Psychoanalysis, attachment, and spirituality Part II: the spiritual stories we live by.

Hall, Todd W.

Psychoanalysis and attachment theory both developed independently as relational traditions within the confines of their own disciplinary walls (see Hall, 2007, this issue). While both traditions were "doing their own thing," in relative ignorance of the other, several other revolutions were occurring that turned out to pave a bridge that was already being built between these two theoretical highways. Recent developments in neuroscience, emotion research, and narrative approaches to human experience have helped to construct this bridge, suggesting that we are hard wired for two fundamentally distinct forms of knowing, one of which exists in storied form. In this article, the second of a two-part series in this special issue, I discuss the rapprochement between these two relational traditions. Following this, I highlight what may capture the common relational metapsychology underlying these converging traditions--a theory of implicit relational meaning--and its implications for "minding" the spiritual stories by which our clients live.

**********

Until relatively recently, attachment theory and psychoanalytic communities have tended to ignore each other, creating a somewhat artificial divide between two theoretical frameworks that share a number of foundational assumptions. In recent years, the rift between the trajectories of attachment theory and psychoanalysis has begun to converge toward relationality, notwithstanding continued differences. This convergence is evident in developments within each theoretical tradition, as well as in recent trends toward interdisciplinary theory building. In all these areas, we see that both theoretical traditions are moving toward a common relational metapsychology--the fundamental idea that personality is structured around implicit relational meanings that are represented in a storied form in our minds.

Convergence Within Psychoanalysis

Contemporary psychoanalysis has evolved toward an increased emphasis on the centrality of relationships. As I briefly highlighted in part I of this series (Hall, 2007, this issue), psychoanalysis took a turn toward relationality in the 1940's and 50's with the work of Fairbairn (1952), Balint (1965), and Winnicott (1958, 1965, 1971) among others. This British "middle school" of object relations spawned several lines of relational theory within psychoanalysis, sometimes referred to loosely as "relational psychoanalysis." This is a broad group of relational theories that includes, among others, object relations theory, interpersonal psychology developed by H.S. Sullivan (1953), self psychology developed by Heinz Kohut (1971, 1977) and theories of intersubjectivity (e.g., Stolorow & Atwood, 1992).

Greenberg and Mitchell (1983), in their landmark book Object Relations in Psychoanalytic Theory, were instrumental in providing a degree of synthesis and cohesion to the growing confusion in this tradition. Many splinters of relational theories were developing, often times couched in the language of drive theory making it difficult to discern the underlying vision of human nature. Greenberg and Mitchell argued that underlying the many theories, two metapsychologies existed: the drive/structure model and the relational/structure model. Ultimately, they argued, all theories fall on one side of the divide or the other. While there were attempts to combine concepts from the two models, the two metapsychologies represent fundamentally incompatible visions of personality. Thus, Greenberg and Mitchell made a cogent case that under the surface of the varied relational theories, a broad relational model exists.

Notwithstanding this broad relational model, many different relational theories continued to abound, often times claiming comprehensive status, even though they typically emphasized a narrow slice of personality (Mitchell, 1988). In a later work, Relational Concepts in Psychoanalysis, Mitchell (1988) presented an integrative relational theory, further articulating his version of this broad relational model. In this work, Mitchell argues that the field has suffered from theorists following in the shadow of Freud by attempting to develop the next grand psychoanalytic theory, and in the process ignoring or discounting compatible relational theories. Mitchell argued, in contrast, that the varied relational theories are in fact complimentary, and that the way forward involves synthesizing complimentary aspects of a relational matrix that includes dimensions of the self, the other, and the space between them. Some theories emphasize the self, others emphasize internal representations of the other, and yet other theories emphasize the space between. Despite their differing emphases, they each comprise parts of an integrated relational matrix.

In his cogent attempt to synthesize relational theories, Mitchell developed a relational-conflict model. He argues that in retiring Freud's drive theory, we need not abandon core aspects of human experience such as sexuality, narcissism, infantile desires, conflict, and illusion. Moreover, we need not choose between the self, internalizations of others, and interpersonal transactions. Rather, each of these major facets of human experience is reinterpreted in a radically different light--that of the relational matrix. Development is not constituted by innate bestial drives as in Freud's theory. Furthermore, development cannot be captured in only one dimension of the self, the other, or interpersonal transactions--the intersubjective space between ourselves and others. The warp and woof of our lives is constituted in our lifelong struggle to maintain a sense of selfhood, autonomy, and self-definition, on the one hand, and a sense of connection to emotionally significant others (both internally and externally) on the other.

Conflict is ubiquitous in our human experience, but the conflict is relational. It is conflict between the fear of losing ourselves, and the fear of losing others and God; between our desiring love and intimacy on the one hand, and our uncanny skill in eliciting others to fulfill our worst relational fears, our gut-level expectations for how relationships work for us, and repeating these patterns over and over again. In the end, Mitchell's contemporary relational psychoanalytic view paints a picture of human experience as being constituted in our own, ever-changing relational matrix--a matrix in which the self, the other, and the space between them cannot be neatly separated. While Mitchell presents one vision of how different relational theories can be integrated, he solidifies a broad relational model within the psychoanalytic tradition, and articulates how differing relational theories can inform each other.

A second trend toward convergence within psychoanalysis is that a number of psychoanalytic authors have begun to interact with attachment theory, and to see the relevance of attachment research for clinical work. For example, Goodman (2002) integrates attachment concepts within the context of object relations theory in his book The Internal World and Attachment. Fonagy (2001), likewise, explores similarities and differences between attachment theory and different relational psychoanalytic theories. Jill and David Savege Scharff (1998) incorporate attachment research in their recent comprehensive text on object relations therapy. Beebe and Lachmann (2002) explore the implications of infant attachment research for adult psychoanalytic therapy. In his last book before his untimely death, Mitchell (2000) interacts with attachment research, and proposes attachment as one of four modes of relating in an integrative synthesis of relational concepts. These are but a few examples that illustrate the substantial ways that psychoanalytic theorists are currently interacting with attachment theory.

