Institute for Integrative Psychotherapy

Artículos de Psicoterapia Integrativa

Transference and Transactions: Critique from an Intrapsychic and Integrative Perspective1

Richard G. Erskine

In Eric Berne’s writings there are two different explanations of psychological functioning: the ego, composed of separate states, with intrapsychic dynamics among the states; and ego state terminology applied to descriptive behavioral roles. Subsequently, throughout the transactional analysis literature, two views of transference and transactions exist that, when applied clinically, are at variance with each other.

One purpose of this article is to draw a distinction between Berne’s two theories of ego states and to describe how each theoretical perspective creates a significantly different concept of transactions and transference. The practice of transactional analysis in psychotherapy is markedly different with each of these two theories.

A second purpose is to demonstrate that consistent use of Berne’s developmental, relational, and intrapsychic theory of ego states in understanding the internal dynamics of transactions can lead to a sensitive and effective response to transactions and transference and to a comprehensive and integrative psychotherapy.

Transference within Psychoanalysis

Freud’s (1905/1955) identification and specification of the transference dimension of the psychotherapeutic relationship is his most fundamental discovery (Langs, 1981). For the past 90 years psychotherapists have struggled with the problem of understanding patients’ communications and clarifying the difference between transactions that are solely in response to the current situation and those that are an expression of archaic relationship conflicts.

In the case of Anna O., Breuer and Freud (1895/1955) discovered the phenomenon of transference when they tried to uncover childhood traumas that were the roots of hysterical symptoms. They first considered transference as resistance to the uncovering of repressed childhood traumas. However, by 1905 Freud described the importance of working with the transference and considered transference and resistance (defenses) as the two main elements of psychoanalysis.

Freud (1905/1955) described transference using the metaphor of new editions or facsimiles of old emotional experiences. In transference patients replace the emotional experience with an earlier person with a similar experience with the psychotherapist. Within psychoanalysis this description of transference remains the basis for treatment. It was echoed by Greenson (1967), who described transference as the emotional experience of a person that does not befit that person and which actually applies to another. A person in the present is inappropriately reacted to as though he or she were a person in the past.

Freud’s hypothesis about the origin of transference was based on the assumption that each individual, through the combined operation of innate disposition and influences brought to bear during early years, acquired a somewhat fixed method or set of methods of living which were evident in all relationships. The patient in analytic treatment was seen as repeating these attitudes and reactions. Freud understood transference as the displacement of behavior and feelings onto the therapist, feelings that were originally experienced and directed toward significant figures from childhood (Freud, 1912/1958, 1915/1958). This early psychoanalytic concept of transference is the one most compatible with Berne’s (1961) original writings on ego states and their application to a theory of transactions and transference.

In the 1910s and 1920s Freud shifted his focus away from a theory of relationship conflicts of early childhood, as represented in his original ideas (1905/1955), to a theory that emphasized innate biological drives. Anna Freud (1965), working within this drive theory model of psychoanalysis, described the defensive, projective aspects of transference as the externalization of instinctual drives. She wrote that many of the transference situations encountered in her work were because the person of the analyst is used to represent one or another aspect of the patient’s personality. In this view, transference and projection are drive theory concepts that describe the defense against awareness of a specific biological drive.

For example, a patient may project a drive of aggression onto the therapist, thus subjectively attributing it to the therapist while experiencing the self as the object of aggression from the therapist. The patient then experiences the disowned and split off drives as being in the other person (Berg, 1977; Novik & Kelly, 1970). This drive theory concept of transference is not compatible with either Berne’s (1961) intrapsychic or descriptive theories of transactional analysis.

Berne’s (1961) descriptions of transference phenomena are more closely linked to those of psychoanalytic object relations theorists such as Bollas (1979), Fairbairn (1952), Guntrip (1971), Khan (1974), and Winnicott (1965). Spotnitz (1969) described the object relations theorists’ view of transference as “the patient’s attempt to reveal the basic maturational needs for objects that were not met in the course of his development” (p. 139).

Greenberg and Mitchell (1983) described in detail the bifurcation of current psychoanalytic theory between a relationship perspective and an instinctual drive perspective and the correspondingly differing views of transference. Anna Ornstein (1989) described transference as “current” resistance: “Transferences contain many elements of the past, but they are not only made of archaic reactions, they also contain a current reaction” to the therapist. When the transference is used to investigate the intersubjective field between patient and therapist, the behavior and unconscious intrapsychic processes of the therapist become an important source of information for use in understanding the patient. From this perspective, what looks like transference is at times a current reaction to the behavior and affect of the therapist (Stolorow, Brandchaft, & Atwood, 1987). Such insight into the meaning of the transference requires an empathic acceptance by therapists of their own childhood experiences and emotions (Brandchaft, 1989).

Kohut (1971) distinguished two types of transference: those based on instinctual drives and those representing early developmental needs such as approval, mirroring, and echoing. Kohut called the transactions that expressed fixated developmental needs “selfobject transferences” (p. 23) and ascribed to them a necessary reparation function within the therapeutic process. In Kohut’s (1977) self psychology the therapeutic goal of working within the transference is the completion of interrupted developmental processes. This is a very different goal than the classical psychoanalytic interpretation of transference as an expression of instinctual drives.

Other psychoanalytic writers have explored the therapeutic relationship, questioning what distinguishes transference from nontransference. Some argue that transference pervades the therapeutic relationship (Brenner, 1979; Friedman, 1969; Langs, 1976), while others argue that there are neutral or rational relationships in therapy (Greenson, 1967; Lipton, 1977).

Baker (1982) described the crucial variable in psychotherapy as “the transference, which involves components of both the real relationship between patient and therapist and the more irrational components displaced, projected and externalized from the patient’s history” (p. 196) of relationships with significant people and their internalized representations.

Greenson (1967) described two types of relationships in therapy that should not be equated with transference. Both the “working alliance” (p. 191) and the “real relationship” (p. 217) are nonarchaic and involve the patient’s reasonable ego. The working alliance is the patient’s cooperation in the therapeutic tasks and may be tinged with elements of archaic motivation (transference). There is, however, an observing ego that can stand back from the experience temporarily and reflect on it. The “real relationship is genuine and reality oriented or undistorted as contrasted to the term ‘transference’ which connotes unrealistic, distorted, and inappropriate” (p. 217). An example of the realistic relationship may be a patient’s concern for or criticism of the therapist. Lipton (1977) used the term “cordial relationship” (p. 255) to describe the nontransference transactions between patient and therapist. In his 1961 theory of transactions Berne implied the ideas of both a transference and nontransference relationship between therapist and patient.

