Interpretation in Child Analysis

Interpretation in Child Analysis Developmental Considerations

Melvin Lewis) M.B.) B.S. (Lond.)) M.R.C. Psycli., D.C.H.

Historically the concept of interpretation in psychoanalysis consists essentially of making the unconscious conscious (Freud, 1907; 1913; 1914; 1926; 1937a; 1937b). It is generally understood that the aims of interpretation are threefold: to encourage the therapeutic alliance; to delineate the transference manifestations; and to effect dynamic (and therefore structural) change. To the extent that these aims are achieved, any communication may be considered an interpretation. Thus, the term "interpretation" may be used as a conceptual designation for a whole hierarchical series of communications. However, the term may also be used to connote a more or less circumscribed portion of that series, representing a partial fulfillment of the total aims. Thus, a common usage is to apply the term to a particular form and aim, namely, the comprehensive communication that provides the patient with the opportunity to experience the link between current behavior and early recall or reconstruction. For the sake of darity, therefore, a qualifying statement should be added when any particular subdivision of the term is meant. Finally, if one postulates a series of such subdivisions, to what extent does the developmental level of the patient modify the range and aims of such interpretations and the means by which they are achieved? To study this question, certain key aspects of development will be discussed within the framework of a proposed subdivision for psychotherapeutic interpretation and particular aspects of any given interpretation.

Dr. Lewis is Projessor of Clinical Pediatrics and Psychiatry, Child Study Center, Yale University, 333 Cedar Street, New Haven, Connecticut 06510, where reprints may be requested from him.

32

Interpretation in Child Analysis

33

PROPOSED SUBDIVISIONS OF INTERPRETATION

Setting Statements Interventions aimed at creating the necessary conditions, or setting, without which the normal analytic procedure would be impossible, have been described previously (Loewenstein, 1951; Bibring, 1954; Greenson, 1967). For example, in inducing the patient to follow the fundamental rule, we may make statements about the recumbent position or the rule of abstinence. Further, an "analytic atmosphere" is created by the analyst in his benevolent understanding or objectivity. Sometimes there is an "educational" effect upon the patient. Then, again, in addition to the general attitude of the analyst, there are certain explanations that are given as to procedure, questions, silences, etc. Lastly, there is the reality of the analytic situation. In a word, there is an understanding between patient and therapist. It is rare for analytic work with children to start in this way. More often, therapeutic work with a child begins with an initial introduction and exploration of the reasons for which the child is coming for treatment. A common practice, for example, is to tell the child that he can do almost anything he wants, but he cannot hurt himself, or the analyst, or break up the place. The child is also told what is not discussed with his parents, although his parents may be seen from time to time. Such introductory statements, it should be said, are fraught with difficulty. Inevitably, the time will come when the child wants to do something which the analyst cannot allow, and the child experiences a sense of disappointment, if not deep hurt, even betrayal. This reaction is especially likely to occur if the introduction focuses on limit-setting statements that are misunderstood by the child, or if the analyst draws attention only to the child's manifest aggressive behavior. A more appropriate preparation is one that focuses on the essence of the analytic work. For example, a child may be told that this time is set aside for him so that he can allow himself to think freely and begin to understand why he sometimes feels troubled; that sometimes the analyst will intervene, but with the understanding that it will always be in the interest of helping the child understand himself better. The recumbent position and rule of abstinence must also be modified for the child. This will be discussed later when the role of gratification in child analysis is considered. Similarly, the child's understanding of the therapeutic situation differs from that of the adult. The child usually does not come for treatment spontaneously, he has no un-

34

Melvin Lewis

derstanding of treatment, and he has fantasies about the therapist which are part of his current fears and wishes . He also establishes an appropriate real child-adult relationship (A. Freud, 1946).

A ttention Statements The aim here is in part to direct the patient's attention to the bare factual content of his actions or verbalizations. Sometimes attention is drawn to a coincidence which he has perceived, but has not, or professes not to have, registered; more frequently, attention is drawn to certain paradoxes. This technical device involves (primarily) secondary thought processes, and its ultimate goal is to lead to the production of some new material. Such statements are used whenever the analyst thinks they will help to consolidate existing gains and elicit new material. Devereux (1951), for example, describes how the analyst may reword a patient's own statements, especially in the form of "calling a spade a spade." In doing so, he may use statements, questions, or "mere inarticulate sounds." In work with children, particular use is made of such attention statements. In the course of the child's play the analyst may at times provide a verbal counterpart to the action being portrayed, or to an affect

that might be present, or , indeed, to the conspicuous absence of certain persons, actions, or affects. An 8-year-old boy with a severe school phobia repeatedly enacted a war scene in which the general was being attacked and almost killed. Many fantasies were contained in this play , but one prominent feature was the absence of any female, not onl y in this pla y item but in any other play. After attention was drawn to this " fact," the child recognized his fear of attack from his mother, his wish to attack her, his resentment that his father was often attacked and offered him no protection, the displacement of his aggression toward his mother to his father, and his anxiety about even mentioning his mother. This is quite different from any kind of direct translation of possible symbolic representation in the play. The play characteristic to which attention is drawn is in bold relief, and capable of being understood by a child. This is emphasized to draw the distinction from more subtle paradoxes which are not readily perceived, at least not by the younger patient. What is important here is that a certain level of cognitive development is required for the child to grasp the essence of such statements.

