Part Ⅰ
Separation-Individuation in Perspective
Chapter 1
Overview
The biological birth of the human infant and the psychological birth of the individual are not coincident in time. The former is a dramatic, observable, and well-circumscribed event; the latter a slowly unfolding intrapsychic process.
For the more or less normal adult, the experience of himself as both fully “in”, and fully separate from, the “world out there”, is taken for granted as a given of life. Consciousness of self and absorption without awareness of self are two polarities between which he moves with varying ease and with varying degrees of alternation or simultaneity. But this, too, is the result of a slowly unfolding process.
We refer to the psychological birth of the individual as the separation-individuation process: the establishment of a sense of separateness from, and relation to, a world of reality, particularly with regard to the experiences of one’s own body and to the principal representative of the world as the infant experiences it, the primary love object. Like any intrapsychic process, this one reverberates throughout the life cycle. It is never finished; it remains always active; new phases of the life cycle see new derivatives of the earliest processes still at work. But the principal psychological achievements of this process take place in the period from about the fourth or fifth month to the thirtieth or thirty-sixth month, a period we refer to as the separation-individuation phase.
The normal separation-individuation process, following upon a developmentally normal symbiotic period, involves the child’s achievement of separate functioning in the presence of, and with the emotional availability of the mother (Mahler, 1963); the child is continually confronted with minimal threats of object loss (which every step of the maturational process seems to entail). In contrast to situations of traumatic separation, however, this normal separation-individuation process takes place in the setting of a developmental readiness for, and pleasure in, independent functioning.
Separation and individuation are conceived of as two complementary developments: separation consists of the child’s emergence from a symbiotic fusion with the mother (Mahler, 1952), and individuation consists of those achievements marking the child’s assumption of his own individual characteristics. These are interwined, but not identical, developmental processes; they may proceed divergently, with a developmental lag or precocity in one or the other. Thus, premature locomotordevelopment, enabling a child to separate physically from the mother, may lead to premature awareness of separateness before internal regulatory mechanisms (cf. Schur, 1966), a component of individuation, provide the means to cope with this awareness. Contrariwise, an omnipresent infantilizing mother who interferes with the child’s innate striving for individuation, usually with the autonomous locomotor function of his ego, may retard the development of the child’s full awareness of self-other differentiation, despite the progressive or even precocious development of his cognitive, perceptual, and affective functions.
From the observable and inferred beginnings of the infant’s primitive cognitive-affective state, with unawareness of self-other differentiation, a major organization of intrapsychic and behavioral life develops around issues of separation and individuation, an organization that we recognize by terming the subsequent period the separation-individuation phase. In Part II we will describe the steps in this process (the subphases), beginning with the earliest signs of differentiation, proceeding through the period of the infant’s absorption in his own autonomous functioning to the near exclusion of mother, then through the all-important period of rapprochement in which the child, precisely because of his more clearly perceived state of separateness from mother, is prompted to redirect his main attention back to mother, and finally to a feeling of a primitive sense of self, of entity and individual identity, and to steps toward constancy of the libidinal object and of the self.
We wish to emphasize our focus on early childhood. We do not mean to imply, as is sometimes loosely done, that every new separation or step toward a revised or expanded feeling of self at any age is part of the separation-individuation process. That would seem to us to dilute the concept and erroneously to direct it away from that early intrapsychic achievement of a sense of separateness that we see as its core. An old, partially unresolved sense of self-identity and of body boundaries, or old conflicts over separation and separateness, can be reactivated (or can remain peripherally or even centrally active) at any and all stages of life; but it is the original infantile process, not the new eliciting events or situations, to which we shall address ourselves. In terms of its place in the larger body of psychoanalytic theory, we consider our work to bear especially on two main issues: adaptation and object relationship.
Adaptation
It was rather late in the developmental history of psychoanalysis that Hartmann (1939) began to bring a perspective on adaptation into psychoanalytic theory. Perhaps that is because, in the clinical psychoanalysis of adults, so much seems to stem from within the patient–from his longstanding character traits and dominating fantasies. But in work with infants and children, adaptation impresses itself forcibly on the observer. From the beginning the child molds and unfolds in the matrix of the mother-infant dual unit. Whatever adaptations the mother may make to the child, and whether she is sensitive and empathic or not, it is our strong conviction that the child’s fresh and pliable adaptive capacity, and his need for adaptation (in order to gain satisfaction), is far greater than that of the mother, whose personality, with all its patterns of character and defense, is firmly and often rigidly set (Mahler, 1963). The infant takes shape in harmony and counterpoint to the mother’s ways and style–whether she herself provides a healthy or a pathological object for such adaptation. Metapsychologically, the focus of the dynamic point of view–the conflict between impulse and defense–is far less important in the earliest months of life than it will come to be later on, when structuralization of the personality will render intra- and intersystemic conflicts of paramount importance. Tension, traumatic anxiety, biological hunger, ego apparatus, and homeostasis are near-biological concepts that are relevant in the earliest months and are the precursors, respectively, ofd anxiety with psychic content, signal anxiety, oral or other drives, ego functions, and internal regulatory mechanisms (defense and character traits). The adaptive point of view is most relevant in early infancy–the infant being born into the very crest of the adaptational demands upon him. Fortunately, these demands are met by the infant’s ability, in the pliability and unformedness of his personality, to be shaped by, and to shape himself to, his environment. The child’s facility for conforming to the shape of his environment is already present in early infancy.
Object Relationship
We feel that our contribution has a special place in the psychoanalytic study of the history of object relationship. Early psychoanalytic writings showed that the development of object relationship was dependent upon the drives (Freud, 1905; Abraham, 1921, 1924; Fenichel, 1945). Concepts such as narcissism (primary and secondary), ambivalence, sadomasochism, oral or anal character, and the oedipal triangle relate simultaneously to problems of drive and of object relationship (cf. also Mahler, 1960). Our contribution should be seen as supplemental to this in showing the growth of object relationship from narcissism in parallel with the early life history of the ego, set in the context of concurrent libidinal development. The cognitive-affective achievement of an awareness of separateness as a precondition of true object relationship, the role of the ego apparatuses (for example, motility, memory, perception) and of more complex ego functions (such as reality testing) in fostering such awareness are at the center of our work. We try to show how object relationship develops from infantile symbiotic or primary narcissism and alters parallel with the achievement of separation and individuation, and how, in turn, ego functioning and secondary narcissism grow in the matrix of the narcissistic and, later, the object relationship to mother.