For the past two decades psychoanalysis has been undergoing a major reevaluation regarding practice and theory. Berne (1961) predated much of the current theoretical reframing of psychoanalysis when he dispensed with a theory based primarily on innate biological drives and instead viewed human functioning as based on relationships. Berne (1961, 1966) continued to acknowledge primary innate human motivations such as stimulus hunger—with its sublimation into recognition hunger, and later structure hunger—but each of these were manifestations of the need for human relationship. Berne’s primary contribution to advancing knowledge of psychotherapy theory was his description of states of the ego and the use of these concepts to identify which transactions were transference and which were nontransference.

As reflected in Transactional Analysis in Psychotherapy (Berne, 1961), transactional analysis began as a reaction to and an advancement of psychoanalytic theory. Today there is much that transactional analysts can gain in theoretical perspective and clinical application by reexamining from an intrapsychic and integrative perspective both Berne’s original theoretical conceptualizations and the current theoretical and methodological debate within psychoanalysis.

Berne’s Original Concept of Ego

Berne’s (1961) original conceptualization of ego states appears to this writer to be a logical and creative extension of psychoanalytic structural theory. He expanded on Federn’s (1953/1977) concept of ego and elaborated the concept of the archaeopsychic and exteropsychic states of the ego. In so doing Berne paved the way for an explanation of intrapsychic conflict that is relational and developmental rather than relying on Freud’s drive model of intrapsychic instinctual-societal conflicts. Berne (1961) eliminated the theoretical concepts of id (pp. 61, 194, 198) and superego (p. 32) by postulating that these psychological dynamics are functions of an ego composed of three states of psychic organization: fixations from childhood, introjections of elements of the personality of others, and an integrating state in full contact with what is currently occurring internally and externally. He hypothesized that “an ego state is the phenomenological and behavioral manifestation of the activity of a certain psychic organ, or organizer” (p. 24).

Based on the references and footnotes found in Transactional Analysis in Psychotherapy (Berne, 1961), one would deduce that Berne was building theoretically on the writings of psychoanalytic authors Breuer and Freud (1895/1955), Fairbairn (1952), Federn (1953/1977), Freud (1949), Klein (1949), and Weiss (1950) and the child developmentalists Piaget (1932, 1951, 1954) and Erikson (1950). Berne (1961) thought of ego function as, in part, composed of archaeopsychic states: “the ego state of the actual child” which “has organization, unified will, logic and, certainly, negation” (p. 198). These archaic ego states consist of fixations of earlier developmental stages. They are the entire personality of a person as he or she was in a previous developmental period of time (pp. 54-55, 192, 1964, p. 23). The archaic ego fixations occurred when critical childhood needs for contact were not met, and the child’s use of defenses against the discomfort of the unmet needs became habitual (Erskine, 1980). These fixations became egotized or, in other words, formed separate ego units or states. The archaic or Child ego states (Berne, 1964, p. 23) are maintained in later life through the current use of defense mechanisms (Erskine & Moursund, 1988).

In Berne’s (1961) words, “The Child ego state is a set of feelings, attitudes, and behavior patterns which are relics of the individual’s own childhood” (p. 77). When functioning in the Child or archaic ego states the person perceives the internal needs and sensations and the external world as he did in a previous developmental age. Although the person may appear to be relating to current reality, he may actually be experiencing what is happening with the perceptual, emotional, intellectual, and social capacities of the child at the time of repression and fixation. It is this theoretical notion of the continuing fixation of Child ego states and the manifestation of a fixated Child ego that serves as one of the cornerstones for a transactional investigation of transferences.

Building on his own clinical observations, Berne extended Federn’s (1953/1977) and Weiss’s (1950) concept of the “psychic presence” (Berne, 1961, p. 19) of parental figures that influence an individual’s current behavior. He postulated the existence of exteropsychic ego states. The exteropsyche or Parent ego states are the manifestations of introjections of the personality of actual people as perceived by the child at the time of introjection (Loria, 1988).

Since the child’s perceptions of the caretaker’s reactions, emotions, and thought processes will differ at various stages of development, so also will the actual content and intrapsychic function of the Parent ego state vary in relation to the developmental age when the introjection occurred. (Erskine, 1988, p. 17)

Introjection is a defense mechanism (involving disavowal, denial, and repression) frequently used when there is a lack of full psychological contact between a child and the adults responsible for his or her psychological needs. The significant other is made part of the self (ego), and the conflict resulting from the lack of need fulfillment is internalized so the conflict can seemingly be managed more easily (Perls, 1978).

Introjected elements of another’s personality may become egotized and theoretically form an exteropsychic ego state. Berne’s theoretical premise of the existence of exteropsychic ego states is a second cornerstone in an intrapsychic and integrative understanding of transactions and transference.

Berne (1961) contrasted the exteropsychic and archaeopsychic ego states with a neopsychic ego state (Adult) that accounts for and integrates: (1) what is occurring moment-by-moment internally and externally, (2) past experiences and their resulting effects, and (3) the psychological influences and identifications with significant people in one’s life. This Adult ego state consists of current, age-related motor behavior; emotional, cognitive, and moral development; the ability to be creative; and the capacity for full contactful engagement in meaningful relationships. This neopsychic state of the ego functions without intrapsychic control by an introjected or archaic ego.

Berne’s original definitions of ego states provide the conceptual basis for an integrating psychotherapy (Clarkson & Gilbert, 1988; Erskine, 1977/1979, 1987, 1988; Erskine & Moursund, 1988; Loria, 1988; Massey, 1989; Moiso, 1985, 1988; Novellino, 1985; Trautmann & Erskine, 1981) that distinguishes nontransference transactions (neopsychic ego in origin) from possible transferential transactions. It is my understanding that transferential transactions are externalized expressions of internal ego conflicts between exteropsychic and archaeopsychic ego states.

