Chapter 9

Silence is golden (usually)

Jay Frankel

Historically, psychoanalysts have thought about their patients’ silences, especially prolonged ones, in different ways. We can group these approaches to into three categories: silence as resistance; silence as benign regression in search of an analytically necessary experience, often requiring silent “holding” (Winnicott, 1955) by the analyst; and silence as a move in an interpersonal negotiation with the analyst, in the underlying hope of renegotiating one’s internalized bad-object relations. The latter two perspectives view silence as facilitating analytic process in some way. In actual clinical practice, a given silence can be complex and multifaceted. Following a discussion of the different analytic perspectives on silence, I will discuss the ways in which silence is essential in the process of the deepening self-reflection and insight in analytic treatment.

Silence as resistance

Since the resistance aspect of silence has been addressed in the previous chapter, I will limit my comments here to noting two essential points that will be important in the discussion to come. First, resistance is a concept that arose from a conceptual framework based on intrapsychic conflict—unconscious wishes opposed by anxieties; and this idea of conflict points us to seeing the patient’s silences as the result of inner conflicts that interfere with his following the fundamental rule of analysis: to say everything that comes to mind. This view invites an analyst to respond to a patient’s silence by exploring or interpreting the conflict behind it. The second aspect of the resistance view that I want to note here is a corollary to the first: that thinking in terms of resistance tends to discourage analysts from understanding the pre-conflictual (i.e. object-relational or developmental-arrest) or intersubjective causes of a silence, or from appreciating its inherent place in the natural process of deepening self-reflection.

Silence as a pathway to a necessary therapeutic object relationship

The object-relational approach developed as analysts began to grasp how early trauma and subsequent regression block patients from fully participating in, and benefiting from, analytic treatment; and consequently, what accommodations regressed patients need from their analysts. This perspective begins with Ferenczi (1929, 1930, 1931). In his 1931 paper on “Child-analysis in the analysis of adults,” he described how analytic treatment, at least during periods of regression, could be understood as a kind of game in which earlier experiences of self emerged. Such experiences of self were expressed in a shift to a “childlike attitude” (p. 131), most likely accompanied by his choosing simpler words, and altered tones of voice, ways of speaking, and bodily attitude—including the presence of silence as an expression of this regressed experience of self.

Ferenczi had recently come to appreciate the therapeutic value of such regressions, which offered the opportunity to work through analytically the earlier traumas that had created these undeveloped or split off “selves,” and for analytic progress to take place. As such, Ferenczi emphasized that analysts must take special care to protect these regressions (1931, p. 130). He adjusted his manner, “entering into a game” (p. 129), speaking gently and simply, as if to a child of the age the patient then seemed to be psychologically. He seemed to be trying to foster the patient’s feeling of safety and to protect the regressed state. Certainly, a resonant “holding” silence would also be one way to remain in tune with a silent, regressed patient. My impression is that a somewhat accommodating approach, and the openness to at least some degree of regression that lies behind it, are now widespread among analysts, whether they do this intuitively or have been directly influenced by Ferenczi’s ideas or by the other thinkers I now turn to.

Winnicott’s (1953) ideas of transitionality—transitional objects and transitional space—can be seen as a development and articulation of Ferenczi’s thinking. (See Boyle Spelman’s elaboration of Winnicott’s contributions to understanding the place of silence in development and in analytic treatment, this volume). Winnicott understood that the baby lives in a world of self-generated illusion—she feels she has created the breast she needs. She will accept external reality and the limits of her own omnipotence only gradually, and only when her natural need to feel omnipotent is first recognized by the mother. Without this, the baby will never embrace the external world, which will remain a constantly threatening entity which she withdraws from behind a “false self,” keeping her “true self” hidden (1960); and she will not achieve the feelings of being integrated, and of living within her own body (1945, p. 140). The child’s omnipotence must never be challenged by the mother’s question: “‘Did you conceive of this or was it presented to you from without?’” (p. 95). This approach—for analysts, not just mothers—emphasizes what must not be said and can easily be understood to include the view that there are times when the analyst must not say anything at all.