In terms of its relationship to clinical psychopathological phenomena, we consider our work to bear on what Anna Freud (1965b) has called developmental disturbances, which the developmental flux of energy (E. Kris, 1955) may even out during later development, or which, in certain instances, may be precursors of infantile neurosis or middle range pathology. In rare cases, in which the subphase development was severely disordered or unsuccessful, we found, as did others such as FrijlingSchreuder (1969), Kernberg (1967), and G. and R. Blanck (1974), that borderline phenomena or borderline states, and even psychosis, may result.
This volume, in contradistinction to the volume on infantile psychosis (Mahier 1968b), deals predominantly with average development and makes contributions to the understanding of, at most, middle-range pathology.
In the study of infantile psychoses, both in the predominantly autistic (Kanner, 1949) and in the predominantly symbiotic syndromes (Mahler, 1952; cf. also Mahler, Furer, and Settlage, 1959), children were observed who seemed either unable to enter or ever to leave the delusional twilight state of a mother-infant symbiotic common orbit (Mahler and Furer,1960; cf. Mahler,1968b). These are children who may never show responsivity to, or the capacity of adapting to, stimuli emanating from the mothering person, that is to say, children who cannot utilize a “mothering principle” (Mahler and Furer,1966). Or, on the other hand, they may show panic at any perception of actual separateness. Even the exercise of autonomous functions (for example, motility or speech) may be renounced or distorted to preserve the delusion of the unconditionally omnipotent symbiotic unity (cf. Ferenczi,1913).
In either event, these children are deficient in the capacity to use the mother as a beacon of orientation in the world of reality (Mahler,1968b). The result is that the infant personality fails to organize itself around the relationship to the mother as an external love object. The ego apparatuses, which usually grow in the matrix of the”ordinary devoted”mothering relationship (see Winnicott,1962) fail to thrive; or, in Glover’s terms (1956), the ego nuclei do not integrate, but secondarily fall apart. The child with predominantly autistic defenses seems to treat the”mother in the flesh”(Bowlby, Robertson, and Rosenbluth,1952) as nonexistent; only if his autistic shell is threatened by penetration from human intrusion does he react with rage and/ or panic. On the other hand,the child with a predominantly symbiotic organization seems to treat the mother as if she were part of the self, that is, as not separate from the self but rather fused with it (Mahler,1968b). These latter children are unable to integrate an image of mother as a distinct and whole external object; instead, they maintain the split between the good and the bad part-objects and alternate between wanting to incorporate the good or expel the bad. In consequence of one or the other of these solutions,adaptation to the outside world (most specifically represented in a developing object relationship to mother [or father]) and individuation leading to the child’s unique personality do not unfold evenly from an early stage onward. Thus, essential human characteristics get blunted and distorted in their rudimentary stage or fall apart later on.
The study of the normal symbiotic period, and of normal separation and individuation, helps make the developmental failures of psychotic children more comprehensible.
Some Definitions
We have found, in discussions and presentations over the years, that three of our basic concepts are misunderstood often enough to warrant clarification. First, we use the term separation or separateness to refer to the intrapsychic achievement of a sense of separateness from mother and,through that, from the world at large. (This very sense of separateness is what the psychotic child is unable to achieve.) This sense of separateness gradually leads to clear intrapsychic representations of the self as distinguished from the representations of the object world (Jacobson,1964).Naturally, in the normal course of developmental events, real physical separations (routine or otherwise) from mother are important contributors to the child’s sense of being a separate person–but it is the sense of being a separate individual, and not the fact of being physically separated from someone, that we will be discussing.(Indeed, in certain aberrant conditions, the physical fact of separation can lead to ever more panic-stricken disavowal of the fact of separateness and to the delusion of symbiotic union.)
Second, we use the term symbiosis (Mahler and Furer, 1966), similarly, to refer to an intrapsychic rather than a behavioral condition; it is thus an inferred state. We do not refer, for example, to clinging behavior,but rather to a feature of primitive cognitive-affective life wherein the differentiation between self and mother has not taken place, or where regression to that self-object undifferentiated state (which characterized the symbiotic phase) has occurred. Indeed this does not necessarily require the physical presence of the mother, but it may be based on primitive images of oneness and/ or scotomatization or disavowal of contradictory perceptions (see also Mahler,1960).
Third, Mahler (1958a and b) has earlier referred to infantile autism and symbiotic psychosis as two extreme disturbances of identity. We use the term identity to refer to the earliest awareness of a sense of being, of entity–a feeling that includes in part, we believe, a cathexis of the body with libidinal energy. It is not a sense of who I am but that I am; as such, this is the earliest step in the process of the unfolding of individuality.
Symbiotic Psychosis and NormalSeparation-Individuation: A Review
Historically, the senior author’s observations of normal development and of the mother-infant dyad gradually led to the study of pathological phenomena, including child psychosis. Of course, the turn from problems of normal development was never complete. Although the immediate predecessor of the current work was the study of symbiotic psychosis of early childhood, we would, at this time, like to show the ways in which that study led naturally to our reconsideration of normal development.
On the Hypothesis of a Normal Separation-Individuation Phase
In our previous research on the natural history of symbiotic child psychosis(with Furer), we reached rock bottom when we tried to understand why those child patients were unable to develop beyond a (distorted) symbiotic phase, why they even had to reach back into bizarre life-maintaining mechanisms of a secondary autistic nature (Mahler and Furer,1960; Mahler,1968b). To understand this, we felt we had to know more about the steps that lead to normal individuation and, in particular, more about coenesthetic, preverbal, and early boundary-forming experiences which prevail in the first two years of life.
We began to ask various questions. What was”the ordinary way” of becoming a separate individual that these psychotic children could not achieve? What was the”hatching process” like in the normal infant? How could we understand in detail the ways in which the mother——as catalyst,activator, or organizer——contributed to these processes? How did the vast majority of infants manage to achieve the second, seemingly very gradual, psychic birth experience which, beginning during the symbiotic phase, gives way to the events of the separation-individuation process?And what, by contrast, were the genetic and structural features that prevented the prepsychotic child from achieving this second birth experience,this hatching from the symbiotic mother-infant”common boundary”?
By 1955(Mahler and Gosliner), we began to be able to articulate a conception of a normal separation-individuation phase.
Let us, for the sake of brevity, call [this] period……the separation-individuation phase of personality development. It is our contention that this separation-individuation phase is a crucial one in regard to the ego and the development of object relationships. It is also our contention that the characteristic fear of this period is separation anxiety.