Interpretation in Child Analysis

35

Reductive Statements Certain statements reduce apparently disparate behavioral patterns to a common form which has hitherto not been noticed by the patient. Thus, a patient may manifest certain kinds of behavior whenever he is, say, angry. He may not have been aware of this anger. In the course of the analysis of a lO-year-old boy, it was noticed that there were recurring episodes of mocking, insulting, or denigrating behavior toward the analyst. Each of these episodes was related in time to one of the frequent trips away from home that the boy's mother would take. His resentment at being left behind, together with his anxiety at being left alone with his father, led to the behavior just described. This type of behavior could be reduced to a single behavioral reaction to underlying rage and anxiety precipitated by the temporary loss of his mother. When the child was told of this relationship, he reacted first of all by an intensification of the behavior pattern, but subsequently was able to recognize for the first time his underlying feelings when he stated, "My parents are always nice to me when we do fun things, but they don't help me with the serious things when I'm unhappy." The child here was manifesting a more or less fixed defense, which, in the example given, was brought into sharper relief. Again, a certain level of cognitive development is required. But more than this, some ego capacity to take distance from and observe affects must be present.

Situational Statements These statements naturally follow from those previously described. For example, the patient, now aware of his anger, can be shown the situations which give rise to his anger and how in certain instances he has repeatedly brought about such situations, either in his current object relationships or in the transference. However, the degree of directness with which such situational statements may be made varies with the cognitive and ego developmental levels of the child. Children who are at the preoperational stage of cognitive development and in the oedipal or preoedipal phase require intermediate steps which indeed may suffice. For example, situational statements at these stages are better perceived and tolerated when they are made in the context of the play, either through dolls and puppets, or indirectly through some other hypothetical child or children. Children who are at the stage of concrete operations and in latency can be approached directly.

36

Melvin Lewis

Transference Interpretation as the Essential Interpretation The problem of the so-called transference interpretation in work with the child is, of course, critical. The issue has two aspects: what can be considered transference in a developing child; and how transference interpretations are to be made to the child. Notwithstanding the difficulties of the introductory phase, a child soon understands one of the goals of analysis: the attempt to gain an understanding of the way he feels and behaves. If, after the initial phase of treatment, the child realizes that everything he says and does is subject to use by the analyst, it would follow that the play of the child is part of the associative process in the context of the analysis. However, in certain respects the developmental difference between the child and the adult influences the form of these associations. The child is much more susceptible to current reality, which exerts a powerful influence on his play. The play is often goal directed, and not freely associative. Again, the child has a tendency to act rather than think, giving at least the impression of an overemphasized aggressive transference (A. Freud, 1965, p. 36). Further, in latency especially, the play is often characterized by organization, reflecting a developmental shift

in many ego functions. Lastly, the child often becomes totally absorbed in his play and is unable to exercise an "observing ego" function. In these ways the associative process in the child is less "free" than it is in adults. However, in addition to play, a child also communicates through most of the other elements that constitute "free association." That is to say, the child talks, pauses, shows affects, exhibits mannerisms, and portrays attitudes. Further, he can participate in associating to parts of a dream. It is this total picture, viewed as a whole over the course of several hours, that enables the analyst to discern an associative thread. And it is this overall connection which can often be grasped by the child. In the course of these modified associations a true transference neurosis can be defined; i.e., the child's previous fixed conflicts and neurotic symptoms are now experienced currently in relation to the analyst (Harley, 1967). Again, the developmental differences between child and adult modify, but do not eradicate, this emerging transference neurosis in three ways: (1) To the extent that the child is normally still dependent, the parents with whom the original conflict was concerned are still with the child and continue to exert their influence on the child, making working through especially difficult. (2) The analysand is not an equal with the analyst: he is still a child relating to an adult

lnterpretation in Child Analysis

37

as well as a patient relating to a doctor. Besides the mutual respect that should exist between analysand and analyst, the child also has certain expectations of the adult which should be met. These might include, for example, birthday gifts and appropriate holiday greetings. (3) The continuing development of the child continues to modify the transference, especially during shifts from the oedipal period to latency and from latency to adolescence. In addition, fixations which occurred earlier may be modified with the increased range, flexibility, and shifts of defenses, for example, that occur with the development of the child. For all these reasons, the transference neurosis in child analysis is incomplete and more unstable than the transference neurosis in the analysis of an adult. Nevertheless, to the extent that there is a transference neurosis, however modified, it is available for interpretation. The more common situation in work with children is one in which a relatively simple current displacement from parent to analyst is recognized and interpreted. However, even within such apparently simple displacements, elements of a transference neurosis can be found, and when they are interpreted, they throw light on the child's current behavior for the analyst and for the patient. Etiological Statements Here a link between the complexities of the patient's current behavior and earlier developmental events and reactions is offered to the patient. Such earlier material is derived from recall or reconstruction. Some degree of maturation of certain developmental functions is required before such statements can be utilized by the patient. Specifically, some degree of autonomy of secondary process is a necessary condition. Yet, at the same time, some developmental achievements interfere with the young patient's ability to utilize such statements. For example, at the stage of normal consolidation of certain defense mechanisms-specifically, reaction formations that include guilt about aggressive impulses, shame about exhibitionistic urges, and disgust with the tendency to mess-the child may offer a strong resistance to the examination of a forbidden impulse. Again, increased resistance to etiological statements may occur if they are perceived as a stress and are made during a vulnerable period. The commonest reason for a period of vulnerability is when a child has but a tenuous hold on a function that has only recently been acquired or is at an interphase in his development. Any additional stress at such times may threaten the relatively undeveloped ego and stimulate a regressive movement which the child at first vociferously resists. This kind of periodic or interphase developmental fragility and

38

Melvin Lewis

vulnerability is most frequently seen in the preoedipal and oedipal child. Vulnerability may also occur, of course, when the child is experiencing other kinds of stresses or strains which threaten to overwhelm his ego. Consequently, reconstruction is often difficult when the patient is close to the reconstructed period. However, one common form of reconstruction almost invariably occurs-the reconstruction that the child makes toward the end of his analysis of the earliest phases of the analysis. In this regard the analysis of a child may be no different from that of the adult, who also tends to recall the opening hours of an analysis during termination. PARTICULAR ASPECTS OF INTERPRETATION

Having outlined this range of form and aim of interventions, I shall now focus more particularly on some of the technical problems of specific aspects of a given interpretation.