Berne’s Illustrations and Descriptions

In each of his writings Berne (1961, 1964, 1966, 1972) augmented his precise theoretical definitions of ego states and intrapsychic function with illustrations and behavioral examples. Evolving from these explanations was a distinctly different theory of ego states which he called “descriptive” (Berne, 1972, p. 13). Although his original definitions of ego states emerged from both clinical experience and an extrapolation of the ideas of psychoanalytic authors, his descriptions of ego states relied not on his theory of “states of mind,” but on metaphors that tended to emphasize “their related patterns of behavior” (Berne, 1961, p. 30).

In providing illustrations of ego state theory in clinical practice, Berne shifted from a relational and developmental theory to a descriptive and behavioral understanding of ego states. He equated ego states with roles or specific behavior typical of those roles. For example, Berne (1961) used the phrasing “a Parental response” (p. 44) and “the parental role of comforting” (p. 95) to imply that the person was transacting from his or her Parent (exteropsychic) ego state. Another time the behavior “rational” (p. 132) was equated with the Adult ego state (p. 132). There are many other examples of ego states descriptively identified (pp. 128-135, 1964, p. 30, 1972, p. 14).

By shifting to a more descriptive and behavioral orientation, it seems that Berne greatly diminished his own creative extension of psychoanalytic theory. He lessened the impact of what his relational understanding of intrapsychic conflicts—as they are manifested in transferential transactions—had to offer. Berne (1972) changed perspectives and created an alternative set of theoretical analogies of ego states as roles and transactions as numerical probabilities of the roles (p. 19).

In articulating his theory Berne (1961) specified: “Ego states must be differentiated from ‘roles’ ” (p. 233); and “Ego states are not roles but phenomena. Therefore ego states and roles have to be distinguished in a formal description” (Berne, 1964, pp. 53-54). Yet throughout his writings he both defined the theory of ego states from a developmental, intrapsychic perspective and also provided illustrations and descriptions of behavioral roles. At one point he acknowledged this theoretical inconsistency: “For the most part, the examples given have concerned the behavioral and social aspects of the Child” (Berne, 1961, p. 235).

Berne’s writings contain several such theoretical inconsistencies as a result of his use of illustrative descriptions as definitions. Moiso (Erskine, Clarkson, Goulding, Groder, & Moiso, 1988) emphasized that Berne was not theoretically consistent: “It isn’t clear when he is giving a definition, a description, or a metaphor” (p. 7). In referring to Berne’s likening the Adult to a computer, Moiso added, “That’s a metaphor for the functioning of the neopsyche” (p. 7).

Three examples of the use of metaphors as definitions follow:

1. Berne (1961) described the Adult as working “deliberately and consciously” (p. 69) as if these two attributes were not possible in a Parent or Child ego state.

2. “The Parent has two main functions. First, it enables the individual to act effectively as the parent of actual children” (Berne, 1964, p. 27). With this description of the function of Parent ego state Berne disregarded that it is the Adult ego state that is in contact now with those in the environment. “Automatic” parenting (Berne, 1961, p. 76), although conserving much time and energy, is not in contact with the child. Instead, it may often be an activation of an introjection related to some other person in another time and place. Effective parenting requires full contact in the present between parent and child.

3. “In the Child reside intuition, creativity and spontaneous drive and enjoyment” (Berne, 1964, p. 27). While this is true of many children, it is not a definition of the Child ego state. Two pages earlier Berne defined Child as “an archaic ego state” (p. 25), and elsewhere he said it was “a warped ego state which has become fixated” (Berne, 1961, p. 54). In cases where the child has been neglected and/or traumatized, the Child ego state of the adult may not be spontaneous or intuitive or joyous. The fixation of the archaic child may be depressed, inhibited, or defended. These symptoms are likely to emerge later in life in transactions with others and in the course of psychotherapy.

Many of Berne’s descriptions sound as if he were reifying his theoretical ideas. His analogies have become specific entities. In his original developmental theory Berne (1961) used “adapted Child” and “natural Child” (p. 77) as adjectives to describe (1) the function of an archaeopsychic ego state under the oppressive internal influence of a Parent ego state and (2) the natural responsiveness of a child in the absence of such critical or controlling parenting. Berne’s descriptive wanderings from his original theoretical definitions (1961) take their final form in his last writings. The adjectives used previously to describe intrapsychic functioning became the nouns, “Adapted Child” and “Natural Child” (Berne, 1972, p. 104).

A review of the Transactional Analysis Journal reveals that most authors have described ego states in behavioral or descriptive terms (Nurturing Parent, Critical Parent, Adapted Child, Natural Child, and Rebellious Child) or as a categorization of psychological processes (equating Parent ego state with values, Adult ego state with thinking, and Child ego state with feelings), or as a mix of these concepts.

When Berne shifted his illustrations of ego states to the descriptive, he ignored his own original definitions and the necessary four-part diagnosis (behavioral, social, historical, and phenomenological) that is required for complete identification of the state cathected (Berne, 1961, pp. 75-76, 225).

When psychotherapists and authors of articles about psychotherapy do not take into account the specific validating procedures that Berne (1961) outlined in his chapter on diagnosis (pp. 68-80), then the validity of Berne’s original definitions and theory is not maintained. Transactional analysis theory loses both its internal and external consistency. Without such a four-part correlation and a consistent use of Berne’s original developmental and relationship-based theoretical definition, ego state theory becomes merely a taxonomy of behaviors. A descriptive taxonomy of behaviors is very useful in a social control or behavioral therapy, but the elegance of Berne’s logical and creative extension of psychoanalytic theory is lost. Transactional analysis becomes less a developmental, phenomenological, intrapsychic psychotherapy and more a therapy of behavioral adjustment. As a result, the meaning and purpose of analyzing transactions and the resolving of transference is uniquely different from the point of view of each of Berne’s two theories of human functioning. Each theory and practical approach has a valid place in psychotherapy. And Berne’s role/communication theory has application in nonpsychotherapy fields. An understanding and appreciation of Berne’s early developmental and intrapsychic theory, however, allows for greater theoretical consistency and a more in-depth psychotherapy.