This separation anxiety is not synonymous with the fear of annihilation through abandonment. It is an anxiety which is less abruptly overwhelming than the anxiety of the previous phase. It is, however, more complex, and later we hope to elaborate on this complexity. For we need to study the strong impetus which drives toward separation, coupled with the fear of separation, if we hope to understand the severe psychopathology of childhood which ever so often begins or reveals itself insidiously or acutely from the second part of the second year onward.
This separation-individuation phase is a kind of second birth experience which one of us described as”a hatching from the symbiotic mother-child common membrane.” This hatching is just as inevitable as is biological birth(Mahler and Gosliner,1955, p.196).
Furthermore:
For purposes of understanding our points, we propose focusing on the position of defense of the eighteen - to thirty-six-month infant, to defend his own evolving, enjoyable, and jealously guarded self-image from infringement by mother and other important figures. This is a clinically important and conspicuous phenomenon during the separation-individuation phase. As Anna Freud [1951b] has pointed out, at the age of two and three a quasinormal negativistic phase of the toddler can be observed. It is the accompanying behavioral reaction marking the process of disengagement from the mother-child symbiosis. The less satisfactory or the more parasitic the symbiotic phase has been, the more prominent and exaggerated will be this negativistic reaction. The fear of reengulfment threatens a recently and barely started individual differentiation which must be defended. Beyond the fifteen - to eighteen-month mark, the primary stage of unity and identity with mother ceases to be constructive for the evolution of an ego and an object world (Mahler and Gosliner,1955, p.200).
Today, we would date the onset of separation-individuation much earlier,and we can add considerably to those early formulations.
On the Hypothesis of Anxiety Attendant upon the Awareness of Separateness
It was hypothesized (Mahler,1952) that in certain toddlers the maturational spurt of locomotor and other autonomous ego functions takes place concomitantly with a lag in emotional readiness to function separately from the mother and produces organismic panic, the mental content of which is not readily discernible because the child (still in the preverbal stage) cannot communicate (cf. also Harrison,1971). This panic never consolidates into an appropriate signal anxiety, but retains the character of acute or insidious organismic distress, with the child’s concomitant inability to utilize the”other” as external organizer or auxiliary ego. This further arrests structuralization of the ego. The very fact that the more or less built-in maturation proceeds while psychological development does not² renders the rudimentary ego extremely brittle.Dedifferentiation and fragmentation may result, and the well-known clinical picture of infantile psychosis then ensues(Mahler,1960).
This view of intrapsychic events remains, of course, a hypothesis–especially in light of the preverbal nature of the phenomena which it is meant to explain. However, it seems to fit very well the observable clinical data–which are not hypothetical but descriptive–regarding loss of already achieved autonomous functions and a halt to subsequent development. This fragmentation may occur any time from the end of the first and in the course of the second year of life. It may follow a painful and unexpected trauma but often follows upon a seemingly minor event, such as a brief separation or a minor loss. Those observations ultimately led us to study the toned-down”panics” in the normal infant and toddler during separation-individuation and the way in which the mother and child, as a unit and as individuals, coped with them. Our increasing knowledge of the developmental tasks that confront the normal infant and later the normal toddler during the separation-individuation phase,and the trials and difficulties and momentary regressions seen in the behavior of these children, provide the basis for formulating our theoretical framework for understanding benign and transient disturbances,and neurotic ones, as well as the rare occurrences of more severe and lasting reactions shown by symbiotic psychotic children at an early or later age.
On the Hypothesis of the Development of a Sense of Identity
A third hypothesis (Mahler,1958a and b) states that normal separation-individuation is the first crucial prerequisite for the development and maintenance of the”sense of identity.” Concern with the problem of identity arose from observing a puzzling clinical phenomenon, namely,that the psychotic child never attains a feeling of wholeness, of individual entity, let alone “a sense of human identity.” Autistic and symbiotic infantile psychoses were seen as two extreme disturbances of the sense of”identity”(Mahler,1958a): it was clear that in those rare conditions something had gone basically astray at the very root, that is, in the very earliest interactions within the mother-infant unit. Briefly, one could summarize the central hypothesis as follows: whereas in primary autism there is a deanimated, frozen wall between the subject and the human object, in symbiotic psychosis there is fusion, melting, and lack of differentiation between the self and the nonself——a complete blurring of boundaries. This hypothesis ultimately led us to the study of the normal formation of separate entity and identity (cf. Mahler,1960).
On the Catalyzing Function of Normal Mothering
A fourth hypothesis grew from an impressive and characteristic observation that the symbiotic psychotic children were unable to use the mother as a real external object as a basis for developing a stable sense of separateness from, and relatedness to, the world of reality. Work with normal mother-child pairs developed our interest in the modalities of contact between mother and infant at differing stages of the separationindividuation process: in the modalities by which contact was maintained even while symbiosis waned; and in the specific role of the mother in facilitating not only the separateness of the child but also the specific patterning of his individuating personality by complementarity,contrast, identification, or disidentification (Greenson,1968).
Thus, the central ideas of the work with symbiotic psychotic children grew and transformed smoothly and with continuity into the organizing ideas of the work with normal mother-infant pairs. So too did our more formal efforts at research, as we shall now describe.
In the late 1950s at the Masters Children’s Center in New York City,Furer and Mahler had started a systematic study of”The Natural History of Symbiotic Child Psychosis.” It was a therapeutic action research in which we used the so-called tripartite design (child, mother, and therapist) first applied by Dr. Paula Elkisch (1953). We attempted to establish what the late Augusta Alpert (1959) would have called a corrective symbiotic relationship between mother and child, with the therapist acting as a bridge between them. Concurrently with the above project, the pilot phase of an observational study of normal mother-child pairs was begun.The latter was a bifocal observational study (that is, with focus on mother and child) of more or less randomly selected mother-infant pairs,in which the mother-child units were compared with one another and with themselves over time. These studies of symbiotic infantile psychosis and of normal mother-infant pairs ran parallel for about 4 years and have continued separately for another 7 years.
The studies of average mother-infant pairs continued on a larger scale and more systematically from 1963 on. The questions to which we Overvierooriginally addressed ourselves were oriented to two main hypotheses: (1)that there exists a normal and universal intrapsychic separation-individuation process which is preceded by a normal symbiotic phase; and (2)that in certain predisposed but extremely rare cases the maturational spurt of locomotion and other autonomous ego functions, when coupled with a concomitant lag in emotional readiness for functioning separately from mother, gives rise to organismic panic. It is this panic that causes ego fragmentation and thus results in the clinical picture of symbiotic infantile psychosis (Mahler,1960). We have since learned that there are innumerable degrees and forms of partial failure of the separation-individuation process.
