Gian Paolo Scano

Phlogiston and the transferential "thing":
An intersubjective formulation of the problem of transference

The object of this essay is first to demonstrate the need to reformulate the theory of transference, and second, to propose an alternative conceptualization that transforms the wealth of traditional assertions into a theoretically coherent perspective that is both epistemologically adequate and empirically verifiable. With this in mind, the author analyzes the recent debate on transference and highlights the limits of both the traditional and the interactional points of view, showing how this discussion is invalidated by a series of misunderstandings that originate from the widespread and erroneous view that transference is a "phenomenon." Having established the conceptual limits of this proposal, the author introduces (for expository reasons) the notion of the "transferential thing" to indicate the object of this conceptualization. Employing Karl Popper's situational logic, he outlines a clearly intersubjective conception. He thus identifies "the transferences" with "the problems within the solving process" that inexorably arise in the therapeutic interaction because of the interpersonal nature of this method. Such problems can be analyzed both from an intrasubjective point of view (in terms of subjective "theories" and repetitive rules used by the patient and the therapist in the construction of contexts) and from an intersubjective one (in identifying repetitive configurations of the co-constructed interaction between the patient and the analyst within each one's context).

For over a century, the transference issue has taken a major role in both literature and in discussions among analysts. It has gained prominence in recent decades because, following the eclipse of metapsychology, the debate has shifted from theoretical aspects of psychoanalysis to clinical and technical ones. For example, Gill's (1982, 1984, 1994) active reevaluation of the concept of transference, conducted during the last twenty years of his life, rekindled interest and gave new impetus to discussions about the so-called blank screen, analytic interaction, the analyst's subjectivity, countertransference as a source of information about the patient, and enactment. In this period of theoretical transition, there has been a growing hope that an awareness of the importance and vitality of transference could act as a lever for both theoretical and technical developments, just as Wallerstein (1990) hoped.
The realization of this hope has not yet materialized. A general recognition of the importance of transference is not based on the solidity or linearity of the concept, but rather on the conviction of clinical self-evidence that also functions to reassure the continuity of the traditional psychoanalytic way of behaving. The debate has not only conferred a heuristic richness on the concept of transference; it has also revealed logical as well as theoretical inconsistencies, the most important being the absence of a universal conception of the transference.(1)
A more determined will to tackle the problem head-on must replace this tacit expectation of a miraculous change in the development of psychoanalytic thinking, if we are to establish the theoretical ground that enables a coherent redefinition of transference and that provides a reasonable degree of empirical support.
In what follows, I will briefly present and explore a view of the transference issue in intersubjective terms, making use of the contextual-situational method (Popper 1963, 1972, 1982, 1983). After a brief reevaluation (parts 1 and 2), I will proceed to reformulate the issue first from a subjective perspective (part 3), next from an intersubjective one (part 4), then in terms of problems related to the process of coming to a resolution (part 5), and, finally, I will conclude by making an initial estimate of risks and benefits (part 6).


Due to the discordance of opinions, it is difficult to briefly summarize the various positions in this debate. In order to simplify matters, one could draw a continuum between a hypothetical point zero---the traditional view---to its "interactional" opposite.
A "traditional conception" of transference is an artifact. It can be described almost in handbook terms,(2) with remarks such as: transference represents a "spontaneous repeating" during therapy of previous life experiences (mostly related to childhood), which distorts some elements of the relationship. Such distortion leads to the creation of "unrealistic" and "fantasized" experiences---perceived by the subject, however, as accurately reflecting the situation. One might say that the use of notions such as repetition, distortion, and spontaneity helps define the traditional position. Neutrality---not only as a technical rule, but also as a logical consequence of theoretical arguments---is often integral to this approach.(3) If transference is a repetitive, spontaneous, and distorted production on the part of the patient, the therapist plays no significant role in its construction. Fundamental axioms---such as the role of interpretation and insight, the therapist's neutrality in regard to the patient's conflicts, and so on---seem to justify the notion of the blank screen, something that accords with the epistemological structure of classical theory and technique.
The interactive position, at the other end of the continuum, is radically different. It negates the characterization of transference as repetition and assigns primary importance to the analyst's participation in the construction of transference. It criticizes the concept of spontaneity, and even more, it questions the notion of distortion. Traditional axioms are interpreted within the context of the relationship between patient and analyst. Such a position must necessarily be defined as intersubjective, and therefore implies the need for a substantial reformulation of both theory and method. This revised definition of transference is not new; it was explicitly described by Gill (1982, 1983, 1994), and developed by Hoffman (1983, 1998), Mitchell (1998), and Renik (1993, 1996), having been first mentioned by Wolstein and Levenson.
At this stage, the status questionis should be clear: there is a contrast between the traditional position, which maintains the earlier conceptual stance, and an innovative one, which moves from the traditional intrapsychic position to a specifically intersubjective one. Progression from one to the other is not easy. The number of so-called constructivists is growing, but they remain a minority. On the other hand, there are not many supporters of the "blank screen theory," left, when we remember that even authors such as Kernberg (1998) and Eagle (2000a) agree with some of the criticisms of the traditional position. The majority of analysts, as a matter of fact, locate themselves somewhere in-between the traditional and the interactive poles. They tend to support basic traditional concepts, including notions such as repetition and distortion, but at the same time, they also favor the analyst's taking a more active role. The issue of spontaneity comes into play in the belief that transferential actions stem from the patient's inferences about clues dropped by the analyst (Eagle 2000a). Within a traditional framework, many analysts try to make use of those elements that recognize the value of interactive needs---modifying, for example, the conception of countertransference, or redefining its connection with transference (as in the transference-countertransference matrix), or by introducing "real-relationship" approaches (alliance, holding, and so on).(4)
Although any compromising synthesis tends to be an easy target for criticism, it may nonetheless have a solid foundation. The analytic position just described offers, first of all, a way of understanding and acting that may be more suited to many clinical situations than the traditional one, but at the same time, it is less risky, less uncertain, and less lacking in supporting references than the radical one.
There is, however, a complication: the abundance of diverse approaches between the traditional and innovative poles. The fundamental thesis of constructivists, for example, is that reality is not an objective presence, but rather, it is something that is constantly being constructed. As in epistemology, different levels of interpretation can be brought to bear on it. A radical position (recently described by Mitchell [1998]; such a position is critical of the traditional belief in the analyst's ability to probe dynamics that are mistakenly viewed as preorganized in the patient's mind) is easily distinguished from a far less radical one, which, by maintaining a distinction between ontology and knowledge, can easily accommodate several levels of organization of the mind. Among traditionalists as well as among interactionists moreover, there are those who remain faithful to the original Freudian vision, dealing primarily with "causes" and "meanings," and those who maintain hermeneutic positions by rejecting the "scientific" setting of the classical model.
Hence, everyone is confused. Unable to distinguish or to critically analyze the countless positions existing between the two poles---with regard both to neutrality/interaction and to cause/meaning---one can only criticize both extremes. The traditional one, characterized by concepts such as repetition, spontaneity, and distortion, appears quite difficult to defend from a logical, as well as a conceptual, point of view; its main weakness lies in the fact that such a position should logically imply the maintenance of the metapsychological framework. This is not, however, always acknowledged by traditionalists. Hence, even authors who explicitly abandon metapsychology somehow remain faithful to traditional axioms. In such cases, presumably, repetition, distortion, and spontaneity are understood as phenomenal characteristics of transference---that is, as "data," easily observed and examined in therapy. And apart from this theoretical and methodological flaw, the traditional position has other serious weaknesses; concepts such as distortion, spontaneity, and the analyst's neutrality have been subjected to serious criticisms.(5) More complex is the concept of repetition, which---though comprehensible, reasonable, and logical within Freudian theory---is neither very comprehensible nor verifiable, and remains overly rigid in its conception of the relationship between past and present. Even considering this conceptual frailty, however, the traditional position has at least one strength: it tries to preserve a certain specificity of transference, which is believed to be (and probably rightly so) endangered by interactive approaches.
The interactive position appears more attractive, comprehensible, and plausible to those who have definitely abandoned the notion of metapsychology. It incorporates a more balanced view of the therapeutic situation and is better suited to the current epistemological moment, because it overcomes the naïve objectualism of the traditional viewpoint. It allows a more reasonable articulation of past and present, of inner and outer reality; it gives credence to the contributions of both parties in the analytic relationship, giving particular attention to the co-construction of transference. Hence, it is both more flexible and more manageable. But despite all these advantages, the interactive position has not succeeded in definitively outlining its criteria to create a formal model. Its epistemological premises imply the need for significant modification not only of the concept of transference, but also of the general conception of psychoanalytic theory and method. Such a position nevertheless does not hide from problems and is ready to face the possibility of reformulating the entire discipline of psychoanalysis via its methodological, theoretical, and technical tenets.
Unfortunately, there are no empirical or theoretical criteria (other than informal accounts of clinical experience) to find one's way through this labyrinth. Metapsychology is now dead, and the interactionists have not yet developed a formal theory. The only reliable instrument that can be useful to orient us in such complicated situations is a historical-critical analysis, which enables us to trace each concept back to its origins and along its theoretical trajectory, avoiding the conceptual transformations that have created so much ambiguity in our terminology. However, with few exceptions,(6) historical-critical analyses have scarcely appeared in the psychoanalytic literature.
With very good intentions, Wallerstein (1988) traced what has become the most-traveled path: having accepted the pluralism of theoretical perspectives and having identified a common ground in the method of handling transference and countertransference, he invited us to turn our attention away from the diversity of "explicative metapsychologies" and toward actual clinical theory and experience. The differing conceptual theories do not necessarily undercut a shared clinical theory that employs a fundamental clinical method based on the support of universally observable data (Wallerstein 1990).