Loria (1988) highlighted these theoretical contradictions and the difficulties inherent in mixing concepts and in deviating from stated theoretical definitions without a supporting explanation of the new theoretical definitions. At the end of his writings Berne predicted the theoretical and methodological confusion inherent in mixing concepts. He recommended the use of “the Conceptual Grid” (Berne, 1972, pp. 409-413) so that theoretical discussions and treatment planning could remain within a given set of concepts and definitions. Berne concluded:

If one takes a structural or biological approach to the Child ego state and another takes a functional and descriptive approach, it is impossible to reconcile the two. . . . One uses structural nouns, the other uses functional adjectives as modifiers, and the nouns and adjectives do not belong to the same framework or come from the same viewpoint. (pp. 411-412)

Ego State Determinants

In 1964 Berne stated that “transactional analysis is concerned with diagnosing which ego state implemented the transactional stimulus, and which executed the transactional response” (p. 29). In order to determine if a particular transaction is transferential or nontransferential, it is necessary to conduct a “careful and systematic analysis of the psychodynamics of . . . transactional stimuli and responses” (Berne, 1966, p. 154), of ego state cathexis and possible intrapsychic conflicts. Verification of which ego state is cathected is only possible with a four-part correlation of the behavioral, social, historical, and phenomenological determinants of ego states. “The complete diagnosis of an ego state requires that all four of these aspects be available for consideration, and the final validity of such a diagnosis is not established until all four have been correlated” [italics added] (Berne, 1961, p. 75).

Berne (1961, pp. 74-76) described the four diagnostic determinants of ego states in the order he saw them in psychotherapy: behavioral, social, historical, and phenomenological. From a perspective of facilitating an integration of the fragmentation of the ego, I have supplementally defined the identifying criteria and listed them in the following order of significance (Erskine & Moursund, 1988):

1. The identifying criterion of the phenomenological determinant is the subjective experience of the person. It includes the sensations, desires and needs, feelings, and beliefs that shape the person’s perspective—the how and what it is like to live in his or her experience. Included in the phenomenological criteria are the physiological, emotional, and cognitive associations of significant life events and the times when elements of the personality of another were introjected. Also included is the subjective experience of the internal defense mechanisms fixated at times of neglect, traumatic experience, or cumulative devaluation.

2. The historical determinant is gleaned primarily from memories of the dynamic events between oneself and others, or the relationship between mother and father or other important family members. These can provide essential information regarding early conflicts. The who and when of early life may reveal memories of similar feelings and behavior in childhood or memories of the parental person who offered the prototype behavior. Included is an inquiry into the distinction between the person’s own fixated childhood defenses and the defense mechanisms possibly introjected from significant others.

3. The behavioral determinant involves a developmental focus (Berne, 1961, p. 154) on gestures, posture, vocabulary, tone of voice, or other mannerisms, and the content of what is communicated. The assessment of the person’s current observable behavior is compared with information about human development regarding early mother-child interaction; motor and language development; emotional, cognitive, and social development; defense mechanisms; moral development; and adult life transitions. All of this comparative information provides a background of data to assist in determining the stage of development at which emotions, behaviors, or interactions have become fixated. Behavior that is not congruent with the current context may have been normal and appropriate for a child at a specific developmental age or may be an indication of how the patient defended himself or herself in a traumatic situation.

Childlike behavior may be an indication of the person’s own active Child ego state, or just as likely, an indication of the Child ego state of an introjected parent. Interweaving the developmental assessment with the historical or phenomenological may be necessary to determine if a specific defensive reaction, behavioral pattern, or emotion is the manifestation of an exteropsychic ego state or of an archaeopsychic fixation.

4. The fourth determinant in verifying ego state cathexis is the social or transactional. The analysis of transactions provides data to indicate which ego state is active, the nature of the intrapsychic dynamics, and what stimulus from the psychotherapist served to trigger the cathexis. The intrapsychic dynamics include the influence of the introjected Parent ego state and the Child’s need for a contactful relationship. Transactions between the person and psychotherapist, or, in group or family psychotherapy, between any two people, may reflect a transference either from an exteropsychic or archaeopsychic ego state. These transferences may take the form of “roles” such as childlike “compliance,” “impertinence,” or “rebelliousness”; adult-like roles of “problem solver” or information exchange; or parental roles of “comforting” or “controlling” (Berne, 1961, pp. 93-96). It is essential in diagnosing ego state cathexis and intrapsychic conflict to evaluate these transactional roles or social entities within the context of a correlated phenomenological, historical, and developmental (behavioral) assessment.

Transference transactions are an expression of the intrapsychic processes and ego state cathexis. To determine which transactions are nontransference and which are transference, it is necessary to validate which ego states are intrapsychically influential and which are active. “Transactional analysis consists of determining which ego state is active at a given moment in the exhibition of a transactional stimulus by the agent, and which ego state is active in the response given by the respondent” (Berne, 1966, p. 223). It is through the careful and systematic use of the four-part correlated diagnosis that it is possible to understand transference transactions and proceed with psychotherapeutic interventions.

An Intrapsychic and Integrative Perspective

An integrative intrapsychic approach to transactional analysis psychotherapy consists of deconfusing the archaeopsychic ego states and relaxing fixated archaic defenses, emending and/or decommissioning the exteropsychic ego states to resolve internal conflicts between archaeopsychic ego states and exteropsychic ego states, and facilitating the integration of one’s life experiences into a neopsychic ego. “It is the process of making whole: taking disowned, unaware, unresolved aspects of the ego and making them part of a cohesive self” (Erskine & Moursund, 1988, p. 40).

This integrative perspective on psychotherapy is an extension and further refinement of Berne’s (1961) original theoretical concepts of ego states, intrapsychic conflicts, and ensuing transferences. These concepts are augmented by the theoretical premise that it is because of the continued fixation of defense mechanisms that the archaic or exteropsychic ego states remain separate states and do not become integrated into neopsychic awareness. Neopsychic ego state awareness of needs, desires, memories, and external influences remains blocked through the fixation of childhood defenses.

Fixation refers to a relatively enduring pattern of organization of affect, behavior, or cognition from an earlier stage of development which persists into and may dominate later life. Defensive patterns of organization are often formed during an interpersonal conflict in which some psychological gain is achieved at the cost of the loss of others. The persistence of these childhood patterns of organization in later stages of development results in an inability to be spontaneous and flexible in problem solving and in relating to people (Erskine, 1980).

