Running head: STALKING AND DOMESTIC VIOLENCE

 

 

 

 

Assessing the Link Between Stalking and Domestic Violence

 

Manuscript under review, please do not cite

 

 

 

 

Kevin S. Douglas1

Simon Fraser University

 

Donald G. Dutton2

University of British Columbia

 

 

 

Abstract

Stalking may be defined as repeated following, communicating, and contacting a person in a threatening manner that causes the person to fear, on a reasonable basis, for his or her safety.  Stalking is a recent legal construct, and social scientific research on stalking is in an early stage.  Given that the most common victim of stalking is an ex-intimate partner, there may be an association between stalking and domestic violence.  This paper evaluates this potential link.  Specifically, the literature on stalking is reviewed by means of comparing it to existing literature on typologies of domestically violent persons.  It is proposed that most stalkers who target ex-intimate partners are characterologically similar to a type of batterer labeled “borderline/cyclical.”  Both domestic stalkers and borderline/cyclical batterers possess traits of Cluster B personality disorders. These traits include emotional volatility, attachment dysfunction, primitive defenses, weak ego strength, jealousy, anger, substance abuse, and early childhood trauma.  Further, both groups have been observed to react with rage to perceived or actual rejection or abandonment.  It is suggested that applying what is known about borderline/cyclical batterers to stalkers may aid in the investigation of this phenomenon.  Implications for research are discussed.

 

Assessing the Link Between Stalking and Domestic Violence

 

            As Meloy (1998) states, “stalking is an old behavior, but a new crime” (p. xix).  This statement indicates that while stalking has likely always been a part of human behaviour, it is only recently that it has received legal recognition as an offensive behaviour.  In fact, the first anti-stalking law in North American was passed less than a decade ago, in California (Penal Code Section 646.9).  Most states followed suit shortly thereafter (Saunders, 1998).  In Canada, anti-stalking legislation, termed criminal harassment, was promulgated in 1993 as section 264 of the Canadian Criminal Code.

            Stalking has been defined similarly in these legal settings. The classification of a stalker generally requires repeated direct or indirect acts of following, communicating, besetting, watching, contacting, and threatening in such a way as to cause the victim to fear, on reasonable grounds, for his or her safety.  Although other terms have been used to denote this behaviour, such as “obsessional following” (Meloy, 1997, 1998; Meloy & Gothard, 1995), and “criminal harassment” (section 264 of the Code), the term “stalking” will be used throughout this paper.

            Stalking is often accompanied by physical violence (Meloy, 1998) and as such, it is recognized that stalkers can seriously effect their victims (Hall, 1998) Further, stalkers’ largest victim group is female ex-intimate partners (Meloy, 1998), establishing, prima facie, an association between stalking and domestic violence.  Although the extant research on stalking is in a very young stage, and little is truly known about it, the research on domestic violence is much more developed.  As such, given this tentative connection between stalking and domestic violence, it may be that research and theory from the domestic violence corpus could be used to contribute to the study of stalking. 

            The purpose of this paper is to evaluate the potential link between stalking and domestic violence.  The literature on stalking will be reviewed with the goal of abstracting patterns of stalking characteristics and juxtaposing these on well-defined typologies of domestically violent people.  The rationale of this approach is to understand the functioning, behavior, emotional constitution, and psychological profile of stalkers.  The goal is to enhance the existing information base concerning the phenomenon of stalking and to suggest testable hypotheses.

           

Nature and Prevalence of Stalking

           

            There are several research projects that have assessed the prevalence of stalking. Kong (1996), in a report for Statistics Canada, reported that in 1994 and 1995 there were 7462 incidents of criminal harassment reported to police.  The survey employed the Revised Uniform Crime Reporting Survey (UCRS).  This method gathers information from participating police agencies.  Close to half (43%) of all police agencies in Canada were involved.  Hence, although this was a large-scale survey, it was neither comprehensive nor random.

            If 100% of police agencies were involved in reporting their statistics for the UCRS, extrapolating from the statistic of 43%, (assuming no selection bias was present for reporting agencies that would serve to differentiate them from non-reporting ones), then the figure of 7462 would be 17353.  Based on a population of approximately 30 million, this represents a prevalence of .06, or 6 people per 1000 people.  If the adult population (approximately 20 million) is used in this calculation, then the Canadian annual prevalence is approximately .09, or 9 people per 1000 (about 1%).

            In the National Crime Victimization Survey (NCVS), sponsored by the U.S. Department of Justice (1997), 8000 men and 8000 women were randomly selected to participate in telephone interviews concerning victimization.  This report found that the U.S. lifetime incidence of stalking is 8.1% for women, and 2.2% for men.  The extrapolated figures, offered by the NCVS, for individuals who have been stalked at some point in their lives are 2.04 million American women and .82 million men. Past year estimates were placed at 1.01 million women and .37 million men. In Canada, based upon the above estimates, approximately 200,000 adults are stalked on an annual basis. Clearly, the scope of the problem of stalking is vast.