Source Material It is generally agreed (Isaacs, 1939) that the material for the work of psychoanalysis is derived from the patient's behavior, associations, dreams, affects, and attitudes. Any anamnesis in children is subject to their fantasies and the process of secondary revision. To be sure, there is often a plethora of information available from the parents, including reports from teachers and pediatrician. However, this information more often reflects the parents' view of events in the life of the child than the child's own experiences and subsequent organization of those experiences. Indeed, sometimes the ease with which certain kinds of information can be obtained from the parents deflects the analyst from the course of data that emerge in the analysis itself. The meaning of analytic data is, of course, subject to the perception of the analyst. Differences in perception between analysts depend upon such factors as the degree of education and experience of the analyst, and the extent to which countertransference phenomena are recognized and resolved. In work with children the analyst is also subject to a much greater "regressive pull" than he experiences with most adults (Bornstein, 1948), although a comparable regressive pull may be experienced in the analysis of adult borderline patients (Lewis, 1965). These modifications of the source of data lead to greater difficulty in making an interpretation with children, since the material may not be clearly defined. The nature of this difficulty will be explored shortly.

Interpretation in Child Analysis

39

General Stages as the Analyst Approaches Making an Interpretation For comparative purposes it might be useful first to outline the approach to making an interpretation to an adult. Loewenstein (1951) described the following steps: 1. Show the patient that certain common elements exist in a series of events. 2. Point out the similar behavior of the patient in each of these situations. 3. Demonstrate that such behavior was manifested in circumstances which all involved, for example, competitive elements and where rivalry might have been expected. 4. Point out that rivalry does exist unconsciously, but is replaced by another kind of behavior, such as avoiding competition. 5. Show that this behavior originates in certain critical events of the patient's life and encompasses reactions and tendencies which could be grouped under the heading of the oedipus complex. Perhaps this account would be more complete if greater emphasis were placed on the interpretation of unconscious, or at least preconscious, material. In all the stages just outlined, the interpretation of mechanism, as opposed to that of content, including affect, is significant. Further, it can also be seen that there is a gradual transition from preparatory intervention, through confrontation, to an interpretation containing a genetic component. Ferenczi (see Kris, 1951) has described the experiences of the analyst as he proceeds in the steps just outlined: One allows oneself to be influenced by the free associations of the patient; simultaneously one permits one's own imagination to play on these associations; intermittently one compares new connections that appear with previous products of the analysis without, for a moment, losing sight of, regard for, and criticism of one's own biases. Essentially, one might speak of an endless process of oscillation between empathy, self-observation, and judgment. This last, wholly spontaneously, declares itself intermittently as a signal that one naturally immediately evaluates for what it is; only on the basis of further evidence may one ultimately decide to make an interpretation [p.29J.

In child analysis, the description given by Erikson (1940) is remarkably similar. Speculations are first derived from the observer's impressions, associations, and recollections; e.g., "It was as if. . . ." The observer also associates past impressions in the same child, from other

40

Melvin Lewis

children, or from data derived from the parents. He reflects on the latent possibilities, that the associations may possibly correspond to a genetic or associative connection in the child's mind, picturing what he is doing under the observer's eyes, and what he is said to have done in other situations. This all leads up to the psychoanalytic interpretation. Erikson then describes three steps in making an interpretation. First, there are observations, feelings, and reflections that lead to interpretational hints. For example, a symbolic equation or metaphor may make it possible to recognize a play act as alluding to and standing for an otherwise manifestly avoided item (a person, an object, an idea). Or a play arrangement may prove to represent a special effort on the part of the child to rearrange in effigy his psychological position in an experienced or expected danger situation. Such an arrangement usually corresponds to the child's defense mechanisms. Second, these hints are then subject to further observations and reflections, and emerge as a conviction in the observer's mind in the form of the reconstruction of a genetic sequence or of a dynamic configuration pertaining to the patient's inner or outer history. Finally, the observer may proceed to convey part of these reconstructions to the child whenever he feels the time has come to do so. Erikson considers the last step to be the therapeutic interpretation.

The significant point here is the step-by-step progression in working with children implied in Erikson's statement that "The observer may proceed to convey parts of these reconstructions to the child whenever he feels the time has come to do so" (p. 589). However, it is important to keep in mind the developmental level of the child in this regard, since massive interpretations given to a young child in analysis are more likely to be heard as interfering noises than helpful statements, with a consequent heightening of resistance and play disruption or, what is worse, a play inhibition. Lastly, as a general rule, interpretations in child analysis as well as in adult analysis are probably more effective when they have "multiple appeal" (Hartmann, 1951). Further, in child analysis as well as in adult analysis the analyst decides clinically when to interpret the past, current reality, the transference-or all three.

