that enables him to cathect his self-representation with neutralized energy. The father is of special importance also during this period.

Shadowing and darting away are two behaviors characteristic of this subphase and indicate the wish for reunion and the fear of reengulfment. Concomitant with the acquisition of skills and perceptual cognitive capacities, there has been an increasingly clear differentiation of the intrapsychic representations of object and self; thus, the toddler must cope with the world of his own, as a small, helpless, separate individual, unable to command relief merely by feeling the need for it.

The rapprochement crisis proper occurs from about 18 to 22 months and is characterized by conflicts deriving from the desire to be separate, grand, and omnipotent, and yet have the mother fulfill wishes without having to recognize that help comes from the outside. The child's mood changes to one characterized by dissatisfaction, insatiability, temper tantrums, and, especially, ambivalence. The mother is used as an extension of the self to deny separateness. The mother, too, shows anxiety about separation, and where mother is dissatisfied with her child, or anxious about him, or aloof from him, the normal rapprochement patterns become exaggerated.

During this period the child shows a realization of his own limitations and relative helplessness. A new capacity for empathy and higher level identifications develops. The child is especially aware of and sensitive to his mother's whereabouts. Piaget's (1952) object permanence is already achieved, so the child can conceive of mother being elsewhere and found again, which is often reassuring; but may demonstrate difficulty in leave taking from their mothers, with clinging behaviors and depression. Splitting as a defense is now possible. Other people become the bad mother; the good mother is longed for, but exists in fantasy only, and the real mother becomes a source of dissatisfaction on her return, leading the child either to ignore the returning mother or to avoid her.

As the rapprochement crisis is resolved, the child finds an optimal distance from mother, exercising autonomy and sociability and avoiding the ambivalence that proximity to mother might bring. By this age, children reveal less phase-specific behavior and more individual differences. Boys and girls seem very different for the first time, with boys being more disengaged from mother and girls being more engaged with her, demanding closeness in an ambivalent fashion and blaming her for the lack of the penis. Boys are less overly concerned with sexual differences.

The developmental tasks for the child at the height of the separation-individuation struggle are enormous. Conflicts about oral, anal, and early genital drives all meet, and in addition there is a need to renounce symbiotic omnipotence. The belief in the mother's omnipotence, too, is shaken. Superego development begins with the intensified vulnerability to the threat of losing the object's love. Where development has been less than optimal, the ambivalence conflict in relation to mother that became discernible during the rapprochement subphase is unresolved, as revealed in rapidly alternating clinging and negativistic behaviors. These reflect an ambitendency not yet internalized. Excessive splitting may be revealed as well.

The Fourth Subphase: Consolidation of Individuality and Beginning of Emotional Object Constancy (24-26 Months and Beyond)

The tasks of this subphase are the achievement of definite, and in some ways lifelong, individuality, the attainment of a certain degree of object constancy, and structuralization of the ego and beginning development of the superego.

The establishment of object constancy depends on the gradual internalization of a constant, positively cathected image of the mother. It depends also on the cognitive attainment of object permanence, a stable sense of the entity (self-boundaries), and a primitive consolidation of gender identity. Necessary attainments include as well unification of good and bad objects into whole object representations, fusion of aggressive and libidinal drives, tempering of hatred for the object when aggression is intense, and not rejecting the love object for another when the former is no longer satisfying.

Separations from the mother during the fourth, open-ended subphase are influenced by the degree of ambivalence in the relationship. When mother's leaving stirs up expressed or unexpressed anger and longing, the child cannot easily maintain her image in his mind. During this period, verbal communication develops rapidly, play becomes more purposeful and constructive, and fantasy play, role play, and make believe are initiated. A sense of time develops, along with the capacity to tolerate delay of gratification. The child is resistant to demands of adults and often reveals an unrealistic wish for autonomy. The re-current mild or moderate negativism of this period is essential to the establishment of identity. Individuation proceeds.