            Coleman (1997), in a survey of 141 female undergraduate students, determined that 13 (9.2%) had been stalked solely in the context of a (past) relationship.  A much higher incidence was reported by Fremouw, Westrup, and Pennypacker (1997), who reported that 30% of 319 female and 17% of 275 male undergraduates had been subjected to stalking.[1]  Similarly, Spitzberg et al. (1998) found that 27% of their respondents “self-defined” themselves as stalking victims.  A survey of counseling centres in the United States revealed that 5.6% of counselors (10/178) had been stalked.  Dempster, Hart, and Boer (1997), in a random sample of 197 federally incarcerated offenders in Canada, found that just under 5% had a history of stalking.  Finally, in a stratified random sample of 721 teachers in British Columbia, Lyon and Douglas (in press) noted a 5.1% rate of stalking victimization, based on definitions derived from Canadian law; most of this stalking was perpetrated by students.

Taken together, the most stable estimate of stalking victimization, for all individuals, seems to be about 5%.  It would also appear that this form of victimization is more common among women than men.  Yet, some studies (Fremouw et al., 1997; Spitzberg et al., 1998) have observed much higher rates.  This finding may be due to the nature of the sample (undergraduate students) or perhaps to differences in definitions of stalking. Nonetheless, a sizeable number of people are stalked.  Given the harmful sequelae of stalking (see, for example, Hall, 1998; Spitzberg et al., 1998), it would seem reasonable to characterize the phenomenon of stalking as a pressing concern.

The Nature of Stalking

            Studies on perpetrators have tended to focus on demographic, criminological, and psychiatric aspects of stalkers and, to a lesser extent, of victims.  The use of threatened as well as actual violence in the stalking process has also received some attention, as has the nature of the relationship between the accused stalker and the victim.  These areas will be reviewed in turn. 

Table 1 presents a summary of some of the research on stalkers and displays some select features of these studies.  As shown, most studies that have addressed this issue have determined that stalkers tend to be in their mid-30s.  On average, research tends to suggest that the vast majority of stalkers are male and that the majority of victims are female.  A high proportion of stalkers target ex-intimate partners, and a high proportion of stalkers are violent to their victims.  Also, psychopathology appears to be quite common among groups of stalkers.

                                               

Insert Table 1 about here

                                               

Demographic information.  A high proportion of stalkers are Caucasian (Burgess et al., 1997; Kienlen et al., 1997; Schwartz-Watts & Morgan, 1998).  In fact, the proportion of Caucasian people in stalking groups may be disproportionately higher than the proportion of Caucasian people in other criminal groups (Harmon et al., 1995).  Stalkers have been found, in some studies, to be fairly well educated (Harmon et al., 1995; Kienlen et al., 1997; Lyon, 1998; Schwartz-Watts, Morgan, & Barnes, 1997); particularly in comparison to persons charged with or convicted of other criminal offences (Harmon et al., 1995; Lyon, 1998; Meloy & Gothard, 1995; Schwartz-Watts et al., 1997).

Most studies have found that stalkers tend to be single, separated, or divorced – more so than other offenders (Harmon et al., 1995; Schwartz-Watts & Morgan, 1998; Schwartz-Watts et al., 1997).  Studies that fail to observe this pattern tend to include offenders that have been charged with domestic assault and who are still involved with their partners (Burgess et al., 1997).  Stalkers are often unemployed at the time of their stalking behaviors and may be more likely than other offenders to be unemployed (Kienlen et al., 1997). (An exception to this finding was the work of Lyon [1998], who found that 80% of 54 stalkers were employed, and were not less likely to be unemployed in comparison to other offenders).  Stalking is a time-consuming activity, and could easily interfere with maintaining steady employment.  Some researchers, however, have failed to observe differences between stalkers and comparison groups on demographic information (Burgess et al., 1997; Coleman, 1997), suggesting that one common “profile” of stalkers has not been identified.

            Criminological information.  This section is meant to summarize the extent to which stalkers tend to have histories of crime and violence.  Most studies have found that a large proportion of stalkers (at least those involved with the criminal justice system) tend to have criminal histories, and that these often involve crimes of violence (British Columbia Attorney General, 1995; Department of Justice Canada, 1996; Harmon et al., 1995; Kienlen et al., 1997; Lyon, 1998; Manitoba Law Reform Commission, 1997; Meloy & Gothard, 1995; Mullen & PathJ, 1994). 