Convergence within Attachment Theory

Within the attachment tradition, this convergence was sparked by Mary Ainsworth's work on secure and insecure attachment patterns. In 1963, Ainsworth developed the well-known Strange Situation experiment, which she used to study the attachment patterns of twelve-month-old infants and their mothers (Ainsworth, Blehar, Waters, & Wall, 1978). The strange situation was designed to assess differences between twelve month-old infants in the organization of their attachment behavior toward their mothers. The procedure consists of a series of three-minute episodes lasting a total of twenty minutes. The infant is observed in a playroom with toys, first with mother, during which time most infants explore their environment while keeping an eye on their mother, and there is limited variability in infants' behavior. However, variability increases dramatically as the procedure continues. Next, mother leaves the room, then returns, then a stranger enters with mother in the room, and mother then leaves the infant with the stranger alone. The entrance of a stranger with mother still in the room reduces exploring behavior in almost all infants. However, when the mother leaves her infant with the stranger, the behavior of over half the children changes abruptly, and clear differences in the organization of attachment emerge (Bowlby, 1982). The strange situation procedure provides the opportunity to observe how an infant uses her caregiver for comfort and as a base for exploration, and the balance between attachment and exploration behaviors as the situation unfolds.

A number of early studies used the criteria of strength of attachment to describe patterns of attachment (Schaffer & Emerson, 1964). Strength of attachment refers to whether, and to what degree, an infant protests when mother leaves for a brief period. This approach emphasizes the amount of a behavior but does not seek to understand the underlying meaning of the behavior. As Ainsworth gained more experience observing infants with their mothers, she came to conclude that this quantitative approach to evaluating attachment patterns was not only insufficient, but could actually be misleading (Bowlby, 1982). For example, anxious-resistant infants exhibit strong protest when their mothers leave--more than secure infants--which could appear to be a sign of a more healthy attachment from a quantitative perspective. However, these infants are not able to derive comfort from their mothers, and are severely disabled in their ability to explore their environment.

The combination of these behaviors reveals an underlying insecurity in the infant's attachment. Thus, within this broader context, Ainsworth postulated that the security of an infant's attachment is the most important dimension in evaluating it. Ainsworth's focus on the evaluation of the meaning of various attachment behaviors was a pivotal first step in moving attachment theory closer to a theory of relational meaning, and hence closer to the relational psychoanalytic trajectory that was developing independently during this same time period. This was further developed by Sroufe and Waters (1977) in a key article emphasizing attachment as an organizational construct. In this article, Sroufe and his colleagues likewise emphasized the subjective meaning of attachment behavior rather than simply the frequency of certain behaviors.

The convergence within the attachment tradition came to a head in the mid-1980s, when attachment theorists shifted from a focus on infant behavior to a focus on the subjective experiences and dynamic internal representations in the infant and parent, largely as a result of the development of the Adult Attachment Interview (Main, Kaplan & Cassidy, 1985). Mary Main, Nancy Kaplan, and Jude Cassidy (1985) reported evidence that the way adults discussed their past relationships with their own parents was reliably associated with the attachment classifications of their own infants in the Strange Situation. For example, adults who idealized their own parents but could not provide clear memories in support of such appraisals (classified as dismissing adults), generally had children who behaved in an avoidant manner in the Strange Situation. Adults who strayed from the AAI questions, or who exhibited intense anger toward their parents (i.e., preoccupied adults) typically had children who were classified as anxious-resistant in the Strange Situation. In contrast, secure adults who produced coherent and believable narratives about their relationships with their parents tended to have children who were secure in the Strange Situation.

What they came to see was that the underlying meaning of early relational experiences are represented and become manifest in the way adults tell their attachment-related stories. They eventually codified these narrative styles in the coding system for the Adult Attachment Interview (see Hesse, 1999). Research in the Adult Attachment Interview tradition has burgeoned in recent years, and has contributed to the growing interest in narrative approaches to human experience, to which we will return shortly. In addition, AAI research has paved the way for exploring the implications of attachment theory for clinical work.

In this vein, several authors have noted a growing sense of "paradigm boundedness" within attachment theory, and the need for theoretical paradigms that enrich clinical work, and clinically relevant research and theory (e.g., Fonagy, 2001; Mitchell, 2000). As a result, a number of attachment researchers, and clinicians influenced by attachment theory, are now integrating their findings with contemporary psychoanalysis and articulating their clinical implications. One area of attachment research that has led to rich clinical implications convergent with relational psychoanalysis is the search to understand how attachment organizations are transmitted intergenerationally. Findings from the Adult Attachment Interview have generated hypotheses about how the primary caregiver's attachment organization is transmitted to the infant. Main (1991) proposed that avoidant and resistant attachment patterns are maintained by a "steady representational state" that is subjectively experienced as a kind of secondary felt security. In other words, consistent with relational psychoanalytic theories, an insecure (but organized) attachment pattern is still the tie that binds one to the other, however insecure it is. This is reminiscent of Fairbairn's (1952) bad objects. It is better to maintain an attachment bond to a rejecting, inconsistent or avoidant attachment figure than to have no bond at all. To sever this tie is to enter a state of the terror of nonexistence. This form of felt security is secondary in the sense that it is not a secure attachment organization (primary felt security), yet it is the only familiar form of attachment within one's relational matrix. It represents our deep, gut-level expectations about how attachment relationships work for us. It is the only relational experience we know, however implicit it may be. We simply cannot imagine anything else within our relational matrix.

A parent's secondary felt security creates in her infant the need to develop certain strategies to maintain connection with the parent. The infant's bids for proximity, emotional closeness, or comfort may challenge the state of secondary felt security of the dismissing (parallel to the avoidant infant) parent that is based on deactivating the attachment system. Conversely, exploration and separation-inducing behaviors may challenge the state of secondary felt security of the preoccupied (parallel to the resistant infant) parent that is based on hyperactivating the attachment system. In both cases, the infant learns a conditional behavioral and emotional strategy to maximize her own sense of self-organization and felt security with the parent. For example, the infant of a dismissing parent will learn to behaviorally avoid the parent when distressed and will develop a general nonconscious strategy to shut down, or deactivate, any seeking of connection and comfort. This strategy paradoxically maximizes the infant's connection to a dismissing parent.

Mary Main also developed the concept of "metacognitive monitoring" as a potential explanation for the transmission of secure attachment (Main, 1991). Metacognitive monitoring is the ability to distinguish between appearance and reality; between one's immediate experience and one's mental state underlying it. She argues that differences in attachment organizations can be linked to the quality of metacognition in the parent. This notion is very similar to much that has been written in psychoanalysis about the importance of the primary caregiver's capacity to view their child as a separate person with her own thoughts, feelings, and desires that are meaningful within her own psychic economy, independent of the caregiver's psychic economy (Fonagy et al., 1995).