Intrapsychic conflict is the result of the cathexis of an influencing Parent ego state and an internal reaction by a Child ego state (Berne, 1961, pp. 32, 42, 75-78, 241, 1964, p. 26, 1966, pp. 222-223). For example, the influencing Parent ego state is sometimes phenomenologically experienced as a hallucinated voice, a compulsion, and/or an inhibition. It may be observable as a childlike adaptation, withdrawal, or dependency. In other situations the fixated Child ego state is defending against the intrapsychic influence of a Parent ego state. It may be phenomenologically experienced either as an overwhelming sense of need or as a lack of sensation and desires, an incapacity to think, or rage. It may also be observable as resistance, defiance, age regression, needy dependence, or a lack of full contact internally and externally. The observable behaviors may provide data for a partial hypothesis of an adapted Child ego state under the intrapsychic influence of a Parent ego state or states. The subjective or phenomenological experiences reported by the person may provide additional supporting data or lead to an alternate hypothesis.

The intrapsychic conflict is in part maintained by the child’s needs for relationship (Fairbairn, 1952), attachment (Bowlby, 1969), or contact (Erskine, 1989) and the fixated archaeopsychic ego state’s defense against full awareness of contact, attachment, and relationship needs. These needs may be manifested as psychological loyalty to the intrapsychically influencing Parent ego state.

Berne (1961) described the intrapsychic dynamics of ego states as representing “the relics of the infant who actually existed, in a struggle with the relics of the parents who once actually existed” for it “reduplicates the actual childhood fights for survival between real people, or at least that is the way the patient experiences it” (p. 66).

When the archaeopsychic ego state is active (either subjectively reportable or behaviorally observable), by theoretical inference the exteropsychic ego state is cathected and intrapsychically influencing (Berne, 1961, p. 42). I am suggesting that all transactions from an active adapted Child ego state—whether described as resistant, rebellious, compliant, or dependent—are aspects of transference. Transference transactions from a Child ego state are one way of obtaining relief from the intrapsychic conflict. Such transferences are theoretically assumed to be accompanied by a projection of elements of either an exteropsychic ego state or of a fantasy of a self-created parental figure (Erskine, 1988; Moiso, 1985). With projection, the intrapsychic conflict is once again externalized and then reacted to as though the stimulus were coming from outside the person. This provides some momentary relief of the intrapsychic conflict. With transference the intrapsychic conflict may once again be as it was in childhood, transactional between at least two people, with the hope of finally mastering the old interpersonal conflict. Projection also serves as a defense against awareness of the intrapsychic conflict and/or the actual historical conflict and the resulting effect on the child.

The active expression of a Parent ego state can also lead to relief from the intrapsychic conflict. The active Parent ego state is a reaction to and expression of an intrapsychic representation of an internally contained historical transaction. This is observable when the person manifests the thoughts, feelings, and behaviors of the introjected person and directs them toward another person. These active Parent ego state transactions are also defined as an aspect of transference.

An essential procedure in an integrative approach is the analysis of transactions to determine which are transferential and which are nontransferential. The purpose of analyzing transactions is to determine which ego states are active and which are intrapsychically influencing as well as to facilitate an amelioration of the fixations and intrapsychic conflicts. Many transactions in psychotherapy do not reflect a transference of early fixations or introjections. Nontransference transactions are an expression of full contact here and now between the patient and therapist or between any two people. Their conversations may include discussion of the life problems of mature adults, reactions to loss or change, existential dilemmas, spiritual searching, and the challenges faced by aware, responsive, and evolving persons.

Transference transactions are an expression of either an archaeopsychic or exteropsychic ego state and, by inference, reflect an intrapsychic conflict between two or more ego states. Nontransference transactions are any expression of a neopsychic ego uncontaminated by fixations of either archaeopsychic or exteropsychic ego states.

Berne’s Analysis of Transactions

Eric Berne parted company with a classical psychoanalytic theory that regarded all transactions from patient to psychotherapist as transference of childhood conflicts or wishes. Berne’s original theoretical concept of neopsychic ego made it possible to understand transactions as Adult-to-Adult—hence, transactional analysis, and not only an analysis of transference. Berne’s diagrams of ego states also made it possible to graphically represent that which is transference and that which is nontransference.

Berne (1961) began his discussion of the analysis of transactions with a case presentation of “transference” (pp. 91-97) within a therapy group. He described an Adult-to-Adult ego state set of transactions between Camellia and Rosita, followed by Camellia’s misperception of Rosita’s questions, and a shift in Camellia to a Child ego state. Berne described this as a “crossed transaction” (p. 93)

in which the stimulus is directed to the Adult while the response originates from the Child, . . . probably the most frequent cause of misunderstanding in marriages and work situations, as well as in social life. Clinically, it is typified by the classical transference reaction. (pp. 93-94)

In Berne’s further writings he began each explanation of crossed transactions with an example of the “classical transference reaction of psychoanalysis” (1964, p. 30, 1966, p. 225, 1972, p. 14) that loosely fit his original theory of ego states. All his other examples of transactions were from a role theory perspective.

A review of Berne’s (1961) group case presentations ( pp. 91-96) shows that he parted company with his own intrapsychic theory (pp. 29-80, 191-210) and related interventions (pp. 224-231, 1966, pp. 233-258). His interventions in this case were “motivated by the ultimate aim of establishing social control” (1961, p. 95) and his assessment that the group members were not “ready to attempt a deconfusion of the Child or a resolution of underlying conflicts” (p. 95). His use of role analysis as an analogy and substitute for his original intrapsychic theory of ego states corresponded with his switch to behavioral therapy.

On the basis of this motivation Berne changed his theoretical concept of ego states and defined transference and transactions significantly differently from what his original ego state theory would have required for consistency. With his theoretical concepts of exteropsychic, neopsychic, and archaeopsychic ego states, the definitions of transactions and transferences would have had to be related to the expression of ego, ego fragmentation, and intrapsychic conflict. However, Berne’s use of roles to describe ego states led to definitions of transactions that described communication from a behavioral perspective. An evaluation of Berne’s role or descriptive theory reveals consistency between the analogy of ego states as roles and subsequent definitions of transactions. With role theory Berne developed a useful taxonomy of behavior and a theory of communication (1961, pp. 128-135) consistent with a social control therapy. Yet there remains a need for definitions of transactions and transferences that are consistent with Berne's original conceptualization of ego states.

