Identifying the Concepts
Although we commonly talk about social deficits in juvenile and adult sexual offenders, as Mulloy and Marshall (1999) have noted, “we need a clearer identification of just what it is we expect when we claim that sexual offenders have social deficits so that we can more definitively examine the hypothesis that they are deficient in these skills” (pp. 95-96). Nevertheless, despite many limitations, reported in most studies, we continue to be sure that “despite the problems... there appears to be no doubt that attachment styles are an important area of dysfunction in sexual offenders” (p. 106). If this is so, and it may be, it is important to take the time to better understand what we actually mean by attachment styles and, indeed, if attachment “style” is actually what we are discussing. In Part I of this article, we explored ideas about attachment, including the difficulty in clearly defining just what we are describing when we discuss and attempt to measure attachment, as well difficulties measuring the construct of attachment in sexual offenders.
Measuring Attachment in Sexual Offenders
It is troubling that as we review the little research conducted with juvenile and adult sexual offenders, we review studies that do not use the same well researched tools (the Adult Attachment Interview, for example) that are used in attachment theory studies of adults, and so we begin to lose the ability to make consistent or replicate research studies, or make strong inferences in contrast to the non-clinical, non-offender population. In fact, despite difficulties and inconsistencies in measuring attachment and possibly related concepts in sexual offenders (empathy, intimacy, loneliness, etc.), there is far greater consistency in measurement among a non-clinical population. Indeed, as a whole, attachment theory and related study is typically conducted in non-clinical settings, and faces a different set of testing circumstances and perhaps pressures. This may reflect not only a lack of usable tools but, even more basic, a problem with definition, meaning what do we mean by attachment, and most certainly disorders of or deficits in attachment?
However, despite measurement problems, if attachment is key in sexually abusive behavior, we would expect to see differences in the measurement of attachment among sexual offenders. Accordingly, as about 67% of the population is considered secure in attachment and about 33% insecurely attached, we would expect to see different percentages within sexual offenders, with a higher percentage of insecure attachment and a correspondingly lower percentage of secure attachment than in the general public. In the few studies so far, this is not the case. Further, if attachment is a central component in the development of adult sexual offending, we would expect to see differences in the attachment classifications of juvenile sexual offenders who do not go on to become adult sexual offenders (as far as we know, the majority of juvenile sexual offenders do not recidivate or develop into adult sexual offenders) when compared to juvenile sexual offenders who do continue to engage in sexually abusive behavior as adults. This obviously requires not only an adequate and valid definition of attachment in adolescents, but the application of tools sensitive enough to measure such attachment and the implementation of prospective attachment research studies that track juvenile sexual offenders over time, including both groups – juvenile sexual offenders who do not sexually recidivate as adults and juvenile sexual offenders who do.
Finally, just as we would expect to see differences between sub groups of juvenile sexual offenders who do not become adult sexual offenders and those who do, we would expect to see differences now between juvenile sexual offenders and juvenile non-sexual offenders, as well as adult sexual offenders and adult non-sexual offenders. In studies of adult sexual offenders, we do not see these differences, not only among groups of sexual offending and non-sexual offending criminals, but among sexual offenders, non-sexual criminal offenders, and the general public.
It is possible that this is because there is no difference, and the postulated role of attachment in the development of coercive or abusive sexual behaviors is not a critical feature. It is also possible that because we have not completed adequately well designed and controlled studies, or because we presently don’t have instruments capable of adequately sensitive measurement, the issue is a problem of design and/or measurement rather than theoretical construct.
The Diagnosis of Attachment Disorders
Attachment theory recognizes that disorganized attachment exists, in adults and children. But in general, attachment theory statistics assert that most individuals are securely attached, and most of the rest are insecurely attached, with relatively few individuals falling into the category of disorganized attachment.
If we take “disorganized” attachment to mean “disordered” attachment, this leaves a much smaller percentage of the population as “disordered” in their attachments, and doesn’t at all fit with the concepts we mean to imply when we describe individuals with attachment disorders or deficits. Based on statistics related to mental disorders and psychiatric diagnoses in the U.S., a far greater number than those conceived as having disorganized attachment, we can either conclude that mental health and psychiatric problems have nothing to do with attachment or that categories of attachment postulated by attachment theory are not sensitive enough to reflect the relationship between attachment (or social connectedness) and mental health and pathological functioning. For instance, according to the U.S. Surgeon General’s report on mental health (U.S. Department of Health and Human Services, 1999), 20% of Americans experience a mental disorder during the course of a single year.
However, the term attachment deficit or attachment disorder is used with increasing frequency, and there are therapeutic models that define themselves as specifically treating attachment disorders. Nevertheless, the term as used in this way remains unsupported by the literature of attachment theory. In fact, most attachment theorists do not apply the concepts of attachment theory to the treatment of clinical (i.e., pathological) populations, although there is recently a greater focus and more literature addressing observations of clinical populations and treatment applications.
