42
BOWLBY AND BEYOND
was unready to claim a decisive role for the quality of the parents' metacognition in determining whether their children were securely or insecurely attached (Main, 1991). Several years later, however, Peter Fonagy—a psychoanalyst researching attachment on the opposite side of the Atlantic— was indeed able to make exactly that claim. He did so while extending Main's conceptualization of metacognition in light of the psychological capacity for what is known as a "theory of mind" (Fonagy et al., 1995).
CHAPTER 4
NOTES
1. Importantly, this "state" has heen shown by Main's research (and that of others) to have such stability over time that it may as accurately be described as a trait—though by no means an immutable one.
2. Recall that, as the second stage of her longitudinal study, Main and her colleagues in 1982 had conducted two-hour-long videotaped assessments of 40 families. Built around separation and reunion, the process for evaluating the six-year-olds and their parents was specifically geared—like the Strange Situation itself—to activate the attachment system. Each family upon arriving at the research site was asked to pose for a Polaroid photograph. Then the whole family was shown a short film that dramatically depicted a two-year-old child's separation from his parents (Robertson & Robertson, 1971). Next came the separation in real time: The six-year-old was led to a playroom, while in separate offices each parents was given the AAI. Meanwhile the six-year-old was participating for 20 minutes in a "warmup" session with a female examiner who then proceeded to show the child a series of six photographs of children about to undergo separation from their parents. After seeing each photograph the child was asked what the children in the picture would feel and what they would do when their parents left. Then the examiner showed the child the Polaroid that had been taken earlier saying "But here is a photograph of yourself and your family, and you see, you arc all together" (Main et al., 1985, p. 89). The child was also asked to draw a picture of the family. Finally, after the child had been engaged in a period of free play in a sandbox, the first parent returned. The reunion lasted three minutes, following which the second parent returned and joined the reunion for as much as three minutes more.
3. Predictions of the six-year-olds' attachment representations from their Strange Situation behavior with mother ranged from 68 to 88%. Interestingly, only reunion behavior and discourse were significantly related to Strange Situation behavior with father (Main, 1995).
Fonagy and Forward
Main has said that she and Fonagy were drawn independently to the research on theory of mind (Main, personal communication, 2004). This term refers to the ways in which all of us, to varying degrees, make sense of our own and others' behavior on the basis of underlying mental states—including beliefs, emotions, and desires. The idea here is that, beginning in childhood, we develop a "theory" that enables us to understand and, to some extent, predict what others will do in light of what we think is going on in their mind. Fonagy may have been inspired by the same literature on theory of mind that influenced Main, but his reading of that literature led to a much broader conceptualization than hers.
PETER FONAGY: MENTALIZING, MODES OF EXPERIENCE, AND THE INTERSUBJECTIVE ORIGINS OF THE SELF
As a junior lecturer at University College London more than a quarter of a century ago, Fonagy had been appointed as liaison to organize the seminars and consultations of the visiting Freud Memorial Professor of Psychoanalysis— who turned out in 1980 to be none other than John Bowlby. Sitting in on Bowlby's teaching that year, Fonagy (who now holds the Freud Memorial Professorship himself) says he was extremely taken not only with Bowlby's ideas but also his profound social concerns:
Bowlby was deeply committed to the welfare of people who were less privileged than he was. I found his vision enormously impressive—of combining science with concern for the individual and individual subjectivity, together
43
BOWLBY AND BEYOND
Fonagy and Forward
45
with looking at social forces and social pressures. That remains for me the aspect of attachment theory that is most important and most intriguing. (Fonagy, personal communication, 2006)
Several years later, Fonagy in collaboration with Miriam and Howard Steele launched a study on the transgenerational transmission of attachment patterns. Throughout their study, Fonagy and his colleagues received consultation from Bowlby. They also received training in the AAI from Mary Main.
Mentalizing and the "Rellective-Functioning" Scale
Inspired by Main's groundbreaking work, Fonagy initially attempted to operationalizc her ideas about individual differences in metacognitivc capacities. But whereas Main had concentrated on the adult's se/^-monitoring of thought and recall during the AAI, Fonagy (taking his cue from the theory of mind) widened that focus to take in the adult's attention to mental states generally, including, in particular, the mental states of others (Fonagy, Steele, 8c Steele, 1991a). Fonagy would later remark that the hallmark of the ability he referred to as mentalizing—that is, "the process by which we realize that having a mind mediates our experience of the world"—was not se/f-knowledge, but rather knowledge of minds in general (Fonagy, Gerge-ley, Jurist, 8c Target, 2002, p. 3). While such knowledge is largely implicit, Fonagy and Target (in press) use the term "mentalization proper" to describe the activity of "thinking explicitly about states of mind" (p. 2). Mentalizing activity (say, a daughter noting that her father's "rejection" of her might have resulted from his depression rather than his hostility) is rooted in what Fonagy called the capacity for reflective function.
Reflective function lets us see ourselves and others as beings with psychological depth. It enables us to respond to our experience on the basis not only of observed behavior, but also of the underlying mental states— desires, feelings, beliefs—that make behavior understandable and give it meaning. As such, reflective function is intimately related to our capacities for insight and empathy.