But what Wallerstein appeared not to consider is that what he described has been put into effect throughout the last century, and that the fundamental clinical method, universally observed data, and a shared clinical theory are the very elements that have led to actual theoretical confusion. It would be unrealistic to expect that, all of a sudden, they should begin to function as the basis of a "scientific" theory. Furthermore, a series of ambiguities has emerged. What Wallerstein identified as general theories or "metapsychologies" (post-ego psychology, Kleinianism, object relations theory, self psychology, post-Sullivanian theory, the work of Schafer, and so on) are, in fact, only clinical theories. This is the first ambiguity. The second is intrinsic to the idea that data or events can be separated from theory.
It was Rapaport (Wallerstein refers to G.S. Klein and Sandler) who, in less suspicious times, pointed out the distinction between general and special theory. The latter, implying both psychodynamics and clinical theory, is significantly less abstract and attempts to relate formal, theoretical axioms to concrete and specific cases. Among the theories mentioned by Wallerstein, only ego psychology and Melanie Klein's theory refer to Freudian metapsychology (the only true metapsychology); the others either explicitly reject it, or, while referring to the so-called object relations theory, fail to maintain a unitary position. There are no other metapsychologies, but only informal clinical theories that are nothing other than generalizations derived from clinical cases laden with unverified interpretations, inferences, and conjectures.
Wallerstein refers to a different experience, however: a sort of "daily practice" that he describes as "common," based on arguments derived from clinical cases. All this leads to the second ambiguity---that is, the possibility of collecting data and discussing and acting on events and experiences without reference to theories. Gill (1994) sharply invalidated this approach, which, however, still finds support in literature and in many analysts' minds, especially concerning "obvious data" such as transference. Even such self-evidence is the result of an ambiguity, in that it is based on the idea that transference is an easily observable and perceptible "phenomenon"---that is, observable by anyone who attends any psychoanalytic session without preconceptions.(7)
There is a long history behind this belief. Freud himself initiated this ambiguity when he began referring to transference as an unavoidable (and irritating) phenomenon that intervenes in therapy. Since then, transference has never gotten rid of this phenomenal attribute. It is easy to comprehend, from a historical-critical viewpoint, the motives behind the Freudian vision; they have to do with its naturalistic framework. Despite Freud's belief, however, transference cannot be classified as a phenomenon. Regardless of the nature of any "phenomenon" it relates to, transference is a theoretical construction, a metapsychological concept whose genesis, topical-dynamic structure,(8) and role within Freudian theory(9) are easily definable. All the unrealistic, distorted, and phantasmatic attributions, interpreted as "phenomenal" in traditional analytic literature, are far from being descriptive characteristics. They are, in fact, connotations required by topical-dynamic theory in order to describe and explain the concept of transference. Therefore, they are actually "characteristics" of the theory, not of the phenomenon.(10)
This is an important matter. If transference is a "phenomenon," the debate should focus on describing it precisely, in which case it would make sense to release observed data in order to facilitate a clearer description of the phenomenon. If it is a concept, on the other hand, descriptions, observations, and the use of related data should be considered irrelevant to the issue. Focus should instead be placed on its conceptual structure, critical validity, and internal coherence, and on its relation to an acceptable theoretical model. To those who see transference as clearly a phenomenon, however, efforts to critically examine it, and even to redefine it, may appear to be incomprehensible, illogical attempts to change the natural course of things---an insane tendency to convince others that pigs can fly.
Unfortunately, the alleged phenomenal nature of transference proves to be a source of ambiguities. If transference is a concept, its conceptual ecosystem is the Freudian model, which is the theory of the mind or of the "psychic machine," described strictly from an intrapsychic viewpoint, from which a phantasmatic and distorted "repetition" appears logical. If one takes into account that supporters of the drive theory are nowadays rarer than pandas, one may conclude that authors do not typically use the concept of transference within the metapsychological framework. So, within which framework are they using it? How is it possible that a concept constructed to explain a single performance of the "psychic machine," from a strictly intrapsychic viewpoint, can circulate within the current psychoanalytic debate, apparently confirming the hypothesis that the interaction between patient and analyst is unavoidable and that references to subjectivity and intersubjectivity inevitably derive from the nature of things?
The concept of transference can be utilized in the debate concerning these problems because of another ambiguity---that is, the unspoken acknowledgment of the relative validity of concepts such as "intrapsychic" and "intrasubjective." In this simple stratagem, the old concept is circulating as a kind of "exchange coin" within a context that, officially, seems to have abandoned the idea of the "Freudian apparatus." Interactionists are obviously aware that their method of analysis proceeds from an intersubjective/intrasubjective viewpoint; the question is, from which viewpoint do those who criticize the assertions of constructivism come from? As we all know, the intrapsychic viewpoint was introduced by Freud to eliminate any "intrasubjective" and "intersubjective" misunderstanding. The intrapsychic viewpoint, indeed, justifies the position of the "neutral" reader of psychic events, because it allows the therapist to go beyond the garish, epiphenomenal world of "subjective" and "intersubjective."
Yet in this mosaic of ambiguities, there is a solid argument for the prospect of a common ground. Two of the most solid types of empirical research data in psychotherapy are: (a) results that are not correlated to theories or to techniques, but rather to variables regarding the therapist; and (b) results that appear to be similar for many types of therapies, regardless of their particular theories and techniques (the so-called "Dodo verdict"). Using this information (solid, but not undisputed), one might think that something universal does indeed take place in all psychotherapies, not only in psychoanalytic ones. However, this conclusion need not be correlated with what Wallerstein calls "clinical theory": it could be something common to all efficacious therapeutic interactions. The uniformity of outcomes in therapy could be related, in fact, to a series of factors and events that are intrinsic to therapeutic interaction itself, and, therefore, of a very different nature from those described by theories. Conversely, however, these factors might also reveal some sort of relationship with transference, countertransference, neutrality, containment, projective identification, and enactment.
If this conjecture is correct, the position of interactionists, rather than that of traditionalists, would seem to be more valid. There is, however, a further ambiguity: it is necessary to determine whether interaction is itself a phenomenon or a concept. It is well known that transferential imperialism was first limited by Greenson's (1969) introduction of a distinction between a transference relationship and a nontransference relationship. In the last twenty years of the previous century, the notions of a real relationship and of a working or therapeutic alliance have become more and more commonplace, to the point that we might say that a juxtaposition has been reached between transference and relationship---and we may even have delineated a contrast between transference and relationship.(11)
Both interaction and relationship, however, have two meanings, which exist on different levels of logic: first of all, the terms indicate, from a descriptive viewpoint, the totality of the spoken and the unspoken, the conscious and unconscious events that take place during therapy, and, therefore, the entire interactive event. If, however, we consider interaction (or relationship), implicitly or explicitly, as a factor or group of factors active in the process of change, then interaction, just like transference, must be considered an explicative concept. This usage of the terms comes into play spontaneously every time we state that meanings are to be "built within the relationship," or whenever a change can be traced back to a new experience, in opposition to the traditional argument in favor of interpretation and insight; and it is also operative each time one attributes therapeutic value to empathy, holding, or containment.
In all the above mentioned cases, interaction (or relationship) has explicative value. Each explicative concept, however, can exist only within its theoretical framework. Transference insists on the topical-dynamic theory, but what theory supports the concepts of relation and interaction? There is no formal theory of such a kind. The contribution made by social constructivists, though useful, does not constitute a theory (and is even less a paradigm). As of today, the claim that changes take place as a consequence of factors intrinsic to therapeutic interaction forms only the foundation of a process of conceptualization, because it necessitates the definition of a list of factors and because it requires some kind of model capable of describing the actions of these factors. Without a precise theory, a concept is generally substantiated with recourse to an implicit theory, or, more often, simply to common sense. In this case, one resorts to a "ready-made product," such as transference. Consequently, transference, born to swim in its topical-dynamic sea of origin, suddenly finds itself in a new, intersubjective sea, being used both by traditionalists---to "transfer" the old intrapsychic world of the "psychic machine" into the new one, and by interactionists---to express the dynamic interaction in which experiences and meanings are built. Ambiguity rules all!
To avoid spiraling down into this confusing vortex, the following discussion is put forth according to these rules:

1. The object of this discussion is therapeutic interaction, which is the only observable and verifiable, phenomenal reality; that is what must be explained.
2. The viewpoint put forward is an intersubjective/intrasubjective one.
3. The theoretical-conceptual framework is that of a work-in-progress theory of therapeutic interaction.
4. The term transference, except when used in the context of classical theory, is used only to describe the problem we are dealing with (also referred to as transference in most of the literature).
5. The temporary notion of a transferential thing is introduced to indicate the specific subject of the concept under discussion.
6. This transferential thing is placed within the structure of interaction. In line with the classical viewpoint, it is supposed to be a factor distinct from the therapeutic relationship, but in accordance with the thinking of constructivists, it is not supposed to exist in a way that contrasts with the relationship.
7. The choice of conjectures must allow for empirical verifiability.
8. Lastly, it is strictly... "forbidden" to introduce elements into people's heads (Bateson 1972), such as images, objects, relationships, unconscious hatred or love, and so on.