Nevertheless, in DSM-IV-TR (American Psychiatric Association, 2000), we have only the single diagnostic of Reactive Attachment Disorder of Childhood or Early Infancy. This diagnosis is oriented toward very young children, and of little use in diagnosing attachment difficulties or deficits in older children, adolescents, or adults, although we may conceive of attachment problems as central to their functioning and the development and enactment of behavioral or psychological pathology. Indeed, this larger focus, without benefit of an age-appropriate or relevant diagnosis (that is, not RAD), has been the focus of the limited work done within the field of research with adult sexual offenders. This work (mostly with adult sexual offenders) suggests/asserts that there is such a thing as disordered attachment, although without a well developed diagnostic system cannot easily define the construct except in other related terms, through scales of emotional loneliness or empathy inventories, for instance. It is thus difficult to say what “disordered attachment” looks like. Work with adult sexual offenders also asserts that early attachment experience and classification does predict, or at least contributes to, later attachment difficulties and the development and enactment of sexually abusive behavior. There is little evidence for this, however, in part because we lack an adequate understanding of what we mean, what to look for, and how to diagnosis such attachment difficulties.
Other than RAD (for use in infants and young children), several diagnostic schemes or descriptions of attachment disorder have been developed to define and recognize attachment disorders in children and adolescents. However, these are usually so broad that they virtually recognize and diagnose all troubled behavior in children and adolescents as attachment disorders. They are thus of very limited utility, and lack any ability to discriminate between troubled behaviors in children and adolescents or provide the means for differential diagnosis. Levy (2000) doesn’t offer a set of related diagnostic categories but does describe disordered attachment: “Children who begin their lives with seriously compromised and disrupted attachment often become impulsive, extremely oppositional, lacking in conscience and empathy, unable to give and receive genuine affection and love, angry, aggressive, and violent” (p. 9). This, obviously subsumes a very wide range of troubled behavior in children as “attachment disordered,” virtually eliminating other alternative diagnoses. In fact, the range is broad that, attachment disorder diagnosis or not, it suggests that the cause of almost all emotional and behavioral problems in children is impaired attachment.
In keeping with this idea, Levy groups the causes of attachment disorder into three categories: (1) parental/caregiver contributions (e.g., abuse and neglect, depression, psychological disorders); (2) child contributions (e.g., difficult temperament, prematurity, fetal alcohol syndrome); and (3) environmental contributions (e.g., poverty, stressful and violent home and/or community), and writes that the most common causes of attachment disorder are abuse, neglect, multiple out-of-home placements, and other prolonged separations from the primary attachment figure. Hence, for Levy, who also uses the term disrupted (and compromised) attachment (Levy & Orlans, 1998), virtually all childhood and adolescent problems are caused by attachment malformation, which is the result of almost any damaging condition within the parent, child, or environment.
However, Levy’s broad description fits well with the ideas of Bowlby and the attachment theorists in that compromised attachment sets the basis (through the internal working model) for many life difficulties. The problem is that in attachment theory, there is no such thing as attachment disorder, and forms of attachment, other than disorganized attachment, are considered to be adaptive. Here, we cross a line between attachment theory, which merely observes individuals as they mature, and a theory of developmental pathology that suggest that even if developing behavior (and internal working models) is adaptive, it can still be judged as unhealthy and even disordered. The major problem with Levy’s model is that, although it may accurately reflect disturbances and deficits in attachment as the underlying cause of most childhood and adolescent psychiatric diagnoses, it is so broad as to not have useful diagnostic value.
Rather than the sweeping and all inclusive model proposed by Levy (and others), Zeanah, Mammen, and Lieberman (1993) propose five specific sub-types of attachment disorders, aimed more at specifically exhibited attachment behaviors rather than a more general set of behavioral or relational symptoms: Type I, non-attached attachment disorder; Type II, indiscriminate attachment disorder; Type III, inhibited attachment disorder; Type IV, aggressive attachment disorder; and Type V, role-reversed attachment disorder. Like RAD, but unlike the diagnostic scheme proposed by Brisch (below), these diagnoses apply only to young children (ages 1-5). Also, unlike Brisch’s diagnostic model, in this model symptoms need be present in only a single attachment relationship. The age range for the diagnosis, however, limits its use to young children. Its value and use is even further weakened by the fact that symptoms need be displayed in only one attachment relationship, rather than across a range of social situations, as well as the lack of specificity in defining for how long symptoms must be displayed before the diagnosis can be made.
Also aimed at specific attachment behaviors, Brisch (1999) writes that an attachment disorder diagnosis cannot be made on the basis of insecure attachment per se, as this attachment category (insecure) falls within the normal range of functioning. Instead Brisch requires disturbed behaviors in interactions with a variety of attachment figures that are manifested as stable patterns across both situations and longitudinally. He describes seven diagnoses of attachment disorders in children, which he asserts can be applied under defined circumstances to both children and adolescents: (1) no signs of attachment behavior, (2) undifferentiated attachment behavior, (3) exaggerated attachment behavior, (4) inhibited attachment behavior, (5) aggressive attachment behavior, (6) attachment behavior with role reversal, and (7) attachment disorder resulting in psychosomatic symptoms. However, although offering more specificity and focus, it is not clear that these same diagnostic symptoms cannot be more adequately explained by or subsumed under other, existing diagnostic (non-attachment) categories. Like other attachment diagnostic schemes, Brisch’s model is more theoretically-driven than having otherwise obvious face value. In addition, of course, it has no relevance in diagnosing attachment difficulties in adults, but presumably should as attachment difficulties are presumed to continue into and throughout adulthood, which is the whole idea behind connecting attachment deficits to adult sexual offending.