During the earlier part of this subphase, the less predictably reliable or more intrusive the love object, a more does this object remain or become an unassimilated foreign body, a bad introject, and child shows an increased tendency to identify and confuse the self-image with this bad introject. In such cases, during the rapprochement, aggression sweeps away the good object and the good self-representation, as is indicated by an increase of temper tantrums and increase of coercion of mother to function as an external ego. Marked ambivalence may ensue, marring development toward object constancy and sound secondary narcissism.

The mother should be available to the child as a mental representation in the mother's absence, the first basis of this availability being the actual mother-child relationship. Threats to the establishment of object constancy and separate individual functioning stem from the pressure of drive maturation, confronting the child with new tasks (for example, toilet training) and new fears (for example, castration fears). Mahler demonstrates that these fears affect budding object constancy and self-development, particularly when there is developmental trauma. The achievement of both object constancy and individuality is easily challenged by struggles around toilet training and by awareness of anatomical differences, which awarenesses pose a threat to the girl's narcissism and the boy's body integrity.

At this point we will discuss Mahler's formulations from the perspective of Kohut's developmental framework and at the same time from the perspective that current infant research provides not only for Mahler but for Kohut as well. We will take in turn each phase and subphase identified by Mahler.

Beginning with a comparison with Mahler's autistic phase, then, it must be said that Kohut never did posit a phase in which there is no psychological connection whatsoever to the outside world or to important others; rather, he (Kohut, 1971) postulated reconstructively, early in his articulation of self psychology, a primitive self-selfobject bond wherein the nuclear self experiences the other as a part of itself in an egocentric fashion. This may be seen as a quasi-autistic phase in that, from the observer's objective point of view, there is a self and an other, but from the subjective view of the self, there is only an impersonal function in this surround that enhances the self.

By 1977, Kohut demurred entirely from such speculation on the self of the pre-symbolic infant, postulating only a "virtual self," that is, the self of the infant as envisioned in the mind of the parent. Other self psychologists speculated more freely about the emergent self in the early months of life. They see, in contrast to Mahler, the infant as active, engaged, and intent upon communication (Basch, 1983), with a healthy cohesiveness (Tolpin, 1980), and immersed within a self-selfobject units.

As to current infant research and Mahler's (1968) concept of autism, we need say no more than there has for decades been strong and increasing evidence that would sweep aside any notion of a normally endowed infant incapable of connection to the outside world. Numerous empirical studies demonstrate that the infant is an active, eager learner, prewired and well-equipped for communications and interactions with the environment from the very beginning. Stern (1985), who has synthesized much current infant observation, puts forward an "observed infant" based on this synthesis. That this observed infant is consistent with "Kohut's baby" (Tolpin, 1971, 1980), much more so than with " Mahler's baby" is no accident. Self psychology has endeavored to remain au courant with infant research and to change its theories in accordance with it. Mahler and classical analysis attempt the same but carry complex theoretical baggage that makes this task more difficult. For example, Mahler incorporates Freud's (1900) postulate of the extremely young infant living initially in solipsistic isolation and employing hallucinatory wish fulfillment to satisfy need. To remain consistent with this Freudian idea, Mahler creates such concepts as delusions of fusion to defend against unbearable separateness. Further, she focuses on Freud's (1911) postulate of fantasy as primary and reality as secondary in the infant. All of this contrasts with infant research findings, which postulate the infant as not only, as previously stated, connected with others from birth on, but also as reality based from the very beginning. Current infant researchers hold that the infant's capacity for fantasy and defense formation, attributed by Mahler to the two-month-old, only originates at the age of 18 months, when the infant has achieved symbolic thought. Mahler's deep fealty to Freud seems to have locked her into a model inconsistent with current developmental observations; Kohut and self psychology, on the other hand, having less fealty, have been free to drop such outdated postulates.

