Introduction

In recent years, models of borderline personality disorder (BPD) have changed from an almost exclusive focus on the internal characteristics of a person to an emphasis on the developmental context. Research suggests that the majority of people diagnosed with BPD are characterized by insecure or disorganized states of mind with respect to attachment, as well as by unresolved trauma (Fonagy et al., 1996). Thus, individuals with BPD have usually themselves experienced early relational trauma and adverse childhood experiences. The long-term psychological sequelae of childhood trauma are profound and include difficulties in affect regulation and interpersonal relationships and failure of the capacity to conceive of how others think or feel – or to realize that people’s behavior is driven by emotional and psychological factors (mentalization). From this perspective, relationships, including the relationship with a child, are complex and compromised. This chapter will briefly outline the characteristics and etiology of BPD with a particular focus on attachment theory and the construct of mentalization (Fonagy et al., 2004). We will outline the impact of BPD on parenting and children development in this context. Drawing on clinical material, we will conclude the chapter by discussing the implications for clinicians working with these families.

BPD – characteristics

BPD is characterized by significant difficulties in maintaining stable attachment relationships and in the regulation of affect, self-image, and impulses. The cluster and behavior associated with borderline personality include fluctuations from periods of self-confidence and buoyancy to periods of absolute despair, unstable self-image, rapid changes in mood, fears of abandonment and rejection, and a tendency towards suicidal thinking and self-harm. Transient psychotic symptoms may also be present. The DSM-5 classification system describes the interpersonal relationships of individuals with BPD as swinging between poles of idealization and devaluation, stressing the individual’s often unrealistic expectations of the other in a relationship and coexisting anxieties in closer interactions. This is a pattern of “oscillating” attachments with the individual experiencing difficulties in maintaining an appropriate “distance” in relationships.

Individuals with BPD may engage in a variety of destructive and impulsive behaviors including self-harm, disordered eating, and excessive use of alcohol and illicit substances. Self-harming behavior in BPD is associated with a variety of different meanings for the person, including relief from acute distress and reconnection with feelings after a period of dissociation. As a result of the frequency of self-harm, people with BPD are at increased risk of suicide, up to 70% attempting suicide at some point in their life (Oldham, 2006). The typical symptoms associated with BPD are therefore likely to affect a parent’s ability to cope with the everyday stresses of caring for a child.

The significance of early attachment relationships

The quality of early care and interactions with the infant has long been seen as central to infant development and later socioemotional functioning. Following Freud, Bowlby (1969, 1973, 1980a) described the relationship with the primary caregiver as shaping children’s sense of self and their relationships with others – the first relationship is essentially a prototype for later relationships. Bowlby posited that infants’ attachment interactions are internalized over time, becoming a mental model that enables them to know what to expect in terms of their interactions with other people. These models emerge from interaction and communication with the carer and, in turn, reflect the carers’ capacity to “read,” process, and respond to the child’s emotional signals. Inner models of relationships are seen as core components of personality functioning, and they influence perception, cognition, and affect about relationships and form the basis for patterns of relating.

Attachment theory has developed a classification of attachment organization in adults which emphasizes the organization of thinking, feeling, and memory around attachment issues and the ways in which different attachment styles represent adaptational or defensive strategies to manage anxieties around significant attachment relationships. Within this framework, parenting style and the quality of interaction between parent and child will reflect the parent’s attachment status and attachment history. Parents may re-enact early attachment experiences with their own child and in some cases attempt to rework or resolve their own attachment-related trauma in the parent–child relationship. The concept of transgenerational transmission of attachment themes is central to understanding parents with BPD. A significant body of research has outlined the manner in which parental “state of mind” with respect to attachment is predictive of child attachment status; for example, insecurely attached parents are more likely to have insecurely attached babies. Similarly, the parent’s own history of being parented and attachment relationships influences emotional interaction and communication with the child. However, there remains a “transmission gap” in explaining the mechanism through which parents’ state of mind affects their child’s attachment.

Attachment-related trauma and development

Attachment is described as a biobehavioral system with the overall aim of maintaining infant proximity to the caregiver, particularly in times of danger or threat. The infant seeks security and comfort from a discriminated attachment figure (usually the mother), and, if safe and secure, will play and explore the world with confidence. Infants who are insecure about the availability of the caretaker may inhibit exploration in order to maintain proximity.

Attachment insecurity and, at its most extreme, attachment disorganization reflect the failure of the caregiver to meet the needs of the infant for containment of anxiety and arousal. Predominately found in high-risk samples, disorganized infants exhibit breakdown or failure to develop a coherent strategy to regulate anxiety with the caregiver. The infant may experience the carer as both frightening and frightened, and these experiences are both unresolvable and anxiety provoking, and have significant effects on the emerging capacity to understand emotions in the self and others. It follows, from a biological stance, that failure of the attachment system constitutes a significant stressor for the infant, and that this is reflected in terms of impact on infant neurodevelopment.