The Relevance of Attachment Deficits in Understanding Sexual Offending
Although attachment, in the sexual offender literature, has been suggested as one key variable in the development of adult sexual offending, less has been written about attachment as a key variable in the development of juvenile sexual offending. However, as attachment is developed in infancy, it is obvious that if it is a factor in adult sexual offending then it must also be a factor of some kind in juvenile sexual offending, or at least in those juvenile sexual offenders who go on to become adult sexual offenders.
This suggests concerns immediately, as many/most juvenile sexual offenders do not go on to become adult sexual offenders. This raises the idea that in reviewing similar cases and outcomes retrospectively (seeking commonalities, for instance, by looking backwards to the possible roots of the problems) one comes up with clusters of similar factors. In this case, we see early attachment and bonding difficulties and challenges as one of these factors, as well as histories of troubled and maltreated childhoods. However, if one starts with those factors postulated to be key and works forward, prospectively, one is not likely to see the same outcomes in every, or even most, cases. This is a misleading trap to fall into, and we must be careful to discuss commonalities in populations of juvenile and adult sexual offenders rather than profiles, and early childhood experiences as risk factors rather than pathologies with certain outcomes. This is as true for classifications of attachment, as for any other childhood risk (or protective) factor.
However, to date, there has been little material written in an in-depth manner on the subject of attachment in sexual offenders and the field has instead relied on generalized descriptions, explanations, and theorizing mostly found in single chapters within larger edited books or a relatively few articles describing research into the area. Even within book chapters and articles, descriptions of attachment difficulties theorized to exist among juvenile and adult sexual offenders do not provide adequate or in-depth exploration of the construct, and rely on poorly defined ideas, frequently paying little or no attention to the large body of related literature that exists outside of the forensic field, with little reference made to the huge literature of attachment theory itself.
Despite the fact that some of the better defined and researched literature within the sexual offender field points to and supports the idea that attachment, or failures in attachment, is key, as noted there is little in the field of attachment theory itself or the larger field of developmental psychology that suggests that attachment is actually related to the development of pathology. However, the premise in the sexual offender literature associates the development of sexually abusive behavior to presumed difficulties in attachment. It is hypothesized that the failure of individuals to form early secure attachments with their parents leads to the development of insecure adult attachment styles. This, in turn, compromises their capacity to form and maintain stable and satisfying romantic and other intimate relationships, as well as contributing or leading directly to other related deficits, such as loneliness, powerlessness, low frustration tolerance, anger, and interpersonal conflicts. Somehow, it is proposed, in some individuals a consequence of deficits in intimacy skills is transformed into coercive sexual activities in order to satisfy emotional needs. In this view, sexual offending is a distorted attempt to build interpersonal closeness in the absence of the social and psychological skills to build emotionally satisfying relationships.
Marshall and his colleagues (for instance, Marshall, Hudson, & Hodkinson, 1993; Marshall & Marshall, 2000; Marshall, Serran,& Cortoni, 2000) argue that insecure attachment style renders people vulnerable to sexual offending, and particularly sexual offenses against children. He recognizes this as a developmental vulnerability, and hence a risk factor rather than the cause (Marshall & Eccles, 1993), but considers that when individuals with this vulnerability are exposed to other predisposing or precipitating factors, they are more likely than securely attached individuals to engage in sexual abuse. Once an offense has been perpetrated, through the establishment of cognitive distortions that support the accompanying ideation, attitudes, emotions, and behavior, the behavior is likely to repeat. Attachment style is thus a significant variable in Marshall’s framework, and has been identified by others as central in the etiology of sexual offending.
Marshall suggests that these problems lead to a primary reliance on sexualized coping, including the early onset of masturbation and sexual acts with others, providing the offender with a way to deal with and avoid difficulties related to a history of family and childhood problems, as well as current difficulties and frustrations. In this model, ongoing sexually active and coercive behavior becomes a conditioned response that builds on sexualized coping, such as excessive masturbation, and aims these individuals toward on-going sexual coercion. In keeping with attachment theory, Marshall suggests that individuals with disturbed attachments experiences do not adequately develop self-regulatory skills, and thus rely on externally based means of self regulation.
A Proposed Disrupted Attachment Pathway to Sexual Offending
Smallbone and Dadds (1998, 2000) argue that the link between attachment style and sexual offending occurs because adult sexual offenders experience an overlap between the attachment, caregiving, and sexual systems of their childhood and early adolescence, and children who experience child abuse may develop a disorganized attachment style that leads to sexual behavior when they experience high levels of stress. Child sexual abuse occurs when individuals with patterns of disorganized attachment experience stress and also have access to a child. Similarly, Burk and Burkhart (2003) focus on the role of disorganized attachment in the development of sexually abusive behavior, with controlling sexual strategies used by the sexual offender as a means to re-establish an internally experienced state of emotional and cognitive organization.