Psychiatric information.  Unfortunately, most of the large-scale studies on stalking that are not affected by the selection biases associated with forensic or criminal samples have not assessed the prevalence of mental and personality disorders within groups of stalkers.  What estimates of psychopathology amongst stalkers do exist derive from samples that may tend to over-estimate psychopathology.  For example, several studies have focused upon samples of stalkers who have been referred from court for psychiatric assessments of criminal responsibility, fitness to stand trial, or pre-sentence reports.  The reader is again referred to Table 1 for summary information on the prevalence of mental disorder within groups of stalkers.

            As can be seen from Table 1, of studies that have been able to assess psychopathology, although there is a range of estimates of such disorder, it is a reasonable estimate that a large proportion of stalkers have some sort of Axis I or Axis II disorder.  Concerning substance abuse, estimates range from 2% (Harmon et al., 1995) to 70% (Meloy & Gothard, 1995).  Some studies have found that stalkers are more likely than comparison groups of other offenders to abuse substances (Burgess et al., 1997).  However, in other studies, no such differences have been observed (Lyon, 1998; Meloy & Gothard, 1995; Sandberg, McNiel, & Binder, 1998; Schwartz-Watts & Morgan, 1998).  In at least one study, stalkers have been found to be significantly less likely to abuse substances (Schwartz-Watts et al., 1997).

 

Axis I Disorders

            Concerning Axis I disorders, clinical lore has suggested that many, if not most stalkers were erotomanic (Meloy, 1998), or, in official parlance, had delusional disorder, erotomanic subtype.  Several studies, despite drawing from psychiatric or forensic settings, have failed to observe a high percent of persons with the delusional disorder erotomania.  The prevalence of delusional disorder appears to be lower than was previously believed.  For instance, Harmon et al. (1995), in a sample of 48 stalkers who were referred from court to a forensic psychiatric facility for assessment, found that 14 (29%) of their sample were diagnosed with erotomania.  Mullen and PathJ (1994), in a non-random sample of 14 stalkers referred for private psychiatric evaluations, reported that 36% (n = 5 of 14) of their sample were erotomanic.  Most other studies have found a lower prevalence.  Kienlen et al. (1997), in another sample of stalkers referred for psychiatric assessment, found that only 1 of 25 (4%) stalkers had erotomania.  Meloy and Gothard (1995) compared 20 stalkers with 30 mentally disordered offenders and found that only 2 (10%) of the stalkers were diagnosed with erotomania.  In a sample of civil psychiatric patients who stalked staff after discharge, Sandberg et al. (1998) noted a 12% (n = 2 of 17) prevalence of erotomania.  Schwartz-Watts et al. (1997) found that 2 of 18 (11%) stalkers had delusional disorder.[2]  Finally, Zona, Sharma, and Lane (1993) found a similar base rate of erotomania — 7 of 74 (9.5%) stalkers were diagnosed with erotomania.

            In terms of other Axis I disorders, studies of stalkers who are assessed in forensic psychiatric settings reveal, not surprisingly, a fairly broad representation of Axis I disorders.  Again, it is important to point out that larger scale community based samples have not been able to estimate mental disorder.  Several studies have found that schizophrenia is fairly common (Harmon et al., 1995; Kienlen et al., 1997; Mullen & PathJ, 1994; Sandberg et al., 1998).  So too are mood disorders such as dysthymia, major depression, or bipolar disorder (Kienlen et al., 1997; Meloy & Gothard, 1995; Mullen & PathJ, 1994; Sandberg et al., 1998; Schwartz-Watts et al., 1997).  As well, adjustment disorders and other Axis I disorders (i.e., psychotic disorder not otherwise specified) seem to occur in these settings with some regularity (Harmon et al., 1995; Meloy & Gothard, 1995).  One study that compared stalkers to non-stalking mentally disordered offenders observed a slight tendency for schizophrenia to occur less often among the stalkers (Meloy & Gothard, 1995).  In another study, there was no difference in occurrence of Axis I disorders between 18 court-referred stalkers attending a forensic psychiatric assessment unit and 18 randomly matched controls at the same institute (Schwartz-Watts et al., 1997).

            Two studies that did not select from court-referred psychiatric settings, although not including diagnostic information per se, collected information related to mental health.  In a study by Canada’s Department of Justice (1996), 14% of 601 persons charged under Canada’s stalking law (criminal harassment) were described in the police reports as having “mental problems.”  In another study by Lyon (1998), 54 persons charged with criminal harassment who were remanded at the Surrey Pretrial Services Centre in British Columbia awaiting first court appearances were compared to three other groups of offenders.  Lyon was able to collect data from the Mental Health Screening Programme (Mental Health, Law, and Policy Institute, 1997), including an expanded version of the Brief Psychiatric Rating Scale (Overall & Gorman, 1962).  The distribution of BPRS scores was positively skewed, with 75% of the group falling between 24 (the lowest score possible) and 33.  The median score on all but two items (anxiety and depression) was one.  Not a single subject scored greater than the minimum on the “bizarre behaviour” item. 