In short, if we look at the broad strokes of developments within psychoanalysis and attachment theory, we see many examples of converging relational concepts. What may capture the common core of these two traditions is a theory of implicit relational meaning--implicit meanings in which our spiritual stories are embedded and enacted.

IMPLICIT RELATIONAL MEANING: THE SPIRITUAL STORIES WE LIVE BY

Christopher Bollas proposed the concept of "unthought knowns"--a deep form of relational knowing that is not formed in thoughts or words (Bollas, 1987). Contemporary neuroscience and emotion research has provided substantial evidence for this notion. There is now strong support for the idea that we are hard wired for two fundamentally distinct ways of knowing and that we are hard wired for stories.

Unthought Knowns: We Know More than We Can Say

In an earlier article, I outlined five central organizing principles of a relational spirituality paradigm based on a theory of implicit relational representations (Hall, 2004), or what we might also refer to as a theory of implicit relational meaning. A core principle of this theory is that our brains have two distinct processing systems, which support two very different ways of knowing. Explicit knowledge--knowledge that is conscious, linear, and exists in images and words--is supported by what neuroscientists refer to as the "high road" brain circuits, based in the pre-frontal cortex (primarily on the left side), that are responsible for analytical and abstract reasoning. (LeDoux, 1996). Implicit knowledge, in contrast, is "gut level" knowledge, or meaning, that is carried in our bodies, emotions, and stories. This form of knowing is supported by what neuroscientists refer to as the "low road" brain circuits (primarily on the right side), involving the amygdala and other circuits responsible for primary emotion (LeDoux, 1996).

In the context of knowledge, or meaning, about ourselves-in-relation-to-others, this latter way of knowing is what Stern et al. (1998) refer to as "implicit relational knowledge;" that is, our gut level sense of "how to be with" attachment figures. This sense of how to be with attachment figures is similar to the general notion that patterns of relational experiences with caregivers become internalized. This idea has been captured by numerous theorists, such as by Stern's (1985) "representations of interactions that are generalized" or RIGS, emotion schemas (Bucci, 1997), mental models (Siegel, 1999), object representations in object relations theory (e.g., Fairbairn, 1952), and internal working models in attachment theory (Bowlby, 1973). Each of these concepts refers to representations of relational meanings that are encoded in implicit memory; thus, we can think of a theory of implicit relational meaning. Implicit memory operates in the limbic and subcortical regions of the brain that are responsible for the processing, expression, and regulation of emotional information and nonverbal communication (Schore, 1997). Implicit memory operates without conscious awareness. In other words, when implicit memory is retrieved, we do not have the experience or sense that something is being remembered (Siegel, 1999).

Because we are not aware when implicit memory is operating, it functions as an "attachment filter" as discussed in the previous article (Hall, 2007, this issue). We do not experience the filter; we experience relationships through the filter of various ways of being attached to caregivers. These attachment filters that influence the way we experience significant relationships turn out to be gut-level expectations of how relational stories will play out. As I mentioned earlier, AAI researchers discovered that the implicit relational meaning of individuals' attachment histories are carried in their attachment-related stories--not in the content per se, but in the way they "tell" their stories. In other words, implicit relational meaning is carried in the emotional communication of our "between-the-lines" stories. Neuroscientists have provided further evidence that our brains are hard wired for stories.

Hard Wired for Stories

Stories, or narratives, are meaning structures that have a certain "grammar" to them, or certain common characteristics that we naturally identify as a "story" (McAdams, 1993). Stories are emotionally meaningful sequences of actions that are causally linked in a particular way. They contain a setting that provides the overall context for the unfolding of a series of emotionally meaningful events. In addition, stories contain characters, human or human-like figures that live within this setting. An initiating event occurs to the central characters, motivating them to strive after certain goals, which in turn leads to a consequence. Multiple episodes of a story, each containing this basic structural sequence, build on each other and provide shape to the story as it unfolds. As the story unfolds, tension builds across the episodes eliciting in us a desire for resolution. This tension typically builds to a climax, or turning point, which is followed shortly by some solution to the plot.

McAdams (1993) argues that human beings are, by their very nature, storytellers. Stories are a natural package for organizing many kinds of information. Furthermore, stories appear to be a fundamental way that we express ourselves and communicate with others. When we want to explain something to someone, we often tell a story to communicate our meaning. Much of our everyday conversations involve some form of storytelling. Psychoanalytic psychotherapy can be aptly viewed, at one level, as an unfolding story between client and psychotherapist, one that involves the client's and therapist's internal relational matrices and interactions between the two. Indeed, a tradition has emerged emphasizing narrative knowing as a framework for interpreting meaning in psychoanalytic psychotherapy (e.g., Spence, 1982).

Stories are a powerful way of knowing because they help us organize, maintain and evaluate our own and others' behavior (Cozolino, 2002). Narratives help us regulate how we experience and express emotion. They are one aspect of our gut level memories that serve as attachment filters. In other words, our attachment filters are stored in the form of stories, and it is through stories that we access them, or bring them online. By the age of two-and-a-half, parents and children create stories together at a rate of 2.2 per hour in everyday conversation (Cozolino, 2002). These stories connect parents with their children, but they also contain within them a grid for evaluating what is discussed, deciding what information to include, how to process and understand that information, and whether the story will have only one subjective center or vantage point, or several subjective centers. Having several subjective centers carries with it the capacity for empathy, which is critical to love and compassion. The way stories are structured contains a gut level, implicit model of how to relate to emotionally significant others, and how many ways there are to do this. To state it differently, internal working models are structured as stories, and in the context of our spirituality, these become the spiritual stories we live by.

Cozolino (2002) suggests that narratives are analogous to a music score for an orchestra. Stories organize and synchronize the participation of many "instruments." The range and complexity of a score determines which instruments are used, how they are coordinated, and the quality of the final performance. Parents and children write this narrative score together in the context of their family and culture. In short, we can see that stories play a crucial role in a child's developing connection with parents, and in their attachment filters.