Order of Interpretation A number of rules for the order of interpretation have been suggested. For example, Loewenstein (1951) states the rule of analyzing the resistances or the defenses before the instinctual derivatives. He also comments on the choice between the interpretations of the transference as opposed to that part of the material which is not included in the

Interpretation in Child Analysis

41

patient's relations to the analyst. A hint is given that one should avoid analyzing an important neurotic symptom at the beginning, and that one should start analyzing still mobile defense traits in preference to the rigid, characterologic defenses of neurotic characters. It is important to note that these so-called rules, especially in the case of children, must be regarded only as guidelines in preparing for a genetic interpretation. Take the case of an 8-year-old child who wanted to take along some tracing paper belonging to his analyst on the eve of his going on a car trip with his mother and father, the anticipation of which had already aroused considerable anxiety. What were the possibilities here for the analyst? 1. The analyst could simply have confronted the child with his wish to take something along. However, at the particular stage in the psychoanalysis of this child, this would have been redundant. 2. He could have shown the child how anxious he was about the trip with his parents, but doing so might have forced the child to face abruptly his anxiety without the analyst's support. 3. He might conceivably have tried to link this wish with his reactions in other similar situations in which he had become anxious, but this could not be done without the previous steps. 4. He might also have interpreted the patient's wishes toward the therapist for protection, but in this patient such a move might have left the patient feeling stripped and defenseless. 5. He might have made the connection for the child between this coming event and earlier events in his life, but this intervention might have had little use for this particular patient, since he did not yet have a clear idea of his feelings about the coming event. 6. It might then be asked whether anything should be said at all. Something should be said, but something that would temporarily buttress the child and offer support, at the same time that both the defense and the fear were being interpreted. For example, the analyst could say to the child: "How nice it will be to have something to take along, especially if you are worried about the trip." There would be no need to interpret the positive transference aspect at this point, since the movement is in a forward direction; or the analyst might also decide to include an aspect of the transference, but in a supportive way. He might say, for example, "I think you would like to have something of mine with you on this trip." He might then allow the child to keep the paper and reserve for a later date any further exploration of the act: "Why don't you take the paper with you? When you come back we can talk again about how you feel when you have to take a trip with your mother and father."

42

Melvin Lewis

These examples are not meant to be recommendations of specific things to say, but rather illustrations of underlying principles. The choice of level and wording depends on such factors as the diagnosis, the stage of the analysis, and the developmental level of the child. A 6-year-old child at the beginning of an analysis might experience his anxiety on the eve of such a trip as ego syntonic and fail to understand an interpretation of his anxiety. On the other hand, a l O-year-old who was more advanced in his analysis might well experience such anxiety as inappropriate and ego alien, and might find such an interpretation useful. Then again, in some analyses it may be important to allow some distance and ambiguity. It is possible that the analyst might wish to identify the paper as the analyst and respond to a younger child's wish in a humorous way by saying something like, "Well, I think the paper would like to go along with you." However, this would not be appropriate for an older child, or for a much younger child who would misunderstand, or for a child with a borderline diagnosis.

Context The relationship aspect between analyst and patient as an important part of the work of psychoanalysis has received special attention (Gitelson, 1962; Nacht, 1962; Segal, 1962). The attitude of the analyst is, in fact, an important part of the context in which an interpretation may be offered: "There is constantly present in the analytic situation an effective ground afforded by the analyst's professional interest, professional understanding, and nonmoralistic attitude" (Paul, 1963, p. 251). The actual relationship is a therapeutic agent in its own right, but it is in addition the soil upon which the seed of an interpretation may flourish. To the extent that the aim is to enable a patient to perceive the interpretation, then to that extent, having regard to the other conditions for psychoanalysis, a positive actual relationship is encouraged. This aspect is particularly important in the case of the child patient, and perhaps also for schizophrenic and borderline patients.

Timing Most statements about timing seem to end up in the general recommendation that interpretation should only be given at the right time. Some interpretations have tactical values, others aim at strategic objectives, but when it comes to timing, it seems again to amount to "not too soon, but not too late." Devereux (195 I) talks of an interpretation being well timed when-

Interpretation in Child Analysis

43

ever the analyst thinks that the statement at that moment will help to consolidate existing gains and elicit new material. Devereux describes an interpretation as being "timely" when it is capable of being utilized by the patient, and this in turn can occur only if the patient understands it. Freud (1937b) stated: "The time and manner in which he [the analyst] conveys his constructions to the person who is being analysed, as well as the explanations with which he accompanies them, constitute the link between the two portions of the work of analysis, between his own part and that of the patient" (p. 259). But again, the question of timing, and indeed of manner, is left indistinct, as perhaps it must be. What is significant here is that the concept of timing is always in association with a particular aim, which in turn is served by a particular part of the range of interventions. The question of timing is clarified to some extent when it is considered in relation to the subdivisions of interpretation and the sequence of interpretations previously outlined. However, in child analysis a further aspect must be considered. Put simply, the question is often asked: "When should an interpretation be made in the play, and when should it be taken out of the play?" Here again, the classification of interpretation into its various subdivisions may be of help. When attention statements are made, the interpretations are clearly made "in the play." Reductive statements imply that the child is receptive to statements about himself, i.e., "out of the play," while transference interpretations in their most effective form are made quite distinctly "out of the play," given the earlier steps of preparation. Dosage Loewenstein (1951) defines dosage as the extent to which optimally a patient should be gratified or frustrated through an interpretation. As can be seen again, this leaves much room for judgment by the analyst. More importantly, dosage is again a function of timing. But it is also a function of the diagnosis, the state of the transference, and the developmental stage. This last consideration is of particular interest here: a young child can probably tolerate only a very small dose of any interpretive statement, except when the act of interpretation is itself a gratification. Children in latency tend to resist particular interpretations that go against the developmental trend. Tact Loewenstein (1951) calls tact that intuitive evaluation of the patient's problems which leads the analyst to choose, among many possible in-

