2
The Psychoanalytic Subject

LACAN'S WORK IN MANY WAYS REVOLVED AROUND THE ISSUE OF THE STATUS of "the subject"—the Cartesian subject, who self-consciously knows he exists. But he was not unique during his time in wanting to refocus psychoanalysis on questions of identity, self, and existence, as so diverse a group as Erikson, Winnicott, and Kohut demonstrates. The commonly voiced criticism that Lacan was behaving more like a philosopher than a psychoanalyst in his seminars suggests the very complacency and banalizing tendencies that motivated his "return to Freud." For him, Freud was not simply a good clinician with old-fashioned scientific ideas, but a revolutionary thinker who exploded certain kinds of received wisdom about human beings, notably the belief in a unified subject or a self-directed conscious ego that was (and is) fundamentally an extrapolation of theological beliefs. Moreover, Freud proposed a model by which conscious experience was the effect of unconscious mental structures, residues of otherness in ourselves (other subjects, inborn patterns of relatedness, moral imperatives, drives, and so forth), an otherness which Lacan separated from an imaginary biology of the Id. Coded biological sequences might determine the behavior of many species, but human life, as philosophers have always puzzled about, seems to function by other rules and motivations with a peculiar status—motives like "pure prestige" or desires for recognition or for self-abnegation.

This is why analytic psychotherapy is mostly an ideographic discipline. It does not follow general rules of treatment applicable across the board to patients of one category or another, but focuses on the unique individual history that has produced a single patient. Cookbook assessments, abstract typologies of patients, or DSM IV-like attempts to match patients to specific treatments all miss the point if they stray too far from this principle. Even, in my opinion, the Lacanian tendency to refer to "the hysteric" or "the obsessive," although full of interesting observations, does not offer much direction in working with real cases.¹ Theory, although necessary for any understanding and modeling of what is happening clinically, has to be applied and worked with. Without theory, however, we are left in the equally precarious position of practicing ritualized techniques or applying some kind of ideology of help. Casting his net as broadly as possible to capture ideas beyond the Freudian paradigm of instinctual pressures, Lacan struggled with understanding important human questions. He probably broke more new ground than any other analyst after Freud (Melanie Klein may be one exception to this generalization), but he did not help us attach his theories to clinical practice. Especially since the matters concerning him were also approached by many others, we need to weigh his contribution with theirs.

In what follows, I trace a few major theoretical issues pertinent to the clinical problem of "having a life." I first discuss Lacan's conception of the mirror stage and its relationship to Winnicott and Kohut's use of the mirror metaphor, as well as his theory of the imaginary and symbolic orders. I then address the value and limitations of Kohut's stress on empathic responsiveness and the selfobject relationship. The symbolic function of a "third" beyond the dyadic relationship is explored through the work of Green. Finally I turn to the place of affect, rather neglected by psychoanalytic theory, and attempt to show its symbolic function, as an extension of a Lacanian model.

The Mirror Stage

One place where Lacan did use biologic references was around the structuring effect of the visual gestalt, for example, on imprinting or mating behavior in birds. An animal is sensitive to a particular configuration of shapes or colors that sets off an automatic response, a neurobiological function virtually absent in humans. Of course there must be preprogrammed sequences in early life, but Lacan generally regarded these as fundamentally irrelevant for psychoanalysis, which is interested in a human subject who arises not out of a biological program but from contact with other subjects and from a cultural system that structures this contact. Yet, in the case of the mirror phase, Lacan (1949) proposed what seems like an inborn developmental stage. Following the observations of infant researchers of his time, he noted the excitement of the 18 month-old infant who identifies with his own visual gestalt in the mirror and hypothesized that this experience of wholeness forms the foundations of the ego. Even in this example, the infant seems already to have emergent psychological properties that grow out of the preceding mother-child matrix. The important point for Lacan to which we return many times was the opposition (later the complementarity) between this mirroring identification of "that is me," as an objective or concrete false self, and the symbolic framework of language, history, and society, which creates the speaking, named subject.

From a Lacanian perspective, subjectivity is an effect of language. Clearly, without the symbolic structure of differentiated signifiers, an individual identity would scarcely be possible. We depend on a proper name and a constellation of abstract terms to locate ourselves in the human world, and we must navigate through language to satisfy our drives and desires. On the other hand, this preeminence of language, of the signifier, without question conveys an abstract and austere conception of human beings that has been repugnant to many readers. The seeming absence of affect in this theorization stands out as a major deficiency of Lacan's work. Of course, it is often said that in Freudian terminology word representations carry the energies that will structure our conscious experience, including those affective expressions central to our subjective coherence and continuity through time. From this perspective, we could say that affect always follows meaning—what moves us is what holds personal significance for us—and that meaning is, therefore, an effect of the signifier. Yet affect can sometimes overwhelm meaning, just as drive expression in Freudian terms can be destructive of boundaries or limits if not modulated by containing structures. Green (1977) insisted on this point, comparing disruptive affect to "a river overflowing its bed" (p. 206). For Freud, affect was the conscious expression of bodily sources of excitation linked with word representations that permit an emotional state to be consciously known and communicated with others. Proper control and "binding" of excitation, as we know, was considered by Freud to be a basic function of the psychic apparatus.

This brief sketch of a Freudian model serves to alert us to some difficulties in understanding the dilemmas of the speaking subject. If affect is important, how can it be integrated with the powerful notion of symbolic systems that seem transbiological (in the sense that culture stands beyond biology)? Is affect inborn like the gestalts to which Lacan referred and to which Pinker (2002), for instance, seems to be alluding from his evolutionary perspective? Does this conception have anything at all to do, as Freud believed, with the notion of aggressive and sexual drives? Or, could affects themselves be considered cultural products? There is an important difference between a limited number of inherited expressive facial configurations and the experience and understanding of affect (see Stern, 1985, pp. 64-68). In this regard, infant researchers infer mediating functions in the mind through which physiologic reactions become subjectivized and translated into communicable experience. As I develop in chapter 3, I regard Freud's construct of the ego ideal as describing this function, specifically by internalization of transgenerational cultural models providing the imagery, metaphors, and narratives through which emotion can be intelligibly or permissibly expressed.