Indeed, Winnicott spoke explicitly about the need for analysts sometimes to be silent with regressed patients.

In almost all our psycho-analytic treatments there come times when the ability to be alone is important to the patient. Clinically this may be represented by a silent phase or a silent session, and this silence, far from being evidence of resistance, turns out to be an achievement on the part of the patient. Perhaps it is here that the patient has been able to be alone for the first time.

(Winnicott, 1958, p. 416)

Indeed, his concept of the “holding” environment (1955, p. 19) emphasizes the times when the analyst must not impinge in any way upon the patient’s transitional state, even by talking at all:

In one vitally important hour near the beginning of the treatment I remained and knew I must remain absolutely still, only breathing. This I found very difficult indeed, especially as I did not yet know the special significance of the silence to my patient. At the end the patient came round from the regressed state and said: “Now I know you can do my analysis.”

(p. 24)

Michael Balint further elaborated the value of the patient’s silence, and of the analyst’s silent response. The background is Balint’s (1979) concept of benign regression: a state in which the patient returns, inwardly, to a simpler approach to life, before his internalized object relations were distorted by attempts to cope with early traumatic experiences. In benign regression, the patient seeks out the particular kind of object relationship he needs with the analyst, that he senses will allow him to “be able to find himself, to accept himself, and to get on with himself” (p. 180)—to find, in Balint’s words, a “new beginning” (e.g. p. 71). In malignant regression, in contrast, the patient chiefly seeks “gratification by an outside object” (p. 187)—the analyst—as an end in itself: the analyst will provide the cure, rather than being a vehicle or medium through which the patient can “reach himself” (p. 142). Malignant regressions lead “to the development of addiction-like states which [are] very difficult to handle, and in some cases … intractable” (p. 138). Consequently, silences during malignant regressions have a different feeling, and a different meaning, than those that occur during benign regressions, and need to be dealt with differently by the analyst. I’ll say more about that when elaborating silence as a form of interpersonal negotiation.

During states of benign regression, in order to facilitate the patient reaching himself, “the analyst must do everything in his power not to become, or to behave as, a separate, sharply-contoured object” (p. 167). “Environment [i.e. the analyst] and individual penetrate into each other, they exist together in a ‘harmonious mix-up’” (p. 66). The patient must be allowed his regressed state and the precise therapeutic object relationship he seeks, and needs, without interference.

Balint illustrates such an approach in response to a patient who was quiet for the first 30 minutes of a session. The analyst, understanding what might be happening, also remained silent. Eventually, the patient started to sob,

relieved, and soon after he was able to speak. He told his analyst that at long last he was able to reach himself; ever since childhood he had never been left alone, there had always been someone telling him what to do. Some sessions later he reported that during the silence he had all sorts of associations, but rejected each of them as irrelevant, as nothing but an annoying superficial nuisance.

(pp. 177–178)

Balint notes that the patient’s silence here could have been understood and interpreted—indeed, correctly—“as resistance, withdrawal, a sign of persecutory fear … etc.” But doing so, he emphasizes,

would have destroyed the silence and the patient would not have been able to ‘reach himself’, at any rate, not on that occasion. [Interpreting] would inevitably reinforce the patient’s strong repetition-compulsion, there would again be someone there, telling him what to feel, to think, in fact what to do.

(p. 178)

In these lines Balint goes straight to the heart of the difference between the intrapsychic-conflict model and a developmental/object-relational model, in their approach to silence. As he elaborates,

The pedestrian analytic attitude is to consider the silence merely as a symptom of resistance to some unconscious material stemming either from the patient’s past or from the actual transference situation. One must add that this interpretation is nearly always correct; the patient is running away from something, usually a conflict, but it is equally correct that he is running towards something, i.e. a state in which he feels relatively safe and can do something about the problem bothering or tormenting him.