Although supporting the overarching premise of Marshall’s model, Burk and Burkhart identify gaps in his framework that fail to describe how stressors that may be attachment related actually lead to the adoption of sexually abusive behavior. They assert the model is incomplete because it fails to provide an explanation of motivation sufficient enough to initiate the use of sexually coercive behavior in the first place. They argue that disorganized attachment is a disorienting and emotionally and cognitively uncomfortable experience, and that the very same conditions that lead to disorganized attachment in the first place also leads to the delayed onset of attachment in some individuals, and the subsequent development of controlling attachment strategies. Such strategies are designed to regain control over and make predictable an otherwise uncomfortable interpersonal environment, and thus serve to stabilize the internal emotional state. Like Marshall, they argue that these individuals come to use sexual behaviors, including masturbation, as pleasurable external coping mechanisms, coupled with an external control mechanism that is virtually always present in a sexual offense. In addition, they note that exposure to maltreatment, aggression, and sexual behaviors, including personal sexual victimization; during childhood is an experience common in the lives of many sexual offenders. Burk and Burkhart thus contend that in sexual offenders these elements combine and result in the use of sexual coercion to fill a self-regulatory need that the attachment disorganized individual is otherwise unable to fill. They add that the physically gratifying properties of the sexual act, as well as the “pseudo-intimacy” provided by the sexual act, reflects a control strategy designed to regain cognitive and emotional equilibrium.
However, although more complex than Marshall’s work, Burk and Burkhart are essentially proposing the same model. Although offering more detail, their model nonetheless remains highly theoretical. It, too, fails to explain why many individuals encountering very similar early life experiences, including disorganizing and disrupted attachment experiences and early forms of maltreatment and abuse, don’t later resort to sexually abusive behaviors. It also raises the question of why and how juvenile sexual offenders, most of who don’t continue on as adult sexual offenders, manage to overcome these very conditions that presumably (in this model) led to their engagement in sexually abusive behavior in the first place.
Smallbone (submitted) also amends Marshall’s model. He asserts that sexual offenders confuse and blend attachment, parenting, and reproductive behaviors/strategies, catalyzed in sexually abusive behaviors by other forces such as psychosocial environment and situational variables. In particular, Smallbone is critical of the way that “attachment” is described as a stable trait in individuals rather than a changing variable that is in constant interplay with the sometimes competing and sometimes overlapping parenting and sexual behavioral systems, as well as other environmental and personal variables.
In fact, it may indeed be that attachment is one highly significant component in the development of sexual aggression, combined with the many other elements that are found so frequently in the lives of juvenile and adult sexual offenders. However, the proponents of the attachment route to sexual offending fail to overcome the contention of well developed attachment theory that insecure attachment itself is not a sufficient cause, and seem to confuse and blend insecure and disorganized attachment styles, describing individuals as disorganized who, in the relatively few studies of attachment in sexual offenders, fail to show as “disorganized” in their attachment patterns, as described below.
Support for the Relationship Between Attachment and Sexual Offending
Of the limited research conducted with sexual offenders, despite assertions that are more hopeful than certain, little has been established that points to any significant relationship between adult sexual offending and attachment deficits, other than on a theoretical basis. For instance, in Marshall, Serran, and Cortoni’s (2000) study of child molesters, non-sexual offenders, and non-offenders, the results showed no difference in attachment styles between groups, and, in fact, indicated “that all subjects reported greater security in their attachments to their mothers than to their fathers.... The only difference observed in characteristic coping showed that child molesters were more likely to engage in emotion focused strategies” (p. 17).
Smallbone and Dadds (2000) noted that their hypothesis that childhood attachment styles predicted adult attachment styles was only partially supported, as was their hypothesis that insecure attachment childhood predicted antisocial, aggressive, and coercive sexual behavior. Their most supported conclusion was that avoidant paternal attachment may influence the development of aggressive and antisocial dispositions, but this was contradicted by their subsequent study (Smallbone & Dadds, 2001) which showed that avoidant maternal, and not paternal, attachment was implicated in the development of coercive sexual behavior. In their study of incarcerated male sexual offenders, Smallbone and McCabe (2003) concluded that, contrary to their expectations, the sexual offenders in this study were no more likely to have experienced insecure attachment than they were to have experienced secure childhood attachment, and the frequency with which the sample reported insecure attachment was no greater than in the general population. In their review of parental attachment, Marshall and Mazzucco (1995) found no significant differences between child molesters and non-offenders in perceived parental rejection, and failed to support their thesis that low self-esteem is related to sexually abusive behavior. Similarly, Marshall, Barbaree, and Fernandez (1995) found that child molesters did not differ from a matched community group of non-offenders with respect to social anxiety, under-assertiveness, or self-esteem. Ward, McCormack, and Hudson (1997) described “a major conclusion” of their study of sexual and non-sexual offenders that intimacy deficits in sexual offenders were shared by non-sexual violent offenders and “therefore, were not specific to sexual offenders” (p. 72).