In any event, it does seem a relatively common clinical finding that many severely damaged or traumatized patients appear to be trapped between powerful emotional urges and the seeming impossibility of their expression or satisfaction. This type of impasse seems to be characteristic in particular of many of those patients who describe something lacking in their existence—the paradoxical feeling of "not being in life," of not existing nor feeling real as embodied, vital beings. They may complain of having no feelings or, conversely, of a panoply of disconnected and senseless emotions that leave them rudderless on an unending turbulence. I include in such cases, among numerous possible examples, Mitchell's (1991) description of "problems of self-definition" of a "postmodern multiplex and discontinuous self" and Bromberg's (1998) evocation of "discontinuous self-states," although their characterizations rely on a Sullivanian paradigm of relational selves, which I find incomplete. Fairbairn's (1944) portrayal of the schizoid sense of futility, elaborated by Guntrip (1971), also addressed this problem, although Fairbairn's description became bogged down in cumbersome internal object mechanics. Guntrip (1971) wrote about "the schizoid problem" concerning "the basic reality and viability of the central core of selfhood in the person" (pp. 150-151). Their wishful postulation of a pristine, "unsplit" self seems problematic to me. Kohut, as I develop, spoke of a lack of self-cohesion as the central deficit in narcissistic patients, but may have conceived of the notion of a "superordinate self" in a too-literal way, perhaps embracing the familiar ideology of an "American way of life" of wholeness, initiative, and mastery. Eigen (1981) most closely approaches my position in this chapter with his powerful interpretation of "faith" as the central concern of Winnicott, Lacan, and Bion in the subject's struggle to be. He wrote, "for Winnicott, the essential battle is over one's sense of realness: does one feel real to oneself or merely a phantom or splinter self?" (p. 425).

Kohut: The Cohesive Self and its Avatars

Among other modern theorists, perhaps Heinz Kohut (1977) came closest to grappling with the kinds of clinical problems presented by patients for whom, as Little (1985) wrote, "anxieties concerning existence, survival, or identity predominate" (p. 15). Through his notion of the cohesive self, Kohut provided a way of understanding the intuitively familiar but hard to grasp qualities of "centeredness," "having it together," and emotional availability which are often notable by their absence in narcissistically damaged patients and address basic issues of agency, relatedness to others, and subjective vitality too often neglected by ego psychologists. Moreover, in the concept of the self object, he recognized the importance of abandoning the model of an encapsulated subject as pure interiority and recognized the necessary role of others in sustaining the self. These features of Kohut's work offer bridges to Lacan, a connection examined by Muller (1989), who has drawn attention to their shared critique of ego psychology and the importance for both thinkers of mirroring phenomena and, in different ways, of desire as a measure of authenticity in the subject. In the following section, I take the position of assuming a common clinical object that both thinkers attempted to describe through different vocabularies. I hope to show why we need to read Lacan to go beyond Kohut, but also how Kohut helps us to rethink Lacanian principles.

In many ways, what Kohut called the cohesive self is best known through its failures or absence, which he addressed in his analysis of narcissistic disorders. For Kohut (1977, 1987), the developing "self" is at risk of fragmentation if not supported by types of interchanges with its objects (or with selfobjects) that he called "mirroring of the self" and "targets for idealization" (p. 185). In most cases, according to him, mirroring relates to maternal acceptance of grandiose ambitions in the child, which provides a kind of container for what might otherwise be disruptive wishes, whereas idealization involves the paternal object. The content of these developmentally primitive, grandiose ambitions is unclear but seemed for Kohut, as I read him, to have mainly to do with narcissistic fantasies of omnipotence and with exhibitionistic needs. Perhaps mirroring implicitly entails a necessary sustaining of illusions or of transitional realities, as Winnicott theorized, a necessary step in the development of an infant's sense of aliveness and spontaneity. The sexual or drive aspects of "ambition," still following Kohut, are secondary to lack of selfobject response as "disintegration products." Idealization or idealizing selfobject relations, on the other hand, involve principally a transfer of infantile narcissism onto (usually) the father, which creates what he termed a "pole of ideals" toward which the self can strive and, in this respect at least, appears closely related to the functions of the ego ideal. From these assumptions or hypotheses flows Kohut's (1977) formulation of the "tension arc" between the two poles of ambition and ideals as the basis for what he termed the bipolar self—a self "driven by ambitions and led by ideals" (p. 180).

There seems to be widespread agreement that the phenomena reported by Kohut are common in the treatment of narcissistically damaged patients, who, he would say, have not established a firmly cohesive nuclear self. These persons have lacked the requisite selfobject experiences and have thereby been obliged to resort to defensive measures to protect themselves against the threat of disintegration anxieties, a fundamental experience for those lacking a secure sense of personal existence. Moreover, Kohut's concept of a tension arc contributes a powerful dynamic metaphor for the process that sustains the sense of having a life. I conceive of the tension arc as an ongoing flow of personal desires toward a tolerable and communicable means of expression that stabilizes or shapes the experience of aliveness. In terms of theory, however, we could question whether Kohut's narrow and vague definition of ambitions is a reduction of a richer psychoanalytic understanding and whether the model of paternal idealization is adequate to address the phenomenology of the ego ideal (not explicitly discussed by him). It is also puzzling that Kohut did not refer to Winnicott, whose notion of the true and false self echoed a long tradition of concern about authenticity in the Western philosophical tradition. Finally, although Kohut's notion of the self has proven useful in analytic practice (he without a doubt was responsible for the attention devoted to the self in North America), it carries a certain false concreteness, as though referring to an entity of some kind (a conception that has its defenders, of course).

In place of Kohut's concept of a cohesive self, suggesting a firmly established homuncular nucleus, with misleading connotations of wholeness and "presence" (Derrida, 1974) or of an essential unity (Khan, 1974),² it may be useful to think in terms of a project of subjective coherence and selfhood. Self might be better considered as an emergent but ill-defined shape, perhaps closer to the Lacanian metaphor of a chain of sustainable discourse through which desire as a free flow of affect can insist. The self has to be told or narrated, as Schafer (1992) has recently insisted, reminding us that there is no "entity" of self existing apart from the discourse that conveys it. Narrations about the self, of course, are subject to the effects of desires and defenses against it, and to all the contradictory demands and pressures of subjective existence, which make any kind of substantial coherence a mythic or idealized belief. Of course, the concept of desire as used by Lacan is not identical with affective expression. No doubt that affect can disguise desire or misdirect it. Yet it seems hard to imagine the vitality of desire as a unique human process animating a state of aliveness without interweaving it with the pulse of emotional arousal that is an inextricable element in every communication. In referring to a sustainable discourse, I employ Lacan's shorthand term for the very complex process of ongoing symbolization and construction that his difficult notion of "full speech" suggests. In full speech, according to Lacan, the subject can come closer to his desire, which above all is to achieve recognition from the other. This conception retains the dynamic structure of Kohut's tension arc, continuously translating desires (we need to develop this concept, as well) "upward" to a more elaborated level of symbolic expression.