The method of study of the normal separation-individuation process approximated the method used in the study of”The Natural History of Symbiotic Child Psychosis”(the tripartite design) and was characterized by the continual presence of the mother, by a physical setup specifically designed for, and uniquely suited to, observation of the infant’s readiness for active experimentation at separation and return, and by the opportunity to observe the infant’s reaction to passive separation experiences.
The work on the normal separation-individuation phase has had in turn considerable feedback to the earlier work on symbiotic child psychosis. Not only has our description of the subphases in the development of separation-individuation made it possible for us to anticipate and conceptualize some of the progressive changes seen in the symbiotic psychotic child during the course of his intensive therapy (cf. Bergman,1971; Furer,1971; Kupferman,1971), but our very formulations(given in part above) about the symbiotic psychotic child bear the mark of our later understanding of the separation-individuation process (Mahler and Furer,1972; Mahler,1969b,1971).
A Preliminary Note on Observation and Inference
The question of the kind of inferences that can be drawn from direct observation of the preverbal period is a most controversial one. The problem is complicated by the fact that not only is the infant preverbal,but that the verbal means of the observer-conceptualizer lend themselves only very poorly to the translation of such material. The problems of psychoanalytic reconstruction here find their parallel in the problem of psychoanalytic construction——the construction of a picture of the inner life of the preverbal child, a task in which coenesthetic empathy, we believe, plays a central role. Although we cannot ultimately prove the correctness of such constructions, we nonetheless believe that they can be useful, and we are committed to attempting their formulation.
Analysts have taken positions that vary along a broad spectrum regarding efforts to understand the preverbal period. At one extreme stand those who believe in innate complex oedipal fantasies—those who, like Melanie Klein and her followers, impute to earliest extrauterine human mental life a quasi-phylogenetic memory, an inborn symbolic process(Mahler,1969; Furer, quoted by Glenn,1966). At the other end of the spectrum stand those Freudian analysts who look with favor on stringent verbal and reconstructive evidence——organized on the basis of Freud’ s metapsychological constructs—yet who seem to accord preverbal material little right to serve as the basis for even the most cautious and tentative extension of our main body of hypotheses. They demand that these hypotheses also be supported by reconstruction——that is to say, by clinical and, of course, predominantly verbal material. We believe that there is a broad middle ground among analysts who, with caution, are ready to explore the contributions to theory that can come from inferences regarding the preverbal period (Mahler,1971).
Generally, in making inferences regarding the preverbal period from clinical psychoanalytic data, analytic theorists are asserting their right always to ask”Why?”“How did it come about?” and to answer by tracing earlier and earlier verbalizable memories, and ultimately to connect these memories to preverbal (but manifestly observable) phenomena of infancy that are isomorphic with the verbalizable clinical phenomena;for example, Freud’s (1900, p.271) comments on dreams of flying and the infant’s experience of being swooped up by adults(cf. also Anthony,1961). That is, we study phenomena of the preverbal period that appear(from the outside) to be the kinds of experience that match up with what patients are only later able to report during analysis, in their verbalizable recollections, that is, free associations, without at that point being aware of their origins.
As in clinical psychoanalysis, our method of working was from beginning to end characterized by”free-floating attention” in order to take in the usual and the expectable, but more particularly the unexpected, surprising, and unusual behaviors and transactional sequences. As the psychoanalytic instrument, especially the ear(see Isakower,1939), functions during analysis, so, in psychoanalytic infant observation, the psychoanalytic eye lets itself be led wherever the actual phenomenological sequences lead(cf. A. Freud,1951b).
But beyond these general modes of psychoanalytically derived observations, the observer of the child in the preverbal period has a special observational opportunity: the opportunity to observe the body in movement. To explain one of our major bases for making inferences from nonverbal behavior, let us briefly refer to the significance of the kinesthetic function and the function of motility in the growing child. As set forth in a number of papers in the 1940s (Mahler,1944; Mahler, Luke, and Daltroff,1945; Mahler and Gross,1945; Mahler,1949a), the observation of motor, kinesthetic, and gestural (affectomotor) phenomena of the entire body can have great value. It permits one to infer what is going on inside the child; that is to say, the motor phenomena are correlated with intrapsychic events. This is particularly true in the first years of life.
Why is this so? Because the motor and kinesthetic pathways are the principal expressive, defensive, and discharge pathways available to the infant (long before verbal communication takes their place). We can make inferences from them to inner states because they are the end products of inner states. One cannot be certain of the inner state, but, in the effort to infer it, multiple, repeated, and consensually validated observations and inferences offer some safeguard against total error. Furthermore, in the preverbal period, by definition, speech has not yet assumed the major expressive function it will later serve, thus leaving the task of communication predominantly to the mimetic, the motor, and the gesturai spheres. And finally, in the very young child, changes like modulation,inhibition, stylization, and defensive distortion of bodily expression have not yet been learned.
The young child’s rich and expressive affectomotor (gestural) behavior of his entire body, as well as the back-and-forth movement of approach and appeal behaviors and distancing behaviors between infant and mother—their frequency, amplitude, timing, and intensity—served as important guidelines, furnishing many clues to phenomena we encounter through verbal communication at later ages. We watched the infant’s expressive motility as it progressed beyond immediate discharge of instinctual drive,by way of the detour functions provided by the primitive ego’s abilities to delay, to learn, and to anticipate. We observed and assessed the infant’s autonomous and conflict-free motor functioning, with special regard to progressive steps in his separation-individuation process. In short, the observation of motor-gestural behaviors gave us important clues to intrapsychic events, and the substantive formulations to which we shall soon turn have been influenced by such observations (see Homburger, 1923; Mahler,1944; Mahler, Luke, and Daltroff,1945).
Instead of entering further into the general controversy regarding ob- service of preverbal infants and the legitimacy of inferences about the evolution of intrapsychic phenomena, we would like to present the history, methods, and tentative results of one such effort.
