DISORGANIZED INFANT ATTACHMENT STRATEGIES AND HELPLESS-FEARFUL PROFILES OF PARENTING: INTEGRATING ATTACHMENT RESEARCH WITH CLINICAL INTERVENTION

Abstract

In this article, recent research on parenting behaviors associated with infant attachment disorganization is summarized and applied to a parent–infant psychotherapy case. Both hostile/self-referential and helpless-fearful patterns of parentingare described and viewed theoretically as alternate aspects of a single hostile-helpless internal working model of attachment relationships. The case material focuses on the more subtle and harder to identify manifestations of a helpless-fearful parental stance. Some attachment-related treatment guidelines for working with a hostile-helpless parenting stance are suggested, including challenging the hostile-helpless model implicitly in the qualities of the therapist's approach to the parent, explicitly articulating the hostile-helpless bind with the parent, increasing the parent's openness to a wider range of affective experience, differentiating attachment-related needs from other communications of the baby, and developing new skills for balancing the needs of the self and the needs of the other in interaction with the baby.

An emphasis on fearful arousal and the relational modulation of that arousal lies at the heart of attachment theory (Bowlby, 1969). This emphasis is clearly a departure from an emphasis on libidinal and aggressive drives as the central motivational systems in traditional psychodynamic theory and regrounds clinical theory in the developmental dynamics of fear. In keeping with this focus on fearful arousal, attachment research has illuminated the development of the infant's defensive adaptations to a caregiver's inability to provide the needed soothing responses to infant fear or distress. In this respect, attachment theory is a two-person theory of conflict and defense. It emphasizes the copingor defensive processes required to deal with fearful arousal within a particular set of attachment relationships. In contrast to an intrapsychic theory of defense, attachment theory and research locate the ontogeny of defenses in an intersubjective field.

In keeping with the focus of this special issue on bridging the gap between research findings and clinical applications, this article is organized around three aims. The first aim is to briefly summarize recent research findings on the parenting contexts associated with disorganized infant attachment behavior, with a particular focus on what we know about the more subtle varieties of parental stances associated with disorganization. The second aim is to propose a clinical/theoretical description of the implicit internalized models associated with hostile or helpless parenting stances. The third aim is to explore the relevance of these research findings for understanding the clinical presentation of a parent–infant psychotherapy case and to draw out some principles of therapeutic engagement with families in the disorganized spectrum that are consistent with attachment research.

PARENTAL AFFECTIVE COMMUNICATION PATTERNS RELATED TO INFANT DISORGANIZATION

The infant defensive adaptations captured in the insecure avoidant and ambivalent attachment patterns involve alterations of both attention and affect expression and are reliably observed by the end of the first year of life, much earlier than previous clinical theory would have predicted (Ainsworth, Blehar, Waters, & Wall, 1978). It has been documented in a variety of studies that parental behavior that is somewhat insensitive, that is, somewhat rejecting and intrusive or somewhat inconsistent and self-preoccupied, is related to the infant's display of avoidant or resistant attachment behavior when needing comfort.

However, it is not avoidant or ambivalent strategies but disorganized attachment strategies in infancy that are most relevant to clinicians. Disorganized attachment strategies in infancy consistently have been shown to be risk factors for later psychopathology in preschool, in middle childhood, and in adolescence (for review, see Lyons-Ruth & Jacobvitz, 1999). These behaviors are shown in Table 1. As disorganized infants acquire further cognitive capacities during the preschool years for taking account of others' states of mind, disorganized strategies often become organized into controlling strategies in which the child takes over the initiative to maintain the parent's involvement through caregiving or punitive behaviors. Parental behavior that is somewhat insensitive is not correlated with infant disorganized attachment behavior since studies using Ainsworth's global rating scale for sensitivity (Ainsworth et al., 1978) have generated only a small association between parental behavior and infant disorganization (van IJzendoorn, 1995). On the other hand, all of the parental factors shown to predict infant disorganization, such as parental psychopathology or parental unresolved loss or trauma on the Adult Attachment Interview (AAI; George, Kaplan, & Main, 1984/1985/1996), suggest that aspects of parent–infant interaction contribute to the development of infant disorganization (for review, see Lyons-Ruth & Jacobvitz, 1999). Frank maltreatment is clearly associated with infant disorganization (Carlson, Cicchetti, Barnett, & Braunwald, 1989). However, that criterion for problematic parental behavior is too extreme since 15% of infants in low-risk families display disorganized attachments (van IJzendoorn, Schuengel, & Bakermans-Kranenburg, 1999), and in longitudinal cohorts where families are known to researchers over long periods of time, it is clear that many disorganized attachments occur outside the context of abuse.

TABLE 1.

Indices of Disorganized-Disoriented Infant Attachment Behavior

Sequential display of contradictory behavior patterns, such as very strong attchment behavior suddenly followed by avoidance, freezing, or dazed behaviors. Simultaneous display of contradictory behaviors, such as strong avoidance with strong contact-seeking, distress, or anger. Undirected, misdirected, incomplete, and interrupted movements and expressions, for example, extensive expressions of distress accompanied by movement away from, rather than toward, the mother. Stereotypies, asymmetrical movements, mistimed movements and anomalous postures, such as stumblingfor no apparent reason and only when the parent is present. Freezing, stilling, and slowed “underwater” movements and expressions. Direct indices of apprehension regarding the parent, such as hunched shoulders, fearful facial expressions. Direct indices of disorganization and disorientation, such as disoriented wandering, confused or dazed expressions, or multiple, rapid changes in affect.