Coming closer, Marsa et al. (2004), in their study of child sex offenders, violent non-sexual offenders, non-violent offenders, and non-offenders, found that a secure adult attachment style was four times less common in the child sex offender group than in any of the other three groups, but found no other significant support for the relationship of insecure or disrupted attachment in the development of sexually abusive behavior, other than ongoing emotional difficulties, emotional loneliness, and antisocial behavior.
Thus far, although it may well be that attachment deficits or styles are significantly related to the onset of sexually abusive behavior in juveniles or adults, little evidence is available to support what must, in the meantime, remain little more than a theory. The assertion of Ward, Hudson, Marshall and Siegert (1995) that different types of sexual offenders will be characterized by different types of adult attachment styles, has not been borne out, nor has any hypothesized difference between sexual and non-sexual offenders been established. Similarly, there is no evidence to support Burk and Burkhart’s or Smallbone and Dadds’ more recent hypotheses that sexual offenders display characteristics of disorganized attachment. In children, there is a relatively low incidence of this attachment pattern, or in the adult equivalent of unresolved/disorganized attachment issues.
Although Proeve (2003) writes that “attachment theory offers a useful perspective for understanding the problems experienced by sexual offenders with intimate relationships” (p. 248), he concludes that “there does not appear to be consistency in findings regarding attachment styles and sexual offending” (p. 258).
As noted in part I of this article, Smallbone and Dadds (2000, p. 13) have similarly written that it is premature “to conclude that prevention and treatment of sexual aggression should adopt attachment concepts and respond to their implications; there is insufficient evidence to support such a broad conclusion,” although they continue to suggest that “notwithstanding these limitations, these results indicate that childhood attachment may play some role in the development of coercive sexual behavior” (2000, p. 13).
Attachment and Juvenile Sexual Offenders
Of the limited work conducted in relating attachment disruptions and disorders to sexual offending, it is clear almost all has been completed with adult sexual offenders, with very little attention paid to juvenile sexual offenders. In fact, the focus of the work, including that of Marshall, Burk and Burkhart, and Smallbone is largely geared towards adult child molesters, rather than adult rapists or juvenile sexual offenders. Further, Smallbone (Smallbone, submitted; Smallbone, in press; Smallbone & Wortley, in press) asserts that contrary to statistics frequently cited, many adult sexual offenders do not engage in sexually coercive behavior until they reach adulthood, thus perhaps skirting the issue of adolescent sexual offenders. This makes the relationship between attachment difficulties of childhood and later adult sexual offending even more tenuous, as adult sexual offenders pass through adolescence en route to adulthood. It makes more sense to hypothesize that those adults who do engage in sexually abusive behavior are engaged in, at least, a variant of sexually troubling behavior as adolescents, thus providing a pathway that directly connects attachment difficulties in childhood to the onset of sexually abusive behavior in adults. Smallbone’s contention that many adult sexual offenders do not engage in sexually abusive behavior until age 30 or later is a further confounding variable.
Furthermore, Smallbone’s assertion that sexual offenders confuse and overlap attachment, parenting, and sexual behavioral systems has far less relevance for adolescents than for adults, as for most adolescents parenting/caregiving systems are presumably not particularly active, although sexual systems may be. Burk and Burkhart’s model, focusing on the development of sexually abusive behavior to re-establish a state of emotional and cognitive organization may well be relevant for adolescents but also focuses on disorganized attachment (as does Smallbone), which is not commonly found in adolescents, at least as described and measured by the Adult Attachment Interview (AAI) and is considered a relatively rare attachment pattern. There is no evidence that supports a link between disorganized or delayed attachment in adolescents and sexually abusive behavior, despite being an interesting idea.
Coming directly from the literature of attachment theory, Tracy, Shaver, Albino, and Cooper’s (2003) contention that childhood attachment style is directly related to adolescent sexual behavior may be far more to the point. They assert that childhood attachment styles are directly related to the development of sexual and romantic behavioral patterns in adolescents and young adults, and sexual attitudes and the use of sex as a tool to get emotional needs met are related to earlier patterns of attachment. Insecure attachments lead to the use of sexual relationships as a means for gaining control and meeting personal needs, rather than sexual and romantic relationships that would be considered emotionally healthy and satisfying, bearing a slight similarity to the Burk and Burkhart model regarding emotional and cognitive equilibrium. In fact, Smallbone’s ideas about the overlap between the sexual, parenting, and attachment behavioral systems appear to be drawn from Shaver’s earlier work in which he and his colleagues assert that romantic love is an amalgam of those three behavioral systems (Shaver, Hazan, & Bradshaw, 1988).
Belsky argues still further (Belsky, 1999; Belsky ,Steinberg & Draper, 1991) that early attachment experiences are directly related to the onset of pubertal experiences and sexual behaviors, serving as an evolutionarily-prompted strategy. In this model, the insecurely attached child grows into late childhood/early adolescence experiencing resources as scarce or unpredictable, other people as untrustworthy, and relationships as transient. This child, he suggests, enters into both puberty and sexual behaviors at an earlier time as a biologically-driven imperative in which sexual promiscuity is a means to ensure reproductive certainty. In this conceptualization, emotional satisfaction is not the goal, but sex is a means for reaching out into the world to ensure reproductive survival under uncertain or adverse conditions. Here, driven by evolution, sexual behavior in young/pre-adolescents is one result of an attachment pattern (secure vs insecure), presumably in combination with other personal and environmental factors.