(Balint, 1979, p. 26)

In Balint’s (and others’) developmental/object-relational model, the way to foster safety and the patient’s natural healing processes generally means a less intrusive clinical approach, often including silence.

Balint (1979) also developed the concept of the “area of creation” (p. 24)—the opaque inner space where new insights and approaches to problems are somehow generated. Protecting this process in patients may also require noninterference, and often silence, by the analyst. “[A]ny intrusion … inevitably destroys for the patient the possibility of creating something out of himself” (p. 176).

Kohut’s (1971) idea that narcissistically disturbed patients must be allowed to experience “selfobject” transferences—feelings of kinship with, idealization of, or admiration and mirroring by the analyst—unchallenged and unexplored, perhaps for extended periods of time, in order to reanimate arrested developmental processes, also belongs to this broad noninterfering clinical approach.

In a wider sense, drawing on these ideas, I (Frankel, 2011) have proposed an “analytic state of consciousness”—a universal state for analytic patients generally, akin to a transitional or play space,

characterized by an increased sensitivity and reactivity to impressions arising from both the inner world and the analyst, a heightened sense of dependence and vulnerability, a permeability of boundaries in regard to the analyst, … a shift toward functioning on the basis of omnipotent fantasy in the analytic relationship [and] a feeling of realness of one’s psychic reality, but without any true loss of reality testing.

(p. 1411)

It not only occurs during special periods of regression or in deeply regressed patients, but is an inherent and basic aspect of analytic process, providing a matrix and foundation for more articulated transferences. If this state is indeed a foundation for analytic process, then the analyst, much of the time, must lean toward a “reticent,” non-intrusive stance of emotional accompaniment that tolerates the patient’s silence as essential in fostering this state of consciousness—and this often requires the analyst’s silence.

Silence as an element of interpersonal negotiation with the analyst

Analysts associated with the relational tradition, like Mitchell (1991a), Pizer (1992), and Frankel (1998), have discussed intersubjective negotiation, not least on the nonverbal level, as an essential therapeutic process in the task of restructuring the patient’s internalized bad-object relations—what relational analysts may refer to as “relational configurations” (Mitchell, 1991b, p. 140). Relational analysts understand that analyst and patient are in a real (though analytic) relationship which includes both conscious and, notably, unconscious intersubjective dialogue — what Ferenczi (1915) called the “dialogue of unconsciouses” (p. 109). In his Clinical Diary (1932), Ferenczi detailed his discovery of how extensive two-way unconscious communication between patient and analyst can be. In this light, everything the patient does, or says, or does not say, can be understood as part of a largely unconscious, intersubjective dialogue with the analyst, intended to communicate something or to influence the analyst in some way. In this, the patient’s silence is no different from his words.

The analyst’s silence, too, whatever its conscious intention—for instance, to provide a holding or non-intrusive space for the patient—may also be doing double duty, communicating or acting on the patient in unintended or even disavowed ways—her “lines” in a “conversation” that may lie largely outside the awareness of both people but are nevertheless “heard” by the patient on some level, and may say something very different from what the analyst says out loud. In such cases, clearly, the analyst’s efforts to remain a nonintrusive, facilitating, and benign presence are undermined (Ferenczi, 1933). Indeed, in the analytic dialogue, it is the nonverbal elements that may speak most forcefully.

Ferenczi’s discoveries about the inevitable unconscious communication in analytic treatment may be the deepest foundation upon which the relational position is built, but the relational viewpoint—especially regarding negotiation—also has more recent conceptual pillars. Among them, Weiss, Sampson, et al. (1986), in their research program, explored the idea that much of what patients do in therapy constitutes behavioral tests of pathogenic beliefs, in the hope of disproving these beliefs. The emphasis on these tests as behavioral indicates that silence is no less part of therapeutic process and therapeutic action than words; both words and silence are actions; and as such, both play a central role.