To evaluate the strength of an individual's mentalizing capacity, Fonagy and his colleagues devised the Reflective-Functioning Scale. Designed for research purposes, this scale can also be used informally to enhance clinical judgments about the kinds of interventions our patients may be able to benefit from. A strong mentalizing capacity—and perhaps in treatment, receptivity to the therapist's interpretations—is likelier to be present when the interviewee (or patient) demonstrates:
• Awareness of the nature of mental states—for example, that our understanding of ourselves and others is invariably incomplete; that
people may modify mental states to minimize pain; that people may deliberately disguise internal states; that certain psychological responses are predictable given certain circumstances.
• Explicit effort to identify mental states underlying behavior—for example, plausibly accounting for behavior in terms of beliefs, feelings, desires; understanding that our interpretations of others may be influenced by our own mental states; realizing that feelings about a situation may be inconsistent with observable aspects of the situation.
• Recognition of the "developmental" aspects of mental states—for example, that what was felt yesterday may be different from what is felt today or tomorrow; that parents' behavior is both shaped by their own parents' behavior and shapes the behavior of their children; that childhood perspectives often need to be revised in light of adult understanding.
• Awareness of mental states in relation to the interviewer {or therapist)—for example, that without being told, the therapist cannot know what the patient knows; that the therapist may have her own distinctive emotional responses to the patient's story; that the therapist's history, and consequently, her mental states may well be different from those of the patient (adapted with permission from Fonagy, Target, Steele, & Steele, 1998).
Fonagy makes the point that what we need to listen for here are not enunciations of principles concerning mental states ("One can never know what someone else feels") but rather evidence that such principles arc implicitly understood ("As a child, I was sure that my mother didn't care for me, but given what I've heard from my father about how she felt that I rejected her, now I'm not really sure what she felt").
In 1987, Fonagy and the Steeles recruited 100 expectant couples and conducted AAI research using their scale. The study was notable for a number of reasons. First, it documented that the parents' state of mind with respect to attachment—assessed before the infant's birth—could predict that infant's Strange Situation classification at 12 months. It also showed that mothers and fathers with a strong reflective capacity were three to four times more likely to have secure children than parents whose capacity for mentalizing was weak. Finally, it demonstrated that a strong reflective capacity could break the "cycle of disadvantage" that ordinarily led parents with adverse attachment histories to raise insecure children.
To test the prediction that an ability to reflect on mental states could function as an "antidote" to the problematic past, the mothers in the study were divided into two groups: the first had experienced severe deprivation (parental mental illness, prolonged separation from parents, and so on); the second had not. It turned out that among the deprived mothers with strong
46
BOWLBY AND BEYOND
Fonagy and Forward
47
reflective functioning, every single one had secure children. In stark contrast, among those whose reflective functioning was weak, only one in 17 had a secure child. Clearly, the strength of the capacity to mentalize was a protective factor that buffered the impact of difficult early experience and diminished the probability of the intcrgenerational transmission of insecurity (Fonagy, Steele, Steele, Moran, &. Higgitt, 1991b; Fonagy etal., 1995; Fonagy, 2001).
Discoveries such as these led Fonagy to regard mentalizing as absolutely central to attachment. In fact, he has gone so far as to contend that "attachment is not an end in itself; rather it exists in order to produce a representational system that has evolved, we may presume, to aid human survival" (Fonagy ct al., 2002, p. 2). This representational system is a mentalizing system that offers the enormous evolutionary survival advantage of enabling individuals to understand, interpret, and predict the behavior of others, as well as their own behavior. As such it is a "cornerstone of social intelligence" and critical to work, play, and collaboration of all kinds (Allen & Fonagy, 2002).
Over more than 15 years, Fonagy and his colleagues' extraordinarily active program of research and theory building has made an increasingly powerful case that mentalizing and attachment play critically important, intertwining roles in development, psychopathology, and psychotherapy. In brief, the parent's mentalizing is crucial to promoting secure attachment in the child and secure attachment provides the key context for activating the child's own mentalizing potential. Much of the psychopathology we encounter in our patients can be seen to reflect either an inhibition of mentalizing or a failure to develop it in the first place. Correspondingly, psychotherapy can be understood as an effort to restore or kindle the patient's capacity to mentalize.
Consistent with Main's theorizing about the internal working model, Fonagy has proposed that what is key about the representation of early attachment experience is less the "template" we register than the depth to which our mentalizing ability allows us to probe our experience, especially experience that is emotionally charged. He has offered the term "mental-ized affectivity" to describe the capacity to simultaneously feel our feelings and reflect upon their meaning. Fonagy's preeminent collaborator, Mary Target, has put it this way: "Reflective function, at its most developed, involves thinking about feeling and feeling about thinking" (personal communication, 2005).
Modes of Experience
Much of Fonagy's exploration has been aimed at helping us to understand the modes of psychological experience that reflect our sense of the relationship between the internal world and external reality. Fonagy describes three such subjective modes: psychic equivalence, pretense, and mentalizing.
