114
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
While Bowlby emphasized the formidable influence upon the developing self of the realities of early attachment experience, Main and Fonagy showed that the stance of the self toward experience past and present can ultimately be even more influential. To the extent that we are able not only to have our experience but also to reflect upon it, our sense of security, flexibility, and internal freedom will be very much enhanced. Beyond a reflective stance that allows us to make sense of the contents of awareness (feelings, thoughts, and the like) is a mindful stance that can potentially afford us a calm and spacious awareness of awareness. To the extent that we are mindful, we can be more fully present, more capable of living from within the center of ourselves, and less vulnerable to confusing our shifting feelings and thoughts ivith who we are. Chapter 9 explores the power in psychotherapy and everyday life of shifting the stance of the self toward experience in a more reflective and mindful direction.
To be able to access our patients' nonverbal experience and strengthen their capacity to reflect and be mindful we need to enlist resources outside the attachment field, because attachment theory is not an explicitly clinical theory. Central among these resources is the clinical research conducted under the banner of intersubjectivity and relational theory—a treatment approach that goes a long way toward fulfilling the clinical promise of attachment theory, as I'll explain in Chapter 10.
CHAPTER 8
Nonverbal Experience and the "Unthought Known"
Accessing the Emotional Core of the Self
I
In his final book on attachment, Bowlby quotes Freud who remarked on the characteristic response of the patient who has become aware of something "forgotten": "As a matter of fact I've always known it; only I've never thought of it" (Bowlby, 1988, p. 101). Perhaps Christopher Bollas (1987) who coined the evocative phrase "the unthought known" was reading the same passage from Freud.
What we "know" but do not (or cannot) think about is also what we cannot talk about. Enormously influential because it registers outside conscious awareness, unverbalized (or unverbalizable) knowledge plays a crucial role in psychotherapy as well as in childhood.
If it is obvious that the therapeutic conversation is always made up of more than words, the case for attending to the nonverbal realm is still vital to make—first, because its clinical centrality is not universally recognized or well understood, and second, because the spell of spoken language can be so hypnotic. We risk allowing the words we exchange in therapy to monopolize our attention when we don't remind ourselves that beneath the words there is a flow of critically important experience that provides the underlying context for the words. Fundamentally emotional and relational, this initially unarticulated experience is often where we find the greatest leverage for therapeutic change.
115
116
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
Having established the crucial importance of the nonverbal subtext, I'll discuss how we might understand it. Finally, I'll begin to explore the research and theory that give us the clinical tools to work with the nonverbal dimension of experience—especially early experience—that investigators of attachment have identified as so central.
THE RESEARCH BRIEF FOR A FOCUS ON NONVERBAL EXPERIENCE
At least two findings from attachment research invite, or perhaps even mandate, attention to experience that our patients are unwilling or unable to put into words. First, there is the fact, established by a multitude of observational and longitudinal attachment studies (see Main et al., 2005), that we learn many of the most significant and lasting lessons about who we are in relation to others by the time we arc 12 months old—or perhaps even earlier, if the split-screen studies of mothers and infants at four months prove as telling as they appear to be (laffe et al., 2001; Beebe et al., 2000). Empirical evidence clearly indicates that the foundations of our internal working models—as well as the habitual attachment and emotion-regulating strategies encoded in these models—are all laid in place well before the acquisition of language. 1 These are the data that underlie Schore's (2003) conclusion that "the core of the self is . . . nonverbal and unconscious and lies in patterns of affect regulation" (p. 46). Because preverbal experience constitutes the basis of the developing self, making room for the reverberation and elaboration of such experience in psychotherapy is absolutely vital.
Second, the parent-child relationships that most successfully foster secure attachment are inclusive (Lyons-Ruth, 1999; Bowlby, 1988), meaning that the parent makes as much space as possible for the full spectrum of the child's subjective experience. To generate a therapeutic relationship that is similarly inclusive—that is, to make room for as much of our patients' experience as we can—we have to attend not just to what patients tell us in words but also to what they show us in other ways. Bowlby's theory that the child will integrate only what her attachment relationship(s) can accommodate implies that the child will exclude from awareness those thoughts, feelings, and behaviors that risk disrupting attachment relationships, with the result that those thoughts, feelings and behaviors will remain not only undeveloped and unintegrated but often impossible to verbalize. Hence the requirement to "listen" to what is communicated nonvcrbally if we are to engage experiences that the patient's original attachment(s) precluded. To integrate what has been defensively dissociated or excluded, we need to access that in the patient that is as yet unspoken, unthought, and, perhaps, unfelt.
I
>
Nonverbal Experience and the "Unthought Known"
117
Neuroscience research both confirms and elaborates the conclusion of attachment research that patients may lack the words to describe crucial experiences for reasons that are either developmental (the experiences occurred prior to the acquisition of language) or defensive (the experiences could not be thought, felt, or talked about without jeopardizing vital relationships). Clearly there are neurophysiological as well as psychodynamic barriers that bar linguistic access to formative (and, especially, traumatic) experience (Fonagy, 2001). Research on neural development has shown that the brain centers that mediate language (left cortex, Broca's area) and autobiographical memory (the hippocampus, in particular) are not effectively "online" until 18 to 36 months of age—hence, the near-universal finding of "infantile amnesia." Moreover, overwhelming emotions of the kind evoked by trauma suppress the functioning of these same brain structures. Evidently we lack verbal access to many of the experiences that shape us most profoundly, either because these experiences occurred before we had the neural equipment to encode them linguistically or because this equipment was temporarily disabled by overwhelmingly intense painful emotion.