In this effort at redefinition, I am allowing myself a general theoretical conjecture. Let me start from the assumption that interaction between the therapist (T) and the patient (P) organizes itself progressively into a complex system, the T/P system, coordinating the operating mode of the two subsets, T and P, and situational features such as place, time, setting, and so forth. In light of this assumption, interaction always represents that of the whole system (the intersubjective viewpoint), which could also be monitored, however, from both T's and P's subset viewpoints (the subjective and intrasubjective viewpoints). Within the analysis of any interaction between two or more subjects, one could moreover distinguish that which effectively happens (the interaction) from that which is said to happen according to any type of perception, account, interpretation, explanation, or theory (the metainteraction) (Scano 2000).
This argument will be elaborated through use of the situational method conceived by Popper, the applicability of which in psychotherapy has been questioned by Cadeddu (1995). According to Popper (1972), knowledge and the evolution of knowledge can be set in the general framework of biological evolution, with which they share the same logic. Knowledge does not derive from observation, but rather from problems that need to be resolved.(12) The process of knowledge, therefore, develops as follows:
PP1 --> TT --> EE --> PP2 -->
Starting from problematic situations (PP), theories are elaborated (TT) and put into practice; as errors (EE) are phased out, a new problem level (PP2) is reached. This procedure, developed through conjectures and subsequent refutations, represents the scheme of biological, psychological, and cultural evolution, as well as the scheme of daily and scientific knowledge acquisition.
The explanation of a subject's action follows the same logic and is expressed as the study, as well as the critical analysis, of the objective content and of the structure of that specific product of the subject's intentions. This explanatory method is known as situational analysis. To give an example, Ms. P, as her wedding day approaches, is suddenly struck by panic attacks, leading to a profound state of fear and a feeling of alienation, which culminates in a serious state of depression. As a result, the young lady seeks counseling and begins psychotherapy.
At this point, we can outline the situation using Popper's method, as follows: The panic attacks represent a condition initiated by the problematic situation (PP1) that Ms. P is facing; they are a manifestation of the patient's unknown theory (TT1). The process leads to a new problem, i.e., panic, depression, and the sensation of living a meaningless life (PP2); hence, she seeks therapy. In order to understand and solve PP2 (her symptomatology), the therapist will have to understand PP1 (the starting problem) and TT1 (the young lady's theory). This represents a second-level problem, that is, a metacomprehension of the young lady's comprehension and all the conjectures that motivate her action; the problem can thus be defined, from the therapist's viewpoint, as a problem of comprehension (PPc). The therapist will then make conjectures in relation to PPc and will also, perhaps, find it necessary to eliminate errors. The therapist thus ends up with a new problem, PPc1, and so forth.
As the therapy progresses, therapist and patient will collaborate to resolve the patient's PP1 problem and the therapist's PPc1, in the concerted effort to ameliorate Ms. P's depression and panic. This process will be fraught with unavoidable obstacles. For example, a particular therapeutic session starts a few minutes late because the therapist is finishing a phone call. After a silence, Ms. P begins to sob, and in a broken voice, says: "You are not really interested in me; you just do your job and I'm only one of your cases." After a respectful pause, the therapist replies, "Maybe you have often felt neglected in the past." The young lady lifts her eyes, bright with tears, and says that just the day before, as has frequently occurred, her mother called her only because she needed to be taken to the doctor.
The understanding of this interaction occurs on a different level than the understanding of PP1, PP2, and PPc, since here the problem originates in the solution process, and hence, has to be resolved in order to maintain the collaboration necessary to eliminate the symptomatology for which therapy began. We are facing a problem of the solution process (PPsn), which follows the same situational logic that directs the actions of both individuals and the process that they have themselves have constructed.
From a subjective viewpoint, the challenge is to explain the experience, the complaint, and the accusations made by the patient in terms of their organization. The classical outlook identifies the transferential "thing" as "unrealistic" and "distorted" elements that emerge from this communication, attempting to explain it in terms of the repetition of old experiences. Obviously, one can define an experience as "unrealistic," from both an abstract and a common-sense point of view, but neither intrasubjective nor intrapsychic considerations have instruments to determine this. Except in extreme cases, it is difficult to determine which criteria to use in formulating this judgment in a way that is both neutral and nonauthoritarian. Furthermore, explaining "distortion" as "repetition" seems to rely on the premise of a "mind" that functions according to the Freudian psychic apparatus, on the one hand; and, on the other hand, this explanation makes use of an unverifiable conjecture.
In the final analysis, in this sample interaction, regardless of any "distortion" concerning T's "interested" or "disinterested" participation, it is true, in any case, that Ms. P is "one of the therapist's cases," and that the therapist, as a matter of fact, is doing the job at hand. If we refrain from resorting to the habitual explicative tools, we might realize that explaining the patient's complaint does not necessarily require the usage of uncertain hypotheses such as repetition and distortion. It is possible, in fact, on the basis of an overall review of the patient's personality, from the knowledge of her history as well as the manner in which she generally builds up her contexts, to make conjectures that quite adequately explain why and how such communication can be considered "reasonable" within her context. In other words, it is possible to make certain hypotheses about the unconscious theories formulated by the patient that render her behavior logically comprehensible, without having to resort to (barely justifiable) "repetition." Within Ms. P's flow of experience and behavior, we can also identify redundancies (such as the tendency to consider herself uncared for), within contexts that may appear unrealistic to the therapist, but that might be entirely comprehensible according to the patient's theories.
We might advance the following formulation, which attempts to explain the transference issue from a purely subjective standpoint, while maintaining the option of incorporating a description of the operational procedures necessary to conjecture a "transference"(13): What has been understood as transference is a modular redundancy in Ms. P's action, and, more specifically, a redundancy in her modality of contextualization---and consequently also in her experience and action. Such modularity is affected by the exercise of rules belonging to the construction of contexts. These rules can be deduced from behavior and, in particular, from the analysis of noticeable isomorphisms, listed below.

(a) in the structure of the systemic organization of P, gathered from the analysis of her story, and, in particular, according to levels that are close to P's problem: the subjective and intrasubjective levels;
(b) within the structure of her intersubjective relationships, gathered by an anamnestic analysis of the vicissitudes of her behavior and of the way she organizes her life: the intersubjective level;
(c) within the structure of the interaction between P and T, from P's contextual viewpoint and from the consequent intentions behind her purposes and actions: the level of the here-and-now interaction.
Such reformulation implies that:

(1) the transference is an inferred modularity that contributes to the regulation and creation of the flow of experience and behaviors of P, in relation to the context;
(2) such modularities, as well as redundancies, represent the result of an organization whose pattern and related rules can be inferred through the analysis of the three previously indicated isomorphic structures;
(3) this isomorphism springs from the historic drift of structural couplings between an organism and its environment.(14) In other words, it is a precipitation of the history of a system's organization, and because of this, tends toward self-confirmation and feeding;
(4) the modular manifestations or the redundancies within the flow of experience cannot be interpreted as "repetitions," but are rather the actions of an organization that, in building up the sequence of present events, gains and learns quite a steady modality of "construction of the present," considered an integral part of its auto-organization and auto-construction;
(5) it is possible to carry out an analysis of the modular activity, both at a micro-event and micro-interaction level, and at a macro-organizational level of events and interactions;
(6) this modular activity is generally unconscious, and, although the result of such a process is generally conscious, it tends to hide several levels, links, or meanings that remain unconscious.

This conception allows us to overcome classical contrapositions such as present/past, real/unreal, spontaneous/reactive, and to explain the role of the past in the construction of the present, without assuming a hardly justifiable functioning of the "mind," avoiding the resort to essentialist explanations and to conceptions of "reality," both "psychic" and "real," which would prove to be not in harmony with post-positivistic epistemologies. Hence, the "present" can be constructed through the exercise of rules that, although built in the past, create every "present." In this sense, a possibly "distorted" effect would not derive from an essentialist persistence of the past, but rather, it would be the result of the unilateralism, rigidity, and closure of the organization's rules. These phenomena would not be autochthonous "projections" of the patient, but they would derive from a functional modality that is typical of every ego system (the therapist's included), which builds a reality using rules learned through the constant exercise of constructing reality. The idiosyncratic aspect of the concept of transference is not completely negligible, as it represents the nature of singularity, typical of every construction of an ego within the framework of a subject's historical drift. Furthermore, this conception, as we will soon see, allows us to evaluate the role of the context and T's role in the construction of transferences.
We now come to the matter of deciding whether a transferential thing can be identified with precision in this kit of rules, expectations, and theories. Ms. P, like every other living being, has organized a series of rules and a hierarchy of answers. One might consider that this organization represents the "thing." We all have a similar repertoire. If our train has stopped at a station, and suddenly the train next to ours begins to move, we could easily make the mistake of believing that our train is moving. To continue along these lines, if we saw the train on the opposite track also moving, we would probably experience a physical sensation of motion. As we are aware, this is because comparison with a motionless point is a tool our brain uses to perceive motion. When a "still" point "moves," the brain perceives this as our own motion. In this sense, our expectation, though often unconscious, is that "if the outside moves and we are inside a vehicle, then we are moving, while if we are still and an object, present within a context of motionless objects, moves, then the object is moving." This expectation is usually confirmed by facts.
In the "still-train-that-moves" case, one can search for another "control data"; in any case, whenever the two trains disappear from our line of vision, our own train will seem to stop, according to our perception. Reliable confirmations such as these are not easily obtainable for most of our expectations and relational attributions related to meaning which, on the contrary, for obvious reasons, tend to have a virtual, at times even real, confirmation resulting not from the truth of our conjectures but from the action that an incorrect conjecture has caused. If I infer that Mr. Brown's behavior is rude and aggressive (and I act accordingly with a similar attitude toward him), Mr. Brown will, probably, prove to be rude and aggressive. What we call transference appears to be imputable to such rules.
This type of approach to the problem, however, instead of identifying a specific transferential thing, seems instead to dilute the transference in a general soup of rules, expectations, and contexts that would seem, by molding the action of each subject, to sacrifice specificity to the notion of transference. The fact is that the basic isomorphism existing between T and P makes them homologous, since it is not only based upon the fact that they have a fundamentally similar brain, but also upon their belonging to the same systems (from the noosphere to their nationality, language, culture, and probable social status), as well as upon their both having grown up in isomorphic systems (i.e., family, elementary school class, teenage group, and so on). What strikes us as important in cognitive as well as in clinical terms, in our effort to delimit transference, is rather that characteristic of the contextualization, together with that part of the rules of expectation and inference that are idiosyncratic of P.
Proceeding in our search for the "thing," we might conclude that, more precisely, it consists of the automatic activation of these expectations, rules, and modalities in P's response. In this case, the transferential "thing" ought to be sought out in the patient's granting of a procedural scheme of relationship with T (in terms of perception, anticipation, and consequent experience). This scheme need not derive from another context (i.e., from the infantile one), but rather from an exercise of learned rules that T automatically applies in therapy. This further identification and explanation of the transferential "thing" appears quite logical and close to the traditional conceptions of transference, but, on close analysis, proves to be useless and with little clinical or cognitive significance. It is not clear, in fact, how it can be possible to identify a "pure" action of T, which can be recognized by P as a neutral stimulus for her perception, action, and "independent and spontaneous" experience. Neither is it possible to calculate the "index of distortion," or deviation, from T's neutral action or communication.
The last option is to identify the transferential "thing" in that part of rules and expectations that have a more direct implication within the building-up process of relational contexts and related experience, emotions, and behavior. This denotation is unjustified unless we can find specific rules for the construction of these contexts, rules that are different from the ones used to build up any other context. By doing so, the one "thing" we have at our disposal is the redundancy in the construction of contexts that results from the existence of relatively invariable rules. One could, therefore, settle the matter once and for all and identify this repetitive modularity with the transferential "thing." The point, however, is that this redundancy is, first of all, an inference, a "theory," or a construction made by those (either the analyst or a judge) witnessing the interaction and, therefore, it is not a "thing"; secondly, intersubjectivity, which has been so far disregarded, needs to be taken into consideration. The first part of this analysis, however, may be concluded with the following remarks:

(1) From a subjective viewpoint, it is possible to trace a pattern in a subject's use of rules to construct contexts and consequently to modulate his or her actions, experience, and interaction modalities. It is also possible to define ways of drawing this map, and there are, in fact, numerous methods to identify such constructions.(15) Nevertheless, the framework of rules, expectations, and theories that can be accessed in this way must not be seen as the transferential "thing," just as the rules of vision are not the thing seen, but aid in determining it.
(2) This inferred modularity could be referred to as the predictive theory of T about P's specific kinds of behavior. It is a sort of transferential prediction from the subject's point of view. It can be formulated as an observer's construction or theory about the functioning of what is observed, which might or might not be confirmed by subsequent interactions. Although primarily of predictive value only, such a prediction is also useful on a clinical level, since research (carried out, for example, by Luborsky [1992]) has revealed that interventions that turn out to be concordant with patients' subjective contexts are more effective.
(3) The same is true for the therapist. Even in his or her attitude, one can detect modularities in the construction of contexts through the exercise of technical, clinical, and very personal rules, by which the analyst formulates the context in which he or she inexorably moves. Psychoanalysis acknowledges this, even if still in a "distorted" way, as the concept of countertransference.
(4) Finally, from an intrasubjective viewpoint, the redundancy, deduced and analyzed through the behavior of both elements of the therapeutic dyad, cannot be forcefully detached from the interaction or the dyad's intersubjectivity.


If we introduce this further consideration, Ms. P's statements do not appear to be an autochthonous product, but rather an element of the interactive process, generated within the "T-P group," and something that cannot be carved out of or taken away from the intersubjective context. As has already been stated, in fact, aside from any other truism, the young lady is undeniably "one of the therapist's cases"; the therapist is doing his job, and on this occasion, he has delayed the beginning of the session. The intersubjective assumption implies that all perceptions, actions, and experiences of both the therapist and the patient need to be viewed as co-determined by the action of both. Hence, even setting aside the difficulties in identifying the transferential "thing," this alternative viewpoint suggests that it cannot be ascribed to something that is owned or internal to P or T; it should instead be considered something that is taking place between them. In doing this, however, the argument seems to lead to a dead end: not only does transference appear to be reduced to an inference of the observer, but it also seems to dissolve and melt in the complexity of the interaction, so that no "thing" can actually be detected. Yet, this dead end is only an apparent one and derives from our habit of searching for the "thing" in the patient's mind. The classic conception, by virtue of the functionality of the "psychic machine" and of the naturalistic framework, pushes intersubjectivity into the intrapsychic, thus transforming it into the fossilized, infantile relationship that represent the transferential "thing" itself. The object relational conception, by contrast, compresses intersubjectivity into the inner object, thus justifying actions pointing toward the outer object.
Having outlined a predictive map of Ms. P's rules, we can again switch on the recording device again and restart the virtual tape that freezes the event---i.e., her interaction with the therapist, where actions follow one another and are co-determined by both. The point is to determine whether these intersubjective considerations allow a more precise identification of the transferential "thing." In light of this wider standpoint, what can be observed is a continuous stream of interactions, which is interrupted at the end of every session. If a transferential "thing" can be detected, it will have to be found within the trajectory of this stream. Consequently, the problem lies in finding sequences which can be referred to as transferential ones, and in determining a criterion to distinguish transferential from nontransferential sequences. This might seem simplistic, but the need for such a determination is also implied in the traditional conception that seeks to identify which of P's statements or actions are distorted repetitions of past experiences. Freud, for example, had to decide which kinds of his patients' words, behaviors, and feelings were to be considered positive transference, and, conversely, which were to be regarded as resistant transference.
Similarly, those who draw a distinction between the therapeutic alliance (or real relationship) and "transference" will have to take a similar decision. In both cases, it is also important to decide which criteria to adopt as indications of transferential behavior and to make decisions accordingly.
As it happens, translation of the transferential issue into both contextual terms and contextualization rules has an unexpected double advantage. First, it succeeds in expressing, without the use of far-fetched interpretations, the real meaning of the traditional assertion that "all is transference" and "transference exists from the very first contact between T and P." In fact, the construction of any context implies a whole system of rules, and particularly important are those rules that determine the meanings that are so important to the subject when deciding, for example, whether to establish a relationship, and what kind of relationship to build.
Such contextualization and meaning-attachment activity are continuous in both partners of the therapeutic dyad, even though they may not always be detected in the same way. Under many circumstances, they could function merely as "background noise," perceivable more as an absence than a presence, coming to the surface only when a perturbing element activates a binding rule for P, thus creating a rift in the shared context.
As contexts are constructed by P and T, respectively, they cross each other in a kind of virtual interface that indicates what we might call the line of the present. In every "present," this interface represents the intercontextual situation: it sometimes highlights a shared context, as in the moment when Ms. P was completely occupied with a dream she had just recounted; and at other times, its focus is an unshared context, such as, for example, the patient's complaint about T's lack of interest in her. There may also be certain situations when a context only appears to be shared---and furthermore, there is always a context of a context, which contains and governs the development of contexts. Since it is possible to perceive redundancies in P's contexts when observed from a subjective viewpoint, it is also possible to gather redundancies and isomorphisms in the shared and unshared contexts of both parties and in their relations with the context of the context.
This, then, is the second advantage of the expression of the transferential issue in both contextual terms and contextualization rules: that is, a contextual outlook allows us to mark a distinction between transferential and nontransferential sequences through observable criteria, thus avoiding the logical contradictions peculiar to the "transference/relation" juxtaposition. From this point of view, it is at last possible to describe the transferential "thing" in a precise way, as something that can be identified within particular and redundant interaction configurations. These configurations can be described and explained both in intersubjective terms, as the co-construction of both actors, and in intrasubjective terms, as contexts and theories peculiar to each subject. Even here, however, we are not dealing with "things" that have to be placed in some recess of the mind, but rather with the application of rules.
In a way that is more concrete and precise, and thus evasive of slippery assumptions about the ontology of "things," the use of the situational method permits the detection of a specific class of problems---problems of the solution process---which, in the context of this method, fulfill the same functions as do those that the classic method attributes to transference. Thanks to this classification, it is possible to assign roles both to transferential factors (rules, predictions, theories) peculiar to P, and to transferential occurrences that materialize in the present, as interwoven components of the intersubjective plot, or in the form of the inevitable precipitation of the problem-solving process, through the interactions of both subjects. This method is also useful in comprehending resistant nuances that, according to tradition, are apt to be assumed by such transferential occurrences.
Following is a brief example of a case brought to supervision that can be used to illustrate the notion of the problem of the solution process:

The patient, Mr. M, is a young psychologist. He begins a session stating that he feels very angry. He gives an account of his dog, which, during rainstorms, seeks protection in the house and becomes calm in a room that was once occupied by the patient's brother. The previous night, during a rainstorm, Mr. M opened the door of this room to find the dog, but his father had already chased the animal out and locked the door. The dog whined most of the night, preventing the patient from sleeping. He is angry because of his father's behavior, which in his view is incomprehensible; he feels that his father treated the dog not as a dog, but as a creature that could follow the father's line of reasoning and his belief that "this is the way it should be."
The analyst states that perhaps the patient recognized in his father's behavior a way of being treated himself. Mr. M accepts this observation and dwells on his relationship with his father, admitting that his own desire to emerge and achieve important results---in addition to his invincible tendency to rebel---might be influenced by this relationship. These themes have often emerged during analysis.
The analyst then points out two aspects of this "transgressive" attitude: one is the patient's ability, curiosity, and general tendency to see things from an unusual point of view (which Mr. M considers a virtue), and the other is that such an attitude often leads him to act in a self-destructive way. Mr. M's expression suddenly becomes tense; he turns pale and falls silent for a while, then says abruptly: "It looks like we don't understand each other anymore."
Although affected by the intensity of this remark, the analyst smiles in a benevolent manner, still believing that he has merely restated som
ething Mr. M is already aware of. But Mr. M shuts himself into a deep and gloomy silence. Finally, the analyst understands that he is missing something, and after a while, he says, "Perhaps I have said something that has upset you." He adds that this situation appears similar to a previous one in which problems occurred in the analytic relationship.
Mr. M remains silent and looks more tense than ever. The analyst remarks that perhaps Mr. M has interpreted the observation regarding his transgression as a negative judgment (to Mr. M, being transgressive is an essential part of his identity). The patient denies that this is the case. He seems to be about to say something else, but is evidently in difficulty. Finally, he says that he is ashamed...because the issue is not very important. It was the analyst's smile that caused his reaction. The feeling of shame, he will explain later, was caused by his fear that the analyst would laugh at him. This explanation does not clarify the issue entirely; it will be investigated further during the next session.

