Attachment, Bonds to Parents, Physical Aggression, and Violence
In Chapter 2, we laid out the foundation for our expectation that lack of attachment to caregivers can have profound implications for social development. Because of the important role that attachment to caregivers seems to play in the early development of children, the fact that early childhood may be a sensitive period for developing social behavior, and the possibility that the “average expectable environment” probably includes a certain amount of sensitive care, we hypothesize that failures to establish attachment relationships may cause violent behavior.
Secure attachment is closely associated with parental warmth. Attachment security is likely to be correlated with and partially a consequence of caregiver warmth but is dependent rather on parental sensitivity. Attachment and warmth are normally treated as distinct constructs in the literature, and we provide separate chapters on warmth and attachment in this book. Measures of attachment are frequently child-centered and indicate that the child is bonded to the parent. Parental warmth, by contrast, is consistently operationalized as behavior emanating exclusively from the parent. Parental warmth includes affection, positive regard, and rewarding emotional expression. Some indicators of parental attachment are parent-centered, such as “parental sensitivity” and “parental responsiveness,” but they tend to emphasize the more “technical” aspects of parenting including noticing when the child has needs and moving to address those needs. The description by Ainsworth, who is credited for introducing the concept, clearly shows this: “... ability and willingness to perceive the infant’s communications as reflected in his behavior, emotional expression, and vocalizations, see and interpret them from the infant’s point of view and respond to them promptly and appropriately according to infant’s developmental and emotional needs” (Grossman & Grossman, 1991, p. 97).While parental sensitivity (one of our indicators of attachment) and parental warmth are likely to be correlated, parental sensitivity and responsiveness could arguably exist without warmth. Obviously, the distinction can be arbitrary in some cases, but we believe it is an important one.
In the field of criminology, the attachment construct has received attention largely due to its centrality in Hirschi’s theory of social control, which emphasizes the role of weak social bonds in the etiology of criminal offending. Criminologists tend to downplay or ignore the timing and quality of the attachment bond and instead emphasize the strength of bonds at the time a crime is being contemplated. Hirschi (1969) saw bonds of affection mainly as a deterrent to crime: “The stronger this bond, the more likely the person is to take it into account when and if he contemplates a criminal act” (p. 83). In some studies of delinquency, the authors have suggested that an important consequence of a good attachment relationship is the acceptance of parent values (Arbona & Power, 2003; Quay, 1987; Stern & Smith, 1995). Some have also posited that the effects of attachment may really be due to increased monitoring or supervision which prevents the formation of deviant friendships (e.g., Vitaro, Brendgen, & Tremblay, 2000).
Insecure attachment has been the object of many studies on the etiology of aggressive, delinquent and criminal behavior. It is empirically associated with aggression (e.g., Buist, Dekovic, & Meeus, 2004; Burgess, Marshall, Rubin, & Fox, 2003; Finzi, Ram, Har-Even, Shnit, & Weizman, 2001; Laible, Carlo, & Raffaelli, 2000; Lyons-Ruth, 1996; Roelofs, Meesters, ter Huurne, Bamelis, & Muris, 2006), antisocial behavior (e.g., Arbona & Power, 2003; Burgess et al., 2003; Kochanska, Barry, Stellern, & O’Bleness, 2009; Marcus & Betzer, 1996), externalizing (e.g., Elgar, Knight, & Worrall, 2003), delinquency (e.g., Buist et al., 2004; Burgess et al., 2003; Dornbusch et al., 2001; Franke, 2000; Gardner & Shoemaker, 1989; Laub & Sampson, 1988; Miller, Esbensen, & Freng, 1999; Owens-Sabir, 2007; Rankin & Kern, 1994; Sommers, Fagan, & Baskin, 1994; Stern & Smith, 1995; Wright & Cullen, 2001), “serious deviance” and “severity of delinquency” (e.g., Anderson et al., 1999; Miller et al., 1999), sex offending (Marsa et al., 2004), and delinquency trajectories (e.g., Allen et al., 2002). It has been associated with persistent criminality into adulthood (Le Blanc, 1994) and chronic offending. Jang (1999) reported that the direct and total effects of family attachment continued to be statistically significant through adolescence, suggesting that the influence of parents does not decline as dramatically as many would predict during this developmental period. In the Dunedin sample, Moffitt and colleagues found that one of the predictors of the life-course-persistent pattern of offending was having had many changes of primary caretaker (Moffitt, 2006).
Many authors now examine indirect effects of attachment relations on antisociality and also test the role of attachment as a protective factor against the impact of other criminogenic risks (e.g., Fonagy, Target, Steele, & Steele, 1997). Ingram, Patchin, and Huebner (2007) conclude that attachment yields an indirect effect on serious delinquency in part because it is associated with supervision, which reduces affiliation with delinquent peers. Attachment relationships have been protective against the adverse effects of low socioeconomic status (SES) and association with delinquent peers (Carswell, 2007). Alink et al. (2009) report that the adverse effects of negative parental discipline were nullified when children experienced high maternal sensitivity.