Attachment disorganization and psychopathology

Most attachment research in the area of psychopathology has turned towards examination of attachment disorganization as a risk factor for mental disorder, particularly personality disorder. Main and Hesse (1990) originally suggested that confusing, “frightened or frightening” behavior by a caretaker left the child with unintegrated representations, and in a situation of insoluble conflict of fearing the figure they need to approach. Put simply, the person who would usually help the child to feel safe – a parent – is unpredictable and likely to be the source of distress or harm. Main and Hesse hypothesized that attachment disorganization may be associated with later borderline or dissociative pathology referring to borderline mechanisms, such as splitting or dissociation, both functioning as defensive mechanisms. The original descriptions of infant disorganized behaviors include what are presumed to be dissociative behaviors: dazed expressions, freezing, lack of completion of behavioral sequences, and contradictory actions.

Several studies find a clear association between childhood disorganization and later aggressive behavior, suggesting that disorganized attachment is a general risk factor for behavioral disturbance. What follows is a brief summary of these. Longitudinal studies have linked infant disorganization with hostile/aggressive behavior towards peers and externalizing behavior problems at 5 and 7 years of age respectively. Children with a history of disorganization are also more likely to develop internalizing problems. Prospective longitudinal studies (Wartner et al., 1994) have demonstrated a shift from disorganized behaviors during infancy to a pattern of controlling behavior with mothers at 6 years of age. Main and Cassidy (1988) developed a rating system for identification of controlling behavior and found that 84% of infants described as disorganized at 12 months displayed controlling behaviors (punitive or caregiving subtypes) at 6 years. Controlling behaviors in children are associated with unresolved loss or trauma – as measured on the Adult Attachment Interview (George et al., 1985) – in mothers, supporting the hypothesis of the transgenerational transmission of unresolved mental states.

Parental mental health problems and substance abuse are also associated with childhood disorganization. While disorganization appears to be predictive of maladaptive behavior, both externalizing and internalizing symptoms have been associated with early disorganization, excluding a simple causal model. There is a need to develop and test more complex models of the development of psychopathology that can look at interactive effects and the second-order effects of attachment variables such as self-esteem, cognitive capacity, and social functioning.

Main and Hesse (1990), following from their original discussion of unresolved states of mind and infant disorganization, also see a clear association between lack of resolution and psychopathology, particularly personality dysfunction and dissociation. They suggest that parents with unresolved trauma or loss experiences may exhibit dissociative symptoms and distortions of interaction with the infant who experiences confusion and contradictory desires to approach and retreat from the caregiver. They describe parental unresolved behaviors as “frightening or frightened” – the parents may be frightening and anxiety provoking for the infant, or they may exhibit fearful avoidance of the infant. The infant’s subsequent confusion is reflected in the contradictory behaviors described as characteristic of disorganized attachment.

An integrative model of BPD and attachment – Fonagy and Target’s mentalization hypothesis

Fonagy and Target (1997) have developed an etiological model of BPD and severe personality disorder integrating attachment, cognitive, and psychoanalytic developmental models. This has contributed in a major way to pulling together disparate theoretical traditions and research models and has generated considerable research. Fonagy and Target’s model is informed by an understanding of the role of trauma in the etiology of BPD and provides an account of the way in which child maltreatment disrupts the development of mental processes needed to understand and regulate socioemotional interactions. The origins of this capacity of “mentalization” or “reflective functioning” are located in the attachment relationship and the parents’ inner working models of attachment.

Fonagy and Target argue that the quality of infant attachment to the parent involves not just sensitivity and “attunement” but also a crucial process whereby the parent acknowledges the infant as having mental states such as desires, feelings, and intentions and communicates her understanding of the “intentional stance” to the infant. The parent needs to reflect upon the inner world of the infant and communicate back to the infant her “reading” or understanding of the infant’s mental state along with an alternate emotional communication that contains the infant’s anxiety. Infants of parents with limited reflective functioning are more likely to be insecurely attached, while, at the most extreme point, infants who are maltreated or abused by attachment figures fail to develop their own reflective capacity, in turn disrupting the development of the representational capacities needed to regulate self and affective states.

BPD, parent–child relationships, and child outcomes

Characteristics of BPD will presumably also operate for the parent with BPD and will influence perception and expectations of the child and parenting behavior. As previously discussed, traumatic attachment issues in the parent’s past will also affect the parenting relationship and quality of early interaction.