In the mode of psychic equivalence, the internal world and external reality are simply equated. There is no differentiating here between beliefs and facts. What we think and feel seems to mirror what occurs to us in the physical world, and vice versa. In this frame of mind, when we are treated badly, for example, we are likely to feel that we are bad—and feeling that we're bad, we "know" that we will be treated badly. In such a closed system, the self as psychological agent tends to be submerged: There is no "I" that interprets or creates experience but only a "me" to whom experience happens.
In the "pretend" mode, the internal world is decoupled from the external one. Here we are unfettered by actualities: Whatever we imagine is felt to be real and whatever we ignore is rendered immaterial. Dissociation, denial, and extreme narcissistic grandiosity are all examples of the "pretend" mode. In this mode, like the one above, the self as interpreter or creator of experience is constrained, because taking reality into account threatens what has been imagined and opens the door to what has been ignored.
In the mentalizing (or reflective) mode, we are able to recognize that the internal world is separate from, but also related to, external reality. Here we can reflect on the ways in which our thoughts, feelings, and fantasies both affect, and are affected by, what actually happens to us. In this mode, our subjective experience is felt to have interpretive depth and thus—because we can grasp the difference between events and our reactions to them—we can enjoy a measure of internal freedom. Mentalizing reveals a world of self and others that is rich, complex, and ambiguous— and one in which we have the potential to revise our mental representations of external reality as our actual realities change.
According to Fonagy, these modes of experience unfold sequentially in the course of development. At first, infants and small children live inescapably in a world of psychic equivalence in which subjective experience is compellingly, and sometimes terrifyingly, real. Then, they find a kind of liberation through the mode of pretense in which subjective experience is decoupled from reality: In play, they can pretend that the constraints of reality simply do not exist. Finally, in normal development, beginning at age four or so, there comes about an integration of these two earlier modes. Now the internal world is neither equated with, nor completely severed from, the external one. With the emergence of the reflective mode comes a growing ability to consider, implicitly and explicitly, the relationship between internal and external reality (Fonagy, 2001; Allen &C Fonagy, 2002; Fonagy et al., 2002).
The patients we see in psychotherapy often have trouble extricating themselves from the modes of psychic equivalence and/or pretense. In the first case, they are bullied by feelings and thoughts that demand to be acted on because they are equated with facts. In the second, they are kept aloft by wishful thoughts, but isolated in the process from their feelings and from the people who might matter to them.
48 BOWLBY AND BEYOND
For psychotherapists and parents, as well as researchers, the key question must be: What fosters the transition out of the experiential modes of psychic equivalence and pretense into a mentalizing mode? Fonagy's answer—an elaboration on the conclusions of Bowlby, Ainsworth, and Main—is: an intersubjcctive relationship of attachment that provides first a full measure of affect regulation and then, not unimportantly, a modicum of play in the presence of a reflective other.
Affect Regulation, Intergenerational Transmission, and Intersubjectivity
Fonagy believes that the bridge to a reflective mode of experience is built on a foundation of affect regulation. While the "biological function" of attachment behavior is protection from predators, babies also need the attachment figure to help ensure their emotional survival. From the moment they arc born, babies are subject to feelings of distress that they are utterly unequipped to manage on their own. To experience the "felt security" that has been described as the set goal of attachment, babies depend on the attachment figure to help them modulate their overwhelming affects.
According to Fonagy, parents who generally succeed in "containing" their infant's distress will usually have a securely attached child with a solid potential for mentalizing. Why should this be so? And what is the tieup between affect regulation, security of attachment, and mentalizing?
Successful Containment and Secure Attachment
The psychoanalyst Wilfrid Bion (1962) proposed that the supportive mother mentally contains emotional experience that the baby cannot manage on his own but manages to evoke in her. Such containment requires the mother to bear within herself, to process, and to re-present to the baby in a tolerable form what was previously the baby's intolerable emotional experience. Drawing on Bion's ideas, Fonagy suggests that parents can contain their infant's distressing affects through communicating affectively and in the language of physical care that (1) they understand the cause of the distress and its emotional impact; (2) they can cope with the distress and alleviate it; and (3) they can recognize the child's emerging intentional stance—by which is meant the child's ability to infer the intentions that underlie behavior, in particular, the behavior of the parent (Dennett, 1987). Strikingly, Fonagy believes that this third element of containment—the parent's recognizing the child as a separate being with a mind of her own, capable potentially of reading her parent's mind as well as her own—"may be the most important in maximizing the likelihood of the child's forming a secure attachment" (Fonagy et al., 1995, p. 248).
Parents who successfully contain their infant's unmanageable emotions
Fonagy and Forward
49
with responses that convey empathy, coping, and appreciation of the child's intentional stance are engaged in a process of interactive affect regulation. Through this process, they are reinforcing their child's confidence in the attachment relationship as a safe haven and secure base. And by acknowledging the child's intentional stance these (mentalizing) parents are providing the building blocks for what will become the child's own mentalizing ability. Note the synergy here of affect regulation, secure attachment, and mentalizing.