In contrast to what might be termed "a well-functioning tension arc," analytic practitioners are familiar with the desubjectifying, disorganizing effect of powerful emotions that sometimes seem to follow a course of their own, as if under the sway of a repetition compulsion, restaging an emotional scenario that seems outside of chronological time. In such cases, in lieu of remembering, the patient enacts a massive transference repetition, collapsing past and present so that space for thought or "making sense" is unavailable. Without the therapeutic alliance Freud (1920) described in "Beyond the Pleasure Principle," in which the analysand retains "some degree of aloofness" (p. 19) and capacity for self-observation, pure reliving of traumatic experience is usually disasterous for analytic work. Affective discharge that is not contained by a structure merely repeats a traumatic situation from the past in a slightly altered form, which only leads to greater subjective disorganization and fragmentation of self-experience. In the worst case, when there is no analytic space in which to talk about such episodes, treatment comes to a premature end via a suicide attempt or hospitalization.

In his early work, Lacan (1953, 1953-1954), like Kohut, was responding to the neglect of the human subject by what he viewed as the mechanistic excesses of ego psychology. He turned to a rich philosophical tradition to find some support for a psychoanalytic theory outside the natural sciences beloved of Freud. In this quest, he was notably influenced by Heidegger's notions of authenticity in his conception of full and empty speech. In full speech, desire apparently flows or "insists" along the chain of signifiers animating the subject (Lacan, 1972-1973), whereas in empty speech we hear the effects of what Loewald (1980) was to call "a lifeless nimbleness" (p. 190), an agile display of representations devoid of affect. Khan (in Winnicott, 1986), elaborating on Winnicott's equation of boredom in the listener with psychopathology, referred to "an interminable verbiage" that petrifies narrative. Khan wrote, "the patient who compels boring narrative on us is not letting language and metaphor elaborate or change his experience" (p. 3). Empty speech presents what Lacan labeled "the wall of words" in his famous schema L, a flow of defensive and closed talk that maintains an imaginary (illusory) sense of self. This seems extremely close to Khan's formulations and to Eigen's (1981) work, which explicitly addressed the issue of authenticity. Lacan sought at this point in his theorizing to find a way around this wall, so that communication could become open to the "otherness" of desire and to the symbolic sources of subjectivity that had been repressed or barred. In fact, seeing the problem of empty speech, perhaps as did Heidegger, as a widespread symptom of the alienation of modern man, he imagined that psychoanalysis might offer a broader help to society than a simple treatment of individual neuroses.

As I have suggested, probably Kohut and Lacan were responding to the relative neglect of the issues of subjectivity, selfhood, and authenticity in classical ego psychology. By proposing the ego of the structural model in place of concepts like self or subject, which it may have mistrusted as excessively philosophical and unscientific, ego psychology lent itself to a mechanistic approach. The ego, in this conception, uses its innate synthesizing capacity to mediate between the different psychic agencies, which are its masters, by constructing an ever-changing kaleidoscope of self and object representations, continuously equilibrating the tensions within the system. From the perspective of agency, this ego functions essentially like the virtual subject of artificial intelligence, which is to say like no subject at all!³ At best, by positing mechanisms of secondary autonomy resulting from the interaction of drives with an average expectable environment, ego psychology offered an ethology or psychobiology of adaptation, rather than a "tension arc" of existential uncertainty and risk.

Lacan and, in different terms, Kohut, believed that a narcissistic object relationship—a recourse to the mirror as a defining structure of the self—was a response to the threat of fragmentation. That is, in the face of a threatening encounter with another, unknown subject that could overwhelm the capacity of the individual to symbolize and contain subjective experience (his affects, desires, fantasies), he or she seeks support for a failing sense of cohesion by restricting the other to the function of a mirror. Here, Lacan was greatly influenced by the philosopher Kojeve's lectures on Hegel, which he attended during his analytic training years.⁴ The encounter with the other as another subject, as portrayed in Hegel's famous parable of the master and the slave, threatens the defensive, narcissistic organization of self by the inevitability of being seen and redefined through an unfamilar gaze. Solomon (1983) presented the apt illustration of a solitary hiker contemplating an inspiring view who is suddenly interrupted by the presence of another person. Suddenly his experience is disrupted and changed. In Lacan's (1949) conception of the mirror relation, the subject attempts to assimilate the other to a familiar internal schema of self and object images in order to master the threat of the unknown other. In the imaginary (mirror) relationship, the object will function as a confirmation of the desired self-definition of the subject (or, conversely, he will identify with the imago of the object). From this narcissistic position, the subject attempts to project (or to assume) a preferred image of wholeness, which will be validated by its reflection in the other. The other (another person, another subject, the analyst) then becomes the means to enact either a reflected confirmation of an ideal self or the reciprocal role that self demands. In Kohut's terms, the other becomes a pure selfobject, propping up a weakly cohsive self. Lacan called this type of relationship "specular" or imaginary, unrelated to the structural and symbolic realities that define the subject and alienate her from her desire by reinforcing an ego whose function is to distort reality and to obscure desire. Who one is, authentically, in "full speech," is not who one appears to be or tries to impress others as being, and the latter "inauthentic self," as we know from extremely narcissistic patients, can manifest itself in either positive (grandiose) or negative (self-critical) forms. In what follows, I present two pared-down case vignettes to illustrate these mirroring concepts.

Two Examples of Mirroring Transferences

Mr. Small described a life-long feeling about himself that he was empty and unworthy. His mother had criticized him as too slow and an academic mediocrity, despite his fairly good scholastic record. He believed that he never did anything well enough and saw this as an intrinsic personality characteristic, an attitude that colored most of his life experience. In his analysis, Mr. Small insisted on his deficiency and believed that he was probably incapable of becoming a good patient who could meet his analyst's requirements. When the analyst (whom I supervised) attempted to encourage Mr. Small to say his own thoughts and feelings instead of striving to meet some imagined standard of performance, he felt admonished for being unoriginal and "coming up short." The analyst's subsequent line of interpretation that Mr. Small carried a fantasy of inadequacy dominating all his relationships was likewise taken as a criticism. He associated these and other interventions to his mother's well-meaning efforts to improve him, which seemed to reinforce his assumption that the analyst knew what he should be saying and doing in the sessions. Here, the analyst seemed stuck in a reciprocal interaction pattern that "mirrored" the patient's self-image. The fantasy of being objectified as defective in mother's eyes sustained Mr. Small's conscious ego identity, while unconsciously, no doubt, involved him in a particular kind of satisfaction in relation to her (in the dual relation characteristic of the mirror). The analyst was participating unwittingly in various ways in an old pattern (a relationship in which the subject was criticized as not good enough), which, at this point in the treatment, was reenacted and not able to be understood. Fortunately, she came to recognize her refusal of his transference projection of a perfectionistic (m)other and began to respond more sensitively to Mr. Small's fantasy of another maternal rejection.