44

Melvin Lewis

terventions or interpretations, the one which is "right" at a given moment. A special form of tact is required when significant developmental differences are considered. A young child may have great difficulty in tolerating ambivalence, and may find it especially difficult to accept a hostile or aggressive wish or fantasy. This reluctance may be overcome by placing within a fuller context the anger, say, that is to be interpreted. For example, one might more tactfully say to a child: "It is very hard to be angry at someone you love." Glover (1930) tries to be more specific with regard to tact and states that the interpretation should be delivered as a plain statement in terms devoid of active emotional stress. The purpose appears to be to prevent an immediate, overwhelming conscious conviction on the patient's part that his analyst is in a state of countertransference. But what is tactful at one time may be undue reticence at another time. Hence, tact, beyond its social appropriateness, is connected with timing and dosage, and subject to the same considerations mentioned earlier.

Wording Glover (1930) cautions the analyst about the use of wit, the exploitation of the comic, and the shelter provided by technical expressions. He also states that the use of the patient's own terms (e.g., ma, pa, mama, papa) is usually preferable. The indication for such usage is said to be that the patient's adult organization is mostly a facade behind which the child makes faces at the analyst. To my mind this almost amounts to an invasion of privacy, is patronizing in tone, and suggests regression in the analyst. Even in child analysis, one usually refers to the child's parents as his "mother" or "father." Only with a very young child would one use such terms as "mommy" and "daddy," and then only because they are developmentally and age appropriate. Loewenstein (l 951) states that the wording should be specific and concrete; the interpretation should also be worded to fit the individual situation. Again, the analyst is cautioned to avoid the same defense mechanism as the patient, e.g., laughing things off and minimizing them. Interpretations appear to gain when they contain an element of time, for example, "now," "before," "at the age of," "after this happened." Wording becomes particularly critical in the case of the child, not only from the point of view of the level of cognitive development, but also from the point of view of what the child can accept in his depen-

dent position with respect to his parents, and his own struggle against

Interpretation in Child Analysis

45

regressive pulls or progressive pushes. The child attaches greater meaning to certain words than does the adult, and it becomes necessary for the analyst to understand these special meanings.

Reconstruction The striking aspect of a reconstruction, besides its creative element, is the genetic link. Why is this so important? Perhaps one reason is that it helps the patient by "making sense" out of what was previously perplexing. Further, it occasionally helps a patient by confirming what was probably an essentially correct perception by the patient at the time, but which has since undergone distortion. This kind of reconstruction is very important and occurs relatively frequently in work with children. Primary process is predominant in the thinking of the young child, and his conceptual abilities are such that he often attributes affects he thinks he perceives in his parents as resulting from thoughts or wishes of his own. Further, the child frequently projects his own fantasies and affects onto his parents, and subsequently acts against the parents whom he now regards as, say, dangerous or angry. Moreover, the preoperational cognitive level makes it almost impossible for the child to distinguish fantasies from reality (Piaget, 1929). Clearly, a certain level of postoedipal ego development and superego formation is required before the child can make maximum use of such reconstructions. That is to say, secondary process must be more available to the child, and the child should at least be in a stage of concrete operations (Inhelder and Piaget, 1958). Since secondary process becomes more prominent during latency, it follows that such reconstructions can be used most effectively from latency onward. The ego of the child can then be presented with important data that will enable the child to arrive at a better solution to his difficulties.

Completeness and Exactness Glover (1931) distinguishes between inexact and incomplete interpretations. An incomplete interpretation is termed by Glover a "preliminary interpretation." For example, one would interpret a genital fantasy before an anal fantasy. He contrasts this with an inexact interpretation when one might never interpret the anal fantasy at all. That is to say, if the interpretation of the genital fantasy is regarded as the complete interpretation, then the interpretation is ipso facto inexact. He goes on to say that an interpretation is never complete until the immediate defensive reactions following on the interpretation are

46

Melvin Lewis

subjected to investigation. The statements made by Loewenstein and Freud quoted earlier are relevant here in one respect-the complete interpretation is really the complete analysis. . . . every construction is an incomplete one. . . . As a rule he [the patient] will not give his assent until he has learnt the whole truthwhich often covers a very great deal of ground. . . . if the analysis is carried out correctly, we produce in him an assured conviction of the truth of the construction which achieves the same therapeutic result as a recaptured memory [Freud, 1937b, pp. 263, 265f.]. What seems to be implied is that no interpretation is complete, and no analysis is complete, until the full genetic sources are revealed. However, in the case of the child one rarely achieves such a perfect state of completion, and neither is it necessary. Often all that is necessary is to bring about sufficient conflict resolution so that development can once again proceed.