MIRRORING THE AFFECTS OF THE CHILD
... the precursor of the mirror is the mother's face.
—I). W. WINNICOTT (1971a, p. Ill)
Fonagy observes that emotionally attuned parents convey their empathy and capacity to cope through affect mirroring that is both "contingent" and "marked." Contingent mirroring is accurate: The parent's facial or vocal displays correspond to the affects of the infant—and thus the parent's affective expressions become the basis for the child's first representations of her own affects. For such contingent displays to be seen as reflections of the child's emotional experience, rather than the parent's, the parent must "mark" these displays as pretend or "as if"—for example, by exaggerating the affect that is mirrored or by intermingling the disturbing affect with one that contradicts it (Fonagy et al., 2002). In these ways, the parent's responses are "giving back to the baby the baby's own self" (Winnicott, 1971a, p. 118).
Such emotionally attuned mirroring is absolutely critical, for it is through "resonating with, reflecting on, and expressing the internal state which the infant displays" that the parents allow the child to gradually discover her own emotions as mental states that can be recognized and shared—a discovery that lays the foundation for affect regulation and impulse control (Allen & Fonagy, 2002, p. 11). In addition, by seeing herself in the mirror of her parents' image of her as a being with an intentional stance, the child can begin to experience herself as a feeling, believing, desiring, and mentalizing individual, who responds to experience not only in terms of physical realities but also in light of mental states (Fonagy et al., 1995). Perhaps most important, it is through the markedness of the parents' mirroring that the child develops the awareness that her mind is her own:
In markedness we deny what we feel while at the same time maintaining our individuality. In effect, we become what the child needs us to be. This is the process at the core of the child's emerging individuality. And if the caregiver is unable to do that—if the caregiver is either too much themselves [noneontingent mirroring] or too much the child |unmarked mirroring]—
50
BOWLBY AND BEYOND
the child cannot develop a sense of separatencss in the same kind of effective way. (Fonagy, personal communication, 2006)
Broadly speaking, then, this is the process—interactive affect regulation made possible by mentalizing at a high level of contingency in a vital relational context—through which secure attachment is transmitted from one generation to the next. Per Fonagy, it is mentalizing that mediates this process and can potentially" enable even parents whose own history of attachment is problematic to raise children who are securely attached.
The Intergenerational Transmission of Insecure Attachment
Fonagy identifies a number of inauspicious scenarios, in all of which the common theme is the inability of the parent to contain the child's unmanageable affects. Preoccupied parents may be able to empathically mirror a child's distress but without being able to deal with it. Dismissing parents may fail to communicate empathy but succeed in conveying a sense of coping and stability. There are also parents whose own vulnerabilities compromise their capacity to respond empathically to the child's intentional stance toward their mental states. Central among these vulnerabilities are the parents' own mentalizing deficits as well as the reverberating anxieties triggered by the child's separateness. For example, an expecting mother told me that she was uncomfortable with babies because she was sure they could "see through her." And what was she afraid they might see? Her scornful intolerance of their crying, their tantrums, their messincss, all the ways they could get in her way.
When parents are chronically unable to contain their child's painful affects, the child tends to behave in ways that reflect an internalization of the parent's characteristic responses to those affects. Tf a dismissing mother, say, responds to her infant's distress by ignoring or suppressing it, that infant may well develop an avoidant strategy for dealing with her own painful affects—in other words, she will avoid or suppress them. Effectively, the children of insecure parents "borrow" their parents' defenses and, thus, the
legacy of parental insecurity is often (Fonagy et al., 1995).
parallel insecurity in the child
PROBLEMATIC MIRRORING
If the mother's face is unresponsive, then a mirror is a thing to be looked at hut not looked into.
—D. W. wINNICOTT (1971a, p. 113)
Fonagy suggests that particular kinds of psychopathology may be associated with particular failures of attunement and mirroring. When the par-
Fonagy and Forward
51
ent's affective mirroring is not "marked" it can lead the child to feel overwhelmed by the contagious nature of his distress—for his upset seems only to provoke an identical emotion in the parent. Repeated exposures to unmarked mirroring are thought to reinforce the psychic equivalence mode because the child's internal experience seems regularly to be matched by his external experience, and there may appear to be no way out. Fonagy theorizes that this may be part of the genesis of borderline pathology.
In contrast, mirroring that is noncontingent can result in a sense of internal emptiness and variations on the false-self theme—because what the child is invited to internalize is an image not of his own emotional self but rather the emotional self of the parent. Because the links here are severed between the child's internal experience and its reflection in the responses of the external world, noncontingent mirroring is thought to reinforce the use of the pretend mode. Thus, according to Fonagy, the child who is regularly-exposed to mirroring that is not contingent may be vulnerable to narcissistic pathology in which imagined grandiosity functions as an anodyne for the empty self (Fonagy et al., 2002).