Mr. Grand came to analysis because of his failure to sustain close relationships with his partners, in whom he always lost interest. Friends told him he was too critical and self-centered in these affairs. Although he did express a high opinion of his talents and accomplishments, he sensed a difficulty in making emotional connections, which he characterized as having been cool or even nonexistent in his family. His parents, in fact, despite his enormous early promise and the fine education and training they had supported, seemed unappreciative of his accomplishments. Initially, he viewed his analyst as a master in the area of human relationships who could show him the royal road to full self-realization and success. Holding an elevated notion of her importance in their community, based on a superlative recommendation by a former classmate, he hoped to learn the secret of her greatness. Unfortunately, these positive anticipations soon ran into feelings that the analyst was not taking him seriously enough and was withholding the direction and advice he needed to progress. This opened the way to some productive exploration of his disappointing relationship with his mother, where the analyst's empathy was reassuring. Unfortunately, he seemed to hear this as confirmation that his mother had withheld something that the analyst could now provide. When he began to press her more and more impatiently to provide these "answers," the atmosphere of the sessions turned more tense. The analyst continued without much success to encourage Mr. Grand to explore his own thoughts, a move he regarded with some validity as an evasive technique. In a moment of exasperation, the analyst observed that he was assuming a knowledge on her part that did not exist, something that could only be worked out by him alone. Eventually she disclosed that she truly did not have the answers he sought. In his frustration, Mr. Grand now began to attack the analyst and the process of treatment itself. After a vacation break, he decided not to return.

With Mr. Grand, the analyst seemed to be doing well in exploring what he had attempted to secure from his unresponsive mother. She did not see the repetition of his powerful idealization of a perfect mother in whose eyes he sought unsuccessfully to be lovable. With the intensification of his unrealistic demands, the analyst retreated to the position of "a real object" from her idealized transference role, which Lacan called the "subject supposed to know" or the "supposed subject of knowledge." According to Lacan, it is this basic transference that elicits a patient's demands and incites a desire that had been repressed. In this case, the analyst's appeal to Mr. Grand's observing ego and a nonexistent therapeutic alliance led to a reversal of the mirror, with the patient now moving into the position of the ideal ego (the imago of the mother), while his analyst was left to represent his devalued self.

Mr. Small and Mr. Grand illustrate the reciprocal relationship characterizing the mirror transference in which the fantasy of a perfect self is at stake for a narcissistically fragile subject. These men sought confirmation of their ideal egos from their analysts, who initially tried to disentangle themselves from the transference position. Mr. Small was encouraged to express his own ideas, whereas Mr. Grand was asked to renounce his unrealistic demands. As we have observed, the dual relationship can vary in tone, depending on the locus of the identification (ego merged with ideal self or facing an idealized object), and can easily be reversed. Thus Mr. Grand could feel either inadequate and small or superior and inflated in the regard of his therapist, depending on which position in the relationship he had assumed. The important thing is the rigidity of the dyadic mode replayed in the here and now, shutting off access to a dynamic and painful personal history, including disappointments and crucial failures of empathic response by important others. For Lacan (1953-1954), if the therapist does not abandon her position as the supposed subject of knowledge (thus representing the Other in the patient's transference), she will uncover the history behind the frozen images of the mirror and, eventually, destabilize this imaginary relation. As we will see, Lacan's conception of therapeutic action at the time revolved around the restoration of the deficient ego ideal and symbolic identification. Mr. Grand's therapist slipped from this position by attempting a "real" relationship (real, of course, only in terms of her needs for a reasonable ally, not necessarily the patient's), which may have left him feeling abandoned and needing to defensively devalue her. We might say that her gift of a caring relationship and offer to identify with her more reasonable ego failed. Had he remained in treatment (the accident of the vacation might have been fatal here), they might have been able to explore this enactment, which was clearly related to the problem for which he had originally sought treatment.

All therapists are tempted from time to time to appeal to the "observing ego," to educate, or to cajole the patient into giving up unrealistic transference expectations and taking responsibility for working together to solve a problem. To state this another way, there is an inevitable oscillation between the analyst's transference position as "subject of knowledge" (the Other who knows the meaning of the patient's symptoms) and her "irreducible subjectivity." When the analyst's subjective response grows out of her own needs or discomfort, it can break the mirror and, as in Mr. Grand's case, lead to a reversal of imagos (a projective identification) that can be hard to interpret or undo. Many times, however, this misattunement can be acknowledged and explored, like any empathic failure, with considerable therapeutic benefit. Conversely, simply remaining within the mirror relationship can perpetuate an endless narcissistic longing that blocks insight and memory. There seems to be the following paradox in the Lacanian theory of the therapeutic mirror: Mirroring creates a transferential tension that must be sustained for deeper developments to take place, but, as an end in itself, it can only stalemate this process.

The Lacanian position relies heavily on an interpretation of mirroring as misrecognition and distortion, supporting an imaginary ego that systematically falsifies reality in order to sustain a narcissistic illusion of wholeness. In Lacan's early writings, he saw mirroring as a trap for both analyst and patient, which had to be subverted by the speaking relationship (in opposition to the specular, fantasy one). Later, the imaginary dimension was accorded more value in the construction of subjective experience, as a link to the "real" that underpins it (Julien, 1985). Reading Lacan through Kohut, as it were, mirroring can also be understood more benignly as part of the normal process of empathy, when the analyst attunes as best he or she can to the patient's demand for recognition. This demand, for Kohut (1987), represents the need for repair of a vulnerable or weakened self by a selfobject function that failed to be supplied at a crucial developmental time (this dynamic seems applicable to both Mr. Small and Mr. Grand, who both failed to elicit the admiring sparkle in the mother's eye that Kohut emphasized). We might define this use of empathy in its best connotation as an attempt to offer accurate recognition, as opposed to Lacans's harsh diagnosis of participation in a narcissistic fantasy. This distinction corresponds in some respects to Winnicott's (1954) separation of needs from wishes, in which, he proposed, need grows out of a regression to the point of failure of a necessary environmental provision. Thus, to oversimplify, Mr. Small was not appreciated nor loved for himself by his mother, whereas Mr. Grand was resentfully an instrument of his parents' narcissistic wishes. A developmental need was not satisfied during a crucial period, and the gap was covered over with a defensive ego organization analogous to a false self. In this sense, the selfobject may supply a real need in the transference, albeit a symbolic (not a literal replacement) version of the missing element.