Two Modes of Storying

Bruner (1990) proposed two different ways of understanding the world--paradigmatic and narratives modes of thought--that parallel explicit and implicit relational forms of knowing, respectively. The narrative mode of knowing is concerned with human wants, needs, and goals, and operates in the implicit relational knowing system. Narrative knowing is not a chronicle of facts that is judged by explicit, logical analysis. Rather, it has more to do with meaning, and is judged by implicit, narrative criteria such as "believability" and "coherence." Moreover, as meaning ultimately derives from relationality (e.g., Mitchell, 2000; Siegel, 1999), narrative knowing, or implicit relational knowing with respect to our fundamental sense of self-in-relationship is appraised by our sense of connection to others, as filtered through the attachment system. Storied knowledge of ourselves and our spirituality exists fundamentally in implicit form, although it can be translated into explicit language and knowledge. Bruner argues that masters of the narrative mode of knowing try to "say no more than they mean" and, through their storied communication, "mean more than they can say." In other words, the narrative mode of knowing about ourselves-in-relationship taps into our unthought knowns (Bollas, 1987): we know, or mean, far more than we can say. Implicit relational meaning, or narrative knowing, cannot be fully captured in paradigmatic or explicit forms of knowing.

However, just as there are two ways of knowing in general, we can think of two modes of storying. Our clients literally tell their stories in the paradigmatic, or explicit sense, using the language of words. This, at one level, is the "stuff" of psychotherapy. Our clients use words to communicate the current episodes of their unfolding life stories. In fact, I have had numerous clients in the early stages of therapy report that they feel responsible to communicate the "right" (verbal) information to me so I can help them. If they miss "right" verbal information in their minds, they fear I am blocked from helping them. In their minds, this is the only story being told during the session. However, our clients also "tell" us a completely different type of story--a story that exists "between-the-lines" and uses the nonverbal "language" of emotion. The narrative mode of knowing, then, cannot be reduced to the verbal content of a story. As Bruner (1990) argues, it is more about meaning; however, the verbal articulation of such meaning in a relational context turns out to be critical for synergizing these two ways of knowing to bring about transformation. Furthermore, "coherent" stories require both types of stories to be working in synchronization with one another.

In order for our clients to tell coherent stories about their relationships (with God and others), it requires a harmonious working relationship between the lower-right (low road) and top-left (high road) circuits of the brain. The "interpretive" top-left brain circuits are predominantly responsible for recounting the logical sequence of events, whereas the lower-right brain circuits are predominantly responsible for the emotional meaning of the events (Siegel, 1999). Bucci (1997) has developed a linguistic method for coding the degree to which emotional meaning is integrated with verbal communication. This coding for "referential activity" assesses the concreteness, specificity, clarity, and imagery level of speech. She has identified patterns in various measures that represent linguistic indicators of different phases of the referential process that involves linking "subsymbolic" emotional meaning to images and ultimately to words. When both of these ways of knowing are working together, the outcome is a logical and emotionally meaningful--or coherent--communication. In other words, our clients' between-the-lines stories will match the stories they tell us in words.

If both of these facets are not woven together, an individual's story will not be coherent. Clients with secure attachments in general are able to tell a coherent story about their relationships. In contrast, clients with insecure attachments tend to be incoherent in their stories, but in different ways. Preoccupied clients are incoherent because they become overwhelmed by painful feelings (Siegel, 1999). Unresolved painful memories have been shown to be associated with a brain activation pattern that is dominated by the right brain (Schiffer, Teicher, & Papanicolaou, 1995). Furthermore, cooperation between the two hemispheres appears to be necessary in order to consolidate memory (Siegel, 1999). Thus, the core problem with unresolved painful memories may be the failure to consolidate memories of such traumatic events. When the right brain is dominant in preoccupied narratives, there is an absence of a verbal, storied version of painful events. There is not a beginning, middle or end; there is no plot that can be detected. It is experienced and communicated as an undifferentiated mass of emotional pain. The interpretive left side of the brain, then, is not able to do its part to place emotionally significant events into a larger network of meaning, and permanent, consolidated memory. Instead, these unresolved painful memories remain in an unstable state of potential implicit activation, which often intrude on the person as if from outside the person.

Dismissing clients, in contrast, tend to be incoherent because they do not integrate the emotional meaning of the events into their verbal stories. We can assume that their narratives tend to be dominated by left-brain activity. If you listen to a dismissing client's story, you will get a precise verbal account of the sequence of events; however, you will not get a good sense of the emotional meaning of the events they are recounting from their emotional communication. In this case, the right brain does not do its part to integrate the emotional meaning of events into the narrative sequence. Thus, you get a sequence of events that has no life to it. However, the incoherence in both types of filters causes the emotional meanings to be clouded. They are very difficult to articulate. This clouding of emotional meaning hinders new relational information from being integrated into their implicit relational meaning structures. Stories turn out to play a critical role in integrating explicit and implicit knowledge.

The Knowledge Spiral: Storying Unthought Knowns

Earlier I proposed the concept of unthought knowns as a picture of our gut level way of knowing (Bollas, 1987). These are things we know, yet they remain unthought, unformed. They are emotional meanings that do not exist in words that can be thought and communicated to others. This is part of why changing our clients' attachment filters is so profoundly difficult. The very nature of our clients' attachment filters is that they are unspeakable, and when they are painful, it becomes even more difficult to be aware of them, and to communicate them to others (Bucci, 1997). When our clients' attachment filters remain unspeakable, they are difficult to transform because it hinders them from coming into relational contact with the therapist, God or others in a fully integrated way. However, while I focus below on the integration of these two ways of knowing, it is important to note that much reparative work in therapy is done at the implicit level. The majority of unthought knowns are never translated. We simply catch out clients' emotions implicitly and communicate empathy to them in a myriad of nonverbal ways. Nonetheless, translating between these two ways of knowing is a core function of what we do in therapy.

The transformation of our clients' implicit attachment filters of God and others is not a function of either implicit or explicit knowledge by itself, but requires a synchronization of both ways of knowing. This process involves building new relational experiences along different dimensions than those in our current implicit memory. Our explicit system then identifies these and analyzes the meaning of them, and the very act of doing this helps us to process and transform our implicit experiences at new, and deeper levels. Our ability to articulate new experiences opens up new connections in our implicit processing, creating a spiral-like process. This process, referred to by Bucci (1997) as "referential activity," happens through images and stories. One of our central roles as therapists is to create a relational environment in which our clients' unthought knowns can become "speakable" through a translation process that links their raw, implicit relational knowledge with words. This can be pictured as a "bottom up" integration process, bringing together our clients' "gut level" experiences with explicit "head" knowledge. Following our discussion of bottom up integration, we will elaborate on the back-and-forth nature of this spiral-like process in which each system refers to the other to develop qualitatively new knowledge.