This fragment of interaction is an example neither of transference nor of the problem of the solution process: it is only a well-filtered portrayal of an event that can be read and explained both in terms of classical theory and of situational logic (but also of projective identification, unconscious expectation, and many other frameworks). At the beginning of the session, both actors behaved in a spirit of collaboration: Mr. M expressed his anger to the analyst, who listened to him. Presumably, the patient felt understood, and in fact agreed to shift his attention to his relationship with his father, thus maintaining the collaboration. For the analyst, the next observation---about the two aspects of the patient's transgression---can still be placed in the same collaborative context, but the patient's response (silence, muscular tension, pallor) symbolized a break verbalized in the patient's comment that "It looks like we don't understand each other any more." The therapist doesn't notice, or rather underestimates, the change of context and continues his work of resolving the problem of understanding until his "smile" causes a sudden deterioration in the situation.
The patient's angry, tense, and gloomy response forces the therapist to enter a new context and play an unforeseen and unexpected role. He is slow in realizing the change that has taken place and risks entering a symmetrical confrontation with the patient; but in the end, he tries hard to save the situation. However, it must be said that it is Mr. M himself who comes to the rescue by bringing up his shame and his fear of another "smile," should he express the motives behind his resentment. It should be noted that, in contrast to the patient's conviction, the change in experience actually took place before the therapist smiled, just as the reestablishment of a context of fragile collaboration does not seem to have been brought about by the therapist's interpretation.
The problem that has suddenly taken place during this session exists at a different level from that which took place the previous night (Mr. M-father-dog), and even from that which the therapist believes is connected (Mr.M[dog]-father). It is a problem that changes the whole context of the situation, so much so that it needs to be resolved before the other two levels can be approached again. This problem concerns the relationship between the two subjects and their emotions and actions more than the contexts defined by their roles.
The notion of the problem of the solution process is an attempt to emphasize that, because of the intersubjective nature of the method (which implies the global participation of the two subjects), this occurrence is inevitably present in any therapeutic process. This means that the technical work cannot take place outside implicit and largely unconscious metacontexts of the actors' relationship. Therefore, in accordance with Freud's theory on defensive transference, it is the significance attached to the encounter of these metacontexts that determines the effectiveness of---and sometimes the very possibility of---analytic work.
Mr. M's emotional and behavioral change could appear to be sudden and unsubstantiated in the therapist's eyes, and might therefore be interpreted as a spontaneous, unrealistic, and unconscious "repetition." Great advantages, however, can be gained from logical, theoretical, conceptual, and even clinical perspectives if we do not treat Mr. M's experience as a "repetition."(16) I will discuss these below.
(1) First of all, the continuous monitoring of the context (and even of the context of the context) might allow the analyst to remain within the process with greater ease and in closer contact with the patient. Therefore, the analyst might be better able to perceive changes in the context and to formulate real-time conjectures about the effects that his or her words and actions are having, in the present moment, on the patient's experience of the therapeutic relationship. Had Mr. M's therapist immediately perceived a change in the context, he might have at least avoided (a) making the situation worst, and (b) increasing the resistant contraposition. More than that, he would have probably been able to produce an efficacious perturbation to the patient's expectation of noncomprehension.(17)
(2) Refusing to search for a hypothetical infantile prototype forces the therapist to explain the patient's experience of the present moment by putting together more precise hypotheses concerning rules that the analyst him- or herself adopts in the construction of repetitive contexts and on the theories the patient formulates about the emotions, experience, intentions, and even the "technical" actions of the therapist.(18) This would help the analyst avoid making stereotyped (Peterfreund 1983) or generic interpretations, and it would also increase the analyst's capacity to perceive those intersubjective configurations known in literature as projective identifications.
(3) Finally, the notion of the problem of the solution process, with its intersubjective presuppositions, allows the analyst to integrate the intrasubjective aspect (of the patient and the therapist) with the intersubjective one in a coherent and balanced manner. These two aspects must be placed in the same framework because they are indivisible. With regard to the first aspect, Mr. M's therapist could have resorted to a more immediate use of the catalogue of inferences available to him concerning the patient's overall behavior. Mr. M had often described similar behavioral reactions in his relationship with friends and partners. Sudden emotional changes with a strong element of anger, hinting at escape as the desired solution, would often arise in those relationships.
A perfect example of this behavior (together with other minor ones) had also taken place in therapy. The therapist could have included this kind of redundant behavior in his "expectancy map" and could have intervened with greater speed on that occasion. Mr. M, in fact, had a tendency to build contexts of danger (for fear of being judged, for example) in his most important relationships. He would make subjective conjectures about the (dangerous) intentions of his partner that would justify negative emotions, his tension, closure, silence, and the possibility of escape. At the same time, the understanding that every action in therapeutic interaction is codetermined stimulates the analyst to look for evidence of his or her participation in the breakup of the context.
In the case of Mr. M, the patient blamed the therapist's smile for his emotional reaction of shutting down. However, this smile occurred only after Mr. M's withdrawal, and can therefore be only a codeterminant in the increasing negative interaction. Had this session been recorded, it would have been possible to identify the moment at which the therapist, in Gill's words, exaggerates with the salt. Prolonged usage of the context of comprehension and the symmetrical proposition of collaboration, manifested in the smile, have certainly done enough to "salt" Mr. M's soup. Even the analyst has his or her own set of rules when building contexts and theories that can automatically trigger emotions leading to situations of explicit or implicit miscomprehension and antagonism. Therefore, the analyst always has a "saltcellar" ready for use. The analyst's understanding and recognition of having participated in the activation of problems of the solution process allows him or her to see the sense of "unrealistic" actions or emotions on the part of the patient. This consciousness can lead to the analyst's acting in ways that produce a powerful and benign perturbation of the patient's mental schemes and negative expectations of misunderstanding; it can act as a lever for drawing attention to what is happening in the other's mind.