Function of Defense Mechanisms

In describing the transactions between Camellia and Rosita, Berne (1961, pp. 91-97) unfortunately did not discuss two significant theoretical and clinical aspects of the transference reaction: what he referred to as the “misperception” and the “shift” (p. 93) in Camellia’s ego states. Throughout his writings Berne seemed to assume that the reader was familiar with the dynamics of defense mechanisms. A missing link in Berne’s concept of ego states is the lack of a definition of how defenses are related to ego state theory, such as in the case of Camellia’s shift to her Child ego state.

An integrative perspective on transactional analysis assumes that it is because of the continued presence of active archaic defenses that Child and Parent ego states remain fixated and separate states of the ego that are not integrated into an Adult ego (Erskine, 1988). Any of the elements of the ego that are not integrated into the neopsychic ego may be denied; if intrapsychic stress increases, the nonintegrated elements are subject to projection. Projection reestablishes a shaky set of defenses, which were originally developed to keep the person somewhat comfortable in a very uncomfortable situation.

Also from an integrative intrapsychic perspective, it is with the dynamics of the misperception and shift that a phenomenological and historical evaluation is assumed to yield psychotherapeutically useful information about Camellia’s ego states, intrapsychic processes, and the function of her misperception of Rosita (Erskine & Moursund, 1988). Berne (1961) only relied on a social role description—“the parental role of comforting and apologizing” (p. 95)—and an all too limited description of the developmental behavior. There is insufficient information with which to make an adequate correlated diagnosis to determine which ego states are involved in the transactions.

Berne did not elaborate on the significance of the misperception. Theoretically, it is a likely projection onto Rosita of elements of an introjected person (Parent ego state) in Camellia’s life. This would provide a concomitant relief of the intrapsychic conflict and a parallel reexperiencing of an external conflict. Camellia can now enact the internal conflict with another person who can play the role of a “parental response” (Berne, 1961, p. 94), that is, one form of transference. The parental response does not require that the person be in the Parent ego state (exteropsychic ego state), but rather, only that she be a suitable projection screen (Joines, 1977; Moiso, 1985; Perls, 1944/1947).

Transference Analysis

Transference transactions of the type described above involve a denial of and a projection of elements of exteropsychic ego states and a reaction from an active archaic ego state. There also may be subsequent transactions from the Child ego state to the misperception of a parental response in the other person. It is also possible to have a transference that involves projection of elements of exteropsychic ego states and a reaction or overt transaction from an exteropsychic ego state.

These transferences from historical relationships provide defensive relief from the discomfort of the intrapsychic conflict. Memories are deflected of the original transactions, where the person(s) with whom the child needed a primary relationship, attachment, and contact were the ones who disappointed, neglected, or abused. In such a transference the interpersonal conflicts of childhood are once again experienced as originating with people in the environment and thus offer the opportunity for resolution.

Relief from intrapsychic conflict may also be achieved through a transference that involves denial of and projection of elements of an archaeopsychic ego state. To avoid the awareness of discomforting or painful feelings, needs, or experiences, the original denial or repression must be maintained. One way of accomplishing this, particularly when these feelings are stimulated, is by projecting elements of the repressed Child onto someone else. The transferential transactions may take two basic forms: (1) projection of elements of archaeopsychic ego states and an overt transaction from an active exteropsychic ego state, or (2) projection of elements of archaeopsychic ego states and a reaction or overt transaction from an archaeopsychic ego state.

A graphic example of Child ego state projection and transference transactions from an active Parent ego state occurs in some cases of multigenerational child abuse. The primary purpose is to diminish intrapsychic conflict: The painful experiences contained in the Child ego state are denied and projected onto a suitable screen, and the verbal or physical cruelty that was historically introjected into a Parent ego state is made externally active and directed at another person.

A second example illustrates the projection of elements of a Child ego state and a reaction by a Child ego state within the same person. In some clinical situations the patient may engage in primary process and magical thinking and project a fantasy onto the psychotherapist. The projection of an archaic fantasy provides an opportunity for the patient to express through the transference with the psychotherapist the Child ego state experiences of intrapsychic conflict. Such early childhood fantasies function as an intrapsychic protection and may be either terrifying and punitive or wonderful and nurturing, similar to Kohut’s (1971, 1977) descriptions of idealizing transferences. Either fantasy serves both to maintain the denial of the caretakers’ effects on the child and to express the need for protection from the intrapsychic conflict (Erskine, 1988). Psychotherapists who regularly confront, define as a game, or attempt to eliminate such a projection of either a terrorizing or idealizing transference inhibit an intrapsychic and integrative therapy process.

Another aspect of transference, the projection of elements of a Child ego state and an overt transaction from a Child ego state within the same person, is evident in those psychotherapists who project their own childhood experiences onto patients. The overt transactions may be an expression of a benevolent, nurturing caretaker fantasy within a Child ego state that functions internally to protect against awareness of Parent ego state influence (Erskine, 1988; Erskine & Moursund, 1988; Moiso, 1985). As long as there is a suitable screen for the projection of a troubled child, the intrapsychic conflict can be transferred and the denial contained. This form of transference is commonly referred to as countertransference.

Ulterior transactions represent those transactions that are at the psychological level of motivation, outside of Adult ego state awareness, and that are a transferential expression of Parent or Child ego state elements (Berne, 1961, pp. 103-105). In 1964 Berne described ulterior transactions as the basis of games (p. 33), and earlier (1961) he defined games as

segments of longer, more complex sets of transactions called scripts. Scripts belong in the realm of transference phenomena, that is they are derivatives, or more precisely, adaptations, of infantile reactions and experiences. But a script does not deal with a mere transference reaction or transference situation; it is an attempt to repeat in derivative form a whole transference drama, often split up into acts, exactly like the theatrical scripts which are intuitive artistic derivatives of these primal dramas of childhood. (p. 116)

Life script is the macro expression of transference; games are a subset of script, ulterior or psychological level transactions are the substance of games, and the analysis of transactions is dependent on the concept of the ego divided into states with ensuing intrapsychic dynamics.