Our gut level knowledge starts off very raw when it is first processed in the lower, right portion of our brains. This information cannot be easily defined or categorized (Bucci, 1997). In order for this kind of information to be verbalized and communicated to others, and to process it using the rules of verbal logic, it must first be translated. There are several steps in this translation process that can be illustrated by a case example. Fred came to see me in the hopes of saving his marriage. Fred and Bonnie had been arguing for several years and things had hit a crisis point. Fred was very angry at Bonnie. For quite a few years, he felt she had been pursuing her own agenda, not being committed to him or their children. Bonnie eventually separated from Fred, saying she wanted to work on the marriage, but Fred did not see any signs that she was following through on this. He became more angry with her and began to express this to her very overtly. Bonnie began to express deep-seated anger and disappointment passively, by withdrawing, and refusing to communicate with Fred, with the exception of an occasional out-of-the-blue outburst. A typical interaction during this time would look something like this: Fred would try to talk with Bonnie about their marriage, probably with an angry tone, and she would withdraw, disagree with him, question him, and eventually refuse to talk about the issue anymore.

A closer look at Fred's attachment filter reveals, in part, how and why he processed these interactions as he did. Fred's attachment filter was a version of the preoccupied filter. His father died when he was very young, and his mother was angry and vacillated between neglect and abuse. At times Fred was left on his own to figure life out even when he was quite young, and at other times, his mother would verbally attack him with tremendous hostility. As a result, Fred developed an attachment filter of himself as being bad and unworthy of care, and an implicit expectation of others as being highly critical of him. On many occasions he said to me: "One thing I've learned in life is that the only person who is going to take care of Fred is Fred." Fred expected this from Bonnie, not consciously, but this filter automatically biased him to experience this in his interactions with her.

When an experience gets processed through our attachment filters, it produces an implicit emotional meaning. When Bonnie would withdraw or disagree with Fred, an emotional response would be activated along with associated components such as sensory and visceral experiences, physiological arousal, tendencies to act in certain ways, and gut level memories with a similar feeling. These sensations are very difficult to express in words.

The first step in the process of connecting these experiences to words is that our right brains chunk information into categories that have similar emotional meanings (Bucci, 1997). Fred's filters chunked his physiological arousal and input from facial expressions into the emotional meaning of rejection, and of himself as being bad and not worthy of love. This emotional meaning was then represented in his right brain as an image that has the structure of a typical episode of needs, desires and actions related to them that lead to a complex sequence of interactions that are expected to occur--that is, as a story. Thus, Fred's gut level sense of badness and rejection gets processed into an image that has the structure of a story in which he tries to connect with someone close, and the other rejects him by either withdrawing or showing hostility. This is Fred's between-the-lines story.

If Fred wants to communicate this emotional meaning to me in therapy (or to anyone else) an important way to do this is to describe an episode in which the emotional meaning was activated. Stories about our relationships can be viewed as metaphors of the emotional meanings associated with our attachment filters (Bucci, 1997). When we verbally tell our stories about our experiences in relationships, it is the closest we can come--through a translation process--to communicating the emotional meanings of our attachment filters in words.

The Knowledge Spiral: Phases in Storying Our Spirituality

Ideally, processing does not stop with linking implicit relational knowledge with explicit, verbal knowledge. Rather, processing continues in a spiral-like back-and-forth fashion, leading to a continuously deepening progression in both ways of knowing. This knowledge spiral operates in parallel in clients and therapists. This is the process we go through in linking our own implicit and explicit relational knowing in order to facilitate our clients storying their unthought knowns about God and everything else.

This knowledge spiral has fascinating parallels in creative scientific work and in the arts (Bucci, 1997). Four phases have been identified in the process of discovery: preparation, incubation, illumination, and reflection/interpretation. These phases overlap and the boundaries between them are fuzzy, much like they are in the underlying process involved in storying our clients' spirituality.

In general, preparation in the process of scientific-discovery is the ongoing, lifelong acquisition of knowledge through which a person develops expertise in his or her field. The specific preparation to solve a particular problem requires the scientist to "back-translate" the problem from the explicit knowledge system to an implicit way of knowing. The scientist hears the problem in its verbal form, which is processed by the top-left circuits of the brain. She then begins to meditate on the problem in a scientific mode in the "back of her mind" so to speak, which is the bottom-right brain circuits. She may work actively on the problem for awhile, and then feel like she is getting nowhere, "stumbling around in the dark." This is what it feels like to work in the implicit system of knowing. You search without any clear direction and without categories that have been defined.

For the psychotherapist, preparation involves the lifelong process of acquiring knowledge, both implicit and explicit, about human nature, psychodynamics, and change processes. A specific client then comes in and articulates a specific problem or set of problems. The therapist then "back-translates" a problem into implicit knowledge. This involves the explicit context of the problem, but more importantly, it involves tuning in to one's implicit sense of one's experience of the client. The therapist translates this into her gut level system by reflecting on, and narrating the issue.

For the client and therapist alike, preparation involves the gradual process of opening oneself up to the other. The client becomes attached to a therapist who ideally provides a safe haven for exploring her internal world with respect to God and others. The therapeutic relationship becomes the context in which deeper aspects of her internal world will be illuminated. A critical aspect that paves the way for this is the therapist opening herself up to being impacted by her client.

The preparation phase flows fluidly into the incubation phase. The scientist's implicit processing occurs predominantly outside of awareness and without intentional control in the incubation phase. The person often times turns their attention away from the problem, but once the implicit processing system of the right brain has been prepared and activated, it continues to work on the problem. It follows its own leads and connections, which the scientist is not aware of, and cannot consciously follow. Scientific breakthroughs often happen after turning away from a problem. In addition, these types of insights involve relationships between questions and concepts that at first seemed entirely unrelated. In other words, these are not logical, linear connections that can be made in the explicit knowledge system. The implicit processing system makes its own connections and builds new dimensions and categories while the problem incubates.