Beatrice Beebe (Beebe and Lachmann, 2014), in her groundbreaking studies of mother–infant interactions, based on frame-by-frame analysis of videotaped communication in a standardized situation, has discovered a high degree of communication—and mutual influence—that occurs very rapidly and outside of conscious awareness. Greatly amplifying Ferenczi’s long-ago insight, Beebe’s mother–infant research can serve as a model of the extent of two-way nonverbal communication and influence that occur in adult analytic treatment (and which interact with verbal communication in complex ways). The implicit, largely nonverbal “unconscious dialogue” exists even despite the limits imposed by most analytic models on what the analyst expresses to the patient. Unavoidably, the timing, rhythm, and nonverbal accompaniments of silences are powerful tools of communication and influence by analyst as well as patient.

If the patient’s silence is a statement, and a forceful one, in the ongoing nonverbal engagement between patient and analyst, what roles can silence play in the ongoing analytic negotiations? I note before addressing this question that in actual clinical situations, roles often overlap and interact.

Silence can be an attempt to find a sense of autonomy or influence when these feel shaky, or to resist a compulsion to be obedient—a silence that may appear defiant. Or through silence, a patient may struggle for a sense of authenticity or honesty when speaking feels false. A patient may fear that others cannot bear the separation that he needs, and his silence expresses. Such patients are likely to pay close attention to the analyst’s reactions: will she accept his silence, and the frightening aspects of himself he expresses through this silence? Does she really want to know this side of him? Is she more able than the inner objects he projects onto her to accept his wishes and needs? Further therapeutic progress will be facilitated by later exploration of these interactions, but it may be necessary, and valuable in its own right (Frankel, 1998), for the patient to live out these interactions with the analyst.

Such motives for, and negotiations around, silence can be seen in the following brief vignette. A young woman emerging from a period of great inner struggle and fragility is uncharacteristically silent in her session—after a friendly hello, not another word for the whole hour. But her face, far from being empty or distracted, seems reflective, attentive to her unspoken thoughts and feelings. Surprisingly, I feel relaxed, engaged, free to observe her and my own thoughts. I even have a vague sense of following the flow of her experience, as revealed through her face and slightly shifting posture. Next session, silence again. Now I am not so relaxed. After a few minutes I ask: “Do you want to say what’s happening?” She asks why she should. I explain, unnecessarily, that saying what’s on her mind could help us better understand it. She says: “Don’t you think my silence has meaning?” We both remain silent for the rest of the hour. The close relationship of the attempt to find necessary conditions for growth, and to renegotiate problematic internalized object relations, is clear in this example.

A patient may also use silences to try to make the therapist feel something the patient feels, perhaps in order to feel understood and less alone. Or he may need the therapist to feel something he has dissociated and cannot face within himself—projective identification (Klein, 1946, Ogden, 1979)—either to find contact with this dissociated experience through the therapist’s experience of it, or to try to get rid of the intolerable feeling.

Silence may be an especially effective way for the patient to feel some kind of closeness to the therapist, or understanding by the therapist. This is true in certain therapeutically necessary narcissistic transferences, like twinship or idealization, where the patient imagines that the therapist simply knows his thoughts, and speaking could spoil this feeling. A silent feeling of closeness may also include romantic feelings; while these may reflect a workable underlying narcissistic transference that is a prerequisite for analytic progress, they may also be a sticking point, if the patient holds onto a sense that love between patient and therapist is the solution to the patient’s problems.

Indeed, not all negotiations move patients toward helpful goals. A patient may use silence to try to control or punish the therapist. We may be sympathetic to a patient’s underlying anxieties, and his fantasies of reversing roles in internalized traumatic scenarios, but this negotiating tactic is ultimately likely to be self-defeating, and to undermine his ability to develop stable feelings of autonomy, trust, calm, and intimacy.

There are undoubtedly more potential meanings of what patients express, and seek, through silence. What are the therapeutic implications of those I have mentioned? What guidelines can help the therapist think about how to respond to the patient’s use of silence as a negotiating tactic to achieve each of these goals?