Working Through It is clear that working through is necessary to sustain any therapeutic effect (Greenacre, 1956). The defensive conflicts remain somewhat structured unless they are dealt with repetitively in relation to various behaviors, events, and feelings. In historical perspective, working through was first stressed from the point of view of being of educative value and compared with mourning and the progressive detachment of the individual libido from the organized tensions and aims which permeated the later life. The concept of the corrective emotional experience is really an aspect of working through, at least in its more modern construction. Originally, the idea involved replenishment of earlier deficiencies through the current relationship. This appealing idea unfortunately proved to be too simplistic. Among other things, it failed to take into account the power of the unconscious repetition compulsion and the need to resolve neurotic conflict. However, if the concept is modified to that of providing the patient in the here-and-now with a different reaction from his previous experiences, one that is now more appropriate and does not perpetuate the malignant interactions to which he has become accustomed, then the concept has more merit. But then it is also a different idea, and one that is more in keeping with the concept of working through. With the rise of ego psychology, the recognition of the need for consistent work with the patterns of defense and the affects related to them once more becomes paramount (Bornstein, 1949).

Interpretation in Child Analysis

47

From a developmental point of view, working through with children poses a special problem. The chief difficulty lies in the fact that the live parents are usually present and may, and usually do, continue to exert a reinforcing influence upon the child's original conflicts. Sometimes this interference can be alleviated by concomitant work with the parents, through analysis, psychotherapy, social casework, or regular meetings between the parents and the child's analyst. Occasionally it becomes clear that the child cannot work through a conflict while in the home, and an alternative plan may become necessary, e.g., boarding school in the case of an older child. Sometimes the difficulty is insuperable at the time, and the analysis must be interrupted until the child is in a more advantageous situation for analysis. Occasionally it is possible to hold the child in analysis until such a situation occurs. In some instances the child can be helped to understand the repetitive and reinforcing behavior of the parents and his involvement in precipitating, or responding to, their behavior. If the influence of the parents is not too strong, the child can be helped to modify his own behavior in this regard, and interrupt the vicious cycle.

Postinterpretative Efforts

Sterba (1944) developed the idea of postinterpretative states. The additional function of the analyst in helping the patient reorient and readapt himself is what Sterba calls the "formative activity," which he recognizes as belonging to working through, but which he thinks goes beyond the formal aspects of working through. He describes this formative activity in a general sense as a continued effort to exert pressure on the patient in order to force him from a state of unhealthy, pathological reactions and fixations into better adaptation and a more satisfactory solution of both his inner and his outer problems. For example, he will show the patient what he did and what he might better have done in certain situations. Or he will use the frustrating effect of interpretations, or use the patient's respect for the analyst's opinion to make the patient feel ashamed of his lapses into infantile behavior. Again, he will show the patient what dangers and frustrations his lapses will lead him into. In summary, such maneuvers seem to amount to persuasion, threat, promise of reward, encouragement, and praise, all reminiscent of techniques used by educators, although not perhaps by the most enlightened ones. Nevertheless, this formative activity is often useful, particularly in the case of the child for whom learning within the context of a good relationship is more readily available. Perhaps another way of viewing "formative activity" is in the context

48

Melvin Lewis

of the real relationship that exists between child and analyst in which the views of the analyst qua adult are important to the child.

The Patient's Reactions The patient's reactions (Devereux, 1951; Loewenstein, 1951) are both a part of the definition of an interpretation and a prelude to the next intervention. Further, the reactions are important data for purposes of validation. In the case of the child, aggressive and testing-out behavior is frequently encountered when anxiety is partially relieved, requiring a tolerant acceptance and further clarification (Schwidder, 1957). Common reactions include the child saying to the analyst, "You're stupid"; complaining that he has no time to play; blaming the analyst for minor tragedies that have befallen the child outside the analysis; or telling him that it's all none of his business. One child repeatedly said to me, "It's for you to find out and for me to know," as she became increasingly angry when the invasion of her privacy at home came into view in the analysis. It is as though the child felt a threat, not only to the control of her aggressive impulses, but to the continuity of the loving aspect of the relationship to the parent; hence the need, mentioned earlier, for tact in making an interpretation to the child. More severe reactions and disruptions occur in child analysis when the child is fixated at an earlier stage or position. Thus, an intense reaction may occur when the demands of the analyst are experienced as a direct threat and frustration to a child who is, say, still at the needfulfilling stage of object relations.

The Problem of Verification Freud (1937b), Isaacs (1939), Waelder (1939, 1962), and others have amply described the types of confirmation that may be observed when an interpretation is correct. Similar manifestations occur in children. However, in work with children a regular finding is the strong resistance encountered when a "correct" interpretation is made. This degree of resistance is of interest, since it appears to be related to an important developmental consideration. In the adult we ally ourselves with the healthy, observing ego, which in turn helps the analysand tolerate the attack upon his relatively stabilized personality structure. In the child the situation is different in several important respects. The observing ego is relatively undeveloped, and the personality structure is relatively unstable and in process of formation. The child reacts sharply and exaggeratedly with a loud signal when he feels he is under