Kohut's concept of the selfobject, as we know, describes an object who is not truly other, nor yet a simple constituent of the self, but functions as a kind of transitional object, mixing up self and other in a way that should not be sorted out. Given the fragility of the human psyche, Kohut argued, the availability of selfobjects is a universal requirement for subjective coherence, analogous to the Winnicottian concept of a necessary zone of illusion in all human experience. Like Winnicott's transitional object, the source or nature of the selfobject must not be questioned to be effective. Similarly, the selfobject is not real but is a construction in the intermediate space between subject and object, with contributions from both sides. It is thereby imaginary (Lacan) and illusory (Winnicott), although these terms do not capture the quality of the mutual construction involved in which the subject's inchoate need encounters a symbolically prepared other. In this sense, the analyst's empathic response does not necessarily or exclusively involve compliance with narcissistic wishes but can be seen as a kind of performative communication offering a point of contact with a structured social order. That is, the analyst's words do not simply gratify the patient but enact a form of social recognition of something that was lacking in his early experience and to which he was entitled. Lacan would not have used the term, need, for this type of situation, although it seems to me legitimate to speak in Lacanian terms of the child's need for a parent to assume a symbolic function. He did speak about a so-called point de capiton (anchoring point, as in upholstery buttons), quilting the impersonal flow of words to a reference point established in the symbolic order. This metaphor represents a halt to the endless sliding of the signifier, binding the speaking subject to particular signifieds—links to a web of social meanings shared by individual subjects that enables communication to take place.⁵ Perhaps we could speak of an empathic point de capiton, in which a specific intersubjective interaction links the subject to a symbolic other. Kohut's work showed that in specific instances of self pathology (the threat of fragmentation, for example), the immediate requirement for the analyst is to accept the role of selfobject and not to attempt interpretation prematurely (there is no possibility of Freud's "aloofness" on either side).

On the other hand, the universal need for selfobjects and the inevitable recourse to the mirror relation both in normal subjects and, as a desperate quest for repair or completion, in patients suffering from a damaged self, exemplifies the very difficulty of sustaining a position as subject, which is the theme of this chapter. Based on fantasies of reciprocal identity and implicit understanding, the dyadic mirror solution leaves the subject extremely vulnerable to the other, who can enslave or exploit him (by manipulating the dependent selfobject transference), and exposes him to risk of the inauthenticity of a false self (by identification of the patient with a partial or distorted image). In the latter case, it could function as the relational equivalent of the "wall of words" referred to by Lacan as a major impediment to treatment, preventing the analyst from finding any freedom to express him or herself or to open a dialogue about anything outside the selfobject relationship, thereby stalemating the analytic process by confining it to a narrow band of experiences already familiar to the patient. It may well be that the contemporary interest in analysts' self-disclosure is a response to this type of impasse, attempting to destabilize a rigid mirror transference by introducing more of the analyst's subjectivity directly into the relationship. As we saw, this was what Mr. Grand's analyst attempted. Lacan, in response to this problem, referred to the possibility for triangulation within the analytic dialogue as a means of getting around the wall and engaging less conscious and narcissistically invested imagos of the patient. He observed that when the analyst interprets, he or she does so from another place than the position the patient anticipates and addresses issues differently from what was expected within the mirror transference. Although cast by Lacan in structural terms, his formulation seems close to the classic Freudian conception of transference interpretation. That is, by interpreting (commenting, clarifying, etc.), the analyst indicates an interest in something a bit outside the conscious field of the patient and addresses less conscious aspects of his or her psychic functioning. Along these same lines, one could say that any interpretation that departs from the most attuned empathic mirroring mode will inevitably reveal something of the subjectivity of the analyst. This expression of the analyst's subjectivity (choosing to respond at a particular time and in a particular way to the patient) undermines the imaginary transference and opens the door toward a different kind of interaction.

The Symbolic Dimension, Kohut to Green

Although his earliest work (1949) focused on the importance of the mirror stage and problems of identification with an image, Lacan soon began to emphasize the symbolic relationship that destabilizes this structure. His major explanatory tool came from his conception of the symbolic order derived from the anthropological structuralism of Levi-Strauss and the linguistic theory of Saussure, which he believed addressed the basic level of psychoanalytic intervention implicit in Freud's writings. In this model, the therapeutic action of psychoanalysis depends principally on a rectification of the mirror relationship, by replacing the aim of completion of a narcissistic image (which he equated with the American technical principle of strengthening the patient's ego by identification with the analyst, see Lacan, 1958) with a recognition of the patient's own desire within his unique historical, autobiographic context. In the analytic setting, the subject's words disrupt his anchoring to a stable image, revealing a succession of identifications—parts of a history that was not historicized and thus not available in the mirror. The well-known designation of Lacan's "return to Freud" was a product of his shift in interest toward the functions of speech and the field of language (the theme of his Rome conference of 1953), which he believed were being ignored by the current generation of ego psychologists ("a technique is being handed on in a cheerless manner ... a formalism pushed to ... ceremonial lengths" [p. 37]). Whether this shift was necessitated by limitations in clinical practice and or by Lacan's reading of structuralist thinkers (the climate of post-war French thought), it led to a major reinterpretation of subjectivity, which now was seen as structural (the effect of structures extrinsic to the individual person) and linguistic (dependant on language). In place of his previous emphasis on imagos like the ideal ego or any other fantasy self-representation, a symbolic order consisting of kinship and status systems, structures of age and gender, and rules of inheritance and succession, operating by a kind of secondary process logic, became the principal determinant of individual identity, and the gestaltic identification as a whole self in the imaginary register of the mirror was seen as a treacherous lure. The analyst was admonished not to think in terms of object relations, which for Lacan consisted of imaginary constructions, but, rather, to pay attention to the patient's words, to the text, not to the psychology of the author. In the treatment situation, the structuring effect of a "third" was now proposed as the crucial element, creating an analytic space in which symbolic meanings could be explored and elaborated.

The concept of the third was developed by French authors influenced by Lacan but had its roots in Freud's emphasis on the structuring effects of the triangular situation of the oedipal phase (Kirshner, 1991). Green (1975) has consistently advanced this perspective, which could be summarized succinctly by stating that the analyst is not exclusively a selfobject or mirror for the patient, but is crucially a new or unknown object speaking from a different place from where the patient expects. That is, the triadic position is fundamentally inherent in the structure of analysis in the sense that it is impossible to remain exclusively within a selfobject mode and not to speak as a separate subject. As we commented, the analyst inevitably addresses the patient's unconscious from this position, as well as implicitly offering an expression of his own unconscious by choice of tone and words and timing of intervention. The patient responds to these analytic symbolizations with a new set of signifiers (of associations). The invisible and rapid verbal interaction of analyst and patient creates a new entity beyond the conscious experience of either participant but one which belongs in some sense to both. As the French psychoanalyst Serge Viderman (1979) wrote in an influential paper, interpretive speech provides a representation for archaic experiences "in a form that exists nowhere in the unconscious of the patient or anywhere else but in the analytic space that provides it with form" (p. 262). Ogden (1994) explored this theme in the English language literature, emphazing the co-construction of the analytic third.