There are clear parallels in the incubation phase among scientists for how we intervene as therapists. This is where we process our clients' relational dynamics outside of awareness, according to the parameters of our implicit processing system, the rules of which cannot be identified (Bucci, 1997). Moreover, therapeutic breakthroughs often come to us when we turn away from the clinical issue at hand. When we are not consciously thinking about it, the words or a metaphor to articulate a client's experience will suddenly come to us. For the client, as Sorenson (2004) noted, incubation occurs not only between sessions, but sometimes while we are making that brilliant intervention we think is going to cause a major transformation: "Recently I offered what I thought was a useful insight for a patient. When I had finished speaking, she paused, looked off to the side with a thoughtful look in her eye, and confessed, 'All kinds of interesting things occurred to me while you were talking.' ... I sometimes bemusedly tell myself that my patients seem to do some of their best work while I am talking because, in these moments, they are often only half listening to me" (p. 12).

Likewise, incubation is a crucial phase in the spiritual transformation process with respect to clients' implicit knowledge of themselves, God and others. This is the beginning of bottom-up integration, in which implicit experiences are in a nonverbal form outside of awareness, and eventually become articulated in a conscious, verbal form in a later phase. In the incubation phase, a clients' implicit processing system processes their relational experiences and filters--their sense of connection to God, their gut level expectations of God and others, and their most deeply held beliefs and values that motivate what they strive to do on a daily basis. The rules that govern this processing are not known, and all this happens behind the scenes, outside of our awareness. It is the place where clients form new connections about the meaning of their experiences, about who they are with and to God and others. It is the place where new story lines are developed. Incubation at some point morphs into the next phase: illumination.

In illumination, the connections that are being forged in the incubation phase make themselves known, as if coming from the outside. Poincare, a mathematician, stated of one of his breakthroughs: "At the moment when I put my foot on the step, the idea came to me, without anything in my former thoughts seeming to have paved the way for it ..." (as cited in Bucci, 1997). Illumination is experienced by the scientist as coming from an external source, but it comes from the implicit processing system. However, a lot of ground work in the preparation and incubation phases has prepared the way for illumination, or turning points, that involve new ways of seeing a problem. When these new connections hit the scientist's awareness, this is the point at which the implicit processing is connected to the explicit, verbal knowledge system (bottom up integration).

For the therapist, illumination is the moment when a previously implicit sense of a client's struggle or experience crystallizes in words or images. This may bring an entirely new way of viewing a client's struggle, and will typically be experienced as if coming from outside of oneself. Understanding how these processes work can help therapists to trust these implicit intuitions more readily.

In the context of transforming clients' attachment filters and capacity to love, illumination is the turning point in which new gut level meanings about themselves, God and others (attachment filters), are crystallized. New story lines begin to take shape in our clients' awareness. This is what changes the very structure of their soul, and transforms how they relate to God and others. These new meanings may have been around awhile in a more fuzzy way, but now they come into a more clear focus in their conscious awareness. A new awareness might start off as an image, or a picture, but the images have stories embedded within them. This then leads into the next phase of reflection and interpretation.

In the reflection and interpretation phase, the scientist capitalizes on her new awareness by interacting with it using her conscious, analytic (explicit) knowledge. This happens primarily within awareness. When she does this, it brings more precision to the new gut level breakthrough. She can now hold this new awareness in her mind's eye (analytic knowledge) as it were, and manipulate it, communicate it to others, look at it from different angles, examine its relationships with other concepts, and sharpen its boundaries.

In our clients' spiritual transformation process, likewise, we capitalize on the implicit processing that has incubated, and led to an illumination, or turning point, regarding their gut level meanings. We do this by helping our clients translate their new gut level experiences into words, concepts, and ultimately a new story line in their relationship with God. We, and our clients, give shape and form to the illuminations, which allows them to tell new stories to God and others, and gives them more access to the gut level meanings within themselves. The very process of translation transforms their gut level meanings by connecting them to a larger network of implicit relational meanings that constitute an ongoing story.

A case example will help illustrate the knowledge spiral in minding our clients' spirituality. I had been seeing a client I will call Kevin for several years when he began to withdraw from his relationship with God, and gradually ceased involvement in his spiritual community. At first he was unaware of the meaning of this behavior, and he avoided discussing it with me. When we did process the issue, Kevin felt extremely sad and became aware of a sense of abandonment by God. This experience of abandonment was operating in his gut level knowledge and is what led to his withdrawal from God and church. It was an unthought known--a primary emotional experience for which he had no words and little awareness. The determination of the meaning of this gut level experience was heavily influenced by his preoccupied attachment filter, which was based on the unique combination of his own needs, and years of experiences of his primary caregivers as inconsistent and abandoning. This gut level experience was driving his attachment filter, and in turn, his pattern of relationship with God.

The general preparation phase involved the attachment we had developed over several years of exploring his experiences--him telling me his story and the two of us beginning to share his story. The story of our work together became intricately intertwined with all the sub-plots of his life. These specific experiences incubated as we discussed them over several months. Part of the incubation process was new experiences with me of being genuinely interested in his emotional experiences. No breakthroughs happened right away, but new connections were being made behind the scenes as we discussed this, and as we turned away to other topics on occasion. Over time, these incubating connections led to a vague sense of abandonment by God, consciously linked at one level to a series of recent emotional wounds in close relationships. He was not yet able to tell this story with words.

Around the same time, Kevin became increasingly guarded with me. At first, I felt something had changed, but I was not sure what it was. I let this incubate in the background for a period of time. After a number of weeks of this guardedness with me, and hearing more about his sense of rejection in several relationships, illumination struck. I told Kevin that I sensed that he felt all his important relationships were "close" on one level, and yet "not close" on another level at the very same time. I commented that I thought this was very confusing for him. This articulated for him a deep experience of himself and his relationships that he had never been able to express in words. The crystallization of this profound meaning into words brought home the power of this deeply painful feeling, evident in the immediate sadness Kevin experienced at this comment. Illumination had occurred for both of us. This was a shared intersubjective experience of grasping something that permeated his experience of himself in the context of emotionally significant relationships--something that helped make sense of his life. As important as this insight was, I believe that Kevin's sense of our sharing this new turn on the therapeutic road was equally important. This experience caused a significant shift in breaking the "close-not close" mold in our relationship. Kevin was beginning to experience a relationship with me that did not have a fundamental disconnect between superficial externals and his sense of feeling known on the inside. This, I believe, paved the way for a different kind of relationship with God that would come quite a bit later.