It is in Mahler's concept of the symbiotic phase that she and Kohut, at least initially, appear to have been most closely linked theoretically. Kohut (1971) writes of the archaic self-selfobject merger in the mirror transference, which is conceptualized to resonate with a stage in normal development, dating back to the first years of life, characterized by a grandiose self in relation to a mirroring selfobject. While that formulation may have appeared inconsistent with the concept of symbiosis, the congruence in theory was weakened somewhat when Kohut (1977) demurred by postulating only the virtual self up to 18 months. Moreover, once self psychology's definition of the selfobject concept was re-formulated so that archaic fusion was no longer viewed as an obligatory aspect of normal development, the two theories were more widely divergent.

Current infant research strongly supports the importance to normal development of the close connectedness between the child and primary caregiver from birth on, consistent with the emotional significance of both Mahler's symbiosis and Kohut's selfobject merger; however, infant researchers, and Stern (1985) in particular, criticized the view of the infant as perceptually unable to distinguish self from other. Rather, Stern considers the human being to be equipped at birth with predesigned emergent structures that prepare him to develop very early totally separate cognitive schemas of self and other. Lichtenberg (1983) adds that the theory of symbiosis is contested by the active role of the infant as behavioral initiator of interaction.

All of this implies that a truly symbiotic phase in normal development à la Mahler is improbable and requires concomitant modification in important aspects of her separation-individuation process. That process postulates a need to separate out of something, to hatch out of something, and the something is conceived of by Mahler as a dual unity. The achievement of connectedness to the other is postulated as an epigenetic given in normal development. In contrast, infant research stipulates a self perceived as separate from the other almost from the start, with the achievement of autonomy being but one of the central tasks of the infant, on a par with, but only on a par with, the equally important task of achieving a capacity for interdependence. The state of interdependence, then, is not seen as something automatically in the infant as a passive acquisition, but rather as a relationship the infant must actively endeavor to construct for himself or herself.

Again, self psychology, because of its comparative freedom to make theoretical change, has more or less adopted the view that infant observation brings to the field. It was for this reason, we believe, that the definition of the selfobject construct was re-thought to eliminate the now discredited (at least by most developmentalists) notion of a normal developmental phase characterized by a perception of self merged with other. Self-selfobject relationships are still seen as imperative throughout the life cycle, but no normal self-other fusion or confusion is postulated as an inevitable phase of normal development.

We will now turn to the separation-individuation phase proper. Rather than contrast Kohut and Mahler specifically in terms of the subphases of separation-individuation as against the unfolding of the bipolar self, with its three sectors and concomitant selfobjects, we will instead focus more globally on a contrast of their respective views of development during that entire time period. We do so because Kohut did not, for the most part, so particularize and date events in the child's developmental course; he was not, after all, an observer of children, as was Mahler, but a reconstructor of the childhood of his patients. We assume a basic familiarity with the overall theoretical constructs of Kohut on the part of the reader, as we did not with Mahler.

To begin with, as we have seen, Mahler points to a gradually increasing independence of the primary caregiver as the important goal for the normally developing child. That is, Mahler identifies as the task of the first three years of life intrapsychic separation from the mother and the attainment of autonomy. Her formulations thus require, as indicated earlier, a separation out of a symbiotic unity. Kohut, on the other hand, conceptualizes the caregiver from the point of view of the self as supplying primitive selfobject functions, which, as maturation proceeds, are maintained increasingly by the self through the development of self-structure via transmuting internalization. In addition, there is a decreasing requirement for physical proximity to the primary selfobject caregiver as self-structure is consolidated, and selfobject need becomes less preemptory. But the main function of this structure building is to permit a developmental progression from primitive to more mature selfobject need and need fulfillment. The requirement for the selfobject function remains.

Thus, while there is an apparent similarity in view between Kohut's concept of improved and consolidated self-structure via transmutting internalization and Mahler's concept of increasing autonomy from caregivers via ego development, there is a difference between them; and it was this difference that was of vital importance to Kohut. Kohut viewed the goal of autonomy from supporting caregivers, so significant to Mahler and to mainstream analysis as well, as carrying with it a hidden moralistic stance, which self psychology, with its emphasis on the need for selfobject experiences throughout life, assiduously avoids. The issue of autonomy from supporting objects, recognized by mainstream analysis as a goal of normal development, versus Kohut's assertion of normal lifelong dependence on selfobjects was for Kohut one indication of an overall difference in world view between his framework and that of classical analysis in general and Mahler in particular that made an integration between their developmental schemas impossible.