Patients with posttraumatic stress disorder (PTSD)—flooded by a chaos of disturbing emotions, somatic sensations, images, and impulses— lack the language to give meaning or context to their fragmented, multi-sensory experience.2 Trauma, which shuts down Broca's area and the hippocampus, can be understood to both cause and result from an "emotional hijacking" (Goleman, 1995) in which the amygdala with its links to the affectively oriented right brain overwhelms the hippocampus and its associated abilities to encode, retrieve, and contextualize memories of the trauma.-1
The fact that trauma's impact registers as it does has implications for our work with many, if not most, of our patients, van der Kolk (1996), arguing that the imprint of trauma is somatic and sensory, advocates using bodily sensations to access experiences that patients lack the words to articulate. While he addresses his very helpful recommendation narrowly to therapists working with PTSD, I would broaden its scope in recognition of the fact that patients with trauma represent anything but a narrow category.
The infant's utter dependence upon the attachment figure means that chronic misattunement, depression, and anger on the part of a caregiver may, in and of themselves, be experienced as traumatic. In this connection, Schore (2002) has referred to "relational rrauma" that arises from experiences of disorganized attachment and may eventuate in borderline and possibly psychotic disorders. I would further suggest that many of our patients (and many of us) suffer from what psychoanalyst Phillip Bromberg (1998a) has described as "islands" of trauma—and dissociation—whose impact and meaning are initially impossible to put into words. Therapists must
118
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
find ways to engage such experiences of trauma if their destructive effects are ever to be mitigated.
Alongside the attachment, neurobiology, and trauma studies, there are findings from cognitive science that help clarify the necessity for a therapeutic focus on nonverbal experience. Cognitive scientists have discovered that memory is not monolithic, and they have identified two distinct systems of memory—explicit and implicit. Here's the shorthand: Explicit memory is roughly coincident with our usual understanding of the term, "memory." As such, it can be consciously retrieved and reflected upon, it is verbalizable and symbolic, and its content is information and images. Implicit memory by contrast is nonverbal, nonsymboiic, and unconscious in the sense that it is not available for conscious reflection. Its content involves emotional responses, patterns of behavior, and skills. Implicit memory entails "knowing how" rather than "knowing that."
Sometimes called early memory because it's available to us even in the womb4 and originally known as procedural memory (involving procedures, like how to dance or ride a bike or be in a relationship, that cannot be conveyed in words), implicit memory has as its subjective hallmark familiarity rather than recollection. (While it's often said that once you've learned to ride a bike you never forget how, the fact is that you never actually "remember": so familiar is the skill that you just do it; this "knowing how" to ride a bike is an example of implicit memory.) The most important implicit memories involve procedures for being with others and being with oneself. Taken together, these remembered procedures make up what has been called implicit relational knowing (Lyons-Ruth, 1998; Stern et al., 1998).
Implicit knowing is expressed not so much in what we say but rather in how we behave and feel, in how we carry ourselves, and in what we expect from relationships. This knowing usually exists outside reflective awareness—not because we can't bear to know but because what we know has registered in an implicit form that is hard to retrieve linguistically.5
Implicit or procedural knowing constitutes the foundation of the internal working model. It has been documented by attachment researchers to emerge early in life as a function of the quality of our first relationships and to persist into adulthood (barring changes in these relationships). An infant may come to know implicitly that his cries of distress will quickly evoke his mother's soothing presence, for example, and this primal knowing will become the enduring expectation that supportive others will be there when he needs them. For many of our patients, however, early interactions have been problematic, registering implicitly as a dispiriting bred-in-the-bone understanding of self and others that they cannot easily articulate but also cannot keep from enacting, often to their own disadvantage.
Paradoxically, perhaps, when these same self-defeating enactments occur in psychotherapy they can be a valuable resource insofar as they enable us to engage and transform the wordless internal representations that hold
Nonverbal Experience and the "Unthought Known"
119
our patients hostage to the past. But reaching patients at these noverbal levels requires of the therapist some ability to grasp the unverbalized subtext of the therapeutic conversation.
UNDERSTANDING THE LANGUAGE OF THE NONVERBAL
The words exchanged in psychotherapy float, so to speak, on the stream of nonverbal communication between patient and therapist. The drift of spoken dialogue—what is and is not addressed, and at what depth—is largely determined by the emotional and relational currents that flow beneath the surface of the therapeutic interaction. These undercurrents shape the experience of patient and therapist very much as the infant and caregiver's experience is shaped by the quality of their (necessarily) nonverbal communication.
There turns out to be a rather extraordinary consistency between the nonverbal behaviors that mark the interactions of infancy and those we can observe in the interactions of adults (Reebc & Lachmann, 2002). Studies of these earliest patterns of prcvcrbal communication and their parallels in later life reveal some of the ways in which—inescapably and usually outside conscious awareness—we affect and are affected by those with whom we interact. It is the quality of these nonverbal interactions that largely determines the impact of attachment relationships on the developing self, whether in childhood or psychotherapy.
Facial expression and tone of voice, posture and gesture, the rhythms and contours of speech and behavior—these are the elements that compose what is essentially a medium of body-to-body communication. Such communication during infancy can be seen as a conversation between the baby's somatic/emotional self and the caregiver's somatic/emotional self— or, from a neuroscience perspective, as a "conversation between limbic systems" (Buck, 1994, quoted by Schore, 2003, p. 49). The subject of this conversation is mainly the infant's internal states—particularly, emotions and intentions. As the conversation unfolds through the bodily expression of internal states, the infant learns about herself and others: What are her own emotions and intentions? Will others recognize and attune to them? Will it "work" for her to take the initiative—independently or with the help of others—to attempt to affect her own internal states?