From M. Main and J. Solomon, 1990.

Several recent studies have focused on exploring the parent–infant interactions associated with disorganized attachment behavior as first observed in infancy. Main and Hesse (1990) advanced the hypothesis that disorganization of infant attachment strategies is related to parental unresolved fear, fear that is transmitted to the infant through parental behavior that appears frightened or that is frightening to the infant. According to Main and Hesse's reasoning, if the parent arouses the infant's fear, this places the infant in an unresolvable paradox regarding whether to approach the parent for comfort because the parent becomes both the source of the infant's fear and the haven of safety.

Several laboratories have explored Main and Hesse's (1990) hypothesis that the parent's frightened or frightening behavior is the distinctive element that is associated with disorganization of infant attachment strategies (Jacobvitz, Hazen, & Riggs, 1997; Schuengel, Bakermans-Kranenburg, & van IJzendoorn, 1999; True, Pisani & Oumar, 2001). In our own lab, prior to the development of the Main and Hesse (1992) coding instrument for frightened or frightening behavior, our pilot work had led us to advance two additional hypotheses regarding the parental behaviors that might be disorganizing to the infant. First, we reasoned that parents might display competing or contradictory caregiving strategies, much as the disorganized infant displays competing or contradictory attachment strategies. Second, we reasoned that the parent's overall regulation of the infant's fearful arousal might be more important than specific behaviors, so that failures to be adequately engaged and structuring with the infant, particularly when attachment needs were aroused, might be as important as more obviously frightened or frightening parental behaviors. In this view, parental withdrawing behaviors or role-confused behaviors that left the infant without adequate parental regulation of fearful affect also would be potentially disorganizing, whether or not the parent's own behaviors were directly frightened or frightening to the infant (Lyons-Ruth, Bronfman, & Parsons, 1999).

Therefore, in addition to coding specific frightened or frightening behavior, five broader aspects of disrupted parental affective communication with the infant also were defined and coded using the AMBIANCE coding instrument (Bronfman, Parsons, & Lyons-Ruth, 1993). These aspects included (a) parental withdrawing responses, (b) negative-intrusive responses, (c) role-confused responses, (d) disoriented responses, and (e) a set of responses we termed affective communication errors, which included both simultaneous conflicting affective cues to the infant and failures to respond to clear affective signals from the infant. Examples of these behaviors are given in Table 2.

TABLE 2.

Dimensions of Disrupted Maternal Affective Communication

Affective Communication Errors Contradictory cues e.g., invites approach verbally then distances Nonresponse or inappropriate or mismatched responses e.g., does not offer comfort to distressed infant; mother smiling while infant angry or distressed Disorientation (includingitems from Main & Hesse, 1992) Confused or frightened by infant e.g., exhibits frightened expression; quavering voice or high, tense voice Disorganized or disoriented e.g., sudden loss of affect unrelated to environment; trancelike states Negative-Intrusive Behavior (including frightening items, Main & Hesse, 1992) Verbal negative-intrusive behavior e.g., mocks or teases infant Physical negative-intrusive behavior e.g., pulls infant by the wrist; bared teeth; looming into infant's face; attacklike posture Role Confusion (includes items from Sroufe, Jacobvitz, Mangelsdorf, DeAngelo, & Ward, 1985; Main & Hesse, 1992) Role reversal e.g., elicits reassurance from infant Sexualization e.g., speaks in hushed, intimate tones to infant Self-referential statements e.g., “Did you miss me?” “Ok, he doesn't want to see me.” Withdrawal Creates physical distance e.g., holds infant away from body with stiff arms Creates verbal distance e.g., does not greet infant after separation

Infants classified as disorganized in their attachment behavior also are subclassified into two subgroups according to standard coding procedures, based on the type of organized attachment strategy their behavior most closely resembles. These two subgroups are usually labeled disorganized–secure and disorganized–insecure. Here we will use the more behaviorally descriptive labels of disorganized–approach and disorganized–avoid/resist.

As we predicted, the frequency of these five aspects of disrupted parental affective communication was significantly related to the extent of the infant's disorganized attachment behaviors. In addition, these disrupted maternal affective communications, which were coded during a series of separations and reunions, demonstrated cross-situational stability in that they also were related to similar behaviors observed at home. Higher levels of disrupted maternal communication in the separation procedure also were associated with increased infant distress at home. Neither infant gender nor cumulative demographic risk were significantly related to maternal disrupted communication (Lyons-Ruth, Bronfman, & Parsons, 1999).

When examined separately, the frightened or frightening behaviors described by Main and Hesse (1992) showed the same relation to infant disorganized attachment classification as did overall disrupted communication (see also Jacobvitz et al., 1997; Schuengel et al., 1999). However, the specific behaviors described by Main and Hesse constituted only 17% of the behaviors included in the larger coding protocol for disrupted affective communication. When all frightened or frightening behaviors were removed from the total disrupted communication score, the remaining disrupted behaviors still reliably distinguished between mothers of organized and disorganized infants. In Schuengel et al.'s (1999) data as well, a broader set of maternal behaviors termed “disorganized behaviors” were more strongly related to infant disorganization. These findings indicate that frightened or frightening behaviors are embedded in a broader context of disrupted affective communication between mother and infant.