Infant research may help us resolve this difference between Mahler and Kohut. Kohut's position regarding the normality of lifelong selfobject needs seems to be confirmed by infant observation. Stern (1985) concludes, on the basis of his synthesis of this research, that from the very beginning, the individual's life is always social. In infancy, with development of the core self taking place from 2 to 6 months, most of what the infant does, feels, and perceives occurs in the subjectively perceived context of human relationships. And with the acquisition of acute memory at that time, henceforth, subjective experiences throughout life are largely social for all of us whether we are alone are not; the experience of self with a self-regulating other and a self-attuning other as a subjective reality is pervasive.

Stern’s stance implies that the normal person, from birth to death, is never completely autonomous, always reliant for self-sustenance either on the internal, psychological presence of an other or on the external presence of an other. The other is not seen as fused with the self; rather, it is perceived as a separate entity that has an all-important role to play in self experience.

It is important to note here the basis for Stern’s concluding from his reading of the newest infant observational research that the infant experiences the psychological presence of important others even when he or she is not in their company. Stern postulates that the infant has the capacity for an evocative like memory, making it possible for interactions with significant others to be laid down in memory, to be retrieved spontaneously when only aspects, or cues, of the remembered experience are present.

For example, the mother and infant play together with the particular rattle, to the infant's delighted excitement. The interaction, occurring many times over, is generalized and laid down in the infant's memory. Subsequently, when the infant, now alone, plays with same rattle, the presence of the mother, along with the excitement experienced with her, is evoked by the rattle, the infant subjectively experiencing the mother's active presence even in her absence. It is this capacity to evoke the mother's presence even in her absence that enables the child to progress without anxious disruption in the face of mother's not being there to sooth, comfort, stimulate, and amuse him at all times. Heretofore it was understood by cognitive psychologists (e.g. Piaget, 1952; Piaget and Inhilder, 1966) and accepted by mainstream analysts that such evocative memory was not possible until 18 months (the advent of object permanence). Positive affective attachment to the image of the mother as a reassurance in her absence, even in the face of subjective feelings of anger and hostility toward her, was not thought to be possible for the infant until 36 months, the advent of libidinal object constancy. Before that age, the infant is viewed by Mahler as unable to retain in his or her mind the picture of the ambivalently loved mother during her absence. And it is perhaps for this reason that the issue of the infant separation from the mother has such high priority in Mahler's theory. Thus for Mahler, once the toddler had achieved object constancy, near the end of the separation-individuation sequence the issue of separation has been for the most part resolved; with libidinal object constancy, the normally developing toddler can endure the mother's absence in relative comfort and without trauma. However, with the new discovery of early, evocativelike memory, the issue of separation ceases to be consigned to a particular time period and becomes, instead, an issue of potential importance throughout life cycle. In self psychology terms, relatively nontraumatic versus traumatic experiences with separation from the other are mainly dependent on how successful or unsuccessful that other has been in supplying selfobject functions, and this is true from early infancy throughout adulthood. Autonomy, too, becomes a lifelong issue, not relegated to any phase or age; moreover, it no longer serves as a criterion for maturity.

Stern (1985) makes another point in the Mahler-Kohut disagreement concerning autonomy as a goal of development. He states that the normal infant, with its capacity for representation of experiences with important others always available to him in time of need, has an inborn ability to adjust to the requirement for either independence or interdependence, and that how much of each is required for adequate adaptation depends on the culture into which he is born. Thus, autonomy from and interdependence with others are best understood as relative issues. A moralistic stance is uncalled for and it is just this avoidance of a moralistic position that will endure as Kohut's contribution to these issues.