Transactional analysis is a theory of personality and social action, and a clinical method of psychotherapy, based on the analysis of all possible transactions between two or more people, on the basis of specifically defined ego states. . . . Any system or approach which is not based on the rigorous analysis of single transactions into their component specific ego states is not transactional analysis. (Berne, 1972, p. 20)

Psychotherapy of Transference

The psychotherapy of transference occurs in part when the therapist does not simply take the patient’s words or behavior at face value but also looks for the unaware meaning of what patients are saying or not saying, doing or not doing through their affective communication and bodily gestures. The understanding of transference from an integrative intrapsychic perspective on transactional analysis requires a multifaceted focus. Transference can be viewed as:

1. the means whereby the patient can demonstrate his or her past, the developmental needs that have been thwarted, and the defenses that were erected to compensate;

2. the resistance to full remembering and, paradoxically, an unaware enactment of childhood experiences;

3. the expression of intrapsychic conflict and the desire to achieve intimacy in relationships; or

4. the expression of the universal psychological striving to organize experience and create meaning.

Novellino (1985) expanded on the importance of understanding the function of transference:

In any psychotherapeutic relationship the unsatisfied childhood need will be projected onto the therapist who will be experienced by the patient as the source of the possible satisfaction of the need (positive pole of transference) as well as its frustration (negative pole of transference). In every case the transference will be characterized by the simultaneous presence of both poles. (p. 204)

Trautmann (1985), in a Transactional Analysis Journal editorial summarizing the transactional analysis literature on transference, said:

Therapy is effective when the internal Parental influence or dialogue is externalized (transferred), allowing for the resolution of childhood impasses and traumas, and the emergence of a stronger, uncontaminated, more integrated Adult. The specific approach used to effect this resolution depends on the level of childhood fixation: the more symbiotic the Child, the more actively the therapist needs to take on the transference relationship. (p. 190)


Berne’s (1972) application of the principle of “Occam’s Razor” (p. 20) gave too close a shave to the theory of analysis of transactions. In his attempt at conceptual “simplicity” (Preface, p. xvi) and theoretical “economy” (p. 21), Berne cut the theoretical concepts to their most simplified explanation and in so doing, I believe, lost the significance and profundity within his own theory. No longer is there either internal or external theoretical consistency.

When Berne redirected the emphasis of ego state theory from the original definitions to behavioral descriptions, he created a fundamental change in the analysis of transactions. With the shift in the theoretical metaphor of ego states the focus of the psychotherapist moved to the effect of the communication (transaction) on the receiver and on the patient’s options for changing behavior to produce more effective communication.

The methodology stemming from this change of theoretical emphasis often resulted in the patient’s improving social skills, but the inherent meaning of the transactions, particularly those that are transferential, was lost. No longer was there a theoretical basis in the psychotherapist’s mind for a sensitivity to the internal psychological message or the desperate communication in the unaware expression of the existential position (Berne, 1964) or script beliefs (Erskine & Zalcman, 1979). Berne’s original theoretical postulates, which led to an understanding of intrapsychic functioning and psychological versus social levels of transacting, was diminished, and a form of transactional analysis as a behavioral therapy emerged. This shift defined the task of the transactional analyst as improving communication and social effectiveness rather than understanding and ameliorating the intrapsychic conflict that is communicated through transference.

Berne developed two distinctly different theories of ego states and transactions. Each has a specific and valuable clinical purpose, and Berne’s descriptive theory has many applications in the social world of human behavior and communication. It has been my goal in this article to show that the use of Berne’s developmental, relational, and intrapsychic theory of ego states and the consistent use of that theory in understanding the internal dynamics of transactions can lead to a sensitive and effective response to transactions and transference and to a comprehensive psychotherapy that results in the integration of ego state fragments.


Baker, E. (1982). The management of transference phenomena in the treatment of primitive states. Psychotherapy: Theory, Research and Practice, 19, 194-197.

Berg, M. (1977). The externalizing transference. International Journal of Psychoanalysis, 58, 235-244.

Berne, E. (1961). Transactional analysis in psychotherapy: A systematic individual and social psychiatry. New York: Grove Press.

Berne, E. (1964). Games people play: The psychology of human relationships. New York: Grove Press.

Berne, E. (1966). Principles of group treatment. New York: Grove Press.

Berne, E. (1972). What do you say after you say hello?: The psychology of human destiny. New York: Grove Press.

Bollas, C. (1979). The transformational object. International Journal of Psychoanalysis, 60, 97-107.

Bowlby, J. (1969). Attachment. Vol. 1 of Attachment and loss. New York: Basic Books.

Brandchaft, B. (1989, October). Countertransference in an intersubjective perspective: A case presentation. Panel discussion, 12th annual conference on the psychology of the self, San Francisco.

Brenner, C. (1979). Working alliance, therapeutic alliance, and transference. Journal of the American Psychoanalytic Association, 27, 137-158.

Breuer, J., & Freud, S. (1955). Studies on hysteria. In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 2, pp. 1-305). London: Hogarth Press. (Original work published 1895)

Clarkson, P., & Gilbert, M. (1988). Berne’s original model of ego states: Some theoretical considerations. Transactional Analysis Journal, 18, 20-29.

Erikson, E. (1950). Childhood and society. New York: Norton.

Erskine, R. G. (1979). Fourth-degree impasse. In C. Moiso (Ed.), T.A. in Europe: Contributions to EATA summer conference 1977/1978. Geneva, Switzerland: European Association for Transactional Analysis. Workshop conducted at the European Association for Transactional Analysis Congress, Seefeld, Austria, July, 1977.

Erskine, R. G. (1980). Script cure: Behavioral, intrapsychic and physiological. Transactional Analysis Journal, 10, 102-106.

Erskine, R. G. (1987). A structural analysis of ego: Eric Berne’s contribution to the theory of psychotherapy. In Keynote speeches: Delivered at the EATA conference, July, 1986, Noordwijkerhout, The Netherlands. Geneva, Switzerland: European Association for Transactional Analysis.