From this perspective, the expression of the analysand's unconscious is linked inseparably with the analyst's function, which depends on the mobilization of his or her own unconscious. The offer of analysis immediately evokes a transference in susceptible subjects who are attracted by the fantasy of "the supposed subject of knowledge," the one who in fantasy holds the secrets of their unconscious desires. The analysand's speech represents an amalgam of signifiers carrying unconscious meanings unknown to her but that will be interpreted or simply responded to, explicitly or implicitly, by the analyst. That response then calls upon further associations from the patient and so on, creating over time a unique conceptual and metonymic structure that belongs to the two participants. This process has been discussed by Green (1975) in relation to the work of Bion for whom the analyst supplies first a container for the patient's content and then a content for the patient's container (via a "metabolized" interpretation). Green's formulation seems to me quite close to Lacan's (1953) well-known statements that the unconscious is "the discourse of the Other" (p. 55) and that "the Other is the locus in which is constituted the I of speech" (1955, p. 41). In essence, both assertions mean that the speech of the subject associating in analysis is determined by structures outside of consciousness—by the field of the Other, the network of linguistic and cultural rules that provides the ground for a subject who will emerge in varying shapes and at different moments as figure. In the transference, the analyst occupies the place of the Other, containing the speech that this arrangement evokes, then continuously interpreting it back to the analysand (providing a new content to be contained). The Other is another way of naming the symbolic order. Of course, the rules of this order (of the cultural matrix) have been learned within a specific familial context, unique to the person, hence conferring an individual and more or less coherent character to each subject. However, there is always inconsistency and ambiguity in these internalized messages from the Other (the mother in the first instance) for each person, so that identity is not fixed. Against this fluidity, according to Lacan, the subject organizes a stable ego, which is based initially, as we have seen, on the identifications of the mirror stage. To summarize: like all relationships, the analytic situation is based to some degree on a kind of primary mirroring or dyadic selfobject arrangement, but it also holds the power to destabilize the ego and its customary modes of interaction by opening an analytic potential space, which permits the play of intersubjective exchanges. Viderman (1979) underlined the importance of affective communication in this process which "is only possible in a specific environment ... in which the affects and counteraffects of the two organizers of analytic space interact" (p. 282).

Perhaps because of intuitions similar to those of Lacan regarding the limits of selfobject relationships, Kohut postulated the "pole of ideals" at one end of the tension arc of his bipolar self. This innovation suggests that the self is not simply hungry for mirroring but requires a set of values or structuring principles external to it to give a shape to its desires. This aspect is made clear in Ornstein's (1980) presentation of the theory of the bipolar self, which emphasizes internalization and structure building. Although Kohut spoke of the pole of ideals in classic Freudian terms as a residue of infantile narcissism via idealization of the parents (an interpretation that from a Lacanian perspective would seem to place it within the imaginary order—as the product of a fantasy), it is equally true that ideals derive from cultural symbolism, which stands as a "third" position beyond the needy, exhibitionistic self and its objects and shapes them. Wolf (1980), speaking of the development of the self, wrote, "as long as a person is securely embedded in a social matrix that provides him with a field in which he can find the needed mirroring responses and the availability of idealizable values, he will feel comfortably affirmed in his total self" (p. 128). Although not explicitly spelled out in Kohut's theory, it seems to me that the self psychologists' distinction between narcissistic grandiosity and idealization of social values (or their representatives) roughly parallels the differentiation Lacan attempted to make between Freud's notions of the ideal ego and the ego ideal.⁶ For Lacan, the ideal ego refers to a grandiose fantasy of the self, whereas the ego ideal derives from an internalization of the symbolic order that bounds and locates the subject.

Freud (1914) observed that people strive to be their own ideal, with a persisting wish for omnipotence, against an ideal imposed from without. Whatever Freud may have had in mind precisely, Lacan wanted to contrast the fantasies of a grandiose ego denying limitations (the ideal ego) with an ego structured by the principle of differentiation within a symbolic system (the ego ideal). Symbolic systems by their nature involve relationships of difference and separateness between their various elements. These qualities were connected by Lacan with the paternal principle of differentiation from the mother, because fatherhood exemplifies a "third" element, a symbolic role, which destabilizes the imaginary oneness of the dyad. What he called symbolic castration involves renunciation of primary process fantasies of oneness and fusion with the mother and of the fantasy of wholeness that accompanies them. These fantasies, which belong to the imaginary register, are corrected by the limiting conditions of a symbolic order based on separateness and finitude. Freud (1914) said that these conditions have a social aspect, as the shared ideals of a family, class, or nation. Here, I believe, one can attempt a rapprochement of Kohut with Lacan. They each saw the human subject as the product of a structure (tension arc to pole of ideals, the symbolic order) outside of conscious experience, although the notions of a third and of the triadic dimension were only minimally developed by Kohut. Because the agency of the ego ideal for Freud arises both from intrapsychic and social sources (infantile narcissism and the parents' transmission of ideals), its function may be to mediate the tensions arising within what Kohut called the bipolar structure of ambitions and ideals. Stated differently, the development of the Kohutian self requires a linkage of private embodied experiences and desires with a structuring social system that sets parameters for the subject.

Attachment to a symbolic framework through the functions of the ego ideal creates meaning and value for individuals and defines a position in the human (cultural) world which is the true bulwark against fragmentation, even in the face of traumatic experience, but it seems hard to achieve for many people. Hence the constant need for recourse to an imaginary selfobject and to the mirror function in relationships to stabilize a vacillating self, especially in modern societies where symbolic roles are much less clearly defined. A recent book by Kilborne (2002), which deals with "invisible people," elaborates the overriding importance of personal appearance and being seen in contemporary ideology. The aim is to create an image that will give a semblance of substance to the self. The counterpart for many is a self-protective fantasy of invisibility, of not being seen or known by the other. Unfortunately, Kilborne (2002) noted, such fantasies can backfire, "since they tend to produce fears of disappearing in reality" (p. 46). Investment in an image that must be strenuously defended against all threats is characteristic of patients labeled narcissistic personalities, but this pathology seems pandemic in contemporary society.

In practice, of course, symbolic attachments and imaginary mirroring are intertwined as inextricable aspects of experience, just as Winnicottian transitional relatedness and the intermediate realm of illusion pervade human life without undermining the distinction between normality and psychosis (while, of course, relativizing this difference). Against Lacan's model, as I have suggested, one could argue that the empathic gesture that responds to a patient's demand partakes of the symbolic role incarnated by the analyst, as it is the gesture of another separate subject. Without such a response, as Ferenczi (1933) argued, the treatment risks retrauma tizing the patient. Only when the analyst colludes in a narcissistic fantasy to deny frustration, loss, and difference out of her or his own wishes for omnipotence or control is the symbolic function breached, usually with bad consequences.