第一部分
分离-个体化视角下的探讨
第一章
概述
人类婴儿的生理诞生与个体的心理诞生在时间上并非同步。前者是一个显著、可观察且界限明确的事件；后者则是一个缓慢展开的内心精神过程。
对于或多或少处于正常状态的成年人而言，既能完全融入“外部世界”，又能与之全然分离的自我体验，被视作生活中理所当然的既定事实。自我意识与无意识的沉浸是两个极端状态，他在这两者之间转换，转换的难易程度不同，交替或同时出现的程度也各异。但这同样也是一个缓慢发展过程的结果。
我们将个体的心理诞生称为分离-个体化过程：即个体确立与现实世界的分离感以及与现实世界的关联，尤其涉及个体自身身体的体验，以及个体在婴儿期所感知到的世界的主要代表——首要爱的客体。与任何心理内部过程一样，这一过程会在整个生命周期中产生持续影响。它永无终结；始终处于活跃状态；在生命周期的新阶段，早期过程的新衍生形式仍在发挥作用。但这一过程的主要心理成就发生在大约第4或第5个月至第30或第36个月这一时期，我们将这一时期称为分离-个体化阶段。
在经历发展正常的共生期后，正常的分离-个体化过程要求儿童在母亲在场且母亲情感可及的情况下实现独立机能（马勒，1963 年）；儿童会不断面临客体丧失的轻微威胁（似乎成熟过程的每一步都会带来这种威胁）。然而，与创伤性分离的情况不同，这一正常的分离-个体化过程是在儿童已具备发展性准备且能从独立机能中获得愉悦感的背景下发生的。
分离与个体化被视为两个相辅相成的发展过程：分离指的是儿童从与母亲的共生融合状态中脱离出来（马勒，1952），而个体化则是指那些标志着儿童形成自身独特个性特征的发展成果。这是两个相互交织但并不完全相同的发展进程；它们的发展可能会出现差异，一方可能会出现发展滞后或早熟的情况。因此，过早的运动能力发展使儿童能够在身体上与母亲分离，但在个体化的一个组成部分——内部调节机制（参见舒尔，1966）尚未提供应对这种分离意识的方法之前，就可能导致儿童过早地意识到与母亲的分离。相反，一个无处不在、过度呵护的母亲若干扰了儿童天生对个体化的追求，通常是干扰了其自我的自主运动功能，那么尽管儿童的认知、感知和情感功能可能在逐步发展甚至早熟，但其对自我与他人差异的充分认知发展仍可能会受到阻碍。
从婴儿原始认知-情感状态可观察和推断的起始阶段，即尚未意识到自我与他人的区别开始，围绕分离与个体化问题，心理内部和行为生活的一个主要组织架构逐渐形成。我们将随后的这一时期称为分离-个体化阶段，以此来识别这一组织架构。在第二部分，我们将描述这一过程中的各个步骤（亚阶段），从最早的分化迹象开始，历经婴儿专注于自身自主功能运作、几乎忽略母亲的阶段，再到极为重要的和解期，在这一时期，正是由于孩子更清晰地感知到自己与母亲的分离状态，才促使他将主要注意力重新转向母亲，最后发展到产生一种原始的自我感、实体感和个体身份感，并朝着力比多客体和自我的恒常性迈进。
我们希望强调我们对幼儿期的关注。我们并非像有时人们随意表述的那样，暗示个体在任何年龄段经历的每一次新的分离，或是在自我认知方面的每一次修正或拓展，都属于分离-个体化过程的一部分。在我们看来，这样的观点会淡化这一概念，并错误地使其偏离我们所认为的该过程的核心，即幼儿早期在内心层面形成的分离感。个体早年未完全解决的自我认同和身体边界感，或是早年在分离和独立方面存在的冲突，可能会在生命的任何阶段被重新激活（或者一直处于边缘活跃甚至核心活跃状态）；但我们关注的是最初的婴儿期过程，而非引发这些现象的新事件或新情境。就其在精神分析理论整体框架中的地位而言，我们认为我们的研究尤其涉及两个主要问题：适应和客体关系。
适应
在精神分析发展史上相当晚的时候，哈特曼（1939 年）才开始将适应视角引入精神分析理论。这或许是因为，在成人临床精神分析中，诸多问题似乎都源于患者自身——源于其长期形成的性格特征和主导性幻想。但在针对婴幼儿的研究工作中，适应现象会强烈地映入观察者的眼帘。从一开始，儿童就在母婴二元关系的母体中塑造和成长。无论母亲会对孩子做出何种适应，也无论她是否敏感且善解人意，我们坚信，儿童新鲜且具有可塑性的适应能力，以及他对适应的需求（以获得满足感），远远超过母亲。母亲的个性，连同其所有的性格模式和防御机制，早已牢固且往往僵化地定型（马勒，1963 年）。婴儿会与母亲的行为方式和风格相协调或形成对比而逐渐成长——无论母亲本身为这种适应提供的是健康的还是病态的对象。从心理玄学角度看，动力学观点的核心——冲动与防御之间的冲突——在生命最初几个月远不如在后期那么重要。在后期，人格的结构化会使人格内部和系统间的冲突变得至关重要。紧张、创伤性焦虑、生理饥饿、自我机制和体内平衡是与生命最初几个月相关的近似生物学概念，它们分别是具有心理内容的焦虑、信号焦虑、口欲或其他驱力、自我功能以及内部调节机制（防御和性格特征）的前身。适应视角在婴儿早期最为适用——婴儿呱呱坠地便面临着巨大的适应需求。幸运的是，婴儿凭借其个性的可塑性和未定型性，有能力受环境塑造，并使自己适应环境。儿童在婴儿早期就已具备顺应周围环境的能力。
客体关系
我们认为，我们的研究成果在客体关系史的精神分析研究领域具有独特的价值。早期的精神分析文献表明，客体关系的发展依赖于本能驱力（弗洛伊德，1905 年；亚伯拉罕，1921 年、1924 年；费尼切尔，1945 年）。诸如（原发性和继发性）自恋、矛盾情感、施受虐、口欲或肛欲型特征以及俄狄浦斯三角等概念，同时与本能驱力问题和客体关系问题相关（另见马勒，1960 年）。我们的研究成果应被视为对这一领域的补充，它展示了在力比多同步发展的背景下，客体关系如何从自恋状态发展而来，且与自我的早期发展历程并行。认识到分离性是真正客体关系的先决条件，这一认知-情感层面的成就，以及自我机制（例如，运动能力、记忆、感知）和更复杂的自我功能（如现实检验）在培养这种认知方面所起的作用，是我们研究工作的核心内容。我们试图阐明客体关系如何从婴儿期的共生状态或原发性自恋状态发展而来，并如何随着分离和个体化的实现而发生改变；同时也试图说明自我功能和继发性自恋如何在自恋关系中发展，以及随后如何在与母亲的客体关系中进一步发展。