More recently, the empirical link between infant disorganization and the parent's disrupted affective communication as coded by the AMBIANCE scales has been replicated in other samples across the economic spectrum (Goldberg, Benoit, Blokland, & Madigan, 2003; Grienenberger & Kelly, 2001; Madigan, 2002). In addition, maternal unresolved states of mind on the AAI also have been related to disrupted affective communication in several studies (Grienenberger & Kelly, 2001; Lyons-Ruth, Melnick, & Yellin, 2001; Madigan, 2002).

HOSTILE AND HELPLESS SUBGROUPS AMONG PARENTS OF DISORGANIZED INFANTS

There were two additional aspects to the study findings that also were quite clinically interesting. The first finding was that of the five classes of disrupted communication coded, parental affective communication errors were particularly strongly related to infant disorganized behaviors. These errors often included simultaneous conflicting affective signals to the infant, such as speaking soothingly but stepping out of reach, so that the parent's attachment-related behaviors toward the infant also displayed some of the same contradictory, unintegrated quality as the infant's disorganized attachment behaviors toward the parent. Maternal affective communication errors also predicted increased infant crying and infant proximity-seeking behavior. The infant's mixture of disorganized conflict behaviors with active signaling and approach behaviors appeared to mirror the mixed affective signals of the parent, which included positive cues mixed with subtle or muted negative cues.

A more surprising finding, however, was that there were substantial differences in maternal behavior within the disorganized infant group, differences that were correlated with the subtypes of infant disorganized behavior mentioned earlier (disorganized–approach, disorganized–avoid/resist). Two subgroups of mothers emerged in the statistical analyses who differed more from one another than they differed from other mothers in the study whose infants were not disorganized. This finding suggests that there is a dynamic operating among parents of disorganized infants that acts to produce polar opposites in parenting profiles.

Mothers in both of these disorganized subgroups displayed elevated levels of affective communication errors and disoriented behaviors. However, mothers in the two subgroups differed dramatically in rates of negative-intrusive, role-confused, and withdrawing behaviors. These differences in maternal behavior were conducted with different forms of disorganized behavior on the part of the infant. In the first subgroup, mothers of disorganized infants who also exhibited avoidance or resistance displayed significantly higher rates of both role-confusion and negative-intrusive behaviors than did mothers of other disorganized infants. Negative-intrusive and role-confused behaviors were strongly correlated as well, so these mothers were displaying a contradictory mix of rejecting behaviors (negative-intrusive) and behaviors that sought attention from their infants (role-confused). Mothers in this group displayed a high rate of ignoring and overriding the infant's clear signals by behavioral initiatives of their own, attributing feeling states to the infant with little rationale, and making a high proportion of self-referential statements, e.g., “Do it for mommy,” “Are you glad to see me?” Infants in this group showed a complex mix of distress and calling for mother combined with marked avoidance, resistant behaviors, or both when in her presence. These parent and infant behaviors are relatively easy to identify clinically. We termed this group “hostile/self-referential regarding attachment.”

In contrast to hostile/self-referential mothers, mothers of disorganized infants who continued to approach their mothers for comfort exhibited significantly higher rates of withdrawal. Mothers in this subgroup were more fearful, withdrawing, and inhibited and sometimes appeared particularly sweet or fragile. They were very unlikely to be overtly hostile or intrusive and usually gave in to the infant's concerted bids for contact; however, they often failed to take the initiative in greeting or approaching the infant and often hesitated, moved away, or tried to deflect the infant's requests for close contact before giving in. We termed this group “helpless-fearful regarding attachment.” Infants of “helpless” mothers also looked different from infants of “hostile” mothers in that they all continued to express their distress, approach their mothers, and gain some physical contact with them even though they also displayed disorganized behaviors such as freezing, huddling on the floor, apprehension, or avoidance while in contact with their mothers.

A helpless-fearful parenting stance may present with either fearful or with drawing elements in the foreground, so that some parents combine responsive behavior with an underlying anxious or tense fearfulness while others appear to manage the tension by attempting to divert the infant's interest away from themselves and interacting only when the infant demands it (for more extended discussion, see Lyons-Ruth, Bronfman, & Atwood, 1999).

Accurate identification of these more fearful and withdrawing patterns of caregiving is important for at least three reasons. First, there are repeated indications in the literature that parents in this group are harder to identify because their more withdrawing and fearful but nonhostile behaviors are harder to discriminate from more confident and structuring parental behaviors. Second, disorganized infants who continue to approach their mothers also are harder to identify, yet large national studies indicate that more than half of all disorganized infants display this pattern (NICHD Child Care Study, K. McCartney, personal communication, October 2003; NICHD Early Child Care Research Network, 2001). Finally, there is repeated evidence that disorganized infants who continue to approach their mothers are at equal risk for a variety of negative outcomes, including elevated cortisol secretion to mild stressors in infancy (Hertsgaard, Gunnar, Erickson, & Nachmias, 1995; Spangler & Grossmann, 1993), elevated hostile-aggressive behaviors toward peers in kindergarten and second grade (Lyons-Ruth, Alpern, & Repacholi, 1993; Lyons-Ruth, Easterbrooks, & Cibelli, 1997), and elevated rates of controlling attachment patterns toward parents by 6 years of age (Main & Cassidy, 1988; Wartner, Grossmann, Fremmer-Bombik, & Suess, 1994).