Perhaps we could propose that ideals (in Kohut's conception) derive from the symbolic order (Lacan's conception) of which they represent an internalized element. Tolpin (1980) referred to the "elevated powerful world" (p. 305) of the idealized selfobjects, offering the example of an admired teacher on the podium nodding to his rapt pupils. Apart from its resemblance to the relationship of Kohut's followers to the master himself, the setup described (teacher, school, auditorium) is essentially a symbolic one, despite the magical or imaginary connotations of the specular "nod." In my view, the self psychologists regularly blur these two functions— no doubt always intertwined, as I have noted—not giving full due to the symbolic Other behind the selfobject (the object in Winnicottian terms that survives all the mirror fantasies of incorporation and projection by the subject). There is a determining symbolic structure beyond the Kohutian self of lived experience, distinct from the imaginary projects of a grandiose self or, in Lacan's vocabulary, of an ideal ego.

In terms of object relations theory, in which one can speak, for example, of the self and the object world, self and objects are constructed mutually and come into being in the course of early development according to a basic set of organizing principles. This "basic set" has to do with the fundamental reality of differentiation and separation through the assumption of language, which Lacan described in his exegesis of the famous fort-da⁷ game reported by Freud (1920). When the child plays this game, he is essentially representing presence and absence of the mother by alternation of two phonemes, which happen to be the German words for here and gone. The two units of sound are distinct, Lacan observed, as basic elements in the system of differentiated sounds of which a language consists. The sounds become signifiers, part of a chain of words that carries meaning, in this case a protoconcept of separation. Mother is gone, leaving the child alone to play out what has happened by use of a new linguistic expression for the preverbal experience of "awayness." Interestingly, the cognitive development alists Gopnik and Meltzoff (1997), in their research, pointed to the importance of this early concept in accelerating the child's intellectual development (seemingly unaware of Freud and Lacan's observations). They were interested in the construction of "theories about objects" (p. 199), extending the Piagettian perspective on the development of object constancy (physical, not human objects), noting a relation between the appearance of words like "gone" and high-level object permanence abilities. "Gone," they observed, "catches a particular contrast between ontology and phenomenology; between the fact that the world looks one way and is another way" (p. 112). Of course, the phonemes are learned (or selectively attended to) as a result of a relationship with the primordial object, the mother, already immersed in the languaged world (the subtle point made by Lacan that all her responses, whatever faint biological sources may be involved, are structured by the symbolic order).

The Symbolic Order: Winnicott to Lacan

Lacan called the entire symbolic field the Other, and he emphasized its structuring of the world into which the child is born. Modell (1968), in speaking about the process of the child's relationship with external reality, observed that, initially, reality is the mother, and it is through the tie to her, through object love, that reality is known. The term reality in this context carries the implications of human or social reality, which Lacan labeled the symbolic order, and, according to Modell, it is identified in the first instance with the mother as the Other. Erikson's (1950) concept of basic trust also seems to refer to more than simply trusting the mother as a reliable person (although that may be a requirement for its establishment), but implies a confident relationship with "reality" (the external object world) in which the child must make his or her way. At this point in our inquiry, we need to follow Modell in returning to Winnicott, who did understand the importance of a symbolic space for establishment of a self. These constitute important bridges to Lacan that were elaborated by Green, as we discuss.

Winnicott's (1971) central interest was the coming to be of the child as a separate entity with an inner world demarcated from the object. Yet he insisted that to be sustained this private inner world depended on a transitional space between self and objective reality. "The use of the transitional object symbolizes the union of two separate things, baby and mother, at a point in time and space of the initiation of their state of separateness" (pp. 96-97). This state of separateness depends paradoxically on the presence of the mother, and it can be damaged, perhaps irreversibly in some cases, by her too-long absence. In the latter situation, Winnicott (1971) wrote, primitive defenses become mobilized in the infant to defend against "madness," defined as the breakup of a personal continuity of existence. The process of the infant's drawing away from the mother and forming a separate self involves the development of internal representations within the mother-child field, in turn depending on the organization of the transitional or cultural space. Winnicott defined the word culture as follows: "I am thinking of something that is in the common pool of humanity, into which individuals and groups may contribute, and from which we may all draw if we have somewhere to put what we find" (p. 99). This is close to Lacan's notion of the symbolic order, if, like his interpretation of the mirror phase, a more humanized version. We put what we find into "the universal discourse" says Lacan (1954-1955, p. 283), into a network of signifiers that is a common possession of the group.

In writing of these matters, Green (1976) followed Winnicott's discussion by referring to the separation of the maternal and subjective spheres in a "state of reunion" (p. 58), which creates a potential cultural space and a feeling of coherence and consistency. The optimal intersection of these spheres produces the "affect of existence", he proposed. "This feeling of coherence and consistency—support for the pleasure in existing—is not self-evident, but must be infused by the object. The destiny of the One," Green concluded, "is to live in conjunction and/or separation from/with the Other" (p. 58; the Other referring here to the mother of infancy). That is, the fragmented incipient subject must have a relationship with the mother to achieve coherence and a sense of existing separately, a function of the maternal holding-presence. The mother becomes internalized not simply as an object but, as Green (1997) later wrote, a container or frame within which representations of objects can be created. This maternal frame (as an internalization of her holding function) might also be considered an anlage of the symbolic (some kind of protostructure in the mind), which derives from (is carried by) the child's relationship with the good object. We could develop this point by suggesting that an important function of the "good enough" mother is to foster basic trust as the basis for internalization of the symbolic order provided by the surrounding culture. To turn this point around to the perspective of the child, the ability and availability of other persons to assume symbolic roles (as mother or father, for example), which are necessary to engender a good internalized object relation, depend on an entire cultural (symbolic) system that defines, validates, and supports those roles. Erikson spoke of a cycle of generations that carries this larger holding function (a holding of the holding), which Lacan called "the big Other" (the Other). The role of mother, beyond whatever biological supports it leans on, depends on beliefs, customs, and institutions maintained by her society. Her connection to this network can be damaged at many levels, from the experience of intense physical pain to the collapse of an entire society, in which cases it falls to the mother's caretakers (or to other survivors of a disaster) to ensure repair and reestablishment of basic trust. Beyond a certain threshhold of damage, repair becomes problematic, and we then enter the territory of severe trauma and its effects.