Despite the clinical and theoretical importance of BPD in parenting and the relevance of this to child protection and mental health outcome, minimal research exists in this area. Most research has focused on the implications of mental illness such as schizophrenia and maternal major depression for parenting. Ironically, Rutter and Quinton (1984) noted that the presence of parental personality disorder was more strongly associated with poor child outcome than was parental depression. Personality disorders are, by definition, chronic conditions, persistently affecting relationship quality. This is in sharp distinction to an episode of disorder and the time-limited nature of those mental disorders with intervening periods of appropriate parenting. Children of parents with BPD are likely to have been exposed to inappropriate and insensitive care as well as a variety of other developmental risk factors such as parental relationship breakdown, domestic violence, parental substance abuse, socioeconomic adversity, and instability of housing and social supports. Given the clustering of risk factors, it is clearly difficult to isolate the direct impact of parental personality dysfunction on child outcome.

Nevertheless, a body of work describing the impact of parenting practices on child outcome has emerged over the past twenty years. Most authors highlight the interaction of factors within the parent (personality), factors within the child, and the sociocultural context of parenting in influencing child outcome. Parents with BPD will typically have vulnerability in neuroregulatory, psychological, and social domains; experience multiple stressors; and have few external supports. Their children will similarly experience multiple developmental risk factors on several levels, including the biological, psychological, and environmental, creating an extremely high-risk developmental climate.

Components of a parent’s personality linked to successful parenting behavior have been described by Heinicke (1984) as including (1) adaptation – conference, or the ability to solve problems calmly and flexibly; (2) capacity for sustained relationships, involving empathy and mutuality; and (3) positive self-development characterized by autonomy and confidence, rather than insecurity. Heinicke argues that the parent’s level of psychosocial functioning before birth is related to overall adjustment to parenting and parental responsiveness in early childhood. Similarly, parental psychological characteristics such as higher self-esteem, internal loss of control, and increased perspective-taking capacity have been associated with improved parenting functioning and increased (self-reported) sense of parenting efficacy. These studies are limited in that they tend to focus on individual personality characteristics as opposed to a broader concept of personality functioning and adaptation.

In a broader model the parent’s psychological resources and subsequent quality of parent–child interaction are central. Clearly, parents with personality disorder would be predicted to have significant difficulties in maintaining an empathic stance towards the child, in sensitive responsiveness, and in self-regulation, particularly in response to the emotional demands of the child. There will also be parents experiencing significant difficulties who do not meet standardized diagnostic criteria for personality disorder, but who may have areas of particular risk (such as high levels of impulsivity), placing them in the category of “high-risk” parents.

The parent’s own history of being parented is seen as having provided a model for later parenting behavior and also shaped the parent’s feelings, beliefs, and expectations in relationships. Parents who have experienced early abuse or rejection may have persistent conflicts over care and control, and difficulty in sustaining intimate relationships. Holmes (2003) describes this as conflict over intimacy versus autonomy, in which the individual struggles to maintain a sense of autonomous self-functioning, while at the same time allowing intimate engagement in relationships. These issues will presumably also pertain to the relationship with the infant where tolerance of dependency and physical intimacy may also be problematic. Parents with personality disorder may have difficulty with a wide range of parenting tasks but particularly with the tasks of sustaining an intimate relationship, maintaining a stable representation of the child, and tolerating negative affect and ambivalence.

The prevailing theoretical model of the “disorganizing” parent as described earlier remains, where the parent with unresolved loss or trauma transmits this to the child via specific interactional patterns characterized by insensitivity, misreading, or frightening interaction. The parents may themselves display contradictory caregiving strategies and simultaneously show the need to position the child as carer, as well as exhibiting avoidance or even active rejection of the child. In these situations the child experiences high levels of attachment anxiety and confusion about the operation of relationships and expression of emotional need in an interpersonal context. Patterns of parent–child interaction, however, also reflect the parent’s current level of stress and social context, sense of parenting confidence and self-efficacy, and representation of the child. The parent’s capacity to reflect on the child and to conceptualize the child as having individuality, mental states, and an internal world is increasingly recognized as fundamental to shaping the child’s attachment organization and inner representation of relationships.

BPD is associated with child maltreatment. The psychological functioning of abusive parents shows some features of BPD, specifically, lack of empathy, poor impulse control, and intolerance of intimacy. Theoretical and clinical data lead to the suspicion that many parents described in the child-protection literature and considered to be high risk or abusive have personality disorders or difficulties that contribute to aberrant parenting behavior. Most of this literature, however, has not adopted a clinical diagnostic frame in regard to parents, making interpretation difficult.

Overall, it appears that there are multiple vulnerabilities in parents with BPD, and a clustering of psychosocial risk factors, suggesting that, as a group, these parents are at risk of replicating dysfunctional interactional patterns with their children. There will be a number of risk factors for child maltreatment in “borderline families” including parental trauma and parental history of attachment disruption. Poor relationship functioning; limited coping strategies, including self-harm and suicidal behavior; and multiple social adversities, including isolation, poor access to services, and poverty, will all contribute to child outcome.