就其与临床精神病理现象的关系而言，我们认为我们的研究与安娜·弗洛伊德（1965b）所称的发育障碍相关。这种发育障碍可能会在后续发育过程中，因能量的发育性流动（E. 克里斯，1955）而得到缓解；在某些情况下，也可能是婴儿神经症或中度病理状态的先兆。在极少数情况下，若亚阶段发育严重紊乱或未获成功，我们与弗里林-施罗德（1969）、克恩伯格（1967）以及G. &R. 布兰克（1974）等人一样发现，可能会导致边缘现象或边缘状态，甚至引发精神病。
与关于婴儿期精神病的那卷著作（马勒，1968b）不同，本卷主要探讨正常发展情况，最多仅对理解中度病理状况有所助益。
在对幼儿精神病的研究中，无论是以自闭症为主的病症（坎纳，1949 年），还是以共生综合征为主的病症（马勒，1952 年；另见马勒、富勒和塞特拉奇，1959 年），研究人员观察到一些儿童，他们似乎既无法进入也无法脱离母婴共生共同轨道的妄想性朦胧状态（马勒和富勒，1960 年；另见马勒，1968b）。这些儿童可能永远不会对养育者发出的刺激做出反应，也不具备适应这些刺激的能力，也就是说，他们无法运用“养育原则”（马勒和富勒，1966 年）。或者，另一方面，他们可能会在察觉到任何实际分离时表现出恐慌。为了维持无条件全能共生统一的妄想，他们甚至可能放弃或扭曲自主功能的行使（例如，运动能力或语言能力）（另见费伦齐，1913 年）。
无论哪种情况，这些儿童都缺乏将母亲作为现实世界中定位指引的能力（马勒，1968b）。其结果是，婴儿的人格无法围绕与作为外部爱的客体的母亲之间的关系来组织自身。通常在“普通尽心”的母婴关系母体中发展的自我机能（见温尼科特，1962）无法良好发展；或者，用格洛弗（1956）的话来说，自我核心未能整合，继而瓦解。主要采用孤独症防御机制的儿童似乎将“真实的母亲”（鲍尔比、罗伯逊和罗森布鲁斯，1952）视为不存在；只有当他的孤独症外壳受到他人侵入的威胁时，才会表现出愤怒和/或恐慌。另一方面，主要处于共生组织状态的儿童似乎将母亲视为自身的一部分，即与自身并非分离而是融合在一起（马勒，1968b）。后一类儿童无法将母亲整合为一个清晰、完整的外部客体形象；相反，他们会维持好的和坏的部分客体之间的分裂状态，并在想要同化好客体和排斥坏客体之间摇摆不定。由于采取了上述两种解决方式中的某一种，儿童对外部世界的适应（最具体地表现为与母亲[或父亲]逐渐发展的客体关系）以及通向独特人格的个体化进程，从早期阶段起就无法平稳推进。因此，人类的基本特征在萌芽阶段就被削弱和扭曲，或者在日后瓦解。
对正常共生期以及正常分离与个体化过程的研究，有助于更深入地理解精神病患儿在发育过程中出现的障碍。
若干定义
多年来，我们在讨论和报告中发现，我们的三个基本概念常被误解，有必要加以澄清。首先，我们使用“分离”或“独立性”这一术语，指的是在心理层面上达成一种与母亲、进而与整个外部世界相分离的感觉。（精神错乱的儿童恰恰无法获得这种分离感。）这种分离感会逐渐形成清晰的、有别于客体世界表征的自我心理表征（雅各布森，1964 年）。当然，在正常的成长过程中，与母亲实际的身体分离（无论是日常性的还是其他情况）对孩子形成独立个体的感觉起着重要作用。但我们要讨论的是独立个体的感觉，而非与他人在身体上分离这一事实。（实际上，在某些异常情况下，身体分离这一事实可能会导致个体更加惊恐地否认分离的事实，并产生共生结合的妄想。）
其次，我们同样采用“共生”这一术语（马勒和富勒，1966 年），用以指代一种内心状态，而非行为状态；因此，这是一种推断出的状态。例如，我们所指并非黏人行为，而是原始认知-情感生活的一种特征，在这种特征下，自我与母亲尚未实现分化，或者出现了向那种自我-对象未分化状态（这是共生阶段的特征）的退行。实际上，这并不一定需要母亲实际在场，而可能基于关于一体性的原始意象，和/或对矛盾感知的盲点化或否认（另见马勒，1960 年）。
第三，马勒（1958a和1958b）曾较早地将婴儿孤独症和共生性精神病视为两种极端的同一性障碍。我们使用“同一性”这一术语来指称对存在之感、实体之感的最初觉知——我们认为，这种感觉在一定程度上包含了力比多能量对身体的投注。这并非是“我是谁”的感觉，而是“我存在”的感觉；就此而言，这是个体性发展过程中的最初阶段。
共生性精神病与正常分离-个体化：一项综述
从历史角度来看，资深作者对正常发育过程以及母婴二元关系的观察，逐渐引导其开展对包括儿童精神病在内的病理现象的研究。当然，研究方向从未完全脱离正常发育问题。尽管当前这项研究的直接前身是对幼儿共生性精神病的研究，但我们在此希望阐明，该研究是如何自然而然地促使我们重新审视正常发育问题的。
论正常分离-个体化阶段假说
在我们先前与富勒（Furer）共同开展的关于共生性儿童精神病自然史的研究中，当试图探究为何那些儿童患者无法超越（扭曲的）共生阶段继续发展，以及为何他们甚至不得不回归到具有继发性孤独症特征的怪异生存机制时，我们陷入了困境（马勒[Mahler]和富勒，1960 年；马勒，1968b）。为了理解这一现象，我们认为必须深入了解通向正常个体化的各个阶段，尤其要更多地了解在生命最初两年中普遍存在的共同感觉、前语言以及早期边界形成的体验。
我们开始提出各种各样的问题。这些精神病患儿无法达成的“常规方式”——即成为独立个体的方式——究竟是什么？正常婴儿的“孵化过程”是怎样的？我们如何才能详细了解母亲作为催化剂、激活者或组织者，是如何促成这些过程的？绝大多数婴儿是如何成功经历第二次心理诞生体验的？这一体验看似进展极为缓慢，始于共生阶段，随后逐渐过渡到分离-个体化进程中的各个事件。相比之下，是哪些遗传和结构特征阻碍了准精神病患儿经历这第二次诞生体验，即从母婴共生的“共同边界”中“孵化”出来呢？
到1955年（马勒和戈斯利纳提出此观点），我们开始能够清晰阐述正常的分离-个体化阶段的概念。