However, there is little more we can say at this time about the relationship between attachment and juvenile sexual offending, other than it is likely to serve as a predisposing factor and link to the onset of many troubled and troubling behaviors in juvenile sexual offenders, including serving as a factor in the development of sexually abusive behavior.
Sexual Offending as a Disorder of Attachment
Unlike other crimes that have victims, in many cases sexual offending, and perhaps especially among juveniles, appears to be directly connected to having a relationship of some kind, rather than simply victimizing another individual. For some sexual offenders, particularly as described by Marshall and colleagues, the sexual abuse itself constitutes a form of social/intimate relationship and moves sexual offenders closer to relationships, however distorted, in which the intention is to engage with someone in a social relationship or derive some social benefit. As shown in Figure 1, in a simple typology one can discern four forms of sexual abuse:
1. Unattached, in which the victim is a pure object of sexual desire in which no relationship is implied or sought.
2. Sadistic, in which the victim is forced to engage in a very direct and distorted relationship with the perpetrator.
3. Victim as Object of Social Connection, in which the victim represents a connection to larger social relationships or social competencies.
4. Victim as Object of Affection, in which there is a clear intention to engage in a genuine relationship with the victim which implies some level of intimacy on the part of the perpetrator.
As shown in Figure 1, if these categories of sexual abuse are laid along a continuum, at one end we see no attachment at all, as in the non-relationship of psychopathy, whereas each of the other three forms represent disturbances in attachment, but not detachment (unattachment). Dividing forms 1 and 2 (unattached and sadistic) and forms 3 and 4 (victim as social connection and victim as object of affection) is an arrow pointing towards disconnected, remorseless, affectless, and detached psychopathy at one end, and connected, socially incompetent, socially inappropriate, but attached pathology at the other. In the case of forms 1 and 2, treatment may be very different as both lie at the psychopathic end of the scale. The dividing line itself separates treatment into two forms: containment and harm reduction at one end, and containment and rehabilitation at the other.
The Treatment of Attachment Problems: “Attachment-Informed” Therapy
There really is no mainstream attachment therapy in the sense of a specific model of therapy or set of therapeutic techniques directly attached to a school of therapy; the closest thing we have is a lens of “attachment theory” through which we can look at and come to understand clinical populations.
The form of therapy that is known as “attachment therapy” is sometimes disparaged by attachment theoreticians and those who incorporate an attachment framework into their therapeutic work, and is not considered substantially relevant to the work of attachment theory. One of the problems with “attachment therapy,” in addition to some controversial or unproven techniques (such as holding therapy, corrective parenting, and corrective attachment therapy), is the view that virtually all problems of childhood are the products of attachment deficits and can be resolved through the application of prescribed techniques, thus rendering the model incapable of differential diagnosis or treatment. Instead, rather than practicing “attachment therapy,” based on a specified model or set of techniques, clinicians recognizing and addressing attachment difficulties use an attachment-informed framework by which psychodynamic and cognitive interventions can be applied with attachment as a target of treatment.
In terms of treatment, attachment-informed therapy doesn’t stand out as a “model.” For the most part, it represents a set of ideas and practices, offering a lens through which client difficulties may be seen and a framework and perspective that individual clinicians bring to bear in their practices. This fits in well with a model of eclectic or integrated therapy in which the clinician is able to easily and freely switch gears in terms of both technique and even perceptual frameworks, as required by a pantheoretical model of treatment. A main emphasis of attachment-informed therapy is to understand insecure attachment and obstructions to secure attachment, and assess whether any of these obstacles can be removed, perhaps through individual, family, or group therapy, or even through medication. A second emphasis is to re-activate deactivated attachment systems. A third is to help the individual re-organize attachment systems, and a fourth is to eliminate ambiguity and incoherence from attachment narratives, or the expression of internal working models. Another is, of course, to improve self esteem, as we discuss the development of the secure personality.
Not surprisingly, as attachment theory is a psychodynamic, interactional model, the therapeutic relationship comes squarely back into the foreground in attachment-informed therapy. Rather than teaching or discussing concepts of attachment in a didactic, psychoeducational, or even cognitive-behavioral mode, the therapist uses interactional techniques imparted through the therapeutic alliance. It is through this relationship, as well as other techniques and practices of treatment, that a treatment environment and relationship is established that can help re-build attachment and develop the sort of coherence and right-left brain integration thought to be reflective of a secure internal working model. Ultimately, the emphasis in an attachment therapy is on the development of an understanding, supportive, and caring relationship, marked by attunement between the therapist and the patient.