We will now turn to specific concepts arising out of particular subphases in the separation-individuation process in order to contrast them first with Kohut's formulations and then with current data from infant observation. During the subphases of differentiation and practicing, Mahler focuses on the checking-back-to-mother pattern, calling it the most important normal pattern of cognitive and emotional development. Checking back is postulated to be a defense against separation anxiety. Kohut, obviously using Mahler's observations to support his own thinking, views checking back, instead, in support of his own concepts of the mirroring self object and the grandiose self, as the infant's way to achieve reassuring affirmation from the mother. The infant's developing capacity to become more and more physically distant from the caregiver is viewed by Kohut not as a diminution of anxiety because of increasing autonomy, as in Mahler's schema, but rather, as an indication of the progression from primitive to more mature selfobject ties. Hence, excessive stranger anxiety, seen by Mahler as a failure to acquire basic trust, becomes for Kohut (1984) an indication of self-fragility requiring a checking-back-to-mother for a mirroring reaffirmation of the selfobject bond. This is a subtle difference, with the emphasis in self psychology on the state of the self; whereas in separation-individuation theory, the emphasis is more on the trust in the new object relationship on the way to the achievement of object constancy and only secondarily on the integrity of the self. However, refueling and perking up—Mahler’s phrases for the practicing subphase child's ability to use the mother as a means for reconstituting when the child feels threatened by new exploration—are similar to Kohut's concept of mirroring, which restores the self. Self-cohesiveness and self-constancy do seem to be closely related concepts, though based on different theoretical formulations (Shane and Shane, 1980).

Stern (1985), in his observation of children in this time. As they move away from mother and check back for refueling or mirroring, emphasizes that the child's need is not always for self-regulation in these instances, as implied by Kohut, but is equally for intersubjective sharing and for verbal affirmation, more consistent with Mahler.

Mahler makes a valuable contribution, too, in terms of the rapprochement subphase, characterized by the child's ambivalent and uneasy return to mother after a period of confidence in turning toward the larger environment, as well as in her description of subphase resolution. Contained in this description are theoretical assumptions about an aggressive drive that create a nidus of divergence between her views and Kohut's theoretical frame of reference.

In writing about fusion of drive energies as prerequisite to resolution of the rapprochement crisis, Mahler seems faithful to the classical Freudian duel drive theory wherein an aggressive drive is postulated as co-equal to the libidinal drive, inborn, and not dependent on environmental frustration for activation. While Mahler herself did not emphasize this aspect of the aggressive drive, Kohut distinguished himself from Mahler in particular, and classical analysis in general, on this basis. Kohut pointed to the fact that the mainstream view of aggression as innately destructive does not fit with data he derived from the clinical situation. He asserted that there was no concurrence between his developmental views and those of Mahler, because, again, he distinguished himself from a worldview that saw aggression as an inevitably destructive impulsion, committing himself instead to a view of aggression as reactive to frustration.

As a matter of fact, infant observation supports Kohut in this view. Parens (1979) conducted a developmental study on infants and children in interaction with their mothers that demonstrated the likelihood that destructive aggression is not inborn but instead it is a reactive response deriving from unempathic interference with the child's assertiveness which assertiveness is seen by Parens as inborn. What is interesting here is that Parens is a classical analyst whose research was organized by Mahlerian thinking. In turn, his work influenced Mahler and her co-workers, creating ultimately a closer fit between Kohut's views and those of Mahler in this regard (Shane and Shane, 1983).

As a final comment regarding these two theories of child development, it is important to note that Mahler conceptualizes attainment of individuality, or self-constancy, as a lifelong task this is not quite the same as Kohut's contention that the self requires selfobject functions throughout life, but perhaps their views are not that far apart after all. If self-constancy continues to require support from the environment, then autonomy, despite its importance to Mahler's developmental progression, is never fully attained. Mahler would postulate that dependency on the environment diminishes; Kohut would postulate that the dependency itself does not diminish, but rather that the quality of the required selfobject tie and selfobject sustenance achieves greater maturity.