Shortly after this, Kevin reported several images that seemed to capture further illumination that was occurring behind the scenes. He told me he saw himself wandering aimlessly in the desert. Around the same time, he had another image of himself drowning with no one around to save him. These images were both disturbing and revealing to Kevin. His gut level processing had led to new connections about how he felt about himself at a deep level. He felt abandoned by God and others in his life, alone, and not worthy of others caring and sticking around. Prior to this, he had not consciously realized this feeling about himself and God. It was just a vague, floating sense of discomfort--background noise in his life that caused him to keep a certain distance from God and some of his friends. Because it was vague and the story was outside his conscious awareness, it was difficult to change the story line. It just floated around in background in an unstable state of potential activation. His gut level sense of abandonment would become activated at times without his knowing the story from which the increased discomfort was coming.

As we processed the issues further, Kevin reflected and interpreted these experiences with my help. During this time period, I suggested to Kevin that he had become more guarded with me, understandably, because he was afraid I would leave. Furthermore, I suggested that he expected everyone to eventually leave. He responded by saying "people only care about me for a season." I said "Yes, and why would it be any different with me? I can understand why getting too close is scary." He was gradually able to put words to his gut level experiences, further articulating the meaning of his sense of abandonment by God and others. The experience of abandonment and subsequent withdrawal from God were in response to a series of perceived rejections of his need for reassurance and comfort in close relationships after he demanded that they be met. However, this filter also reflected his gut level sense of how relationships work--that eventually people leave, emotionally if not physically. He became more aware of this story that ran through every thread of his life. I remember numerous occasions in which Kevin reported "ah ha" experiences of seeing the themes we had been talking about in multiple aspects of his life. He started to be able to tell this story in words. The very nature of this knowledge spiral of linking his gut level experiences to words and a story, transformed the experience by simultaneously providing more access to the experience within himself, and by bringing it into relationship with me, and eventually God and others. Over the next year, the patterns of his relationships with me, his family, his co-workers, and his romantic relationships changed substantially. This period represented a major turning point of growth in his journey.

Reflective Spiritual Functioning

As this case example illustrates, this knowledge spiral, or referential process, in which implicit and explicit knowledge about our sense of self each refer to the other, is critical for both client and therapist in revising the spiritual stories by which our clients live. For the client, this is the process that undergirds the transformation of their implicit sense of themselves in relation to God, as well as others. A safe haven in the therapeutic relationship provides the relational environment for clients to capture their gut level sense of "how to be with" God and others in a more explicit dimension. The very act of making implicit relational knowledge explicit--of telling their story in words--transforms it. Furthermore, this linkage between the implicit and explicit facilitates a more full and coherent communication, and more secure between-the-lines story in the therapeutic relationship. This produces what Siegel (1999) refers to as "contingent communication" in which clients "feel felt" or seen by the therapist. This is a form of communication in which there is a resonance that "means more than we can say" to use Bruner's (1990) terminology.

For the therapist, fostering such contingent communication with our clients is also the result of this knowledge spiral involved in the referential process. It is the process by which the therapist tunes into her implicit sense of the client's experience, including spiritual experience, and translates this into words, images and metaphors for the client. It is also the process by which the therapist translates a sense of what the client needs at any given moment into emotional and verbal communication. Contingent communication promotes what Fonagy et al. (1995) have called "reflective self-functioning" in clients. This derives from William James' (1890) usage of the term "reflective" to describe the developmental achievement in which the mental state of an individual becomes a subject of thought. In other words, the individual is able to observe and reflect on her own and others' mental states about attachment-related issues.

Reflective self-functioning seems to explain part of how attachment security is passed down intergenerationally. Fonagy and his colleagues (1995) found evidence that it buffers the negative effects of deprivation and high stress among mothers on the attachment security of their children. It thus seems to reduce the likelihood of insecure attachment being passed down intergenerationally. Moreover, there appears to be a reciprocal relationship here--secure attachment seems to be an important precursor to the development of reflective self-functioning in children. A "secure base" in the attachment relationship allows the child to explore the minds of her caregivers, which facilitates the development of an appreciation of mental states. There are clear parallels in the context of the therapeutic relationship. As therapists provide a secure attachment relationship, this fosters the development of reflective self-functioning in their clients. This in turn helps them appreciate and process their own and others' experiences.

Reflective self-functioning bears resemblance to James Fowler's fourth stage of faith development (Individuative-Reflective), in which one becomes aware that one has a worldview, and is able to step outside of it and interact with it (Fowler, 1981). Similarly, reflective self-functioning allows an individual to recognize that he and others have mental states that are not identical to "the way things are." It is a form of epistemological humility in the realm of interpersonal relationships. Reflective self-functioning allows me to take my mental states about attachment as objects of scrutiny to be processed, analyzed, constructed, and reconstructed. It puts a wedge, as it were, between my subjective experience of myself and my mental states.

This reflective capacity requires a secure attachment relationship to be fully developed, and has enormous implications for relational and spiritual development. In the context of spirituality, this fosters what we might think of as "reflective spiritual functioning." Lacking this reflective capacity will make it difficult for clients to take their subjective experiences of God as an object of scrutiny in order to process them. In other words, it will make it difficult to tell and enact new stories about their spirituality. Helping our clients become secure enough to appreciate all their "spiritual" experiences (which could be viewed as all encompassing, but certainly including their direct experiences of God through Scripture, prayer, etc) as something they have; something they can hold in order to process, analyze, construct, and reconstruct, may be one of the most important things we can do to mind our clients' spirituality--both in the sense of attending to moments of transition between different levels of experience, and of cultivating it (Sorenson, 2004). Randy (Sorenson, 2004) encouraged us to "... take an interest in our patients' spirituality that is respectful but not diffident, curious but not reductionistic, welcoming but not indoctrinating" (p. 1). As we heed his words, we will model for our clients a way they can begin to approach their own spirituality, thereby creating new spiritual stories to live by.

CONCLUSION

As relational branches of psychoanalysis and attachment theory increasingly converge, a common metapsychology is beginning to emerge. This metapsychology may be captured by a theory of implicit relational meaning that distinguishes between implicit and explicit forms of knowing. The implicit relational meaning of our sense of self-in-relation is carried in individuals' attachment-related stories; that is, in the emotional communication of the "between-the-lines" stories. Clients both tell their stories in words, and enact the spiritual stories by which they live. Helping them transform these implicit spiritual stories involves a spiral-like back and forth process of linking implicit and explicit knowledge for both client and therapist alike.