DeKlyen and Greenberg (2008) review the literature on attachment and psychopathology in childhood. They reason that attachment may inversely affect behavior disorder through several processes; we will highlight those that we think apply to violence. In a later chapter, we will discuss neglect as a separate line of research, though neglect and attachment problems are likely to be strongly linked.
First, attachment may influence psychopathology and behavior through its impact on emotion regulation (e.g., Thompson & Meyer, 2007). In Chapter 2 we discussed the importance of negative emotionality and emotion regulation in the development of antisocial behavior. Some authors have argued that secure attachment allows children to develop the ability to “tolerate and manage affect” (p. 642) and Landy and Peters (2007) list several important components of parent-child interaction that they see as important for the development of adequate selfregulation. Sensitive responding is the first one on their list. Lyons-Ruth and colleagues also make a strong statement about attachment and negative affect in summarizing their study results: “The strongest single predictor of deviant levels of hostile behavior towards peers in the classroom was earlier disorganized/ disoriented attachment status” (Lyons-Ruth, Alpern, & Repacholi, 1993, (p. 572). Dutton (1999) points to insecure attachment as a common traumatic stressor in the early lives of abusive men which he proposes leads to intimate rage later in life.
We believe that this line of reasoning is particularly applicable to significant disruptions in attachment relations, such as separation from parents and parent death. Klevens and Roca (1999) explain that persistent offenders in their Colombian sample had similar risk factors to their less persistent counterparts, but theirs were more serious. The example they provide is that the more serious offenders had experienced a parental death, while the resilient offenders had experienced parental abandonment. To Leibman (1992), the loss of a parent in the child’s view is a deliberate act of abandonment to which the child is likely to respond with sadness, rage, fear, shame, and guilt. Ongoing uncertainty and anxiety prevent the child from developing into a secure and emotionally stable adult. Leibman reasons that abandonment fosters suspiciousness and mistrust which, in conjunction with the lack of coping skills that the individual would have learned had he developed a secure attachment in early life lead to easy provocation and lack of control over rage.
Second, attachment problems may be directly linked to disruptive behaviors in early childhood. Attachment strategies, such as attention seeking, could be seen as caused by insensitive parenting if we accept that children naturally seek attention and will employ whatever means necessary for obtaining it: “Theoretically, a child who has received less contingent caregiving might act more disruptively to obtain parental attention and have less to lose by disobeying parental requests (i.e., loss of love). Such interaction patterns could promote the beginning of coercive interaction patterns. ...” (Shaw & Winslow, 1997, p. 153). Studies suggest that maternal lack of responding is associated with comorbid oppositional behavior problems (Pfiffner, McBurnett, Rathouz, & Judice, 2005). It is likely that emotions play an important role in this process. Other authors have observed less anger in securely attached children than insecurely attached children (e.g., Kochanska, 2001).
Summarizing data on aggressive children in Canada in their article, “Bad, Sad, and Rejected" Sprott and Doob (2000) write: “From all perspectives, regardless of gender, the ‘most aggressive’ children were more likely to describe themselves in terms that simply made them sound unhappy when compared to the ‘other’ children” (p. 129). In the Minnesota Study of Risk and Adaptation from Birth to Adulthood, seen as the most comprehensive study of early attachment and its later effects, the authors report significant associations between early secure attachment and later positive affect. Preschoolers characterized as “securely attached” in infancy fussed and whined less and brought “enjoyment and enthusiasm” to their interactions with others (Sroufe et al., 2005a). Those ranked high in self-esteem were virtually the same group as those previously identified as securely attached in this study.
Finally, like Bowlby and others, DeKlyen and Greenberg (2008) make the case that attachment is likely to affect working models of intimate relations. Their discussion implies that insensitive parenting could result in a child whose expectations of care and trust will be low. If we follow this line of reasoning, it also implies that the insecurely attached individual is likely to care less about harming others. Research suggests that securely attached children show greater empathy toward others (Smith et al., 2003). Magid and McKelvey (1987) imply that, at extremes of maternal deprivation, “affectionless psychopathy” may be caused due to absence of early attachment relationships. Others have drawn links between the inability to form attachment bonds and psychopathy (Saltaris, 2002), though it is unclear whether this deficit is due to genetics or parental behavior.
The fact that there exist different forms of an attachment bond is a sign that attachment is a navigational tool to help the child orient to the world he or she is likely to inhabit (Belsky, 2005). This idea puts one in mind of our discussion of sensitive periods and “programming” that were discussed in Chapter 2. This perspective implies that the nature of the early relationship “calibrates” expectations for social relationships. Because there is “clear evidence” for a critical or sensitive period for neural development (Dutton, 2011) and because structures related to emotion regulation are developing rapidly in early life, the proposition that the effects of attachment are long-term or even permanent becomes highly plausible. A large body of evidence documents the long-term stability of attachment from infancy through later phases of development (Ammaniti, Speranza, & Fedele, 2005).