For Green, the fundamental threats to basic trust are intrusion (excess of presence: impingement, penetration, castration) and object loss (separation, abandonment, the experientially dead mother), which become psychically organized very early in life. In his writings on this process, Green (1986) developed and clarified Lacan's and Winnicott's conceptions of the organizing function of the symbol, a process he saw as beginning with a very early triangulation between subject and object, inherently involving the father. Early triangulation is implicit in the maternal role and in speech, which structures her role, creating a space between the real of the infant-mother biological relationship and the imaginary of earliest fantasies (the Kleinian object relations of the infant, for example) within which experience can be symbolically represented.

Winnicott (1971) commented that beyond a certain limit of absence the mother is dead to the child. "This is what dead means" (p. 22), he said. In his conception of the dead mother in "La Mere Morte," Green (1980a) expanded on these few lines from Playing and Reality. He wrote of the unbearable separation that leads to the disinvestment of the maternal representation and thus to the psychic death of the mother (for the child). In this situation, the contents of the holding frame consist of a negative hallucination or void. By this, Green meant that the mother, even when in fact physically present, is no longer represented as a symbolic internal object. He saw the dead mother not as an object representation but as a "cold nucleus" (p. 230), one with which the subject eventually identifies. Winnicott stated that failure of the external object in some important function leads either to deadness or to a persecutory quality of the internal object, which resembles the bad object of Fairbairn (1943). Recall that Fairbairn saw neurotic guilt and so-called superego anxiety as defensive against the emergence of a primitive, destructive bad object, which attacks, as it were, the very right of the self to be. In addition to these possible outcomes, Green spoke of damage to the framing structure itself, producing disintegration anxiety.

Perhaps it is worth pausing here to reemphasize the symbolic aspects of the framing structure and of the holding and containing functions of the mother. This is an important point because it bears upon certain technical disagreements and even more fundamental conceptual ones in contemporary psychoanalysis. No doubt there has been a correct appreciation of the revolution wrought by Winnicott in bringing attention to the importance of these early functions within a psychoanalytic treatment. Green (1975) alluded to the movement in psychoanalysis away from an analysis of the content "to the analysis of the container, to the analysis of the setting itself" (p. 45). This implies a more rigorous attention to the countertransference as an inherent and necessary part of the analytic work. Certainly the emotional climate of the setting and the patient's reliance on the selfobject aspects of the relationship are now recognized as central features of any psychoanalytically informed treatment. The disagreement arises when the focus on the analyst's affective responsiveness, empathy, and openness to the patient's needs obscures or displaces the analyst's symbolic functions. Green (1999) criticized a tendency toward a kind of overprotective mothering in contemporary treatment by its emphasis on the patient's need for the analyst's availability and responsiveness, at the expense of attention to the effects of the presence of the differentiated other and to the anxieties of intrusion. For Winnicott, Green, and Lacan (of course, with important differences), the intuitive emotional response of the mother is made possible by the symbolic situation of the human family setup, which provides a space for development of a self or of a separate speaking subject. Although the analyst is not ever in precisely the same position, the maternal analogy seems to have taken firm hold, both in terms of explicitly repeating a maternal role (at times this becomes more one of a nongendered parent), and in the widely used metaphor of establishment of a relational matrix for a reconstruction of self. The important critique raised by Green is not necessarily one of overmothering, nor even of a possible overpersonalization of the analytic relationship, but rather one of neglect of the symbolizing and structuralizing functions of the mother and, by corollary, a lack of attention to these dimensions of psychoanalytic work.

As Green (1999) reiterates, returning to the Lacanian conception of a space for representation opened by a third out of the infant's immersion in the "real" of the primary object, the symbolic father (or another symbolic third) is aleady present from the very first as a major component of the psychic world of the mother. She carries "him" with her as part of the frame that holds the infant. We might say that this is a virtual space that is not able to be precisely situated in chronology (hence Green's critique of infant developmental research) but that it appears retroactively once the incipient subject can say "I." So there is an opening—an undefined zone of transitionality and "harmonious mix-up" (to use Balint's phrase)—and a language (a symbolizing structure) that fills it. To these two necessary ingredients of psychic life, Green (1999) proposed a third component incorporating the drive, the link with the body, and affect, which he criticized as missing in Lacan. His assessment of Lacan in this regard seems on the mark, especially in view of the evolution of Lacanian theory toward an almost exclusive emphasis on the signifier, in the wake of which every other human phenomenon follows. Language, Green (1973) reminded us, is empty without affect, and he has labored to develop a useful psychoanalytic theory of affect. As he would be the first to admit, however, the theory is lacking in many respects. Clearly, for a sense of vitality and existence, there must be the emotional resonance Green evokes. Yet, for affect to be bearable and vivifying, it must be to some degree expressed or expressible. It must not simply carry communicative functions (as a sign), but symbolic ones, so that a living language can inhabit the transitional space shared by the self with others.

In the preceding pages, we have traveled a difficult and circuitous route from Lacan's concepts of the imaginary order, the mirror stage, and the symbolic, through their links to Kohut's innovations involving the mirroring function of a selfobject and the bipolar self, all of these terms in some sense building on Winnicott's crucial insights about self-formation. Pursuing Modell's hypothesis abut the noncommunication of affects and Green's insistence on affect as a third constituent element of mental life, we must now turn to a closer look at the symbolic dimension of affect and its role in sustaining subjective existence.

¹Van Haute (2002) gives an interesting summary of some of these formulations, which are scattered throughout Lacan's writings. Classical analysis has been guilty of similar attempts to find a common structure or etiology for classes of patients, which suggests a temptation to make the discipline "scientific."

²Derrida's (1974) work of postmodern deconstruction was a study of how certain metaphors or ideologies become enshrined in the way language is used and thereby perpetuate illusions—like the "full presence" of a discrete subject. He felt that Lacan was vulnerable to the same criticisms, although remaining quite affirmative about Lacan's overall project. Khan (1974) was, if anything, a disciple of Winnicott's in the British society, but who yet raised questions around how the "true self" was conceptualized.

³I have discussed some of these variant conceptions of self and self-representation in my paper, "The Concept of the Self in Psychoanalytic Theory and Its Philosophical Foundations" (Kirshner, 1991).

⁴"The Hegel-Kojeve influence has been addressed by many authors. See, for example, Van Haute (2002), Borch-Jacobsen (1990), and Ver Eecke (1983).

⁵The point de capiton was discussed by Wilden (1968) and Van Haute (2002). It was mentioned by Lacan (1955-1956) in his seminar on the psychoses in which he proposed that a minimum number of such quiltings of signifer to signified is required to enable a personal (nonpsychotic) discourse (p. 269).

⁶Very few commentators accept Lacan's reading of Freud on this point (see Diatkine, 2000), which depends on only a few early references. Probably, Lacan himself must take credit for this construction.

7The fort-da refers to the back and forth, "here and gone" of a spool on a string thrown and recaptured by the child.