We would predict that mothers in the helpless-fearful group would show more appropriate caregiving behavior when the child's attachment system is not aroused and would become more fearful, hesitant, contradictory, or withdrawn when the infant's fearful and distressed attachment affects are more directly aroused and expressed. At such times, one would expect the mother's own underlying sense of helplessness to become more pronounced. As the infant begins to react with conflict and apprehension to the mother's hesitancy and fear in responding to attachment affects, the mother's sense of helplessness would likely increase. This transactional process might lead to the more obvious dysregulation in the relationship and compensatory controlling/caregiving behavior on the part of the child that is evident by the time the child is seen during the preschool period (e.g., Solomon, George, & DeJong, 1995). Table 3 summarizes the behavioral profile of helpless-fearful parents.

TABLE 3.

Helpless/Fearful Profile of Parental Behavior

Delayed responsiveness or ignored cues, usually followed by compliance to infant's continued demands (e.g., greets or hugs infant only after persistent bids by the infant) Withdrawing/distancing (e.g., fails to greet, interacts from a distance, stands at a distance, circles around the infant, holds infant facing out) Cursory Responsiveness (e.g., gives quick hug then moves away; “hot potato” quality to treatment of infant) Directs infant away from self to toys; uses toys to soothe Hesitation or tension at moments of heightened attachment, such as greetings or contact-seeking by the infant (e.g., parent hesitates, freezes, passes by infant, backs away, teases infant, or voice quavers, cracks, stutters) Little physical contact between mother and infant unless infant demands Vacatingparental role Little sense of authority Little sustained collaboration of mother with baby's initiative Little parental direction/protection May seem subtly fearful, submissive, or placating with regard to infant (e.g., high frightened voice when greeting; hesitation, then compliance with infant's cues) May show little overt negative affect or intrusiveness; not ominous or threatening

A CLINICAL/THEORETICAL CHARACTERIZATION OF HOSTILE-HELPLESS INTERNAL WORKING MODELS OF RELATIONSHIPS

Although the two polarized behavioral profiles observed among mothers whose infants were disorganized appear superficially quite different, we have advanced the theory that these two different constellations of parenting behavior can be meaningfully explained as alternate behavioral expressions of a single underlying hostile-helpless dyadic internal model (Lyons-Ruth, Bronfman, & Atwood, 1999; Lyons-Ruth, Melnick, Bronfman, Sherry, & Llanas, 2003). Our thoughts about the role of contradictory hostile-helpless internal models began with our data on the severity of trauma in the mother's childhood and its relation to her own interactive behaviors with her infant at home (Lyons-Ruth & Block, 1996). Those data revealed that physical violence or abuse in the mother's childhood predicted increased covert hostility and intrusive behavior toward her infant. However, this increased subtle hostility and interference were not displayed by mothers who had experienced sexual abuse only. Those mothers displayed only emotional and physical withdrawal. Because clinical treatment of sexual abuse survivors clearly reveals both the underlying fear and rage of those who have been victimized (see Terr, 1991), we felt that sexual abuse survivors were equally likely to be experiencing negative affects. However, mothers who had been exposed to violence or physical abuse appeared to handle their underlying fear and anger by identifying with an aggressive style of interaction while sexually abused mothers appeared more likely to manage their negative affects by more passive withdrawing from interaction with the infant, perhaps partially due to the difficulty in identifying with a male sexual aggressor.

These findings made clear that both hostile-intrusive and withdrawn parenting profiles can emerge from unbalanced victim/aggressor relational patterns in the parent's history. These research findings led us to theorize that hostile or helpless parental stances reflect alternative expressions of a single unbalanced, controlling-controlled relationship prototype experienced in the parent's own attachment history. As a single dyadic prototype, we would expect that parents displaying either the hostile or helpless profile alone also would maintain an implicit internal model of the complementary relational position in the dyad. The degree to which either the hostile or helpless position in this dyadic organization is identified with the self may depend on situational, temperamental, and cultural factors, including their own gender and its relation to the gender of their more domineering or more passive parent.

Converging with this formulation, George and Solomon (1996) interviewed mothers of 6-year-olds with controlling attachment strategies and found that mothers of both punitive and caregiving children communicate a helpless attitude toward the child, either describing the child as unrealistically supercompetent or as out of control in some way. We would expect that a single individual could display either or both of these relational stances at different times or in different situations or relationships. For example, a parent who is very submissive toward an authority at work could be very explosive toward the family at home, or a parent who is locked into a coercive struggle with one child may have an idealizing, submissive relationship to another child. In addition, a parent with a hostile-helpless internal model may show contradictory behaviors that reflect both hostile and helpless relational roles. The parental contradictory affective communications that were more frequent among both subgroups of parents of disorganized infants are viewed as evidence of such contradictory and unintegrated representations of parenting. Other parents may not display the contradictory elements of the internal model, acting more consistently in the role of either the hostile or the helpless party in the relationship. Importantly, what is represented mentally is the entire dyadic relational pattern of controlled self/controlling other.