Consider the following account (Sander, 2002) of a filmed interaction of an eight-day-old baby who, having grown fussy in her mother's arms, has just been handed off to her father:
One sees the father glance down momentarily at the baby's face. Strangely enough, in the same frames, the infant looks up at the father's face. Then the infant's left arm, which had been hanging down over the father's left arm, begins to move upward. Miraculously, in the same frame, the father's right
120
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
Nonverbal Experience and the "Unthought Known"
121
arm, which had been hanging down at his side, begins to move upward. Frame by frame by frame, the baby's hand and the father's hand move upward simultaneously. Finally, just as they meet over the baby's tummy, the baby's left hand grasps the little finger of the father's right hand. At that moment, the infant's eyes close and she falls asleep, while the father continues talking, apparently totally unaware of the little miracle of specificity in time, place, and movement that had taken place in his arms. (p. 20)
In this "action dialogue" of facial expressions and bodily movements— accompanied by the "lullaby" of the father's talk—we can see an exquisite relational choreography. The baby's nonverbal communication of her needs for soothing and sleep evoked a series of unconsciously coordinated and attuned responses from her father. We might infer that such an experience of attunement registered as a tiny but formative influence on this newborn's dawning implicit knowledge about herself in relation to others.
There is a comparable nonverbal choreography that influences the experience of the patient in psychotherapy and shapes, ideally for the better, his evolving sense of himself in relation to others. Not long ago, for example, while speaking to a patient with whom I had been meeting for several months—call him Eliot—I noticed that my voice sounded louder than usual and that the pace of my speech was accelerated. I realized that I was trying to stimulate myself in order to avoid succumbing to the sleepiness I was just now becoming aware of. After privately asking myself what might be going on (a self-inquiry doubtless hobbled by my heavy-lidded state of mind), 1 decided to enlist F.liot's participation.
What emerged as I let him in on my experience was that he too had felt sleepy—but, beyond that, he was "gone" emotionally, had withdrawn from me, had (as he put it) "dissociated." It was, he said, his familiar response to feeling anxious, angry, or despairing, and he revealed that he had felt crowded out by me—my chair was too close to him for comfort, I was leaning too far forward, I was talking too much. Note that these troubling concerns only came to the surface through a focus on my own nonverbal (or, rather, paravcrbal) behavior and experience, while in parallel fashion Eliot's initially undisclosed distress with me was rooted in the physical facts of our relationship.
Attending to the nonverbal subtext had a therapeutic yield related to several of the developmental desirables discussed earlier. With regard to in-clusiveness, we were able to contain within our relationship what Eliot had previously had to leave out: namely, disturbing feelings in relation to me (as well as others) involving boundary issues, closeness, safety, and self-definition—not to mention his self-protective dissociation. With regard to attunement, learning together how easily Eliot could feel crowded out and intruded upon led me to draw back and tone myself down in ways that permitted him to feel safer, closer to me, and more in charge of his own ther-
apy. Overall, our original interaction and the subsequent adjustments it spawned gave my patient an experience of disruption repaired—of "misalignment and re-alignment" (Schore, 2003)—that both of us found quite moving.
It seems plausible to infer that, like the episode of the infant with her father, this one might register internally. Although unlike the baby, Eliot and I of course had the benefit of words in our effort to get in synch, I suspect it was less the content of our verbal exchange than the relational process that had an impact on my patient. In this process, through an initial focus on my own voice and body, I found a way to access and meaningfully respond to emotions that Eliot had previously been unable to articulate. Rooted in the nonverbal subtext, our shared experience—an experience more inclusive, collaborative, and attuned to the patient's needs than his history would have predicted—may well have contributed to a shift in his "implicit relational knowing."
As I have explained, this kind of implicit knowing is always enormously influential and usually very difficult to put into words. Certainly the preverbal or traumatic origins of such implicit knowledge can be impossible to retrieve linguistically. Yet, what we cannot recall explicitly—and cannot put into words—is almost invariably expressed in other ways.
In this connection, I would propose the following shorthand: That which we cannot verbalize, we tend to enact with others, to evoke in others, and/or to embody. Before going into more detail, let me illustrate what I have in mind by returning briefly to my experience with Eliot.
Eliot enacted with me a scenario that was simultaneously all too familiar to him and yet impossible for him to recognize—or object to—as it was unfolding. In this jointly created enactment, I found myself talking fast and loudly, as if to drown out the stupefying silence that seemed to hang in the air between us. Only in disclosing my drowsiness did I begin to realize how frustrated 1 had felt at my failed attempts to have an impact on him. For his part, Eliot was initially mainly aware of my clumsy intrusions on his physical and mental space. As we spoke, however, Eliot began to connect with his guilty anger as he recognized that my efforts to reach him had been feeling misattuned, insufficiently respectful of his vulnerability, and as having more to do with my need to feel effective than his need to feel understood. His emotional reflex had been to withdraw from me, very much as he had pulled away (or tried to) from his intrusively seductive mother.