Beyond Containment
If the pathway out of psychic equivalence and pretense begins with containment and affect regulation, it leads to the terrain of reflection in part though the medium of play. When the child is completely absorbed in his or her own play, the worlds of imagination and reality can seem to be entirely separate. But if that play is watched by a parent, an older child, or, for that matter, a therapist, then the pretend world and the real world can start to overlap. With a comment, a look, or an "interpretation" the observer makes links between internal experience and external reality, such that the two can begin to be related—rather than equated or dissociated. Thus, the groundwork is laid for mentalizing and, further, for "metarepresentation"—the ability to reflect on one's own internal experience and that of others in light of the merely representational nature of that experience (Fonagy et al., 2002).
Fonagy stresses that such development occurs only in a relational and intersubjective context. The affective mirroring discussed earlier is a core element of the intersubjective process through which we find aspects of ourselves in the minds of others. Whether in the course of childhood development or psychotherapy, it appears that the psychological, emotional, reflective self is discovered (or perhaps created) primarily as it is recognized and understood by others. The optimal setting for such recognition and understanding is, of course, a relationship of attachment.
52
BOWLBY AND BEYOND
Fonagy and Forward
53
FROM ATTACHMENT TO INTERSUBJECTIVITY
The first extended tutorial in intersubjective awareness is usually conducted with an attachment figure whose presence and participation are necessary for the child's survival. . ..
—Karlen Lyons-ruth (1999, p. 605)
Intersubjectivity has become a pivotal concept for infancy researchers and psychoanalytic theorists alike, yet there is no consensus regarding its meaning. Intersubjectivity is variously seen to be a feature of the human condition from birth, a developmental achievement, and/or a theory of therapy. Given the multiple uses of the word, Beebe, Knoblauch, Rustin, and Sorter (2003) recommend that we think in terms of "forms of intersubjectivity." Because intersubjectivity, broadly speaking, describes the interaction between two subjectivities—the interface of two minds—it is clearly central to the developmental relationships of childhood and psychotherapy that are the subjects of this book.
Innate Intersubjectivity
Infancy research suggests that rudimentary forms of intersubjectivity are present virtually from birth. We appear to be preprogrammed—neurologically "hard-wired"—for intersubjectivity. (The discovery of mirror neurons is a key part of this story that will be taken up in Chapter 5's discussion of the neurobiology of attachment.) Andrew Meltzoff and Colwyn Trevarthen are the researchers who first detected evidence among newborns for what Trevarthen (1979) calls primary intersubjectivity.
Meltzoff and his colleagues discovered that at 42 minutes of age, infants intentionally imitate the facial expressions of an adult. While initially sucking on a "pacifier" and thus unable to imitate, the infants watched the adult's face—mouth wide open or tongue protruding. Then for the next two-and-a-half minutes, with the pacifier removed, they were observed making progressively closer approximations of the expression they had seen on the adult's face. At six weeks, infants exposed to the facial gestures of an adult on one day will imitate those gestures on the following day when confronted with that adult's "neutral" face. Evidently very young infants can detect—and preserve a mental representation of—the correspondence between what they see on someone else's face and what they feel in their own face. This deliberate cross-modal matching indicates an extraordinarily early capacity for interconnectedness of self to others—and for the sense that "I am like you" (Meltzoff, 1985, 1990; Meltzoff & Moore, 1998). Through imitation and the perception of correspondence, infants discover aspects of themselves in others.
Trevarthen, like Meltzoff, found evidence for innate intersubjectivity in neonatal imitation. But beyond imitation, Trevarthan observed mutually reg-
ulated communication between infant and adult: "Each partner can mirror the motivations and purposes of companions, immediately. Infants and their partners are thus in immediate sympathetic contact" (Trevarthen, quoted in Beebe etal., 2003, p. 782). The basis for this contact is the baby's capacity virtually from birth to detect contingent effects—that is, to recognize when the partner's behavior is (or is not) contingent upon the actions of the baby, and vice versa. Babies strongly prefer contingent responsiveness and their ability to recognize such responsiveness makes mutual regulation of joint action possible. According to Trevarthen, it is specifically through "reading" the matches in timing, form, and intensity of each other's communicative behavior that infants and their partners conduct nonverbal "protoconversations." These action dialogues reveal an exquisite mutual intercoordination that seems to reflect a matching not only of behavior but of the corresponding inner states—particularly the motives and intentions—of self and other (Trevarthen, 1998). Fonagy concurs with Trevarthen that human beings are preadapted for intersubjective relatedness: "There is primary intersubjectivity. We're born believing that our mind exists in other people. We turn to other people to see what's in our mind and to find out what things mean" (Fonagy, personal communication, 2006).
In much the same vein, Daniel Stern (1985) notes that babies turn to their mothers in order to find out about themselves—but also to find out about others and about the world. At roughly nine to twelve months of age, according to Stern, the infant makes the momentous discovery that he has a mind, that his mother has a mind, and that the subject matter of the mind—inner subjective experience—can be shared. In a classic experiment, for example, a one-year-old is invited to step across a mildly frightening glass surface in order to reach a tempting toy that lies on the other side of the "visual cliff." Confronting this ambiguous situation the baby invariably looks to the mother's face—and the subjective appraisal it conveys—to determine what his own subjective experience ought to be: When mother smiles, the baby appears to feel safe enough to boldly venture forth; when she looks frightened, the baby seems upset and he retreats. This kind of "social referencing" involves a form of joint attention in which the child makes up his own mind by looking into the mind of an other.