Caregivers who display a hostile interaction pattern appear to be attempting to master unbearable feelings of vulnerability by denying their own feelings of fear and helplessness and identifying with a hostile or controlling parent. This denial may be accomplished through suppression of conscious experience of vulnerable emotions and through consistently controlling others in relationships through punitive behavior. Behaviorally, parents in this group may respond by suppressing children's distress and anger and using coercive disciplinary techniques.

Caregivers who display a “helpless-fearful” pattern may have adopted a lifelong caregiving adaptation characterized by paying hypervigilant attention to the moods or needs of others (e.g., their own parent) at the expense of having their own attachment needs met. Clinically, mothers in this category appear to be fearful and easily overwhelmed by the demands of others. Their long-time focus on others may be based on a coping strategy of dissociating from their own affect life and withdrawing from closer emotional contacts with others, so that they feel powerless to assert themselves more adaptively to manage their own affects as well as their infant's affects. Their anxiety and fear of close emotional contact may be noted by their young children, who in turn gradually develop either caregiving strategies or punitive, coercive strategies to maintain emotional contact with their parent. However, mothers in the “helpless” subgroup do not always seem to display the more pervasive suppression of emotional life and adoption of a “tough,” invulnerable stance that characterizes the “hostile” subgroup. Instead, they may appear anxious, tense, timid, or vulnerable themselves.

HELPLESS-FEARFUL PARENTING MODELS IN CLINICAL CONTEXT: THE CASE OF JANIE AND BRAD

What can the developing understanding outlined here of the helpless-fearful parenting profile offer to the therapist treating a mother and baby in distress? How do these research findings enhance our appreciation of the challenges faced both by the mother–baby dyad and by the therapy triad? Parents with young children who present for treatment are often at the beginning of an escalating controlling process; the child is becoming more and more demanding in an increasingly frantic attempt to feel comforted and safe with the parent, and the parent is feeling increasingly helpless and angry. The clinical material offered here describes one presentation of this dynamic within the context of a mother–baby psychotherapy case. In a separate case report, Lyons-Ruth et al. (2003) discussed how attachment theory and research can inform the clinical approach to a toddler with escalating coercive behaviors. Related parent–infant case material also is available in Lyons-Ruth (2003).

The clinical material discussed here captures a frequent presentation of issues and dynamics seen in the Early Connections Program at Jewish Family and Children's Service of Greater Boston, although the case details represent a composite of several families to best insure confidentiality. The dynamics and treatment process described here, conducted by the second author, are limited to the mother and baby and to only certain aspects of the treatment for the sake of clarifying as simply as possible the utility of the model from attachment theory. Other significant factors, such as the role of the baby's father and the family's community context, are left aside in this discussion.

The Attachment Context of New Parenthood

The experience of becoming a parent is extremely challenging physically and psychologically. The baby is in almost-constant need of a parent and expresses distress in strong, but often hard-to-read, ways. The new mother is in a relationship of forced intimacy with her baby and is dealing with intimate bodily functions and physical contact.

For the mother who has experienced relationships of harm and fear in her own early development, the responsibility for the well-being of another can be especially overwhelming. As described in the earlier research sections, the new mother's response to her baby's distress can be impeded by her own memories of punitive attacks or abandonment from her own parents. The relationship carries both sides of a potentially highly polarized internal working model. The mother may experience herself as both the angry or unavailable parent and the vulnerable baby; her baby may be experienced as both controlling and helpless. The parent is often caught between opposing fears: Either she will assert limits and become the uncaring, domineering parent of her own childhood who then will emotionally abandon her child, or she will withdraw and fail to set limits and the child will become the coercive, emotionally abusive figure from the past.

Attachment research situates this bind around limit-setting and assertive behavior within a deeper dilemma around attachment and abandonment. Parenting behavior that is either hostile–self-referential or helpless-fearful is not adequately responsive to the infant's attachment cues, and by definition involves a serious degree of emotional abandonment of the child with accompanying disorganization of attachment behavior. For the mother who has experienced such emotional abandonment in her own childhood, the fear both of losing the love of her own child and of her child's similarly experiencing abandonment by her if she responds with more anger and self-assertion represents a significant vulnerability. She often also is afraid of her own anger at her child that might accompany her frustrated attachment needs. As these emotional conflicts clash out of the mother's awareness, responses to the baby's developing relational initiatives can be understandably unpredictable and contradictory.

Janie and Brad: The First Visit

Thirty-five-year-old Janie was the first-time mother of Brad, now 9 months old. On the phone before the first home visit, Janie spoke softly and pleadingly about her current experience. She was utterly exhausted by caring for Brad, who was still breastfeeding every 2 to 3 hours during the day and waking several times at night.

At the initial home visit, Janie greeted the therapist at the door with Brad in her arms; she smiled briefly while Brad smiled broadly. Janie put Brad on the floor with toys and, although she sat at some distance from him, she talked to him sweetly about his play. Janie described the months of mothering as increasingly difficult for her. Sleep deprivation, a bad back, and Brad's frequent ear infections had left her depleted and questioning her decision to stay at home full time. She wearily summed up her experience: “I love him to bits, but I feel that he's in control of everything I do.”