Eliot evoked several distinct experiences in me, as a function, I suspect, both of what he was communicating nonverbally and of what I was receptive (or vulnerable) enough to be internally affected by. In retrospect, I read my own drowsiness not only as a bodily echo of his (yawn and the world yawns with you) but as a defensive reaction of mine to the feeling of being frustrated and potentially angry—as if I were about to be told that what I had to offer was not merely ineffectual but hurtful. Unsurprisingly, these
122
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
experiences of mine, by virtue of their connections to Eliot's own, enabled me to know him in an emotionally direct way—-through identification, as it were, rather than through the information conveyed by his words. It was as if rather than hearing what Eliot felt, I simply felt it. Patients who succeed in evoking such subjective responses offer their therapists the opportunity to know them "from the inside out" (Kromberg, 1998a).
Patients may also embody—or induce their therapists to embody— what they cannot or will not communicate in words. Eliot was unable to tell me that he needed to be "gone"—to dissociate or, in effect, to leave his body; 1 was unable to recognize that Eliot was becoming distant and sleepy, but my body apparently "knew" what my mind did not.s With Eliot, the deactivating, parasympathetic branch of my autonomic nervous system was triggered and I became drowsy—as a response to his deactivation and/or as a defense against the feelings our interactions were evoking in me.
Enactment, evocation, and embodiment are the primary means by which patients communicate what they know but have not thought—and, therefore, cannot talk about. As such, these channels for conveying the un-thought known are absolutely essential for therapists to understand. While the developmental centrality of nonverbal experience has been empirically documented by Bowlby's heirs, it is necessary to turn to contemporary clinical theory—specifically, to intersubjective and relational theory—to fully capitalize on this particular finding of attachment research. In subsequent chapters I explore in more detail how therapists can work with these ways of knowing and being known that arc largely unmediated by language. What follows should therefore be considered a first pass at some crucial ideas about psychotherapy that will be elaborated as we go on.
WORKING WITH ENACTMENTS OF THE UNTHOUGHT KNOWN
Lyons-Ruth (1999) coined the term "enactive representations" to describe the prcsymbolic internalizations of early experience that provide the foundations of our internal working models. The term seems apt precisely because of the intense focus among intersubjective and relational theorists on enactments. As illustrated in my work with Eliot, enactments are the jointly created scenarios that reflect the initially unconscious, overlapping vulnerabilities and needs of patient and therapist.
Enactments in psychotherapy can be seen as the here-and-now behavioral manifestation of implicit relational knowings whose first (but not only) roots lie in what we—patients and therapists—"enacted" with our attachment figure(s) as infants. When, for example, our earliest overtures for comfort were regularly welcomed, we probably learned the advantages of turning to others to soothe our distress; when such early overtures evoked rejection, we probably learned the necessity of concealing our distress from
Nonverbal Experience and the "Unthought Known"
123
others whenever possible. Such primal lessons about self and others are learned—remembered, represented, internalized—as they are enacted. Later, rather than being recognized with a sensation of recollection ("Aha! Now I remember what happened!") these hard-to-verbalize representations of formative experience are recognized—generally by third parties, if at all— mainly as they are enacted ("Can't you see you're treating our kids exactly as you complain your mother treated you?"). Usually, however, what is known implicitly remains implicit. Rather than become available for conscious reflection, it is simply enacted, automatically and reflexively.
Freud (1958) was insightful on this point: "The patient does not remember anything of what he has forgotten and repressed, but acts it out. He reproduces it not as a memory but as an action" (p. 147). Freud's discovery that patients repeat the past rather than remember it is the cornerstone of his conception of transference. From the intersubjective perspective, what Freud overlooked was the fact that the therapist is never merely a blank screen onto which the patient projects the past. Instead, the patient's transference arises from selective perceptions of the current actualities of the therapist's character and behavior. From this angle, what is enacted in the therapeutic relationship always reflects a spiral of mutual reciprocal influence in which the therapist's contribution is no less significant than the patient's.
As I later elaborate, contemporary intersubjective and relational theories offer clinicians the most powerful tools available for working effectively with transference-countertransference enactments. These theories require that we consider questions such as the following: What is most emotionally compelling in the immediacy of the here-and-now interaction with the patient? What is the interpersonal pattern that is presently being played out—and, in particular, what is the nature of our participation in it? How can the jointly created enactment be understood? Usually questions like these are answerable only in dialogue with the patient. Such dialogue sometimes demands that the therapist "go first": To make the latent enactment manifest, it may be necessary for us to put our own experience of the interaction into words.
Not long ago, for example, in the setting of a treatment newly begun, I was attempting to convey my empathy to a recently separated female patient who was unhappily recounting the difficulties she was having with her husband. While I felt that I was "tracking" her in quite an attuned way, she seemed to find my carefully chosen expressions of understanding consistently useless.
Responding with irritation, the patient ("Carol") dismissed most of what I had to say with words that sounded entirely reasonable but left me feeling thwarted, impatient, and increasingly frustrated. Finally, I told her that I was beginning to feel quite irritated myself, adding that I usually had the feeling here that the two of us were on the same side, as I suspected she
i 2-1
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
did, but that somehow our conversation today seemed to have turned adversarial. This caught her attention.
I realized as I was speaking to her that, in keeping with my own psychological makeup, I had been ignoring her implicitly provocative devaluation long enough that my exasperation—when I finally felt it—was experienced and expressed with a surplus of intensity. As happens not infrequently, this kind of inadvertent participation on therapist's part can turn out to be a blessing in disguise.