When the patient uses silence as a way to establish some kind of relationship with the therapist as an end in itself, rather than as a way to establish conditions for safety and self-reflection—for instance controlling or punishing the therapist, or expressing romantic feelings as the final answer to one’s problems, we should look to Balint’s (1979) idea of malignant regression; in such cases, the analyst’s silent response may be taken as an indication that she submits to control or accepts guilt—a collusion (Frankel, 1993) with the patient that can intensify the patient’s resistance and entrench his pathology (Frankel, 2018).

When the patient is trying to negotiate what feel to him like necessary therapeutic conditions—establishing trust and safety, a sense of being thought about, cared for, or understood, a feeling of authenticity, or an analytically necessary transitional selfobject transference—the analyst should not interfere with the silence. The analyst may need to be silent herself (often alongside her own inner identification with some aspect of the patient’s conscious or unconscious experience). Here, Winnicott’s concepts of transitional space and holding environment, and Balint’s concept of benign regression, are likely to be good guides for the analyst.

Integrative: The roles played by silence in the process of symbolizing: The work of Norbert Freedman and his colleagues

Psychoanalyst and clinical researcher Norbert Freedman, along with Wilma Bucci and other colleagues, conducted a research program in which they developed a particular idea about the role of silence as facilitating the patient’s symbolizing activity, and thus the analytic process. Freedman briefly summarized his later thoughts about this in a 2011 proposal for a conference panel; but he died suddenly before this panel could be further developed. I think it is important that these ideas find a hearing among clinical psychoanalytic thinkers, so I will go into some detail about them. In terms of the current chapter, Freedman’s ideas can be seen as intersecting with, and integrating, all of the above ways of viewing silence: as resistance; as facilitating a needed, safe object relationship with the analyst so that the treatment may deepen; and as interpersonal negotiation of unresolved internalized object-relational issues.

In earlier empirical research, Freedman and his colleagues (Gilani et al., 1985) demonstrated that analytic silences can be signs of consolidation—a “pre-narrative activation period” (p. 100). “Analyzing the observable external kinetic and movement behavior” (p. 101) during a silence can provide a sense of the internal processes, as judged by the narrative activity that follows the silence. “The pattern of discharge movements followed by self-stimulating body-focused movements [was observed to lead to] more organized and cohesive verbal performance during the [ensuing] monologue” (p. 101, and see pp. 106 and 108).

In his later panel proposal (Freedman, 2011), based on his subsequent clinical-empirical research program, which included systematic analysis, on a number of dimensions, of audio recordings of a complete psychoanalytic treatment, Freedman suggested that “moments of silence at critical junctures during an analytic hour—the concurrent muteness by both patient and analyst—represent a particular state of consciousness that offers an opportunity for the deepening of the analytic process.”

Freedman elaborated, proposing three different dimensions, or dynamic structures, in relation to silence in analytic treatment. Working in a coordinated way, these dynamic structures promote the progressive symbolization of intolerable or dissociated experience, and thus of transformation, in psychoanalytic treatment. Freedman proposed that a

symbolizing silence [the first of these states] arises from a sense of implicit relational connectedness to the analyst—in some sense it mirrors the transference. The symbolizing silence, thus, is a kind of communication that invites analyst participation—a signal for closer interaction. Further, the consequent of the silence—when speech resumes—reveals a progression toward metaphoric thought.¹ Thus, the symbolizing silence is one step in a linear process of transformation involving both thinking and object-relational aspects—a fact that places the inherent facilitating function of ‘silent thought’ within the transference.

Freedman also talked about a defensive

desymbolizing silence [which] is part of a fragmented and frozen psychic constellation. It arises in the experience of a barren object relation, in which the analyst in fact withdraws from interaction (whether a reaction to the patient or analyst-initiated). There is minimal object relation to nourish reciprocal silent exploration (e.g. sleepiness). The desymbolizing silence reflects a mode of ego organization marked not only by fragmentation of thought, but also a defense against intolerable affect. … Whereas in the symbolizing silence, there is a wish for the analyst’s participation, here discharge of intolerable affect can result in alienating the present other—the analyst. However, in spite of momentary paralysis, a desymbolizing silence is necessary in order for the patient to confront intolerable affects and bring them into the analytic dialogue.