When asked about her past history, Janie's voice and body became tense. She described an upbringing by an emotionally distant mother and a father who was sometimes loving, but often critical and angry. In recalling her sense of fear and loneliness, she spoke haltingly and tears filled her eyes. Brad looked up at her from his play with a sober expression; although initially unaware of his reaction, Janie eventually moved down to the floor and pulled him onto her lap, stroking his face. Her voice dropped almost to a whisper: “Don't worry, my baby, I'm not going to yell at you.” Addressing the therapist, she elaborated on her commitment to be a different kind of parent to Brad than she had experienced, planning to be caring and available, shielding him from pain and anger. What Janie had not noticed was that Brad had already registered her tearful upset and fearful affect, different from the angry upset that she felt determined to hide from him, yet with its own meanings.

As Janie talked about the current stress in her life, she acknowledged that being a parent was harder than she had ever anticipated. She was finding herself overwhelmed with the constant demands of daily care for Brad and was confused about how to meet his needs when he was unhappy. During this visit, Janie offered Brad her breast at several points when he was fussing; the first time, he settled into nursing for a brief time, but at subsequent times his fussing intensified. Janie did not appear to have other ways of relating to Brad to understand and alleviate his fussing. She looked imploringly at the therapist, who then acknowledged how hard it could be when your baby was crying and did not respond to attempts to calm him. As the visit ended, Janie agreed to continue meeting, although she voiced skepticism that anything could really help.

The Emerging Clinical Picture

Certain impressions emerged in this first encounter that were confirmed in future visits. Janie was working very hard at being a good mother, albeit with a palpable sense of anxiety and an increasing sense of defeat and withdrawal. She was drained both by the effort to be the all-giving mother and by the fearful affects related to past memories triggered by the mothering experience.

Janie's memories of her growing-up years emerged in bits and pieces, telling a choppy story of hurt and sadness. She could recall her own childhood history with emotional immediacy, although the incidents she shared sounded discontinuous, as though a lot remained unsaid and unintegrated into her awareness. Early on, she shared one particularly vivid incident in her relationship with her father when she was 6 or 7 years. They were having fun playing in a pool when he suddenly held her under water for a frighteningly long period; when he finally let her go, he was laughing. Her mother's distance seemed best expressed by the image of a closed bedroom door whenever Janie would look for her. Her current relationship with her parents was distant, although she was vague about this; they had only met Brad a few times despite living close by.

In her interactions with Brad, Janie was generally concerned and caring, although she often seemed passive and rarely initiated play or other activities. The difficulties they were having showed themselves most clearly when Brad was unhappy and not immediately soothed by her efforts. While her words were always sweet and placating, she was increasingly tense in her movements and agitated in her voice tone.

When talking about these moments, Janie would acknowledge frustration and complained that as Brad was growing, she faced these times more frequently. In the early months, she had felt pretty competent in following his lead and meeting his relatively simple needs for feeding and soothing, but now that he was beginning to express himself in more forceful ways and needed a more complex and active parental stance from her, she often felt like a failure in being able to “keep him happy.” As Janie became more trusting in the therapy, she was able to share more of the depth of her sense of helplessness, saying at one point that when Brad was inconsolable, she felt he was “enjoying torturing her.”

The View From Attachment Research: The Helpless-Hostile Model

Although neither Janie nor Brad was assessed with the AAI or the Strange Situation (Ainsworth et al., 1978), the clinical picture had many features in common with the helpless-fearful parental profile. Janie's history suggested a childhood where her experiences of fearfulness were not responded to by her parents. The swimming pool incident between Janie and her father captured the overall dynamics of their relationship. Not only was her feelingof terror not comforted but it was provoked and enjoyed in a perverse way by her father. At the same time, her mother's withdrawal made her unavailable for comfort in the face of Janie's fright.

Janie was acutely aware of some of the hurt she felt growing up and was consciously committed to never inflicting fear or abandonment on Brad; however, she had no inner representations available to her of a middle ground between her father's frightening and sadistic anger and her mother's passivity and abandonment. Additionally, she was unaware of the way her signals to Brad were subtly contradictory, so that her feeling of always submitting to his needs was overtly caring but tinged with aggravation, resentment, and fear.

As noted, during the early months, Janie had felt relatively successful; nursing had been a comfortable way of consoling Brad and feeling good about herself as a mother. This included nursing him to sleep and nursing again at his frequent night awakenings. Although others had recommended she let him cry when he awoke at night, she was convinced he would feel frightened and lonely if she did not respond immediately to his fussing. At 9 months, however, the dysregulation around sleep had become pervasive, and Janie's confusion and frustration were mounting. She experienced Brad as being in complete control with his crying for her. Her stance echoed strongly the findings on helpless-fearful parenting behaviors related to vacating the parental role, in that she had little sense of authority, direction, or sustained collaboration with her son.

She could imagine only two equally unsatisfactory possibilities: that of refusing to respond to him altogether or completely “giving in” to his demands. Janie felt compelled to “choose” a position of helpless compliance with her infant because the only other model available in her experience was both sadistically aggressive and abandoning. To feel angry at her child was a frightening experience, associated with the angry tirades directed at her by her father when she was a child. In disavowing her own aggression, she could only be its opposite—submissive to her baby no matter what the cost. She had difficulty helping him regulate his sleep because she experienced anything that involved letting him fuss as repeating her own history of hurt and abandonment. With no internal models of balanced and structuring nurturance available to her, she could not offer him help in the form of more modulated initiatives or responses. Attachment research helps frame Janie's extremely submissive behavior toward her baby in terms of these polarized internal models of relationship, and the conflicted and unintegrated mental and behavioral tendencies they engender.