Before long Carol was engaged in a rather troubling consideration of the ways our interaction mirrored that with her husband, with whom at times she felt irresistibly drawn to pick a fight. Hearing this, I shared my thought that perhaps her contentiousness today might be related to her telling me at the end of the last session that therapy was finally starting to help. And now began an exploration—still ongoing—of her fear of her feelings, her strategy of self-sufficiency, and her terror of dependency and rejection. The emotionally charged interaction in this session marked a turning point in our therapy. 1 believe it illustrates the fact that making optimal use of enactments often has as much to do with the therapist's authentic responsiveness and deliberate self-disclosure as it has with interpretation.
The understanding that enactments in psychotherapy are co-created is entirely consistent with the research suggesting that early attachment relationships are themselves co-constructed. As I mentioned earlier, among the most significant conceptual bridges between the relationships of infancy and psychotherapy are those that have been generated by Daniel Stern, Karlen Lyons-Ruth, and the Change Process Study Group (CPSG). Though Stern et al. might define enactments more narrowly than I do, they consistently emphasize the essentially enactive, rather than verbal, processes that make the most significant psychological development—and therapeutic change— possible. In illuminating the healing impact of changes in the "shared implicit relationship" between patient and therapist, their unique approach makes an invaluable clinical contribution (Stern et al., 1998; Lyons-Ruth, 1999; Lyons-Ruth & Boston Change Process Study Group, 2001).
The shared implicit relationship reflects the relatively stable but nonetheless evolving sense in each partner of who the other is, who each is to the other, and who they are together. While it is the product of the actual ongoing personal engagement of patient and therapist, it is also necessarily influenced by the implicit relational knowing—the internal working model, if you will—of each partner. I suspect it is this intersubjective meeting of the self and the other, the internal and the interpersonal, the anticipated experience and the lived experience, that makes the shared implicit relationship a fulcrum of potential change.
In 1998, the CPSG published a landmark paper whose subtitle—"The 'Something More' than Interpretation"—hints at the nonverbal experiences
I
Nonverbal Experience and the "Unthought Known"
125
in therapy that effect change. Specifically, they observed that changes in implicit relational knowing occur as a function primarily of what is enacted in the intersubjective field of patient and therapist. When their relationship is altered, it shifts the patient's sense of who the therapist is, who he or she is to the therapist, and who they are to each other.
Stern and his colleagues note that therapy unfolds through a series of present moments ("beats" in the terms of drama), each of which embodies a distinct subjective sense of "what is happening now between us." At times these present moments become charged with intense feeling, pulling patient and therapist irresistibly into the immediacy and emotional heat of the here and now: The CPSG refers to these moments as now moments.
When a now moment evokes an authentic personal response from the therapist that resonates deeply with the patient, the therapeutic couple may experience a memorable moment of meeting that transforms the shared implicit relationship. A moment of meeting offers the patient a glimpse into new ways of being, beyond the constraints of preexisting transference predispositions or implicit relational knowing. Such a corrective relational experience can open the door to sudden, dramatic change.
After initially spotlighting these transformative encounters, the CPSG shifted focus to the ongoing therapeutic relationship—the larger context for the high-impact moments of meeting. As previously mentioned, development in therapy—just as in childhood—is facilitated by a relationship involving collaborative, attuned, and contingent communication. Such communication depends much more on the implicit, affective, interactive process between patient and therapist than on the explicit content of the words they exchange. Lyons-Ruth puts it this way: "Process leads content in this conception, so that no particular content needs to be pursued; rather the enlarging of the domain and fluency of the dialogue is primary and will lead to increasingly integrated and complex content" (Lyons-Ruth & Boston Change Process Study Group, 2001, p. 15).
Enlarging the affective as well as linguistic dialogue occurs through the therapeutic couple's trial-by-error "improvisation of relational moves" (Lyons-Ruth & Boston Change Process Study Group, 2001) rather than the therapist's deliberate attempt to structure the treatment. When patient and therapist both sense that they are fitting together in moving toward mutually held goals, the result is often an experience of vitalization that reinforces a growing sense that their shared relationship is a valuable and helpful one. Repeated rounds of relational improvisation tend to create increasingly effective patterns of fitting together that eventually come to compete with the patient's old predispositions and to destabilize them— thus generating the experience of possibility, flux, and disorder that is the (often disquieting) subjective precursor to change.
Whether such shifts in the patient's implicit relational knowing occur
126
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE Nonverbal Experience and the "Unthought Known
127
suddenly (in a moment of meeting) or gradually (through ongoing dialogue that is progressively more inclusive and collaborative than the patient expects) their context is always an enactive and intersubjective one. Years ago, Frieda Fromm-Reichmann remarked that the patient needs an experience, not an explanation. One might say the patient needs a relationship more than a reason why.
What is enacted in the therapeutic relationship will be a function of the interaction of the therapist's implicit relational knowing with that of the patient. Focusing on the quality of our participation as therapists is essential in order that we recognize what we may be unwittingly contributing. As we have seen, co-constructed enactments have the potential to provide some of the most vital contexts for transforming our patients' sense of themselves, of others, and of relationships.
But when clinicians unconsciously collude in replaying the patterns encoded in their patient's internal representations, co-constructed enactments can be obstacles to realizing the goals of treatment. Old learning can be locked in place, familiar expectations confirmed, the problematic past repeated; the result can be a therapeutic impasse. Worse still, the patient can be retraumatized. In addition, there is the issue of our stance toward experience: To the extent that enactments fail to attract our thoughtful attention, it's as if we're on automatic pilot—sleepwalking through our role in the interaction rather than awake to it, embedded in the experience rather than reflective or mindful about it.