Treatment Applied Through the Attachment Framework
For the therapist focusing on building attachment, coherence, and a more secure internal working model,
interaction and behaviors are understood through an attachment lens. Behaviors are re-framed and seen as attachment seeking rather than attention seeking, for instance, and healthy, non-pathological behaviors are recognized as possible only in light of a sense of security, in which the exploration behaviors we wish to stimulate in our clients are activated only when attachment behaviors, or their neurological counterparts, sparked by insecurity, anxiety, and fear are deactivated. Behavioral episodes that otherwise appear irrational are recognized and understood as emotional, and even neurobiological, episodes fueled by poorly processed and unintegrated perceptions in the patient. Seen through this attachment lens, the development of a “secure base” injected into the internal working model of the juvenile sexual offender is paramount. Simply put, built on an attachment framework, in our clients the goals of treatment include developing:
$ a sense of experienced security (secure base) from which to explore and grow
$ confidence (security) in and connection to important figures
$ a secure and coherent sense of self
$ balance in the use of affective and cognitive problem solving strategies
$ cooperative and non-coercive strategies
$ perspective taking and the unlocking of empathy for others
$ the capacity to tolerate and regulate frustration and disappointment
This brief list will not be new to anyone who provides a form of holistic, integrated treatment to troubled young people, built on treating the juvenile as a “whole” person rather than a “sexual offender.” The difference may simply be in the application of a framework that recognizes many of these features as attachment-dependent, based upon earlier experiences that have limited the capacity of the youth to engage in healthy and satisfying social relationships. This framework, coupled with an understanding of the neurobiology of attachment, allows the practitioner to recognize and understand behavioral, social, and even sexual difficulties as the sequelae, at least in part, of earlier attachment experiences. The goal of this treatment, in the final analysis, cannot be the re-establishment of childhood attachment experiences with a primary caregiver (as that is impossible), but is instead the rehabilitation of the internal working model, providing for the juvenile the capacity for self regulation and a sense of self efficacy and security that will serve as the basis for all future experiences of self and others, and hence relationships and behaviors in the world.
This requires an understanding, definition, and application of the concept of attachment that makes it relevant to adolescents, rather than simply applying ideas about attachment in infancy or early childhood, or even the attachment needs of adults. Through an attachment perspective, we recognize that in treatment:
$ there is a need for empathic attunement to the child,
$ the child must see the value of self in the minds of other people,
$ the child must discover important others as capable and competent,
$ seemingly irrational behaviors can be understood as vacillations between variants of primitive attachment and exploratory behaviors, or variants of insecure or disorganized attachment strategies,
$ the brain is plastic and can change, and neural pathways reflect experience,
$ change requires giving up prior adaptive strategies,
$ change comes slowly,
$ “healthy” attachment requires a secure base, and
$ the development of a secure base results from life experience
“Attachment” is not well understood in older children and adolescents, and attachment needs and behaviors change during the course of human development. With this in mind, there is little concrete (empirical) support for any connection between attachment difficulties and sexual offending, and we have little real understanding of what role early and on-going attachment experiences have on the development of sexually abusive behavior in children and adolescents.
However, we do recognize that attachment experiences are quite likely to have some definite effect on the development of sexually abusive behaviors, just as attachment experiences will have an impact of other aspects of human behavior and relationships. In this regard, attachment is not a unidimensional “trait,” but takes its shape over time with respect to cognition, affect, behavior, and relationships in conjunction with other life experiences and developing personality traits. In this regard, it seems unlikely that attachment alone has a substantial impact of juvenile sexual offending. Nevertheless, the ability to feel connected to others in a mutually healthy manner seems critical to behavior that is both personally rewarding and safe for other people.
Just as patterns of attachment and experiences of the world are neurologically “hard wired” into the brain, and this hard wiring in turn establishes a pattern for further behaviors and social interactions, so too is the personality embodied in the mental map of the brain. It is this internal working model, built upon a foundation of attachment, and which is continually re-shaped by on-going social connections, that is the target of treatment in both young people and adults as well. This is not so different than treatment found in more traditional models of psychotherapy, but certainly is borne out in an age of brain-based learning. It is the “working” part of the internal working model that we’re looking to change in terms of internal representations. How we understand the formation of the internal working model, and therefore how we understand and approach change and select treatment interventions, both cognitive and relational, will be key in both understanding and treating juvenile sexual offending.
On a final note, if we recognize that juvenile sexual offenders are a heterogeneous, rather than a homogenous group then we recognize that there are different pathways to sexual offenses. This reminds us, then, that although attachment difficulties may be a serious concern for many, and perhaps on some level for all, it is certainly not the be all and end all for understanding the nature of or treating juvenile sexual offending.
Citations & References
American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (4th ed. text revision). Washington, DC: Author.
Belsky, J. (1999). Modern evolutionary theory and patterns of attachment. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical application (pp.141-161). New York: Guilford.
Belsky, J., Steinberg, L., & Draper, P. (1991). Childhood experience, interpersonal development, and reproductive strategy: An evolutionary theory of socialization. Child Development, 62, 647-670.
Brisch, K. H. (1999). Treating attachment disorders: From theory to therapy. New York: Guilford.
Burk, L. R., & Burkhart, B. R. (2003). Disorganized attachment as a diathesis for sexual deviance: Developmental experience and the motivation for sexual offending. Aggression and Violent Behavior, 8 (487-511).