Attachment Challenges in the Therapeutic Relationship

Janie's polarized internal working model of attachment was in evidence in the therapeutic relationship as well. From the outset, she had voiced fears that the therapist would not support her parenting choices; in the home visits, she was exquisitely sensitive to any hint of difference between the therapist and herself as reflecting criticism and rejection. The relationship felt constrained by her feeling that the therapist was either in complete agreement with her beliefs or hostile to them. In the early meetings, there was little room for exploration of the issues related to Brad's sleep, for example, because Janie responded to the therapist's interest in this as another instance of condemnation of her choices.

In holding fast to her rigid view of good parenting, Janie also was initially closed to any discussion and integration of her own needs into the relationship with Brad. To think about her need for adequate sleep threatened her equation of selflessness with closeness to her baby. She was suffering, yet she could not allow herself to consider alternatives that involved a more self-assertive parental stance on her part or any frustration for her baby. Instead, she would frequently collapse into a sense of resigned helplessness.

Treatment Guidelines That Emerge From the Application of Attachment Research to Clinical Practice

The goal of treatment with this mother–baby pair can be conceptualized as preventing the consolidation of a hostile-helpless parenting dynamic in Janie's relationship with Brad. Use of attachment theory shapes several general principles to guide the work with this pair:

1. Establishing security in the therapeutic relationship: The relationship with the therapist offers the possibility of a healing attachment experience in which differences and negative affects do not lead to attack or psychological abandonment. Janie needed to feel the therapist's attentiveness to and acceptance of the entire range of her affective experiences, including her negative feelings about the therapy. The treatment frame needed to both explicitly name the hostile-helpless bind and implicitly challenge Janie's polarized internal workingmodel by beingneither intrusive nor abandoning. From this base of security, which took a longperiod to establish, she could then begin to explore the meanings of the feelings mobilized in the relationship with her infant and develop a more coherent understanding of the constraints on her relationship with Brad.

2. Creating room for openness to a wider range of affective experience: Janie had access to certain memories about her past, but those had not yet been integrated in a way that allowed her both distance and flexibility in connecting with a broad range of feeling experiences. In her either/or view of relational possibilities, anger and related negative affects were associated with abusiveness, and she had to disavow those. This need to avoid negative affect was a likely contributor to Janie's passive stance in parenting, which in turn was a likely contributor to Brad's fussiness. Helping Janie to acknowledge, and even comfortably inhabit, the inevitable frustrations and angers of motherhood was important in allowing her to develop beyond her helplessness and resignation. For the parent caught in the internal bind of hostility versus helplessness, broadening the space for affective experience seems an essential therapeutic step.

3. Differentiating attachment needs from other emotional communications of the baby: Parents with difficult attachment histories often misread the attachment cues of their children. Janie's unintegrated memories of hurt and abandonment created a vulnerability to experiencing all Brad's discomfort as communicating fear and a need for her immediate response. In this aroused state, she could not differentiate between distress that signaled an acute attachment need and fussiness that was part of a more moderated frustration in learning something new or developing coping on his own. Conversely, when Brad was not distressed, Janie had little sense of what his needs might be for positive parental structuring, affect, and involvement. As the research suggests, Janie's internal model left her confused about responding to her child's attachment needs and unable to distinguish when a more structuring response could be helpful rather than depriving. Helping her to sort out Brad's central attachment needs for protection, affection, and predictable responsiveness and discriminate these from less central discomforts was an important treatment goal.

4. Developing new models of balancing the needs of self and baby: One attachment bind that an unbalanced hostile-helpless model may contain is the belief that the needs of the parent and child are in contradiction. Although the tension between self and other underlies the dynamics of much of human interaction, for a parent like Janie this dilemma was greatly intensified by her past experiences. The fear of repeating the self-involved and domineering behaviors of her father or the unprotective and withdrawing behaviors of her mother kept her unable to claim any needs of her own as legitimate. With her own needs so often overridden or unseen while growing up, Jane was left with little sense of knowing herself as an individual. Parent–infant therapy needed to provide her with an experience of coming to know and feel valued in her own subjectivity so that she could develop more flexible and balanced models of negotiating self–other tensions.

Janie and Brad: The Treatment Process

The early work with Janie and Brad had a quality of tentativeness as the therapist found her way into the hazardous emotional terrain Janie occupied. Her presentation of defeated helplessness was intermingled with rigid convictions about what she should expect of herself as a mother. Many of the early therapeutic interactions involved bumping up against one or another pole of this rigid and contradictory attacker–victim dynamic. If the therapist empathized with the difficulty of giving up so much of herself to being a mother, Janie would feel that her commitment to putting Brad's needs before her own was being attacked. Connecting this dynamic in the therapeutic dialogue to her childhood past of polarized relationship choices—i.e., one of us must be victim and the other victimizer—allowed the therapist to name the challenge to the relationship as finding a “third way,” a space that was neither critical nor abandoning and neither hostile nor helpless.