All this makes it imperative for us to be consistently attentive to the nature of the implicit relationship we enact with our patients. To engage with what they cannot verbalize, we have to tune in as much to the music as the words: How do we feel we're affecting and being affected by them? How does the patient feel he is affecting and being affected by us? What can we infer about his subjective and intersubjective experience—and what do we sense about our own? We must keep in mind that every verbal exchange, every interpretation, every intervention is an interpersonal event; each one influences the shared implicit relationship in ways large or small, and often unanticipated—such that our attempts to be helpful (like my overeager effort to "reach" Eliot) can have an impact very much at odds with what we intend or expect.
WORKING WITH EVOCATIONS OF THE UNTHOUGHT KNOWN
A high-powered, rather intellectual executive came to see me, ostensibly at his wife's insistence. She complained that he was tense, distracted, and emotionally unavailable. The patient, whom I'll call Gordon, wasn't sure about this, or about his need for therapy, but he seemed willing to give it a
(brief) try. Three or four sessions into a treatment that appeared to have an increasingly dubious future, I found myself noticing the care with which I was choosing my words. I realized that I was feeling unaccountably anxious, almost as if I were being threatened by a prosecutor and needed my language to be bulletproof.
After some hesitation, I chose to share this experience with my patient. Hearing what I had to say, Gordon was astonished. He said that 1 was describing his experience, not only when he was here with me but also more broadly. He had never before found the words to describe it, but now it seemed to him that he was bringing what he called his "internal landscape" into the interaction between us. In this connection, he disclosed to me that he had a pattern at work of compulsively "goldplating" his performance, out of a vague sense of threat, and added that his mother, a Holocaust survivor, had recently asked him, "Don't you feel anxious? You must be the only Jew there."
As we explored his experience, and ours, over the course of several sessions, Gordon became aware that what drove his own bulletproofing was specifically a fear of judgment and attack, especially in environments he didn't trust to be rational. He thought his mother "had this same anxiety from her experiences" and had somehow transferred it to him. Now it appeared that he had transferred it to me. My vulnerability to feeling threatened allowed Gordon to evoke in me an experience he was at pains to avoid in himself. By "relocating" his unconscious sense of danger, he enabled the two of us to identify/articulate it and then to begin to grasp that, at some level, the feelings he was reluctant to claim as his own were, in fact, originally his mother's.
In suggesting that this patient evoked in me what he was reluctant to know and, hence, unable to tell me, I'm referring to projective identification. Conventionally understood, this is the process through which we project onto (or into) another what we cannot bear in ourselves. Then we relate to the other in such a way that he comes to identify with what we have projected. Projective identification, while usually considered a mechanism of defense, is also a mode of nonverbal communication.7
As Melanic Klein originally conceived it, projective identification was essentially the fantasy in the minds of infants and psychologically primitive adults that they could somehow relocate parts of themselves in others. The psychoanalysts Winnicott and Bion are generally credited with "interper-sonalizing" Klein's insight. They realized that what Klein had regarded as an exclusively internal phenomenon was in fact an interpersonal one: All of us—from birth onward—actually evoke in others experiences that we are unable or unwilling to claim as our own.
Bion (1962) theorized that "normal projective identification" was the single most important medium of communication in infancy. Affects that
128
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
Nonverbal Experience and the "Unthought Known"
129
were overwhelming were projected by the infant into the receptive mother, who contained and processed them before returning them to the child in a modulated and "digestible" form. The observations of infant-parent research tend to confirm Bion's theory—while adding a crucial emphasis on bidirectional influence and co-construction.
Stephen Seligman (1999), a psychoanalyst and researcher at the University of California, has suggested that a realistic understanding of infant-parent relationships must take into account the parent's projections as well as those of the infant. In these and other close relationships—such as marriage and psychotherapy—adults clearly make use of projective identification. Bion (1967) actually argued that projective identification was the most significant form of interaction between patients and therapists. What I emphasize in subsequent chapters is the complexity of projective identification: first, the fact that it is bidirectional; and second, that as therapists, we must be wary of assuming too readily that what we feel the patient has evoked in us belongs to the patient alone. Usually human beings need a hook to hang their hat on.
Precisely how we evoke own experiences in others is a matter that has been clarified somewhat by contemporary research in a variety of fields. It now appears that the transfer of internal states from infant to parent (and vice versa) as well as from patient to therapist (and vice versa) is mainly accomplished through the medium of body-to-body communication. You might say that we become what we behold: When wc perceive emotions in others, we feel these emotions in ourselves.
As previously mentioned, newborns as young as 42 minutes will imitate the facial gesture of a model who opens her mouth or sticks out her tongue (Meltzoff & Moore, 1998). At two-and-a-half months, infants will react to their mothers' displays of emotion with corresponding affects of their own (Haviland & Lelwica, 1987).8 In related research, Dimberg et al. (2000) showed adult subjects a neutral video in which 30-millisecond sequences of a smiling face and an angry face had been embedded. When exposed to these subliminal segments, the experimental subjects reflexively shifted the micromusculature of their own faces to conform to the expressions they (unconsciously) beheld on the faces in the video.
Apparently we are constructed by evolution to reflexively imitate the facial behavior of those with whom we interact. But what does imitation have to do with the transfer of internal states? Duplicating the facial behavior of another is not the same as participating in his emotional experience. Or is it?