And he talked about a transformative silence that encompasses both symbolizing and desymbolizing silences:

the transformative silence involves the confrontation with both states in sequence. … while the symbolizing silence is the mirror of the transference, the transformative silence is the mirror of the countertransference. It is through the intrusion of countertransference that a new direction is sparked and a new course set. … the transformative silence can be spelled out in successive phases. It begins with a symbolizing silence (reflecting initial analyst engagement), then a paralysis of meaning, followed by resymbolization. It is the unexpected confrontation with ‘newness’—a discordant countertransference reaction—that triggers a transitory regression, followed by resymbolization. Transformation always involves the confrontation with the dialectic of meeting an incompatible experience needing resolution.

Freedman credits Glover with the idea that “silence may mirror the impact of the countertransference—the latter acting as a discordant element that must be accommodated and integrated,” thus contributing to the patient’s symbolizing activity.

Conclusion

What I have discussed as different functions of silence are often, in fact, different aspects of one and the same dynamic. While one function may be most visible at any given moment, others are likely to be operating in the background; which one seems clearest may simply reflect one’s angle of view. For instance, to label a silence as resistant—as avoiding self-exploration: the very task that the patient has set for him- or herself—indicates that we see the silence as a manifestation of an internal conflict; in other words, the silence reflects the current state of a negotiation between the patient’s wishes related to disclosing his inner life to the analyst, and his fears about doing so—a conflict that has become externalized and enacted with the analyst, and is being fought out in the transference relationship. It is partly because of, and through, this enacted silence, in tandem with self-reflection about the enactment, that these unresolved inner conflicts can be reworked in relation to the analyst. We can paraphrase Michael Balint’s (1979) idea of “regression for the sake of progression” (p. 132) here and talk about resistance for the sake of progress.

Further, a patient’s wish and attempt to renegotiate internalized bad-object relations through his relationship with the analyst—to free himself from his internalized bad objects—is the same as saying that the patient is hoping to find facilitating conditions, for instance a sense of safety, that will allow greater inner freedom and authentic analytic self-exploration. And when the analyst provides a safe, holding environment she may, from a different angle, be disconfirming the patient’s pathogenic beliefs and helping the patient to renegotiate his internalized bad-object relations.

However, what the analyst feels outside of her own full awareness can be very different from the facilitating presence the analyst wishes to have, or believes she is communicating; and these disavowed feelings can be unwittingly communicated to the patient. Analysts, therefore, regardless of their preferred clinical stance, must always try to be attentive to, and reflect on, what they may be feeling and expressing outside of what they intend, and how this may be affecting the patient. Analysts’ self-reflection can be aided by listening for disguised observations of themselves in their patients’ associations (Ferenczi, 1933).

Finally, Freedman’s work underlines the multiple essential roles that silence plays in the process of analytic symbolizing and working through—roles related to all the views of silence I have described.