Such a sharing of the focus of attention is one of the hallmarks of intersubjective relatedness. There are two others: the sharing of intentions and, perhaps most important, the sharing of feeling states. Affect attunement is Stern's term for the double process of emotional resonance and communication that allows us both to feel something closely akin to what someone else is feeling and to convey to the other the fact of that shared experience.
Stern observes that in order for the baby to feel that his mother is attuned to his internal state, her imitation of his affects is insufficient because it communicates a response to his behavior rather than his experience. To convey her sharing of the baby's experience, the mother must offer a
54
BOWLBY AND BEYOND
"cross-modal" response to the baby' emotionally expressive behavior. For example, the mother might vocalize in rhythm with the baby's bodily expressions of excitement or she might convey attunement to the baby's disappointment by making audible sighs that have the same temporal profile as the baby's unhappy cries. The timing, intensity, and form of communicative behaviors (here Stern echoes Trevarthen) are the nonverbal dimensions through which attunement is expressed.
For Stern, it is the sharing of intentions, feelings, and focus of attention that define the nature of the child's intersubjecrive experience and, in due course, the child's experience of him- or herself. Here the crucial developmental question is what can, and cannot, be shared. The answer to this question determines "which experiences are within and which are beyond the pale of mutual consideration and acceptance" (Stern, 1985, p. 208). An intention, feeling, or attentional focus that evokes an attuned response is one that is sharable and thus can be integrated into the child's sense of self, whereas one that fails to evoke such a response can neither be shared nor integrated. From this angle, intersubjective relationships are the crucial context within which subjective experience takes shape. As Stern (2004) puts it, "Two minds create intersubjectivity. But equally, intersubjectivity shapes the two minds. The center of gravity has shifted from the intrapsychic to the intersubjective" (p. 78).
Attachment, Mentalizing, and Intersubjectivity
Stern believes that attachment and intersubjectivity are separate and complementary motivational systems. The attachment system balances our related needs for the security of physical proximity and the learning that exploration makes possible. The intersubjective system is driven by our need to know and be known by others. If attachment exists to foster felt security, intersubjectivity exists to promote the experience of psychic intimacy and belonging. Like attachment, intersubjective experience confers an evolutionary survival advantage. For one thing, it facilitates the formation and effective functioning of groups (including the group of two who fall in love). It also contributes to the formation and maintenance of self-identity. While it is possible to be attached without intersubjective relatedness (think of autism) and intersubjectively related without attachment (think of a chance meeting with a stranger who seems a soulmate), it is generally true that attachment and intersubjectivity are mutually enhancing (think of psychotherapy).
Stern sees mentalizing as a manifestation of intersubjectivity, but also regards the two, I would guess, as quite distinct. Mentalizing, as the word's connotations may suggest, is a process that allows us to understand and make meaningful sense of our own experience and that of others. Intersubjective relatedness, as the connotations of the phrase may suggest, has less
Fonagy and Forward
55
to do with understanding and meaning than it does with resonance, alignment, and the "sharing of mental landscapes" between ourselves and others. It is the permeability or "interpenetrability" of personal boundaries that allows us to participate in the subjective experience of other people. In this light, the affect attunement that is a signature of intersubjectivity can be seen as a matter not just of communication but also of what Stern calls "interpersonal communion"—that is, joining in, being with, or sharing the subjective experience of another person with no attempt to change it (Stern, 1985, 2004).
Intersubjectivity as a Developmental Achievement
Trevarthen and Stern highlight not only the innateness of intersubjectivity and its origins in the perception of correspondence between the experience of self and other, but also the fact that intersubjective experience is mutually regulated. Taking the concept of intersubjectivity a step further, the psychoanalytic theorist Jessica Benjamin proposes that intersubjectivity in the fullest sense depends upon not only correspondence but also difference, not only mutual regulation but also mutual recognition.
Benjamin (1990/1999) asks, "How is the meeting of two subjects different from one in which a subject meets object?" (p. 35). In answering the question, she suggests that the capacity for intersubjective relatedness is an evolving and imperfectly acquired one that, at its zenith, makes possible an encounter between two subjects—two separate but equivalent centers of initiative and experience. This is a meeting of minds in which, as Benjamin puts it, "you know what I feel, even when I want or feel the opposite of what you want or feel" and "we can share feelings without my fearing that my feelings are simply your feelings" (p. 40). This sort of intersubjective relatedness exemplifies what Benjamin calls mutual recognition—that is, the ability both to recognize and be recognized by another.
Such recognition is fundamental to an "intersubjective" as against an "intrapsychic" experience of others. The first depends on our perceiving the other as a separate subject who primarily exists outside our mental field of operations. The second involves our responding to the other primarily through projection, identification, and other intrapsychic processes—in which case the other is essentially an object in our representational world, to be idealized or devalued, perhaps, but not experienced as a real person. In the terms of Martin Buber's (1923/1970) "interhuman" philosophy of dialogue, intersubjective relatedness makes for an "I-Thou" relationship marked by mutuality, dialogue, and the ability to experience others in their own terms. By contrast, intrapsychic relating confines us to an "I—It" relationship in which mutuality is absent, imposition supersedes negotiation, and preexisting categories dominate our experience of other people.