Paul Ekman,9 the world's preeminent investigator of the phenomenology and psychophysiology of facial expression, has discovered that facial muscle reactions not only express emotions but also activate them. When we deliberately assume the facial expression associated with a particular
emotion, our physiology and pattern of brain activation change in conformity with it.10 Ekman's research coupled with the imitation studies suggests that in fact we often have access to such states in others whether we choose to or not. For when—unconsciously and involuntarily—we duplicate the facial expression of another person, we also set up within ourselves an emotional response that resonates with, matches up, or corresponds to the emotional experience of that person (Ekman, 2003; Ekman, Levenson, & Friesen, 1983).
This may be how as therapists we have the potential to know "from the inside out" what our patients are experiencing. What they cannot put into words, they convey through face-to-face communications that evoke their emotion in us. Not surprisingly, Ekman believes that the "music" of the voice (tone, rhythm, contour) both communicates and activates emotion just as the expression of the face does. Call it projective identification or nonverbal communication, the fact is that our patients will activate inside us resonances of their own experience.
The crucial point here is one at the very center of the intersubjectivc, relational approach to psychotherapy: To access what our patients cannot put into words, we must tune in to our own subjective experience. Later, I explore in detail how contemporary relational theory can help us to utilize our own subjectivity to identify, understand, and make good use of the patient's evocative influence. For now, I will only say that to receive the patient's nonverbal communications we must learn to recognize their reverberations inside ourselves.
Once we do so, it may be vital at certain times to deliberately disclose to our patient what we believe has been evoked in us. At other times we may use our awareness of what has been evoked to develop and convey a deeper understanding of the patient's unspoken experience. At still other times, patients may need to see us successfully struggle to bear experiences they have found unbearable. None of this is possible without understanding that our patients will often evoke in us what they cannot communicate through language, except perhaps the language of the body.
WORKING WITH EMBODIMENTS OF THE UNTHOUGHT KNOWN
A woman patient was acutely uncomfortable whenever the possibility arose of a silence between us. In exploring the details of this experience, she said that if there were no conversation then we would simply be looking at each other. And if that were to occur? I inquired. Then we'd each just be looking at the other's body, she answered. It'd be like we were just two bodies here. This apparently provoked for her the very troubling question of the relationship between our bodies—also known as the issue of sexuality.
130
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
We cannot exclude the body if psychotherapy is to make room for as much of the patient's experience as possible. The "talking cure" is likely to be significantly less inclusive, less integrative if it is only a conversation between talking heads. Bodily sensation is always the substrate of emotion: To a considerable extent what we feel physically is what we feel emotionally.
Preverbal experience, identified by attachment research as so influential, is largely, of course, bodily experience. And as I've indicated previously, it is body-to-body communication that provides the evocative subtext of the spoken dialogue in psychotherapy. Although much of the impact of this communication registers outside conscious awareness, it is also true that it can be very hard to find what we fail to look for. Clinicians cannot afford to ignore the body—neither their patient's nor their own—because the body often receives and transmits what has not or cannot be put into words.
The impacts of acute trauma as well as disorganized attachment are frequently somatic. A patient of mine who was chronically traumatized in childhood finds herself alternately overtaken by physical pain or anesthetized to it—as if the internal signals were either deafeningly intense or barely audible. She feels at times a prisoner of her own body and at times as if she has no body of her own. She has difficulty knowing whether her physical suffering might not really be a stand-in for her emotional suffering.
Invariably patients like this—who oscillate between overwhelming hyperarousal and numbing dissociation—have enormous difficulty with affect regulation. They have trouble translating somatic sensations into feelings they can articulate and use to guide appropriate action. The ease with which they can be autonomically triggered makes it difficult for them to think and feel; instead, they deny and dissociate. Unsurprisingly, neuro-biological studies show that patients with a history of trauma have increased reactivity in the amygdala and correspondingly diminished activity in the prefrontal cortex (Rauch et al., 2000; Shin et al., 2004). For such patients, "the body keeps the score" (van der Kolk, 1996, p. 214).
It's as if the body remembers too well the agonies it registered in the past—and now reacts as if everyday difficulties might be life-threatening disasters. Much of the work of psychotherapy with such patients involves the effort to recognize, tolerate, and label somatic states so that bodily sensations can be linked with emotions and emotions with the contexts that provoke them. With these patients the path to affect regulation and the integration of dissociated experience usually begins with the body.
In contrast to patients who switch between sympathetic nervous system hyperactivation and parasympathetic deactivation are those in a dismissing state of mind with respect to attachment. Such patients can truly seem like talking heads and rather unexpressive ones at that, appearing stiff in posture and impassive of demeanor, their voices having little inflection.11 Therapists with inhibited, deactivating patients like these may need to be
Nonverbal Experience and the "Unthought Known"
131
especially deliberate about attuning to their own bodily sensations—of tension, constriction, sleepiness, and so on. Often the reverberations of the patients' own disavowed emotions, or the defenses against them, will register first in the body of the therapist.
A clinician colleague told me that once he had felt a sharp ache in his chest while working with a "real Marlboro man of a patient." This patient had shown little affect over the course of several years of treatment. The therapist sat silently with the aching sensation in his chest and realized that it was a bodily echo of feelings he had had as a lonely adolescent. He decided to share his experience with the patient. When he did so and then asked if the patient had ever felt a similar sensation, tears welled up in the man's eyes and he began to talk for the first time about his own boyhood feelings of aching loneliness—feelings he had never been able to share or overcome.