Identifying "transferences" through use of the notion of problems of the solution process may appear simplistic at first sight, and perhaps too "cognitive" (according to Freudian clinical heritage), since it disregards the role of what is referred to as the unconscious, albeit imprecisely. This likely but erroneous impression stems from that aspect of the concept that, for expositive reasons, emphasizes the more external aspects of the method. An analysis of the notion of context and its relation with consciousness may provide deeper insight.
The rules governing the construction of contexts work mostly unconsciously, and the contexts themselves---except when they are the object of direct attention---are part of the background and do not appear on the front stage of the theater of consciousness. Emotions, feelings, meanings, and behaviors activated by such rules and contexts, in contrast, are generally conscious. This, however, does not prevent a particular part of the process, or the whole process, from appearing to be further removed from consciousness, because of the more specific rules governed by defensive mechanisms as classically conceived. Nonetheless, a particular type of contexts has a very different relationship with consciousness: those which have been referred to as "contexts of contexts." In the continuously operating tape of our experience, and also in that of therapeutic interaction, we find not only a sequence of regularly aligned contexts that follow one another, but also higher-level contexts that control new ones that are constructed as experiences unfold.
Let us suppose that Ms. P has a rule in her repertoire specifying that: "X knows what is good for me better than I do." Not only does such a rule invariably contextualize an interaction or communication with a person playing the role of X, but it also acts subtly on a diachronically wider scale, constructing higher-level contexts (contexts of contexts or metacontexts). Higher-level contexts, besides ruling the contexts related to single occurrences, can also considerably modify both their direction and meaning. Imagine that one fine day, the zealous Dr. T succeeds in identifying and understanding this rule and communicates it to Ms. P in such a way that the enlightenment she achieves allows her to read the chapters of her own life with fresh eyes. Well, all this does not stop this metacontextual rule from continuing to guide the interpretative context, the therapist's attempt to disarticulate the rule, or even the contexts resulting from the patient's new insight (once again, in fact, "X has known better than I did"). Such metacontexts govern a variety of possible relational contexts and, paradoxically, even those in which the rule is put on record. Mitchell (1996) gave an excellent illustration of this common occurrence in one of his recent articles, demonstrating how the therapist may behave as though he is trying to elevate himself by pulling up his bootlaces.
It is usually quite easy to identify ordinary contexts, whereas it is hard to uncover the contexts of contexts, because often their existence is not even suspected. Ultimately, on a still deeper level, we find an additional category of contexts, which are completely left out of consciousness and which act not only as microcontexts, but also, in most cases, as a sort of starting key and a production chain of contexts. On the basis of a set of reliable data, Damasio (1994) conjectured the existence of a meaning-attached mechanism called a somatic marker, which, basing itself on the memory of emotions concerning people, objects, situations---and, therefore, on general emotional expectations---chooses from among the options of "go," "stop," or "go carefully." The functioning of this mechanism may be seen as a key part of the construction process of higher-level contexts and, in particular, of those concerning relational situations and the "to-be-with" modalities. These different levels of contextualization may be thought of as fulfilling the functions that traditional psychoanalysis has assigned to unconscious processes, in both the subjective and intersubjective domains.
Apart from that, however, the impression of simplification in the notion of problems of the solution process derives mainly from the fact that one necessary distinction has not yet been taken into account in this discussion. In consideration of both the experience and the study of intersubjective transactions, it would be appropriate to distinguish a domain that we might call "interactive," encompassing what happens in the subject or among subjects, from a "meta-interactive" standpoint, which relates, conversely, to what is reported to happen by means of whatever perceptions, accounts, explanations, and theories are put forth. This distinction refers neither to intersubjective nor to intrasubjective entities. It is a conjecture that carves distinct domains out of the therapeutic interaction, in which different sets of factors active in the process can be summarized. The fundamental property of the interaction is that it does take place and cannot be erased or modified by meta-interaction (which can distort it but cannot make it not happen), but it can be experienced by and related (either to oneself or to someone else) only through a meta-interactive operation.
The essential aspect of meta-interaction, then, is that it always implies an interaction; in fact, even an interpretation, regardless of its contents, intervenes in the context as an action with its own meanings---which are not necessarily those intended by the agent. These features suggest that, instead of the linear causality we have become accustomed to as a result of the Freudian theoretical attitude, we must envision a never-ending circularity between the two domains and their respective range of factors, which seems to rule and dominate both the subject and the subject's system. These concepts have been introduced in a previous work (Scano 2000); here, it is sufficient to say that technique and technical factors need to be placed in the domain of what is reported to happen (meta-interaction), whereas the most important factors in the process of change are probably interactive ones. These latter factors are the only ones capable of promoting change in expectations and emotional dynamics, as well as of opening the way to new forms of experiencing.
Transference and countertransference issues need to be accurately placed in this complex field, where the circularity between what happens and what is reported to happen is silently organized into a sieve of contexts, from the elementary predetermined contexts of emotional expectation (the somatic marker), to those regulating the scene of the conscious experience here and now, or to those which organize hierarchically the relational strategies with regard to the two fundamental needs of every subject: to maintain one's own oneness and identity and to change what is possible and necessary in order to maintain that unique oneness and identity.
An analysis of the transference issue from the viewpoint of interactive, meta-interactive, and contexts stratifications lies outside the limits of this discussion and must be undertaken on another occasion. What has been mentioned so far is enough to dispel the impression of simplification. These ideas actually describe a complex organization in which rules operate silently, and the more they are unnoticed and unnoticeable, the more they appear to be in line with the "to-be-with" patterns or with the nonverbal scenes representing the patterns of our early experiences.
Before drawing conclusions, it is necessary to proceed to the execution of a homunculus, which has previously been accepted without question. This is the notion of the transferential "thing," which has been introduced and employed merely for expositive reasons, in order to create the necessary maneuvering room between transference as a concept and transference as a "thing" to which the concept refers. In the foregoing discussion, I have identified the "thing" in redundant interactional occurrences, the specifics of which are established through predetermined criteria. Moreover, the "thing" has been identified with the problems of the solution process, according to the terminology of the contextual method. In reality (and here is the execution!), no transferential "thing" exists, at least in terms of the meaning we normally attribute to existence. In fact, it cannot be expressed without resorting to the concepts, the delimitations, and the descriptions we give of it, which do not draw its physis, but only outline our way of understanding and describing it.
In light of this, the transferential "thing" (like the "change-thing," the "resistance-thing," and so on) consists, more precisely, in the process of delimitation, definition, and construction of the "thing-problem-transference" itself, which belongs to the observer and not the observed. In other words, the transferential "thing" corresponds to the conceptual operation that starts from a theoretical-clinical problem (i.e., the problem of transference in the current discussion) and its universe of conceptualization, and from a practical problem ("Why does Ms. P have the X problem and how can it be solved?"), and leads straight to a redefinition of the "thing" based on a different epistemological and conceptual universe. According to Freudian theory, the "thing" is the description of redundancies in terms of both the functioning of the psychic apparatus and the nonhistorical repetition of a phantasmatic image. In the formulation discussed here, it is the exercise of procedural rules of contextualization that can, in the intersubjective situation, be identified with the problems of the solution process.