Note

1 See the chapter by Dreyer and Franzen in the research section of this volume (editors’ note).

References

1. Balint, M. (1979). The Basic Fault. Therapeutic Aspects of Regression. London: Tavistock.
2. Beebe, B., & Lachmann, F. (2014). The Origins of Attachment: Infant Research and Adult Treatment. New York: Routledge.
3. Ferenczi, S. (1915). Psychogenic anomalies of voice production. In: J. Rickman (Ed.), Further Contributions to the Theory and Technique of Psycho-Analysis (J. I. Suttie & others, Trans., pp. 105–109). London: Hogarth.
4. Ferenczi, S. (1929). The unwelcome child and his death instinct. In: M. Balint (Ed.), Final Contributions to the Problems and Methods of Psycho-Analysis (E. Mosbacher & others, Trans., pp. 102–107). London: Hogarth.
5. Ferenczi, S. (1930). The principles of relaxation and neocatharsis. In: M. Balint (Ed.), Final Contributions to the Problems and Methods of Psycho-Analysis (E. Mosbacher & others, Trans., pp. 108–125). London: Hogarth.
6. Ferenczi, S. (1931). Child-analysis in the analysis of adults. In: M. Balint (Ed.), Final Contributions to the Problems and Methods of Psycho-Analysis (E. Mosbacher & others, Trans., pp. 126–142). London: Hogarth.
7. Ferenczi, S. (1932). The Clinical Diary of Sandor Ferenczi (J. Dupont, Ed., M. Balint & N. Z. Jackson, Trans.). Harvard University Press.
8. Ferenczi, S. (1933). Confusion of tongues between adults and the child. In: M. Balint (Ed.), Final Contributions to the Problems and Methods of Psycho-Analysis (E. Mosbacher & others, Trans., pp. 156–167). London: Hogarth.
9. Frankel, J. (1993). Collusion and intimacy in the analytic relationship: Ferenczi's legacy. In: L. Aron & A. Harris (Eds.), The Legacy of Sandor Ferenczi (pp. 227–247). Hillsdale, NJ: The Analytic Press.
10. Frankel, J. (1998). The play's the thing: How the essential processes of therapy are seen most clearly in child therapy. Psychoanalytic Dialogues, 8(1), 149–182.
11. Frankel, J. (2011). The analytic state of consciousness as a form of play and a foundational transference. International Journal of Psycho-Analysis, 92(6), 1411–1436.
12. Frankel, J. (2018). Thoughts on the limits of a mutual technique. The American Journal of Psychoanalysis, 78(4), 350–360.
13. Freedman, N. (2011). Silence and the deepening of the analytic process: Psychoanalytic and convergent empirical evidence Proposal for panel presentations. In: IPA clinical research conference on analytic evidence, at that time scheduled to be held in San Francisco.
14. Gilani, Z. H., Bucci, W., & Freedman, N. (1985). The structure and language of a silence. Semiotica, 56(1–2), 99–113.
15. Klein, M. (1946). Notes on some schizoid mechanisms. International Journal of Psycho-Analysis, 27(3–4), 99–110.
16. Kohut, H. (1971). The Analysis of the Self. New York: International Universities Press.
17. Mitchell, S. A. (1991a). Wishes, needs, and interpersonal negotiations. Psychoanalytic Inquiry, 11(1–2), 147–170.
18. Mitchell, S. A. (1991b). Contemporary perspectives on self: Toward an integration. Psychoanalytic Dialogues, 1(2), 121–147.
19. Ogden, T. H. (1979). On projective identification. International Journal of Psycho-Analysis, 60(3), 357–373.
20. Pizer, S. A. (1992). The negotiation of paradox in the analytic process. Psychoanalytic Dialogues, 2(2), 215–240.
21. Weiss, J., Sampson, H., & The Mount Zion Psychotherapy Research Group. (1986). The Psychoanalytic Process: Theory, Clinical Observation, and Empirical Research. New York: Guilford.
22. Winnicott, D. W. (1945). Primitive emotional development. International Journal of Psycho-Analysis, 26(3–4), 137–143.
23. Winnicott, D. W. (1953). Transitional objects and transitional phenomena — A study of the first not-me possession. International Journal of Psycho-Analysis, 34(2), 89–97.
24. Winnicott, D. W. (1955). Metapsychological and clinical aspects of regression within the psycho-analytical set-up. International Journal of Psycho-Analysis, 36(1), 16–26.
25. Winnicott, D. W. (1958). The capacity to be alone. International Journal of Psycho-Analysis, 39(5), 416–420.
26. Winnicott, D. W. (1960). Ego distortion in terms of true self and false self. In: The Maturational Process and the Facilitating Environment (pp. 140–157). New York: International Universities Press, 1965.