The body, to paraphrase the psychoanalyst Otto Kernberg, is a geography of personal meanings. To get at these meanings we have to make room for the somatic self, bring our attention both to what the patient's body reveals and to the patient's relationship to her body. We also have to attend to the sensations of our own body—for often they represent resonant physiological responses to what is occurring within the patient. Finally, as indicated in the vignette that opened this reflection on the body in psychotherapy, we have to attend to the relationship between the two bodies in the room. These arc all routes to recognizing, engaging, and, if all goes well, modifying the impact of problematic formative experiences for which as yet our patients have no words.
Focusing on the nonverbal domain can allow us to connect with facets of the patient's self that have never been integrated and cannot be articulated. By becoming aware of what the patient enacts with us, evokes in us, or embodies, we have the opportunity to begin to know something about the patient's "unthought known" while often learning about ourselves in the process. Attending to the relational, intersubjective, and affective undercurrents of the therapeutic interaction can help make room for experience and awareness the patient has previously had to disown. And ideally our responses to the patient can enable that experience and awareness to deepen.
NOTES
t. Note that the research evidence should not be misconstrued to suggest that the trajectory set by our first relationships is unalterable. Working models have the potential to be updated; early experience establishes stable patterns, not rigid structures.
2. Thus deprived of the capacity for expressive language, people who are traumatized may-experience "speechless terror" (van der Kolk et al., 1996).
3. Individuals traumatized in childhood have been found, in some studies, to have smaller
132
FROM ATTACHMENT THEORY TO CLINICAL PRACTICE
left hippocampi and diminished left-brain development, in general, when compared with healthy control subjects. Correspondingly, when adults with a history of abuse were asked to recall a disturbing early memory, their hemispheric activity showed a radical skew to the right; when recalling a neutral memory, the skew was way to the left. In a control group, by contrast, hemispheric activity was balanced, regardless of whether the memory was disturbing or neutral. Moreover, the volume of the corpus callosum—the hrain's main channel of information exchange between its two hemispheres—was found to be significantly smaller among traumatized individuals than among controls without such a history. Trauma thus appears to impede neural as well as psychological integration—isolating the emotional right brain from the verbal resources of the left.
4. It perhaps determines the usual choice of mother as the primary attachment figure because, as Main (1999) notes, the maternal voice to which the newborn orients was such a familiar feature of the intrauterine "soundtrack."
5. Together with the Freudian unconscious produced by repression, this implicit nnncan-scious (Stern, 2004; Siegel, 1999) is probably what Hollas had in mind when he coined the term the "unthought known."
6. The self psychologist Michael Basch (1992) has written that "the patient subtly causes the therapist to resonate autonomically with the patient's unconscious" (p. 179).
7. In this connection, Schore (2003) writes: "Freud began to model the state of mind of 'evenly suspended attention' in which one could receive the unconscious communications of others. I suggest that if Freud was describing how the unconscious can act as 'a receptive organ,' Klein's conception of projective identification attempts to model how an unconscious system acts as a 'transmitter,'and how these transmissions will then influence the receptive functions of another unconscious mind" (p. 59).
8. Mothers in facc-to-face interactions with their infants were asked to adopt a variety of facial expressions: In response to a show of joy from their mothers, the infants' own joy appeared to be heightened and their "mouthing" movements decreased; in response to the mothers' sad face, the infants appeared subdued and mouthing increased; in response to their mothers' angry expression, the infants showed anger and their bodies stilled (Haviland &c Lelwica, 1987).
9. Ekman initially won renown for demonstrating cross-culturally that every basic emotion (sadness, angei, fear, surprise, etc.) is associated with a signature pattern of facial muscle reactions: enjoyment, for example, is reflected in the smile but also crucially in the involuntary engagement of the muscles around the eyes.
10. Illuminating the process, Ekman quotes from The Purloined Letter in which Edgar Allan Poe, writing in the persona of a detective, explains how he deliberately gains access to the internal states of others: "When I wish to find out how wise or how stupid or how good or how wicked is anyone, or what are his thoughts at the moment, I fashion the expression of my face, as accurately as possible, in accordance with the expression of his, and then wait to see what thoughts or sentiments arise in my mind or heart, as if to match or correspond with the expression" (Poe, cited in Ekman, 2003, p. 37).
11. Ekman's research suggests that such inhibition of the body actually inhibits the subjective experience of emotion in infants as well as adults.
CHAPTER 9
The Stance of the Self toward Experience
Embeddedness, Mentalizing, and Mindfulness
The investigations of Main, Fonagy, and others confirm that the capacity to reflect coherently upon experience—rather than being embedded in it or defensively dissociated from it—is a marker of both our own attachment security and our ability to raise children (and perhaps patients) who will also be secure. As in the Adult Attachment Interview of a secure adult, this capacity for a "reflective" or "mentalizing " stance (1 use the terms interchangeably) is manifested in a coherent account of experience that, in turn, reveals a coherent self. By that I mean a self that (1) makes sense rather than being riddled with inconsistencies; (2) hangs together as an integrated whole rather than being fractured by dissociations and disavowals; and (3) is capable of collaboration with other selves. Following the lead of Daniel Siegel (2006), I would suggest that a coherent self is also one that is stable, adaptive, flexible, and energized.
As psychotherapists, we aim to help our patients to live, more and more, from within such a coherent self. Our task is to co-create a relationship with our patients that allows them to make sense of their experience, to feel more "together," and to relate to others more deeply and with greater satisfaction. Central to this task in the model of psychotherapy I am proposing is the therapist's mentalizing stance that fosters the patient's own
133