At this point, one might ask how the simple notion of problems of the solution process can account for such a complex problem as transference. We might first remember that, if the notion turns out in fact to be "simple," this does not in and of itself imply a reduction of complexity in what it describes. The simplicity of a given concept depends upon its construction, formal configuration, and the possibility of its being translated in controllable operations. The complexity of the concept of transference, however, is not the result of its richness or versatility, but of the formal features of its construction, in which the contents have played an important role, and a more and more decisive one, since the time of Freud. Indeed, within the framework of Freudian theory, transference, despite its complex construction---and thanks to its connection with the topical-dynamic theory and with the stages of psychosexual development---maintains its own linearity, even though determined by an excess of reductionism. Mainly because of the need to overcome this reductionism, the role of contents later on became predominant in an almost exponential manner, doubling the psychodynamic bases which have mainly been constructed on the basis of "new" transferential contents, as documented by "new" clinical data. Consequently, because of the lack of simple and clear theoretical, clinical, and technical criteria, even the identification of multifaceted transferential behaviors has become more complicated, and has had to increasingly rely on the capabilities of each single therapist (which are not easily definable or transferable in training).
The simplicity of the notion of problems of the solution process, together with its capacity to take complexity into account, ultimately derives from the fact that contextual analysis is not only a problem-solving method; it is also one of internal situational logic fostering each developmental process, and therefore, this internal logic governs both the development of the problem for which a solution is being sought, and the solution process that T and P bring forward.
The reason why therapy ostensibly triggers this process remains a more important but obscure point. This issue is also relevant to the classical method and can be formulated in more general terms: Why does therapy set in motion those problems and processes that tradition invariably sees as transference? To the bitter end, Freud defended the belief that transference is a universal phenomenon that may be found in all human relationships, not merely an artifact of the psychoanalytic setting. His position was motivated by simple assumption, by the view of transference as a "phenomenon," and by the fact that his explanations utilized the general operative modalities of the psychic apparatus.
However, even if nowadays we reject the Freudian explanation, we can still agree with the essence of Freud's beliefs. The therapeutic setting does not generate transference, but rather brings it to the surface where it can be observed. Human subjects contextualize in all the scenes in which they act and fashion their "history," but, presumably, the "background noise" turns to perceivable music only in certain situations. This background noise seems to reveal itself in palpable configurations in the scenes in which the context of relationships, and the context of the context, become more important. On the basis of this conjecture, it should be possible to describe how and why therapy manages to trigger the process conceptualized by tradition as transference.
An important role is played by the setting, which is not merely a "container" technically suited for carrying out the job; it also seems to act as the right ecosystem for the promotion of a structural coupling. This binds P and T into a single system, as a consequence of the patient's needs, the therapeutic helping situation, the prospect of a constant duration of interaction, and the therapist's availability. The same process is seen in other human interactional systems (mother--son, teacher--pupil, husband--wife, and so on). Since the therapeutic setting is suited to facilitate structural coupling, it should be possible to identify it as a promoter of transference. It is well known, however, that structuring a setting is not always enough to initiate an efficient structural coupling. At times, therapeutic dyads crystallize into extrinsic bonds, or a kind of structural coupling develops that fails to trigger positive change processes. It is, therefore, up to empirical research to cast light on such an important issue for diagnosis and prognosis, through use of the intersubjective view to analyze the typologies of T-P systems and the range of their possible configurations.
On the basis of positively evolving processes, it is possible to assume that the promotion of structural coupling results from an essential similarity between the two systems, by virtue of a similar history of structural couplings in which these systems have developed. This similarity presupposes a substantial structural and functional isomorphism. More specifically and profoundly, however, this isomorphism can be traced back to the biopsychological systems of attachment that form the primary entry point into the semantic system for a newborn of the species, and the mold into which the various "egos" that replicate this system take shape. From this viewpoint, we can put forth a hypothesis for the "law" responsible for the activation of these deeper contexts in the therapeutic process: By virtue of its genesis within the ambit of structural couplings, the human ego learns and codifies procedural modalities and rules, binding the construction of contexts according to a "range of possibilities" determined by its "to-be-with" experiences. Such rules mold the "to-be-in" factor in every structural coupling in a relatively binding way. Consequently, in every coupling, only contexts and contexts of contexts that are compatible with the above-mentioned range will be activated. A clue to the likelihood of this hypothesis may lie in the fact that therapy seems to be characterized by different events related to attachment/detachment, love/hate, acceptance/refusal, comprehension/incomprehension, confirmation/refutation, dependence/independence, passivity/activity, collaboration/rivalry, and so forth.
Hence, all "technical" operations inevitably involve the activation of an interactive domain that places every technical activity into a context and a context of a context that are consistent with the range of possibilities of the subject or subjects involved in the interaction. In this sense, the unavoidable activation of transference is not an artifact of the psychoanalytic setting, but rather, a consequence of the impossibility of applying the analytic method in a nonintersubjective situation. The unavoidable activation of procedural codifications, therefore, does not allow both subjects to indefinitely maintain a neutral position or to evade the context of context. In this way, every "technical" action is necessarily placed in the domain of rules governing the interaction, as can be observed in the structure and dynamics of the problems of the solution process. From a contextual-method viewpoint, this placement occurs because situational logic governs not only the actions of the individual systems in their synchronic and diachronic functioning, but it also rules their actions as a subset of the T-P system and of the whole dual system.
This conception does not include any "transfer" or anything likely to be "transferred." What we must now decide is whether employing the term transference is still useful, or whether, after a long and glorious career, it can be relegated to history. There are many reasons to shelved it, some of which are based on theoretical-critical arguments, logical considerations, and concerns about research and development of both clinical theory and practice. A less in-depth analysis of the concept of transference reveals its metapsychological substance: transference for its genesis, nature, and function is strongly marked by the conceptual background from which it sprang. It lays the foundations, indeed, for the concepts of displacement and repression, as well as for the conceptualization of the relationship between primary and secondary processes, as described in chapter VII of the Traumdeutung. Consequently, transference incorporates topical, economic, and dynamic assertions that outline the "psychical machine" and its functioning. In addition, the epistemological background has changed so much in the last hundred years that we are now forced to interpret the essential ingredients underpinning the concept, in a way Freud would not have expected---ingredients such as the notion of reality itself, its "construction," the past--present connection, the genetic determination, and the meanings of "psychic" and "mental."
In light of these considerations, what is surprising is not that the idea of shelving transference has emerged, but, rather, the fact that it has not already happened. Its extraordinary longevity, of course, is due in part to the enduring survival of metapsychology, but even more to the belief among analysts that the metapsychological armor of the concept is a kind of outer suit that can be easily removed from transference, and that, because of its indubitable occurrence in every therapy, transference remains alive and florid in its reiteration. This widespread belief denies the fact that transference is a metapsychological concept, and therefore, not only does its application appear to be utterly logical and understandable, but one may also be surprised that anyone could think the opposite. As previously mentioned, the assumption that transference is a "phenomenon" carries with it the belief that it has its own existence and essence independent of its description and theoretical explanation.
A concept, however, is not a garment that can be donned or removed. It presupposes the segmentation and delimitation of a problem, and therefore, the construction of the "problem-thing" must necessarily take place within a preexisting, conceptual universe. A concept thus presupposes a definition and an understanding of the problem of which it is the explanation, and these determine the description of the "phenomenon," the coordinates, the point of view from which the problem is monitored, the kind of questions that can be posed and the direction in which answers ought to be found. From this viewpoint, we can see that transference continues to filter its metapsychological nature everywhere, paradoxically even into the minds and words of those who believe it to be dead.
This way of dealing with concepts and theories is, unfortunately, typical in psychoanalysis. Scientific practice tends to throw away those concepts that are falsified; this is why physics no longer has anything to do with "phlogiston." In psychoanalytic literature, in contrast, dozens of "phlogistons" survive because of the habit of redefining terms rather than eliminating them. This gives birth to an ungovernable transformism that makes any semantic agreement extremely difficult, filling the literature with undefined functions/words that invariably require a debate merely to clarify their meaning. As has been stated, the intrapsychic viewpoint represents a significant example of this habit. It demonstrates the difficulty that psychoanalysis has in combining identity with change. In fact, transference, in all probability, owes its longevity to the resistance to change shown by the psychoanalytic movement in the theoretical---and above all, in the clinical and technical---field. The reasons for this resistance probably lie outside the strictly theoretical framework. They originate in and draw on the strength of motivations, necessities, and dynamics peculiar to the analytic institution, as well as in the ties and bonds that psychoanalysis establishes with subjects who use theories. Such theories, indeed, are not just nets for fishing "truth-fishes" (Popper 1982) or instrumental extensions of one's arm or hand, in Piaget's words; unfortunately, they also turn out to be flags and identification cards.
For all these reasons, transference is not only an outdated concept, but also a serious obstacle to research. It implies a description and precomprehension of facts in terms that have been predetermined by an invalidated theory. On the basis of these considerations and because of its precise theoretical, historical, and technical connotations, we may be justified in hoping for the elimination of transference and for its replacement with a more operational concept based on the determination of empirically definable factors. The view described in this essay is meant as a proposal to move in this direction. I have defined the "transference-problem-thing" in a way that may turn out to be advantageous from logical, epistemological, and theoretical viewpoints, as summarized below.
(1) Acknowledgment of the conceptual and nonphenomenal nature of transference allows us to escape from the logical incongruities of the traditional conception related to the "reality-fantasy" contraposition, and above all, from notions of repetition and distortion. The "distorted" effect can be explained in terms of the unilateralism and the rigidity of the rules of organization of the subject.
(2) We must avoid repeating the irritating and unwarranted example of "upside-down epistemology" to which psychoanalytic theorizing often resorts to explain a behavior, an experience, or an action by creating a causal duplicate of what it is explaining (such as hate, dependence, object relation) as an ever-present homunculus constructed in our heads.
(3) Not only does this reformulation avoid the causal explanations of contact peculiar to positivistic epistemologies, but it also circumvents linear explanations peculiar to preevolutionist epistemologies, thus favoring a global vision of an assumption of complexity.
(4) From a point of view closer to theoretical assumptions, this reformulation safeguards the autonomy and heteronomy of systems. It explains, from an inner point of view, the so-called transferential behaviors in terms of the system's organization and closure in its unity and singularity. This, however, does not detract from intersubjective considerations and does not force the exclusion of the patient's actions from the stream of intersubjective occurrences or from the joint codetermination of events.
(5) From a strictly theoretical viewpoint, the determination of the present by the past can be explained, more generally, in terms of procedures and learned rules valid for all experiences, expectations, and actions, thus avoiding our need to categorize a specific class of actions and experiences within the framework of a subject's behavioral ability. The need to detect transactions that are specifically "transferential" would turn out to be essentially an empirical problem, both from the viewpoint of the subject---in terms of rules, anticipated actions, and theories more closely associated with P's problem---and from a therapeutic interaction point of view, in terms of redundancies which should be viewed as problems of the solution process.
(6) There is no more need to set transference against relationship. In this way, one avoids the tension in this binomial and all the logical and theoretical problems it involves, while at the same time safeguarding the positions of both the advocates of the specificity of transference and the defenders of the relationship.(19 )The conflict between insight and experience-related factors would also cease to exist, thus creating a conceptual space in which to understand the notion of new experience, as expressed by Alexander fifty years ago in terms that the transference theory could not incorporate, and later put forward by Gill (1994), even though in a theoretically undefined way.
(7) This outlook ultimately seems to reformulate many traditional statements about the transferential issue in a different and less troublesome way. Freudian assumptions about positive transference, resistant transference, transferences as neoproduction, and neuroses such as "new illnesses" and the "compulsion to repeat" can be placed within the framework of structural coupling, systemic closure, and contextualization rules. Therefore, they are easily translated into the modularity and redundancy in the construction of contexts and the contexts of contexts, thus transforming Freud's extraordinary intuitions into a coherent conception of intersubjective interaction.
(8) Finally, this viewpoint appears promising in considering the so-called "Rashomon effect" as no longer an insurmountable limit to the scientific nature of psychotherapy, but rather as an effective instrument for clinical practice and a worthy target for research. Such a direction can be undertaken on condition that: (a) one renounces the idea of identifying the transferential "thing" as something belonging to and within P; (b) one defines transference as a specific class of interactive events co-constructed within the T-P group as an effect of the encounter of both contextualizations; (c) one describes the contribution of P and T in terms of rules, predictions, and subjective theories; and (d) one deletes from the agenda a not so easily conceivable "transference resolution," in order to achieve more operative objectives of modification of procedures for the construction of contexts, which can be generated both through meta-interactive feedback about the experience of the event and through the process of constructing new meanings within the interactive domain.