We must conclude that the greatest difference between Kohut and Mahler, then, lies in their underlying theoretical frameworks, not in the particulars of their observations on the developing individual. We have followed closely the creative elaboration of these two psychoanalytic innovators; each has influenced as well as antagonized the other, and both have enriched the psychoanalytic world.

References

Basch, M. (1983), Empathic understanding: A review of the concept and some theoretical considerations. J. Amer. Psychoanal. Assn., 31: 101-126.

Freud, S. (1900) The interpretation of dreams. Standard Edition, 4 & 5. London: Hogarth Press, 1953.

_____ (1905), Three essays on the theory of sexuality. Standard Edition, 7:125-243. London: Hogarth Press, 1953.

_____ (1911), Formulations on the two principles of mental functioning. Standard Edition, 12: 215. London: Hogarth Press, 1958. To

_____ (1915), Observations on transference love. Standard Edition, 12: 157-171. London: Hogarth Press, 1958.

_____ (1920), Beyond the pleasure principal. Standard Edition, 18:7-64. London: Hogarth Press, 1955.

_____ (1923), The ego and the id. Standard Edition, 19:3-66. London: Hogarth Press, 1961.

_____ (1926), Innovations, symptoms and anxiety. Standard Edition, 20: 77-175. London: Hogarth Press, 1959.

Greenacre, P. (1957). The childhood of the artist: Libidinal phase development and giftedness. The Psychoanalytic Study of the Child, 12: 27-72. New York: International Universities Press.

Harley, M. & Wiel, A. (1979), The Selected Papers of Margaret S. Mahler, Vol. 1. New York: Aronson.

Kohut, H. (1971), The Analysis of the Self. New York: International Universities Press.

_____ (1977), The Restoration of the Self. New York: International Universities Press.

_____ (1980), Reflection. In: Advances in Self Psychology, ed. A. Goldberg. New York: International Universities Press.

_____ (1984), How Does Analysis Cure? Ed. A. Goldberg & P. E. Stepansky. Chicago: University of Chicago Press.

Lichtenberg, J. D. (1983), Psychoanalysis and Infant Research. Hillsdale, N. J.: The Analytic Press.

Mahler, M. (1979), Selected Papers of Margaret S. Mahler. New York: Aronson.

_____ & Furer, and Am (1968), On Human Symbiosis and the Vicissitudes of Individuation. New York: International Universities Press.

_____ Pine F. & Bergman, A. (1975), The Psychological Birth of the Human Infant. New York: Basic Books.

Parens, H. (1979), Development of Aggression in Early Childhood. New York & London: Jason Aronson.

Piaget, J. (1952), The Origins of Intelligence and Children. New York: International Universities Press.

_____ & Inhelder, B. (1966), The Psychology of the Child. New York: Basic Books, 1969.

Shane, M. & Shane, E. (1980), Psychoanalytic developmental theories of the self: An integration. In: Advances in Self Psychology, Ed. A. Goldberg. New York: International Universities Press.

_____ & _____ (1983). The strands of aggression: A confluence of data. Psychoanalytic Inquiry, 4:263-281.

_____ & _____ (1985), Change and integration in psychoanalytic developmental theory. New Ideas in Psychoanalysis, ed. C. F. Settlege & R. Brockbank. Hillsdale, NJ: The Analytic Press, pp. 169-82.

_____ & _____ (1988), Pathways to integration: Adding to the self psychology model. Learning from Kohut: Progress in Self Psychology, Vol. 4, Ed. A. Goldberg. Hillsdale, NJ: The Analytic Press, pp. 71-77.

Spitz, R. (1965), The First Year of Life. New York: International Universities Press.

Stern, D. (1985), The Interpersonal World of the Infant. New York: Basic Books.

Tolpin, M. (1971), On the beginnings of the cohesive self. The Psychoanalytic Study of the Child, 26: 316-352. New York: Quadrangle.

_____ (1980), Discussion of psychoanalytic developmental theories of the self: An integration, by M. Shane and E. Shane. In: Advances in Self Psychology, ed. A. Goldberg. New York: International Universities Press.