为简洁起见，我们不妨将这一时期……称为人格发展的分离-个体化阶段。我们认为，这一分离-个体化阶段对于自我以及客体关系的发展而言至关重要。我们还认为，这一时期特有的恐惧是分离焦虑。
这种分离焦虑并不等同于因被遗弃而产生的对自我消亡的恐惧。与前一阶段的焦虑相比，这种焦虑不会突然让人难以承受。然而，它更为复杂，我们希望后续能详细阐述这种复杂性。因为如果我们希望理解儿童严重的精神病理学问题（这类问题往往从孩子两岁半起就会以隐蔽或急性的方式开始显现），就需要研究推动分离的强大动力以及与之相伴的分离恐惧。
这一分离-个体化阶段是一种类似二次诞生的体验，我们中的一位将其描述为“从母婴共生的共同膜层中孵化而出”。如同生物意义上的诞生一样，这种“孵化”过程不可避免（马勒与戈斯利纳，1955年，第196页）。
此外：
为了便于理解我们的观点，我们建议聚焦于18至36个月大婴儿的防御状态，即保护其自身不断发展、令人愉悦且被悉心维护的自我形象，使其免受母亲及其他重要人物的侵犯。这是分离-个体化阶段一个在临床上重要且显著的现象。正如安娜·弗洛伊德（Anna Freud，1951b）所指出的，在两三岁时，可以观察到幼儿会经历一个近似正常的逆反阶段。这一阶段伴随的行为反应标志着其从母婴共生状态中脱离的过程。共生阶段越不尽如人意或越具寄生性，这种逆反反应就会越突出、越夸张。对再次被吞噬的恐惧威胁着刚刚开始且尚不稳定的个体分化，因此必须加以防御。超过15至18个月这个节点后，与母亲的统一和同一的初始阶段对于自我和客体世界的发展便不再具有建设性意义（马勒和戈斯利纳，1955年，第200页）。
如今，我们会将分离-个体化阶段的起始时间追溯到更早，并且能够对早期的相关理论阐述进行大幅补充。
论因意识到个体分离性而伴生焦虑的假说
有研究提出假设（马勒，1952 年），在某些幼儿中，运动能力及其他自主自我功能的成熟突增，与情感上准备好脱离母亲独立运作的发展滞后同时发生，这会引发机体性恐慌。由于孩子（仍处于前语言阶段）无法表达，这种恐慌的心理内容难以察觉（另见哈里森，1971 年）。这种恐慌不会演变成适当的信号性焦虑，而是持续呈现出急性或隐匿性的机体痛苦特征，同时孩子无法将“他人”作为外部组织者或辅助自我。这进一步阻碍了自我结构的形成。生理上或多或少的固有成熟进程持续推进，而心理发展却停滞不前，这一事实使得初始的自我极为脆弱。这可能导致去分化和碎片化，进而出现众所周知的婴儿精神病临床症状（马勒，1960 年）。
当然，这种关于内心心理活动的观点仍然只是一种假设，尤其是考虑到其所试图解释的现象具有前语言的本质特征。然而，它似乎与可观察到的临床数据高度契合。这些临床数据并非假设性的，而是描述性的，涉及已获得的自主功能丧失以及后续发展停滞的情况。这种功能碎片化现象可能在婴儿出生后第一年结束至第二年期间的任何时间出现。它可能由痛苦且意外的创伤引发，但通常也会因看似微不足道的事件而出现，比如短暂的分离或轻微的失落。这些观察结果最终促使我们去研究正常婴幼儿在分离-个体化阶段出现的程度较轻的“恐慌”情绪，以及母婴作为一个整体和个体应对这些情绪的方式。我们对正常婴幼儿在分离-个体化阶段所面临的发展任务的认识不断加深，同时也了解到这些儿童在行为中出现的考验、困难和短暂的倒退现象。这些认识为我们构建理论框架提供了基础，有助于我们理解良性和短暂性的心理障碍、神经症性障碍，以及共生性精神病患儿在不同年龄段出现的更为严重和持久的罕见反应。
论身份认同感发展的假设
第三个假设（马勒，1958年a和b）指出，正常的分离 - 个体化过程是形成和维持“身份认同感”的首个关键前提。对身份问题的关注源于对一种令人困惑的临床现象的观察，即精神病患儿永远无法获得完整感和个体存在感，更不用说“人类身份认同感”了。婴儿期的自闭症和共生性精神病被视为“身份认同感”的两种极端紊乱情况（马勒，1958年a）：显然，在这些罕见情况下，问题从根源上就出现了偏差，也就是说，在母婴单元最初的互动中就已出现问题。简而言之，核心假设可以总结如下：在原发性自闭症中，主体与他人之间存在一道死寂、凝固的屏障；而在共生性精神病中，自我与非我之间相互融合、界限模糊，缺乏区分——界限完全消失。这一假设最终促使我们开展对独立个体和身份正常形成过程的研究（参见马勒，1960年）。
论正常母性的催化作用
第四个假设源于一项令人印象深刻且极具特点的观察结果，即共生性精神病患儿无法将母亲视为真实的外部客体，而这一客体本应是他们形成与现实世界稳定的分离感和关联感的基础。通过对正常母婴组合的研究，我们对母婴在分离 - 个体化进程不同阶段的接触模式产生了兴趣，包括在共生关系逐渐减弱时维持接触的模式，以及母亲在促进孩子的分离感、并通过互补、对比、认同或反向认同等方式塑造其独特个性方面所发挥的特定作用（格林森，1968 年）。
因此，针对共生性精神病患儿开展工作的核心理念得以顺利且持续地发展演变，进而成为针对正常母婴伴侣开展工作的组织性理念。正如我们接下来将阐述的那样，我们在研究方面更为正式的努力亦是如此。
20世纪50年代末，在纽约市的马斯特斯儿童中心，富勒和马勒开展了一项关于“共生性儿童精神病自然史”的系统研究。这是一项治疗性行动研究，采用了宝拉·埃尔基施博士（1953年）首次应用的所谓三方设计（儿童、母亲和治疗师）。我们试图建立奥古斯塔·阿尔珀特（1959年）所说的母亲与儿童之间的矫正性共生关系，治疗师则在二者之间起到桥梁作用。与此同时，针对正常母婴对的观察性研究进入了试点阶段。后者是一项双焦点观察性研究（即聚焦于母亲和儿童），研究对象为大致随机选取的母婴对，研究过程中对母婴单元进行相互比较，并观察其随时间的变化情况。这些关于共生性婴儿精神病和正常母婴对的研究并行开展了约4年，之后又分别持续了7年。