Bateson G. (1972), Steps to an Ecology of Mind, San Francisco: Chandler Publishing Company.
Beutler L.E., Machado P.P., Neufeldt S.A. (1994). Therapist variables. In A.E. Bergin, S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change, New York: Wiley, 4th ed., 1994, pp. 229-269.
Cadeddu G. (1995). Per ripensare la psicoanalisi: inconscio e intenzionalità. In Scano, Mastroianni, & Cadeddu, 1995, pp. 229-308.
Damasio A. (1994). Descartes' Error. New York: Grosset\Putnam.
Eagle M. N. (2000a). A critical evaluation of current concepts of transference and countertransference. Psychoanalytic Psychology, 17, 1: 24-37.
Eagle M.N. (2000b). "A Critique of the Postmodern Turn in Psychoanalysis: Recent Work of Mitchell & Renik." Paper read at the Rapaport-Klein Study Group Annual Meeting, Stockbridge, Mass., June 10, 2000. Internet edition: (A later version of this paper, authored by Eagle M.N., Wolitzky D.L., & Wakefield J.C., and titled "The analyst's knowledge and authority: a critique of the 'new view' in psychoanalysis", appeared in the J. Am. Psychoanal. Ass. 2001, 49, 2: 457-89).
Gabbard G.O. (1995). Countertransference: the emerging common ground. Int. J. Psychoanal., 76: 475-485.
Gill M.M., (1982), The analysis of Transference, vol I: Theory and Technique. New York: Int. Univ. Press.
Gill M.M., (1983). The interpersonal paradigm and the degree of the therapist's involvement. Contemp. Psychoanal., 19, 2: 202-237.
Gill M.M. (1984). Psychoanalysis and psychotherapy: a revision. Int. Rev. Psychoanal., 11: 161-179. Internet edition:
Gill M.M., (1994). Psychoanalysis in Transition. Hillsdale, NJ.: Analytic Press.
Greenson R.R., (1969). The Nontransference Relationship in the Psychoanalytic Situation. Int. J. Psychoanal, 50: 27-39.
Etchegoyen R.H. (1986). Los fundamentos de la tecnica psicoanalitica. Buenos Aires: Amorrortu editores .
Hoffman I. Z. (1983). The patient as interpreter of the analyst's experience. Contemp. Psychoan., 19, 389-422.
Hoffman I. Z. (1998). Ritual and Spontaneity in the Psychoanalytic Process: A Dialectical-Constructivist View. Hillsdale, NJ: The Analytic Press .
Kernberg O.F. (1998). The influence of the gender of patient and analyst in the psychoanalytic relationship. J. Am. Psychoanal. Ass., 48: 859-883.
Laplanche J., Pontalis J.B. (1967). The Language of Psychoanalysis. New York: Norton, 1973.
Luborsky L., (1992) Teoria e tecnica nella psicoterapia dinamica: i fattori terapeutici e la formazione per incrementarli. Psicoterapia e scienze umane, XXVI, 3, pp. 43-56.
Luborsky L., Diguer L., Luborsky E., Singer B., Dichter B., Schmidt K.A. (1993). The efficacy of dynamic psychotherapy: is it true that "everyone has won and all must have prizes"? In Miller N., Luborsky L., Barberj.P., & Docherty J. (editors), Psychodynamic Treatment Research. A Handbook for Clinical practice, New York: Basic Books.
Maturana H., Varela F. (1980). The Tree of Knowledge. Boston: New Science Library.
Mitchell S.A. (1996). When interpretations fail: a new look at the therapeutic action of psychoanalysis. In L. Lifson (editor), The therapeutic action of psychodynamic psychotherapy, Hillsdale, NJ: Analytic Press.
Mitchell S.A. (1998). The analyst's knowledge and authority. Psychoanal. Q., 57: 1-31.
Peterfreund E. (1983). The Process of Psychoanalytic Therapy. Models and Strategies. Hillsdale, NJ: Analytic Press.
Popper K.R. (1963). Conjectures and Confutations. London: Routledge & Kegan Paul.
Popper K.R. (1972). Objective Knowledge. Oxford: Clarendon Press.
Popper K.R. (1982). The Open Universe. An Argument for Indeterminism, from the Postscript to the Logic of Scientific Discovery. London: Hutchinson.
Popper K.R. (1983). Realism and the Aim of Science, from the Postscript to the Logic of Scientific Discovery. London: Hutchinson.
Rapaport D. (1967). The Collected Papers of David Rapaport (M.M. Gill Ed.). New York: Basic Books.
Rapaport, D. (1977). Il modello concettuale della psicoanalisi. Milan, Italy: Feltrinelli.
Renik O. (1993). Analytic interaction: Conceptualizing technique in light of the analyst's irreducible subjectivity. Psychoanal. Q., 62: 553-574.
Renik O. (1996). The perils of neutrality. Psychoanal. Q., 65: 495-517.
Saraval A. (1988). La tecnica classica e la sua evoluzione. In A.A. Semi (editor), Trattato di Psicoanalisi, Milano: Cortina, vol. I.
Scano, G.P. (1995). Transfert o relazione? Note critiche sul concetto di transfert, in G.P. Scano, A. Mastroianni, & G. Cadeddu, Psicoterapia psicoanalitica: verso una rifondazione, Milano: Angeli, pp. 150-166.
Scano G.P. (2000). La nozione di interazione e il suo significato per il metodo e la teoria. Psychomedia,
Scano G.P. (2003). Metafora e costruzione del senso. Psychomedia,
Scano G. P., Mastroianni A., & Cadeddu G. (1995). Psicoterapia psicoanalitica: verso una rifondazione. Milano: Angeli.
Wallerstein R.S. (1990). Psychoanalysis: The common ground. Int. J. Psychoanal., 71: 3-19.
Wallerstein R.S. (1988). One Psychoanalysis or many? Int. J. Psychoanal., 69: 5-21.


* The author thanks Owen Renik and Paolo Migone for having read the manuscript and given helpful suggestions.

1 Gill (1982) underlines the lack of clarity in the concept of transference and the variety of definitions existing in our literature---where no distinction is made, for example, between a facilitative and hindering transference, or between resistance against awareness of transference and resistance against resolution of transference. He discusses other important divergences, such as those about the importance of transference, the role of reality in its development, its omnipresence, and so on.

2 In a recent historical-critical study (Scano, 1995), I discussed definitions put forward by Laplanche and Pontalis (1967), Saraval (1988), and Etchegoyen (1986).

3 Gill (1982) demonstrated that this is not always the case---that is, that a traditional position is in fact compatible with various methods of clinical practice.

4 An early proponent of this compromise position was Greenson (1969).

5 See, especially, Gill (1982, 1983, 1994), Hoffman (1983, 1998), and Renik (1996).

6 See the results of a study I made a few years ago (Scano 1995), in which I conducted a historical-critical and theoretical-critical analysis of the concept of transference (at least in its original Freudian iteration).

7 Rapaport (1977), in his underrated study on methodology, wrote: "Transference and resistance are not empirical observations, even though these two terms can be used to describe empirical observations. The problem is that the terms transference and resistance, in psychoanalytic theory, are not made to describe empirical observations; they are concepts that draw together a multitude of dynamic variations of phenomena in a theoretical framework" (1977, p. 122, my translation). And more emphatically: "Transference we have hereby defined is not a phenomenon; it is a theoretical explanatory construction derived from the method of interpersonal relationship adopted in psychoanalytic theory" (1977, p. 119, my translation).

8 The concept of transference, together with the 1914 concept of narcissism, is the main offspring of psychological-clinical needs that are less compatible with a metapsychological framework. Transference and narcissism have guaranteed the survival of metapsychology, avoiding its implosion by introducing other needs into the system. Transference has carried out this role mainly in the clinical sphere, while nonetheless allowing for the consideration of intersubjectivity, while narcissism has guaranteed its survival in the theoretical sphere due to the problem of the subject-object relationship, which had to be addressed between 1910 and 1915 in considering the problem of psychosis (Scano 1995).

9 Because metapsychology did not have the theoretical equipment necessary to explain the current concept of intersubjectivity, Freud considered the latter an epiphenomenal "symptom," like a "fossilized" relationship that is reactivated in the analytic setting. This description is reductive not only from a general epistemological point of view, but also in the sense that it reduces the dyadic to the monadic and the present to the past. The Freudian conceptualization of transference, however, not only brilliantly resolved a difficult theoretical and clinical issue, but also revealed itself to be extremely rich in heuristic terms, by managing to deal with (for the first time, in scientific circles) the formation of subjectivity within intersubjectivity---a problematic development, even though a natural one.

10 Even the concept of repetition, which has always been considered "descriptive," should gives us grounds for doubt. Considering an event as a repetition of the past, or even attempting to explain it in terms of the past, is a logical operation that should not be confused with the observation of a phenomenon. We are talking about an explanation that requires a theory capable of demystifying this extraordinary psychic performance, both in terms of the general functioning of the mind, and, more specifically, in the way past and present are articulated in the formation of the subject. This is not something that is perceptible and verifiable, but rather, a conceptual operation that requires a theory of the mind and presupposes a framework of Freudian metapsychology.

11 Today, we may tend to prefer the term interaction to relationship.

12 This does not imply that observation is irrelevant, but simply that it is impossible to observe without making preconstituted, implicit conjectures.

13 Because this formulation considers the concept of transference from a purely intrasubjective point of view, it cannot in any way be considered a definition per se, but is rather an approximate description of what can be identified as an expectation of transference, built upon an analysis of past redundancies.

14 The notion of structural coupling (Maturana and Varela 1980) is used in biology to indicate the relation of structural congruency between an organism and the environment in which its ontogenesis takes place, or that between a cell and a multicellular unit, or that which takes place when two or more organisms constitute a network of recurrent and stable interactions during their ontogenesis, subsequently forming systems or units of a superior, specifically "social" level.

15 There are at least seventeen ways of studying and developing constructions similar to the above-mentioned one. The most well-known are the CCRT (Core Conflictual Relationship Theme) by Luborsky, the Plan Formulation Method by Weiss and Sampson, the Frame Method by Dahl, and the PERT (Patient's Experience of the Relationship with the Therapist) by Gill and Hoffman.

16 This does not imply either the negation of the importance of the past in determining the present, or the usefulness of rebuilding the past in order to understand the present, but only the refusal to accept an essentialist and traumatic conception of this determination. Rules and negative expectations are built in the past and tend to construct the present according to an internal logic of their functioning, in a much more complicated and indirect way than can be explained by the notion of repetition.

17 I cannot make a more detailed clinical analysis of this fragment of interaction here, but I would like to note that, apart from the meaning and specific motives that determine a change in the context, even the lack of perception or underestimation of a change by the therapist can be considered an "intervention" that can justify an experience of noncomprehension and confirm a patient's negative expectations.

18 Classical theory, in general, does not help the therapist in this job. If transference is a "repetition," in fact, the crucial element for its understanding ought to be "that which is repeated," and the therapist should concentrate on this. The increase of unrealistic, phantasmatic, and spontaneous aspects within the transference may make one believe that the sudden breakup of the shared context is proof of the existence of an irrational, preconceived transference existing in the subject. This is actually a circular argumentation: one identifies the explanation that creates the proof that confirms the explanation, i.e., there is a transference that causes irrational behavior and irrational behavior is proof of the existence of transference.

19 If we distinguish a "nontransferential relationship" (a work alliance, a therapeutic alliance, or the like) from a "transferential" one, we find ourselves unable to distinguish this relationship, conceptually and genetically, from transference. If, alternatively, one decides that "everything is transference," we do not have to define what transference is not, but instead to explain how change is possible and how transference can overtake itself---for if the present is all transference or repetition of the past, how can there be a future that is not of the past?