On the one hand, Benjamin (1990/1999) (rephrasing Freud) asserts in
56
BOWLBY AND BEYOND
Fonagy and Forward
57
regard to the aims of psychotherapy, "Where objects were, there subjects must be" (p. 34). On the other hand, she believes (as does Buber) that both kinds of relationship—subject/subject and subject/object—arc necessary and that, paradoxically, each makes the other possible.
In clarifying the paradox, Benjamin (1990/1999) traces the development of intersubjectivity from the infant's earliest awareness of correspondence and similarity to the toddlet's rapprochement conflicts between separation/ individuation and relatedness starting in the middle of the second year:
We could say that it begins with "we are feeling this feeling," and then moves to "1 know that you, who are an other mind, share this same feeling." In rapprochement, however, a crisis occurs as the child begins to confront difference—"You and I don't want or feel the same thing." The initial response ro this discovery is a breakdown of recognition between self and other: I insist on my way, I begin to try to coerce you, and therefore I experience your refusal as a reversal: you are coercing me." (p. 40)
Repairing the "breakdown of recognition" is crucial if the child is to begin to acquire the capacity for mutual recognition. Developing this capacity is, of course, not a one-time event, but rather a process that can be initiated early but continues lifelong. It requires that the child become capable first, of asserting her own reality and second, of accepting the opposing reality of the other.
Following Winnicott, Benjamin explains that it is only the child who forcefully expresses her anger and finds that the other "survives destruction"— neither retaliating nor withdrawing—who has the opportunity to learn that the other is, in fact, a separate subject rather than an object. Here is the resolution of the paradox that in being treated as an object, the other may be found as a subject. Through the child's failed attempt to impose his or her will and negate that of the other, the child discovers that differences need not be a barrier to shared experience and that dialogue can obviate the requirement that one must dominate and one submit. This discovery allows the child to take fuller possession of his or her own subjectivity, realizing that a relationship can potentially make room for two—two wills, two views of reality, two subjects.
Note that mutual recognition and mcntalizing arc conceptual cousins. Without referring to Benjamin's discussion of subjects and objects, Fonagy makes a distinction that parallels hers, suggesting that when we mentalize we respond to others not as objects but as persons.
Intersubjectivity as a Theory of Psychotherapy
Benjamin gives credit to Robert Stolorow and his colleagues for introducing the term "intersubjectivity" into the arena of clinical theory. Characterizing their own work as an expansion of psychoanalytic self psychology,
they describe intersubjectivity in the analytic context as "the specific psychological field constituted by the intersection of two subjectivities—that of the patient and that of the analyst" (Stolorow, Brandschaft, &C Atwood, 1987, p. 1). Similarly, in A Primer of Clinical Intersubjectivity, Natterson 'and Friedman (1995) state, "Intersubjectivity is the over-arching term that refers to the reciprocal influence of the conscious and unconscious subjectivities of two people in a relationship" (p. 1).
While these writers use the word "intersubjectivity" in connection with a specific approach to treatment, I regard it more broadly as the best umbrella term for an invaluable body of clinical research that has taken shape in the last 20 years, that both echoes and extends the clinically fertile insights of attachment theory and infant-parent research. Variously described with the terms "intersubjectivity theory," "relational psychoanalysis," and "social constructivism," these findings from the consulting room have brought about a paradigm shift in the clinical realm that gives us new tools to work with the "unverbalizable" experience attachment research identifies as so central. At the heart of this shift is a rejection of the ideal of the anonymous therapist as objective observer. In its place is recognition of our inevitable involvement in enactments that reflect the interlocking vulnerabilities of patient and therapist—and the challenge of turning these potential obstacles into opportunities for healing.
Attachment and Intersubjectivity in the Healing Relationship
Bowlby's seminal insights as well as the research of Ainsworth, Main, and Fonagy point to the absolute ccntrality of attachment relationships as the crucibles within which our personalities take shape. By implication, attachment relationships may also be the setting in which—whether in love or psychotherapy—our early emotional injuries arc most likely to be healed. The therapist, then, may be a new attachment figure in relation to whom the patient can develop fresh patterns of attachment.
Fonagy, in attempting to make the bridge from attachment tesearch to clinical practice, emphasizes the importance of enabling the patient to find an image of him- or herself in the mind of the therapist as a thinking and feeling being. In this view, psychotherapy "works" by generating a relationship of secure attachment within which the patient's mentalizing and affect regulating capacities can develop. For Fonagy, such a relationship must be an intersubjective one in which the patient comes to know him- or herself in the process of being known by another.