With increasing understanding of Janie's past, Janie and the therapist were able to make links between her experiences of fearful submission to her abusive father, the emotional absence and lack of protection on the part of her mother, and her present responses to her baby. She was able to accept that it was understandably confusing to become a parent of a helpless, dependent baby in the face of the painful and unbalanced relational patterns she had experienced as a child. From acknowledging frustration and confusion, she began to give more specific words to other elements of her maternal experience, including her feeling that Brad was in charge and controlling her, her resentment of his waking up so much, her anger that he was so needy, and her jealousy that he was getting love from her that she had never gotten. Being able to share and understand these feelings in the context of a therapeutic relationship that was curious, accepting, and nonretaliatory allowed Janie to be less afraid of the feelings aroused in her relationship with Brad and to find her way towards a more involved and active mothering stance that was less frozen in a position of fearful helplessness.

As Janie became more able to acknowledge and integrate a broader range of her own feelings in the relationship with the therapist, she could increasingly tolerate a broader range of feelings in her relationship with her child without fearing that the loving attachment would be lost. A pivotal step in this process came when Brad was 15 months old and still involved in a pattern of frequent night nursings. The therapeutic relationship was then strong enough for the therapist to encourage her to question her assumptions of what Brad's night crying meant. Whereas she had thought he was afraid when he cried at night and needed her to comfort him with nursing, she was beginning to think that his cry might be his wish rather than his fear. With this newly developing capacity to see that the positive attachment relationship could accommodate a range of negative feelings and survive, she began to wean Brad gradually from night nursings and see that their relationship was still strong and that they both slept better.

The absence of her mother's emotional availability in Janie's childhood also had handicapped her in being able to offer Brad an experience of consistent emotional presence. In a session around the same time as the night weaning, Brad became quite unsettled. As his fussiness became more urgent, Janie was clearly feeling increasingly frustrated and struggling not to withdraw. She pleaded with Brad to calm down, seeming to need him to regulate his own behavior so she could feel better. When asked what she was feeling and what she thought Brad was feeling, she said, “I think he's angry at me … when he looks like that, I think he doesn't love me.” The therapist wondered whether she thought that might be her fear from past experiences rather than what Brad was expressing. Over the next several months, Janie and her therapist returned to this incident as central in untangling the dynamics underlying Janie's collapse into helplessness. Janie was vulnerable to experiencing Brad's expressions of negativity—anger, turning away, upset—in a role-inverted way; in those moments, he felt like her angry father or withdrawing mother and she became the wounded child. She was convinced that their relationship was headed for conflict and irreparable disruption. Janie and her therapist came to recognize when this confusion was happening during their meetings and would try together to make sense of Brad's communications and behavior. When Janie could hold onto seeing Brad as a toddler, herself as his mother, and have confidence in the basic strength of their relationship, she could begin to respond to his distress states with more differentiated and active behavior, balancing nurturing and limits.

DISCUSSION

The clinical material presented here offers a window into the intergenerational cycle of helpless-hostile parenting patterns and possible directions for intervention. Janie's presentation as a conscientious, but quite anxious, new mother strongly committed on a conscious level to being available to Brad but increasingly defeated in the face of his emotional demands describes one manifestation of such a helpless-fearful intergenerational dynamic. Janie's polarized internal working model left her limited in her caregiving responses to Brad. The therapeutic work revealed that Brad was simultaneously experienced as her own baby self who was vulnerable to being controlled, hurt, or left and as the one who is now controlling or rejecting her just as her parents did.

In this particular version of the intergenerational cycle, Janie was stirred by her infant's vulnerability and committed herself to respond compassionately as she would have wanted her parents to respond to her. The baby's continued demands, however, led understandably to feelings of tension, frustration, and irritation, which were frightening and unacceptable both because of their identification with a malevolent parent in her own past and because of their association with being emotionally abandoned by an attachment figure. When anger has been so strongly disavowed, as in Janie's case, the mother is left in a paralyzed and helpless position, with very unintegrated internal models of how to proceed, feeling forced to comply with the distressed demands of her baby at some moments and withdrawing unpredictably at others.

These findings have implications for clinical work at the level of the mother–baby relationship and at the level of the therapy relationship. In both relationships, attachment needs are mobilized in the face of intimacy and the experience of distress on both the mother's part and the baby's part. For a mother struggling with the dynamics of a hostile-helpless relational model, the relationship with the therapist is particularly vulnerable to feelings of power imbalance and polarization. Identifying the enacting of the hostile-helpless internal working model in the therapeutic interaction was helpful with Janie because it tied together past and present in the context of the current relationship; the therapeutic work involved finding a “third way,” i.e., a space for relating that was neither hostile nor helpless.

In the year of therapeutic work, Janie became increasingly able to feel the therapy relationship as a base of security for her exploration. The therapy focused on creating room for a full range of affective expression, with particular openness to the negative aspects of the experience of mothering. Moving then to the dynamics of the mother–baby relationship, the therapist helped Janie to understand Brad's communication of his needs, encouraging her to reflect on the full range of her son's affective experiences. The work facilitated finding ways for Janie to take initiatives with Brad that were involved, parental, and structuring without feeling hostile or abandoning. With an expanding sense of flexibility and freedom, although no doubt with many challenges ahead, Janie and Brad's relationship gradually freed itself of the rigid contradictions and consequences of the helpless-fearful cycle.