自1963年起，针对普通母婴对的研究在更大规模上且更系统地持续开展。我们最初着手研究的问题围绕两个主要假设展开：（1）存在一种正常且普遍的心理内部分离 - 个体化过程，该过程之前有一个正常的共生阶段；（2）在某些有易患倾向但极为罕见的案例中，运动能力及其他自我自主功能的成熟突增，若与在情感上准备好与母亲分开独立运作方面的同步滞后相结合，会引发机体恐慌。正是这种恐慌导致自我分裂，进而呈现出共生性婴儿精神病的临床症状（马勒，1960）。此后我们了解到，分离 - 个体化过程存在无数程度和形式的部分失败情况。
对正常分离 - 个体化过程的研究方法与“共生性儿童精神病自然史”研究中采用的方法（三方设计）相近，其特点包括：母亲持续在场；设置专门用于且特别适合观察婴儿在分离和返回时进行主动探索的准备状态的物理环境；以及有机会观察婴儿对被动分离经历的反应。
关于正常分离 - 个体化阶段的研究反过来又为早期关于共生性儿童精神病的研究提供了相当多的反馈。我们对分离 - 个体化发展各亚阶段的描述，不仅使我们能够预测并从概念上理解共生性精神病患儿在强化治疗过程中出现的一些渐进性变化（参见伯格曼，1971年；富勒，1971年；库普费尔曼，1971年），而且我们（部分内容已在上文提及）针对共生性精神病患儿所提出的理论阐述，也体现了我们后期对分离 - 个体化过程的理解（马勒和富勒，1972年；马勒，1969b，1971年）。
关于观察与推理的初步说明
关于从对前语言阶段的直接观察中能够得出何种推论的问题，是极具争议性的。这一问题因以下事实而变得更为复杂：不仅婴儿处于前语言阶段，而且观察者兼概念构建者所使用的语言手段，极难用于解读此类观察材料。在此，精神分析重建的问题与精神分析建构的问题具有相似性——即构建前语言阶段儿童内心生活图景的任务，我们认为，共同感觉共情在这一任务中起着核心作用。尽管我们最终无法证明此类建构的正确性，但我们仍然相信它们是有用的，并且致力于尝试对其进行阐述。
在理解前语言阶段的研究工作方面，分析学家们的立场呈现出广泛的差异。在一个极端，是那些坚信存在先天复杂俄狄浦斯幻想的人，比如梅兰妮·克莱因及其追随者，他们认为人类子宫外的早期精神生活具有一种类似种系发生的记忆，即一种与生俱来的象征过程（马勒，1969 年；富勒，格伦 1966 年引述）。在另一个极端，则是那些弗洛伊德学派的分析学家，他们青睐基于弗洛伊德元心理学构建的严格言语和重建证据，但似乎认为前语言材料几乎没有资格作为哪怕是最谨慎、最试探性地拓展我们主要假设体系的依据。他们要求这些假设也能得到重建的支持，也就是说，要得到临床材料，当然主要是言语材料的支持。我们认为，在分析学家中存在着一个广泛的中间群体，他们谨慎地准备探索从前语言阶段的推断中能为理论发展做出的贡献（马勒，1971 年）。
一般而言，在依据临床精神分析数据对前语言期进行推断时，分析理论家们主张自己始终有权追问“为什么？”“这是如何发生的？”，并通过追溯早期可言语化的记忆来给出解答，最终将这些记忆与婴儿期前语言（但明显可观察到）的现象联系起来，这些现象与可言语化的临床现象具有同构性；例如，弗洛伊德（1900年，第271页）对飞行梦以及婴儿被成人抱起的体验的评论（另见安东尼，1961年）。也就是说，我们研究前语言期的现象，这些现象（从外部来看）似乎是与患者在分析过程中稍后才能以可言语化的回忆（即自由联想）报告出来的体验相匹配的类型，而此时患者并未意识到这些体验的根源。
如同在临床精神分析中一样，我们的工作方法自始至终都以“均匀悬浮式注意”为特征，旨在捕捉常见和可预期的情况，但更着重于捕捉意外、惊人且不同寻常的行为及互动序列。正如精神分析工具，尤其是耳朵（参见伊萨科夫，1939 年）在分析过程中发挥作用一样，在精神分析式婴儿观察中，精神分析的视角会任由实际现象学序列引导（参见 A·弗洛伊德，1951b）。
然而，除了这些源自精神分析的常规观察模式之外，对前语言期儿童进行观察的研究者拥有一个特殊的观察契机：观察儿童身体运动的契机。为了阐释我们从非语言行为进行推断的一个主要依据，让我们简要探讨一下运动觉功能和运动功能在成长中儿童身上的重要意义。正如20世纪40年代的多篇论文（马勒，1944年；马勒、卢克和达尔特罗夫，1945年；马勒和格罗斯，1945年；马勒，1949a）所阐述的那样，对整个身体的运动、运动觉和姿态（情感运动）现象进行观察具有重大价值。这种观察能够让人推断出儿童内心正在发生的状况；也就是说，运动现象与心理内部事件存在关联。这一点在儿童生命的最初几年尤为明显。
为何会如此？原因在于，运动和动觉通路是婴儿主要的表达、防御和释放途径（远在言语交流发挥作用之前便是如此）。我们能够依据这些通路推断婴儿的内心状态，因为它们是内心状态的最终外在表现。虽然我们无法确切知晓婴儿的内心状态，但在尝试推断时，通过多次、反复且经共识验证的观察与推断，能在一定程度上避免出现完全错误的判断。此外，从定义上来说，在婴儿的前语言阶段，言语尚未承担起其日后将发挥的主要表达功能，因此交流的任务主要落在了模仿、运动和手势等领域。最后，对于年幼的儿童而言，他们尚未学会对身体表达进行调节、抑制、风格化处理以及防御性扭曲等改变。
幼儿全身丰富且富有表现力的情感运动（手势）行为，以及婴儿与母亲之间趋近、吸引行为和疏离行为的往复互动——包括这些行为的频率、幅度、时机和强度——均为重要的参考指标，为我们在日后通过言语交流所遇到的现象提供了诸多线索。我们观察到，婴儿富有表现力的动作行为超越了本能冲动的即时释放，这得益于原始自我所具备的延迟、学习和预判能力所发挥的迂回调节作用。我们对婴儿自主且无冲突的运动机能进行了观察和评估，尤其关注其在分离 - 个体化进程中的阶段性进展。简而言之，对运动 - 手势行为的观察为我们了解心理内部活动提供了重要线索，而我们即将探讨的实质性理论阐述也受到了这些观察结果的影响（参见洪堡格，1923 年；马勒，1944 年；马勒、卢克和达尔托夫，1945 年）。
我们无意深入探讨有关对前语言期婴儿的观察以及对内心精神现象演变进行推断的合理性这一普遍争议，而是希望介绍此类研究工作的历史、方法以及初步成果。