Fonagy is not the only clinician/researcher attempting to bridge the fealms of attachment, intersubjectivity, and psychotherapy. In addition, Beebe (2004), Beebe and Lachmann (2002), Beebe et al. (2003), Fosha (2000, 2003), Holmes (1996, 2001), Seligman (2000, 2003), and Slade (1999, 2000), have all made significant contributions to our understanding
58
BOWLBY AND BEYOND
of the developmental role of intersubjective attachment relationships. Deserving special note, perhaps, are the contributions that have come from a collaborative effort launched by Daniel Stern in 1995.
At that time, Stern brought together a number of developmental researchers and psychoanalysts (of whom the most prominent were Karlen Lyons-Ruth, Louis Sander, and Edward Tronick) to form what is now referred to as the Change Process Study Group (CPSG) (Stern et al., 1998; Lyons-Ruth &c Boston Change Process Study Group, 2001; Boston Change Process Study Group, 1998, 2002, 2005). Leaning on Stern's conclusion that human beings are "hard-wired" for intersubjectivity, the work of this group has emphasized the importance in childhood and psychotherapy of attending to the nuances of interaction in order to understand development. In particular, they have focused on the co-created, implicit, and often unverbalized process through which change occurs in both infant-parent and patient-therapist relationships. From their perspective, the therapeutic relationship as lived rather than analyzed is the primary therapeutic intervention. Stern himself, perhaps harking back to his earlier attention to affect attunement as a kind of intersubjective communion, has recently made the case that the pursuit of meaning may short-circuit the deepening of experience—in particular, the experience of intersubjective meeting between patient and clinician that can sometimes produce, in and of itself, the most significant therapeutic change. Without referring directly to mindfulness, Stern highlights in this connection the absolute centrality of a focus on the present moment (Stern, 2004).
Surveying the range of these contributions reveals the outlines of an integrated approach to treatment that fills in the largely missing clinical dimension of attachment theory. As elaborated in the chapters to follow, this approach places the new attachment relationship at the center of therapeutic work. In the context of this co-created intersubjective relationship, implicit and unverbalized communication plays a crucial role in enabling us to recognize, and allowing the patient to revisit, his or her unfinished developmental business. Just as important, we can structure the therapeutic relationship in such a way that it comes to function as an incubator of the patient's more or less undeveloped capacities to freely feel, think, and love. Winnicott (1965) wrote, "The characteristic of the maturational process is the drive towards integration" (p. 239). When we can somehow enable that drive in the patient to be reengaged, the new attachment relationship becomes transformative.
PART II
I    ATTACHMENT RELATIONSHIPS AND THE DEVELOPMENT OF THE SELF
11 is in the crucible of the child's first relationships that, for better or worse, the self is originally shaped. For patients whose healthy development was derailed by the shortcomings of these formative relationships, psychotherapy may recreate an interactive matrix of attachment in which the self can potentially be healed. In light of the overlap between what is provided by the sensitively responsive parent and the empathically attuned therapist, our work as clinicians can be strengthened to the extent that we understand the change processes of childhood.
Such understanding requires a somewhat detailed account of the facets of the self—neurobiological as well as psychological—that emerge in the course of development and set the stage for the life experience that will follow. As we are about to see, the trajectory of later development is shaped fundamentally by the ways in which the child first learns (or fails to learn) to manage difficult emotions. Attachment relationships are the school in which this emotional learning originally occurs. In identifying early patterns of interactive affect regulation, both secure and insecure, attachment research enhances our ability as therapists to generate a developmentally
59
60 ATTACHMENT RELATIONSHIPS AND THE DEVELOPMENT OF THE SELF
facilitative relationship with our patients in which we are at once reworking old experiences and co-creating new ones.
Chapter 5 describes the multiple dimensions of the self that are first shaped—and can subsequently be reshaped—in an attachment context. Chapter 6 summarizes research that documents the varieties of attachment experience from infancy through adulthood and gives us a uniquely evocative basis for grasping the origin, meaning, and impact of each of the four primary states of mind with respect to attachment. Chapter 7 begins to build bridges between attachment research and clinical practice, with the focus on affect regulation, the co-created attachment relationship, and the qualities of collaborative communication that most effectively foster security in both childhood and psychotherapy.
CHAPTER 5
The Multiple Dimensions of the Self
n each of us, the ensemble of body proper, brain, and mind interact to generate the relatively stable internal point of reference known as the self. This is the part of the human being that experiences life—but also shapes it—nonconsciously as well as consciously. The self can also be described in terms of the particular cluster of more or less enduring patterns that determines the character of our unique responsiveness to experience. The present character of the self will vary depending on both our history and current context. The history that is usually most influential here is, of course, the history of the attachment relationships that have formed the self.
THE DOMAINS OF SELF-EXPERIENCE
The impact of attachment relationships registers in the interrelated, indeed overlapping domains of the body, the emotions, and the representational world, shaping the stance of the self toward experience in each. These domains or dimensions of the self arc linked and mutually influential to such a degree that distinguishing one from the other can be a difficult matter: Somatic sensations are a component of emotions which shape representations that affect, in turn, the quality of subsequent sensations, emotions, and representations. Yet, despite this kind of loop or overlap, identifying the separate domains of self has considerable clinical value, in part because
61