REFERENCES

Ainsworth, M.D.S., Blehar, M.C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the Strange Situation. Hillsdale, NJ: Erlbaum.

Balint, M. (1965). Primary love and psycho-analytic technique. New York: Liveright.

Beebe, B. & Lachmann, F. (2002). Infant research and adult treatment. Hillsdale, NJ: The Analytic Press.

Bollas, C. (1987). The shadow of the object: Psychoanalysis of the unthought known. New York: Columbia University Press.

Bowlby, J. (1973). Attachment and loss, Volume II: Separation, New York: Basic Books.

Bowlby, J. (1982). Attachment and loss, Volume I: Attachment, New York: Basic Books, Second Edition.

Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard University Press.

Bucci, W. (1997). Psychoanalysis and cognitive science: A multiple code theory. New York: Guilford Press.

Cozolino., L. (2002). The neuroscience of psychotherapy. New York: Norton Press, 2002.

Fairbairn, W.R.D. (1952). Psychoanalytic studies of the personality. New York: Brunner-Routledge.

Fonagy, P. (2001). Attachment and Psychoanalysis. New York: Other Press.

Fonagy, P., Steele, M., Steele, H., Leigh, T. Kennedy, R., Mattoon, G., & Target, M. (1995). Attachment, the reflective self, and borderline states. In S. Goldberg, R. Muir, & J. Kerr (Eds.) Attachment Theory: Social, Developmental, and Clinical Perspectives (pp. 233-278). Hillsdale, NJ: The Analytic Press.

Fowler, J. (1981). Stages of faith: The psychology of human development and the quest for meaning. San Francisco: HarperSanFranciso.

Goodman, G. (2002). The internal world and attachment. Hillsdale, NJ: The Analytic Press.

Greenberg, J. R., & Mitchell, S. A. (1983). Object relations in psychoanalytic theory. Cambridge: Harvard University Press.

Hall, T.W. (2004). Christian spirituality and mental health: A relational spirituality framework for empirical research. Journal of Psychology and Christianity, 23(1), 66-81.

Hall, T.W. (2007; this issue). Psychoanalysis, attachment, and spirituality Part I: The Emergence of Two Relational Traditions. Journal of Psychology and Theology, 35(1).

Hesse, E. (1999). The adult attachment interview: Historical and current perspectives. In Handbook of attachment: Theory, research, and clinical applications. Cassidy, J., Shaver, P.R., New York: Guilford Press.

James, W. (1890). Principles of psychology. New York: Holt.

Kohut, K. (1971). The analysis of the self. New York: International University Press.

Kohut, K. (1977). The restoration of the self. New York: International University Press.

LeDoux, J. (1996). The emotional brain. New York: Simon & Schuster.

Main, M. (1991). Metacognitive knowledge, metacognitive monitoring, and singular (coherent) vs. multiple (incoherent) models of attachment: Findings and directions for future research. In P. Harris, J. Stevenson-Hinde & C. Parkes (Eds.). Attachment Across the Lifecycle, (pp. 127-159). New York: Routledge.

Main, M., Kaplan, N. & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move to the level of representation. In I. Bretherton & E. Waters (Eds.), Growing points in attachment theory and research. Monographs of the Society for Research in Child Development, 50, 66-104.

McAdams, D.P. (1993). The stories we live by. New York: Guilford Press.

Mitchell, S.A. (1988). Relational concepts in psychoanalysis. Cambridge: Harvard University Press.

Mitchell, S.A. (2000). Relationality: From attachment to inter-subjectivity. Hillsdale, NJ: The Analytic Press.

Schaffer, H.R. & Emerson, PE. (1964). The development of social attachments in infancy. Monograph of Social Research and Child Development, 29(3), 1-77.

Scharff, J. & Scharff, D. (1998). Object relations individual therapy. Northvale, NJ: Jason Aronson.

Schiffer, F., Teicher, M.H., & Papanicolaou, A.C. (1995). Evoked potential evidence for right brain activity during recall of traumatic memories. Journal of Neuropsychiatry, 7, 187-250.

Schore, A.N. (1997). Interdisciplinary developmental research as a source of clinical models. In M. Moskowitz, C. Monk, C. Kaye, & S. Ellman (Eds.). The neurobiological and developmental basis for psychotherapeutic intervention, (pp. 1-71). North-vale, N.J.: Jason Aronson.

Siegel, D.J. (1999). The developing mind. New York: Guilford Press.

Sorenson, R. (2004). Minding spirituality. New York: The Analytic Press.

Spence, D.P. (1982). Narrative truth and historical truth: Meaning and interpretation in psychoanalysis. New York: WW. Norton.

Sroufe, L.A. & Waters, E. (1977). Attachment as an organizational construct. Child Development, 48, 1184-1199.

Stern, D. (1985). The interpersonal world of the infant. New York: Basic Books.

Stern, D.N., Sander, L.W., Nahum, J.P., Harrison, A.M., Lyons-Ruth, K., Morgan, A.C., Bruschweiler-Stern, N., & Tronick, E.Z. (1998). Non-interpretive mechanisms in psychoanalytic therapy: The 'something more' than interpretation. International Journal of Psychoanalysis, 79, 903-921.

Stolorow, R. & Atwood, R. (1992). Contexts of being: The intersubjective foundations of psychological life. Hillsdale, NJ: The Analytic Press.

Sullivan, H.S. (1953). The interpersonal theory of psychiatry. New York: Norton.

Winnicott, D.W. (1958). Through paediatrics to psycho-analysis. London: Hogarth Press.

Winnicott, D.W. (1965). The maturational process and the facilitating environment. New York: International Universities Press.

Winnicott, D. W. (1971). Playing and reality. London: Tavistock Publications.

AUTHOR

HALL, TODD. Address: 13800 Biola Ave, La Mirada, California, 90639. Title: Associate Professor of Psychology; Editor, Journal of Psychology and Theology; Director, Institute for Research on Psychology and Spirituality. Degrees: Ph.D., M.A., Rosemead School of Psychology, Biola University; M.A., Doctoral specialization, University of California, Los Angeles; B.A., Biola University. Specializations: Spiritual Development, Attachment theory, Relational Psychoanalysis.

TODD W. HALL

Rosemead School of Psychology

Biola University

Correspondence concerning this issue may be sent to Todd W. Hall, PhD, Rosemead School of Psychology, Biola University, 13800 Biola Ave, La Mirada, California, 90639.