an attack, which indeed is how an interpretation is frequently per-

Interpretation in Child Analysis

49

ceived. Further, particularly in the young child, an interpretation may initially be perceived as an invitation to act upon the impulse. Therapeutic Action and Theory Many attempts have been made to describe the therapeutic effects of interpretation in terms of the insight achieved. The particular type of interpretation which will produce a therapeutic insight, i.e., a structural change, has been given different names, as has the actual therapeutic insight. For example, "transference" (Fenichel, 1945) or "mutative" (Strachey, 1934) interpretation may produce "emotional," "psychological," "ostensive" (Richfield, 1954), or "dynamic" insights as opposed to "intellectual," "descriptive," or "neutral" (Reid and Finesinger, 1952) insights. The significant point is that shifts in cathexis from the unconscious to conscious awareness occur only after derivatives of the original feeling are recognized with an experiential sense of conviction and worked through. Bergler (1945) considers that the whole process of working through is centered chiefly in the correct handling and mobilization of the feeling of guilt. The decisive point in therapy, he believes, next only to the strengthening of the ego, is the transformation of the superego. The crucial factor is the disposition of the patient's feelings of guilt, the potential dynamic or static disposition of this feeling of guilt. Devereux (1951) feels that an interpretation (which consists of supplying an unconscious closure element) is effective when practically all conscious and preconscious material pertaining to the neurotic Gestalt has been produced. Loewenstein (1951) believes that the therapeutic effect of the analysis is due to a psychic process in which each of the following parts has its respective place: (1) the overcoming of resistances; (2) the working through; (3) the remembering and reliving of repressed material; and (4) the effect of the analytic reconstruction. They represent steps in the process of solving pathogenic conflicts. Speech, being between emotional expression and action, partaking of both, is an essential prerequisite of those dynamic changes which are produced by psychoanalytic treatment. Consequently, if speech as a function is not fully autonomous, as may be the case in a very young child, it is very difficult to achieve a therapeutic result through direct analysis. In considering the desirable changes just mentioned, i.e., the shifts in cathexis from the unconscious to the conscious, the strengthening of the ego, and the transformation of the superego, we may note several important developmental factors which may facilitate or handicap the

50

Melvin Lewis

accomplishment. To some extent, the developmentally appropriate reaction formations of childhood may act against change. However, the strength of these defenses, as well as the severity of the superego, also is modified in the course of development. Indeed, we are accustomed to thinking of an early latency and a late latency, a crucial distinction being the integration of the superego. Thus, the anticipated changes resulting from psychoanalysis are also to some extent a function of normal development. In this way, too, all of the factors that facilitate normal development will also facilitate the desired changes in psychoanalysis. What are some of these factors? In the case of a child, the partial gratifications appropriate to a child-adult relationship give rise to a feeling of being understood. Indeed, Lebovici (1957) specifically recommends partial gratification of the child's neurotic or realistic needs to establish a positive transference in the introductory phase. In some cases, analysis cannot proceed without such gratification even when it is clearly neurotic. Throughout the 3~ years of analysis of a 9-year-old boy, it was necessary from time to time to provide the boy with candy. Curiously, the boy would aggressively insist that the analyst share the candies with him! It was as though an early gratification was constantly being sought through a symbiotic merging in which both the patient and the analyst had to be simultaneously gratified. At the same time, if this gratification did not occur, the boy would become enraged and act in a dangerous way. For example, he would suddenly bolt out of the room and climb a tree just outside the office building. Once in the tree he would perch precariously and tauntingly in full view of the analyst. Further, the experience with the analyst not only is corrective in the sense mentioned earlier, but represents an oasis phenomenon for the child who now finds himself, at least temporarily, in a relatively protected, facilitating environment that enables development to proceed by reducing the impact of trauma and the acting-out tendencies in the child. Another way of putting this last point is to say that the child in analysis finds himself at least momentarily at a distance from the acute upset of the current developmental turmoil, whether it be a too exciting oedipal love affair, a tormenting anal-sadistic relationship with a parent, a grief reaction at the birth of a sibling, or frightening castration fantasies at the height of the phallic phase. The child is also afforded an opportunity for confirmation of his essentially correct perception of his parents, leading to a strengthening of reality testing and an increase in self-esteem. The play of the child, as well as the use of words, provides a handle with which he grasps affects and encourages delay, inasmuch as it represents an intermediate stage between action

Interpretation in Child Analysis

51

and thought. Most important, the play itself undergoes a development that carries with it a significant shift from primary process to secondary process. The dyadic therapeutic relationship provides an introspective opportunity for the child, and also serves to foster an identification with an appropriate adult model. Indeed, the child may "borrow" from the strength of the auxiliary ego of the analyst in his struggle to deal with an acute developmental crisis. Concomitant work with other family members leads to shifts in the dynamic equilibrium within the family, releasing the child and parents from their locked-in, fixed positions and allowing development to proceed. A further significant therapeutic force is the change from despair to hope, leading to a therapeutic optimism. Lastly, the extended moment of time, the opportunity to examine in detail, itself contributes to clearer, more direct, communications. In child analysis these aspects may be considered parameters inasmuch as they are not sufficient factors for the analysis, yet may be necessary factors for the analysis to proceed at all. In this respect parameters are more frequently used in child analysis. SUMMARY

Communications to the patient by the therapist take many forms. These forms range from silence to more or less comprehensive statements involving current behavior and genetic formulations. The interventions may be classified according to their type and aim. Six types of intervention are described: setting intervention; attention, reductive, and situational statements; transference interpretations; and etiological statements. In working toward the ultimate goal of structural change, these interventions have various intermediate aims, including: fostering the therapeutic alliance; overcoming resistance; clarifying the transference; facilitating remembering; convincing insight; and dynamic shifts. Within this framework, modifications in technique, based on the developmental level of the patient, are described. The true goal of psychoanalysis, namely, that previous pathological solutions be corrected through new perspectives, still obtains with a child, in spite of, or perhaps because of, the modifications described.

REFERENCES

BALINT, M. (1936), The final goal of psycho-analytic treatment. Int. ]. Psycho-A nal., 17:206-216. BERGLER, E. (1945), 'Working through' in psychoanalysis. Psychoanal. Reo., 32:449-480.