Levy, T. M. (2000). Handbook of Attachment Interventions. San Diego: CA: Academic Press.
Levy, T. M., & Orlans, M. (1998). Attachment, trauma, and healing: Understanding and treating attachment disorder in children and families. Washington DC: CWLA Press.
Marsa, F., O'Reilly, G., Carr, A., Murphy, P., O'Sullivan, M., Cotter, A., & Hevey, D. (2004). Attachment Styles and Psychological Profiles of Child Sex Offenders in Ireland. Journal of Interpersonal Violence, 19, 228-251.
Marshall, W. L., Barbaree, H. E., & Fernandez, Y. M. (1995). Some aspects of social competence in sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 7, 113-127.
Marshall, W. L., & Eccles, A. (1993). Pavlovian conditioning processes in adolescent sex offenders. In H. E. Barbaree, W. L. Marshall, & S. M. Hudson (Eds.), The juvenile sex offender (pp. 118-142). New York: Guilford Press.
Marshall, W. L., Hudson, S. M., & Hodkinson, S. (1993). The importance of attachment bonds in the development of juvenile sexual offending. In H. E. Barbaree, H. E., W. L. Marshall, & S. M. Hudson (Eds.), The juvenile sexual offender (pp. 164-181). New York: Guilford.
Marshall, W. L, & Marshall, L. E. (2000). The origins of sexual offending. Trauma, Violence, & Abuse, 1, 250-263.
Marshall, W. L., Serran, G. A., & Cortoni, F. A. (2000). Childhood attachments, sexual abuse, and their relationship to adult coping in child molesters. Sexual Abuse: A Journal of Research and Treatment, 12, 17-26
Marshall, W. L., & Mazzucco, A. (1995). Self-esteem and parental attachments in child molesters. Sexual Abuse: A Journal of Research and Treatment, 7, 279-285.
Mulloy, R., & Marshall, W. L. (1999). Social functioning. In W. L. Marshall, D. Anderson & Y. Fernandez (Eds.), Cognitive-behavioral treatment of sexual offenders (pp. 93-109). Chichester, England: John Wiley.
Proeve, M. J. (2003). Responsivity factors in sexual offender treatment. In T. Ward., D. R. Laws., & S. M. Hudson (Eds.), Sexual devianbce: Issues and controversies (pp. 244-261). Thousand Oaks, CA: Sage.
Shaver, P. R., Hazan, C., & Bradshaw, D. (1988). Love as attachment: The integration of three behavioral systems. In R. J. Sternberg & M. Barnes (Eds.), The psychology of love (pp. 68-99). New Haven, CT: Yale University Press.
Smallbone, S.W. (in press). Social and psychological factors in the development of delinquency and sexual deviance. In H.E. Barbaree, & W.L. Marshall (Eds.). The juvenile sex offender (2nd ed.) New York: Guilford Press.
Smallbone, S.W. (submitted). An attachment-theoretical revision of Marshall and Barbaree’’s Integrated Theory of the Etiology of Sexual Offending. In W.L. Marshall, Y.M. Fernandez, and L.E. Marshall (Eds.). Controversies and Issues in sexual offender management. New York: John Wiley and Sons.
Smallbone, S., & Dadds, M. (1998). Childhood attachment and adult attachment in incarcerated adult male sex offenders. Journal of Interpersonal Violence, 13, 555-573.
Smallbone, S. W., & Dadds, M. R. (2000). Attachment and coercive sexual behavior. Sexual Abuse: A Journal of Research and Treatment, 12, 3-15.
Smallbone, S.W. & Dadds, M.R. (2001). Further evidence for a relationship between attachment insecurity and coercive sexual behavior. Journal of Interpersonal Violence, 16, 22-35.
Smallbone, S. W., & McCabe, B. (2003). Child attachment, childhood sexual abuse, and onset of masturbation among adult sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 15, 1-9.
Smallbone, S.W., & Wortley, R.K. (in press). Onset, persistence and versatility of offending among adult males convicted of sexual offenses against children. Sexual Abuse: A Journal of Research and Treatment.
Tracy, J. L., Shaver, P. R., Albino, A. W., & Cooper, M. L. (2003). Attachment styles and adolescent sexuality. In P. Florsheim (Ed.), Adolescent romance and sexual behavior: Theory, research, and practical implications (pp. 137-159). Mahwah, NJ: Erlbaum.
U.S. Department of Health and Human Services (1999). Mental health: A report of the Surgeon General. Rockville, MD: U.S. Department of Health and Human Services.
Ward, T., Hudson, S., Marshall, W., & Siegert, R. (1995). Attachment style and intimacy deficits in sexual offenders: A theoretical framework. Sexual Abuse: A Journal of Research and Treatment, 7, 317-335.
Ward, T., McCormack, J., & Hudson, S. M. (1997). Sexual offenders’ perceptions of their intimate relationships. Sexual Abuse: A Journal of Research and Treatment, 9, 57-74.
Zeanah, C. H. Jr, Mammen, O. K., & Lieberman, A. F. (1993). Disorders of attachment. In C. H Zeanah, Jr. (Ed.), Handbook of infant mental health (pp. 332-349). New York: Guilford.