Erskine, R. G. (1988). Ego structure, intrapsychic function, and defense mechanisms: A commentary on Eric Berne’s original theoretical concepts. Transactional Analysis Journal, 18, 15-19.

Erskine, R. G. (1989). A relationship therapy: Developmental perspectives. In B. R. Loria (Ed.), Developmental theories and the clinical process: Conference proceedings of the Eastern Regional Transactional Analysis conference (pp. 123-135). Madison, WI: Omnipress.

Erskine, R. G., Clarkson, P., Goulding, R. L., Groder, M. G., & Moiso, C. (1988). Ego state theory: Definitions, descriptions, and points of view. Transactional Analysis Journal, 18, 6-14.

Erskine, R. G., & Moursund, J. (1988). Integrative psychotherapy in action. Newbury Park, CA: Sage Publications.

Erskine, R. G., & Zalcman, M. (1979). The racket system: A model for racket analysis. Transactional Analysis Journal, 9, 51-59.

Fairbairn, W. R. D. (1952). An object-relations theory of the personality. New York: Basic Books.

Federn, P. (1977). Ego personality and the psychoses. London: Maresfield Reprints. (Original work published 1953)

Freud, A. (1965). Normality and pathology in childhood: Assessments of development. New York: International Universities Press.

Freud, S. (1949). An outline of psychoanalysis. New York: Norton.

Freud, S. (1955). Fragments of an analysis of a case of hysteria. In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 7, pp. 1-122). London: Hogarth Press. (Original work published 1905)

Freud, S. (1958). The dynamics of transference. In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 12, pp. 97-108). London: Hogarth Press. (Original work published 1912)

Freud, S. (1958). Observations on transference—love: Further recommendations on the technique of psychoanalysis, III. In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 12, pp. 157-173). London: Hogarth Press. (Original work published 1915)

Friedman, L. (1969). The therapeutic alliance. International Journal of Psychoanalysis, 50, 139-159.

Greenberg, J. R., & Mitchell, S. A. (1983). Object relations in psychoanalytic theory. Cambridge, MA: Harvard University Press.

Greenson, R. R. (1967). The technique and practice of psychoanalysis. New York: International Universities Press.

Guntrip, H. (1971). Psychoanalytic theory, therapy and the self. New York: Basic Books.

Joines, V. (1977). An integrated systems perspective. In G. Barnes, (Ed.), Transactional analysis after Eric Berne: Teaching and practice of three TA schools (pp. 252-272). New York: Harper & Row.

Khan, M. M. R. (1974). The privacy of the self. London: Hogarth Press.

Klein, M. (1949). The psychoanalysis of children. London: Hogarth Press.

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

Kohut, H. (1977). The restoration of the self: A systematic approach to the psychoanalytic treatment of narcissistic personality disorder. New York: International Universities Press.

Langs, R. (1976). The therapeutic intervention: Vol. II. A critical overview and synthesis. New York: Jason Aronson.

Langs, R. (1981). Classics in psychoanalytic techniques. New York: Plenum.

Lipton, S. (1977). The advantages of Freud’s technique as shown in his analysis of the rat man. International Journal of Psychoanalysis, 58, 255-273.

Loria, B. R. (1988). The parent ego state: Theoretical foundations and alterations. Transactional Analysis Journal, 18, 39-46.

Massey, R. (1989). Script theory synthesized systemically. Transactional Analysis Journal, 19, 14-25.

Moiso, C. (1985). Ego states and transference. Transactional Analysis Journal, 15, 194-201.

Moiso, C. (1988). Eric Berne memorial scientific award acceptance speech: Ego states, transference and the TA psychodynamic approach—an overview. Transactional Analysis Journal, 18, 4-5.

Novellino, M. (1985). Redecision analysis of transference: A TA approach to transference neurosis. Transactional Analysis Journal, 15, 202-206.

Novik, J., & Kelley, K. (1970). Projection and externalization. Psychoanalytic Study of the Child, 25, 69-95.

Ornstein, A. (1989, October). Countertransference in an intersubjective perspective: A case presentation. Panel discussion, 12th annual conference on the psychology of the self, San Francisco.

Perls, F. S. (1947). Ego, hunger and aggression: The beginnings of Gestalt therapy. New York: Vintage Books. (Original work published 1944 as Ego, hunger and aggression: A revision of Freud’s theory and method. Durban: Knox Publishing)

Perls, L. (1978). An oral history of Gestalt therapy. Part I: A conversation with Laura Perls, by Edward Rosenfeld. The Gestalt Journal, 1(1), 8-31.

Piaget, J. (1932). The moral judgment of the child. New York: Harcourt Press.

Piaget, J. (1951). Play, dreams and imitation in childhood. New York: Norton.

Piaget, J. (1954). The construction of reality in the child. New York: Basic Books.

Spotnitz, H. (1969). Modern psychoanalysis of the schizophrenic patient. New York: Grune & Stratton.

Stolorow, R. D., Brandchaft, B., & Atwood, G. (1987). Psychoanalytic treatment: An intersubjective approach. Hillsdale, NJ: Analytic Press.

Trautmann, R. (1985). Letter from the editor. Transactional Analysis Journal, 15, 188-191.

Trautmann, R. L., & Erskine, R. G. (1981). Ego state analysis: A comparative view. Transactional Analysis Journal, 11, 178-185.

Weiss, E. (1950). Principles of psychodynamics. New York: Grune & Stratton.

Winnicott, D. W. (1965). The maturational processes and the facilitating environment: Studies in the theory of emotional development. New York: International Universities Press.




1This article was originally published in the Transactional Analysis Journal, Volume 21, Number 2, April 1991, pp. 63-76.The author wishes to gratefully acknowledge the members of the Professional Development Seminar of the Institute for Integrative Psychotherapy, New York, for their valuable suggestions in the formulation of this article.

El Instituto de Psicoterapia Integrativa está aprobado por la Asociación Americana de Psicología para impartir formación continua a psicólogos; por el National Board of Certified Counselors para los counselors y por el American Board of Examiners in Pastoral Counseling para los asesores pastorales. El Instituto de Psicoterapia Integrativa se hace responsable de este programa y su contenido.