52

Melvin Lewis

- - - (1946), Use and misuse of analytic interpretations by the patient. Psychoanal. Rev., 33:416-441. . BIBRINC, E. (1954), Psychoanalysis and the dynamic psychotherapies. ]. Amer, Psychoanal. Assn., 2:745-770. BORNSTEIN, B. (1948), Emotional barriers in the understanding and treatment of young children. Amer. ]. Orthopsychiat., 18:691-697. - - - (1949), The analysis of a phobic child. The Psychoanalytic Study of the Child, 3/4:181-226. New York: International Universities Press. DEGROOT, A. D. (1948-49), The problem of interpretation in psychology. In: Proc. 10th Int. CongL Philos., pp. 934-936. DEVEREUX, G. (1951), Some criteria for the timing of confrontations and interpretations. Int.]. Psycho-Anal., 32:19-24. ERIKSON, E. H. (1940), Studies in the interpretation of play. Genet. Psychol. Monogr., 22:557-671. }'ENICHEL, O. (1945), The Psychoanalytic Theory of Neurosis. New York: Norton, p. 573. FREUD, A. (1946), The Psycho-Analytical Treatment of Children. New York: International Universities Press, 1950. - - - (1965), Normality and Pathology in Childhood. New York: International Universities Press. FREUD, S. (1907), Delusions and dreams in jensen's Gradiua. Standard Edition, 9:7-95. London: Hogarth Press, 1959. - - - (1913), On psycho-analysis. Standard Edition, 12:205-211. London: Hogarth Press, 1958. - - - (1914), Remembering, repeating and working-through. Standard Edition, 12:145156. London: Hogarth Press, 1958. - - - (1926), Inhibitions, symptoms and anxiety. Standard Edition, 20:77-175. London: Hogarth Press, 1959. - - - (1937a), Analysis terminable and interminable. Standard Edition, 23:216-253. London: Hogarth Press, 1964. - - - (1937b), Constructions in analysis. Standard Edition, 23:257-269. London: Hogarth Press, 1964. GITELSON, M. (1962), The curative factors in psycho-analysis. Int. ]. Psycho-Anal., 43:194205. GLOVER, E. (1930), The 'vehicle' of interpretations. Int. ]. Psycho-Anal., 11:340-344. - - - (1931), The therapeutic effect of inexact interpretation. Int. ]. Psycho-Anal., 12:397-411. GREENACRE, P. (1956), Re-evaluation of the process of working through .. Int. ]. PsychoAnal., 37:439-444. GREENSON, R. R. (1967), The Technique and Practice of Psychoanalysis. New York: International Universities Press. HARLEY, M. (1967), Transference developments in a five-year-old child. In: The Child Analyst at WOTk, ed. E. R. Gelcerd. New York: International Universities Press, pp. 115-141. HARTMANN, H. (1951), Technical implications of ego psychology. Psychoanal, Quart., 20:31-43. HEIMANN, P. (1956), Dynamics of transference interpretations. Int. ]. Psychol-Anal., 37:303-310. INHELDER, B. & PlACET, J. (1958), The Growth Of Logical Thinking from Childhood to Adolescence. New York: Basic Books. ISAACS, S. (1939), Criteria for interpretation. Int.]. Psycho-Anal., 20:148-160. KRIS, E. (1951), Ego psychology and interpretation in psychoanalytic therapy. Psychoanal. QuaTt., 20: 15-30. LEBOVICI, S. (1957), Der Platz der Psychoanalyse in der Psychotherapie des Kindes. Prax. Kinderpsychol. &- Kinderpsychiat., 6:38-41. LEWIS, M. (1965), Identification, language, and early object relations in a borderline patient. Unpublished thesis presented to the Western New England Institute for Psychoanalysis.

Interpretation in Child Analysis

53

LOEWENSTEIN, R. M. (1951), The problem of interpretation. Psychoanal. Quart., 20:1-14. NACIlT, S. (1962), The curative factors in psycho-analysis. Int. ]. Psycho-Anal., 43:206-211. PAUL, L. (1963), The logic of psychoanalytic interpretation. In: Psychoanalytic Clinical Interpretation, ed. L. Paul. New York: Free Press of Glencoe, pp. 249-272. PlACET, J. (1929), The Child's Conception ot the Wodel. New York: Harcourt Brace. REID, J. R. & FINESINGER, J. E. (1952), The role of insight in psychotherapy. Arne,'. ]. Psychiat., 108:726-734. RICHFIELD, J. (1954), An analysis of the concept of insight. Psychoanal. Quart., 23:390-408. SCHWIDDER, W. (1957), The meaning of psychoanalysis and of treatment methods derived from psychoanalysis for psychotherapy in childhood and adolescence. Prax. Kinderpsychol. <& Kinderpsychiat., 6:41-47. SECAL, H. (1962), The curative factors in psycho-analysis. Int. ]. Psycho-Anal., 43:212-217. SLUTSKY, A. (1932), Interpretation of a resistance. Psychoanal. Quart., 1:345-348. SPERLlNC, O. E. (1950), The interpretation of the trauma as a command. Psychoanal. Quart., 19:352-370. STERBA. R. (1944), The formative activity of the analyst. Int. ]. Psycho-Anal., 25:146-150. STRACHEY, J. (1934), The nature of the therapeutic action of psycho-analysis. Int. ]. Psycho-A nal., 15: 127-159. WAELDER, R. (1939), Kriterien der Deutung. Int. Z. Psychoanal., 24: 136-145. - - - (1962), Review of Psychoanalysis, Scientific Method, and Philosophy. ]. Amer. Psychoanal. Assn., 10:617-637.