            However, stalkers did have more elevated scores in comparison with other offenders on several items, including anxiety, depression, and grandiosity.  Similarly, total scores were significantly, though slightly, higher in the stalking group compared to other offenders.  These data suggest that there was not much psychopathology present in this sample of stalkers, but that stalkers may have slightly more symptoms of psychological maladjustment than other offenders.

            It is somewhat difficult to discern a pattern of results across studies concerning Axis I disorders.  It is reasonable to state that stalkers tend to abuse substances, although this may not be to any greater degree than other offenders.  Although it appears that stalkers may commonly display other Axis I disorders, most of the studies that have investigated this have drawn from forensic psychiatric settings.  Studies with less of a selection bias (i.e., Lyon, 1998) show that stalkers may have slightly more symptoms of psychopathology than other offenders, but that the absolute incidence and level of severity is not high.  The types of Axis I disorders that appear to predominate, in order of frequency, are substance abuse and dependence, mood disorders (depression and dysphoria), and psychotic disorders.

 

Axis II Disorders

            In terms of Axis II disorders, and personality disorders in particular, somewhat of a clearer clinical picture emerges.  Again, it is important to note that diagnostic information is not present in larger scale community studies, and hence the estimates of Axis II disorders derived from smaller, selected samples may be inflated.  As shown in Table 1, several studies have evaluated the role of personality disorders in stalking.  Estimates of prevalence range from 19% to 100%, with an average in the 50% to 60% range.  While this is a high prevalence rate, especially in comparison to epidemiological community studies (Bland & Orn, 1986; Robins & Reiger, 1991), it should be pointed out that compared to offender samples, this is actually low.  That is, in samples of federally incarcerated offenders, somewhere between 70% and 80% of offenders may have personality disorders, particularly antisocial personality disorder (APD) (Hare, 1991, 1996).

            What is interesting among samples of stalkers is the types of personality disorders and characteristics that are manifest.  Typically, Cluster B disorders and traits predominate, apart from APD.  A possible explanation for this finding lies in conceptualizing stalking as a dysfunctional or abnormal attachment behaviour, and the construal of APD as a disorder of detachment (Meloy, 1998).  As such, persons with APD tend not to become invested in others, and would rather dismiss a rejecting partner than obsess over and stalk them.  More will be said about these matters in a subsequent part of this paper.

            Studies that have been able to evaluate personality disorder include Harmon et al. (1995), Kienlen et al. (1997), Meloy and Gothard (1995), Mullen and PathJ (1994), Romans et al. (1996), and Sandberg et al. (1998).  Some of these studies are more informative than others.  In a few of them, little more than descriptive information is provided.  As such, in Sandberg et al. (1998), it is reported that 8 of 17 (47%) of stalkers from a civil psychiatric sample were diagnosed with personality disorder.  This compared to 35 of 326 (11%) comparison inpatients, a significant difference.  In Romans et al. (1996), the researchers sampled from counselors across the United States, and reported that of the 10 who had been stalked, 6 indicated that among the stalkers there were “presenting problems related to Axis II personality disorders” (p. 596).  Mullen and PathJ (1994) indicated that all of their 14 stalkers had “pre-existing personality disorders in which intense self consciousness and sensitive ideas of reference were prominent” (p. 471). 

            Other studies provide somewhat more information.  Meloy and Gothard (1995) reported that of their 20 stalkers, all court-referred for psychiatric examination, 17 had personality disorders.  Stalkers were significantly less likely to be diagnosed with APD in comparison to non-stalking mentally disordered offenders, but significantly more likely to be diagnosed with other personality disorders.  Two of the 20 stalkers (10%) had APD.  The most common Axis II diagnosis was personality disorder not otherwise specified (NOS), followed by developmental disorder NOS (n = 2, 10%), borderline (n = 2; 10%), avoidant (n = 1; 5%), paranoid (n = 1; 5%), and schizoid (n = 1; 5%) personality disorders.  Meloy and Gothard reported that of the personality disorder NOS diagnoses, most involved traits of Cluster B disorders, such as histrionic, narcissistic, anti-social, and borderline.  However, traits of schizoid, passive-aggressive, and obsessive-compulsive disorders also were present.

            Harmon et al. (1995) reported that 9 of 48 stalkers (19%) had personality disorders.  Of these, six were diagnosed with personality disorder NOS, and one each was diagnosed with borderline, schizotypal, and narcissistic personality disorders.  In the study by Kienlen et al. (1997), the researchers divided their sample into stalkers receiving psychotic diagnoses and those not receiving such diagnoses.  None of the psychotic stalkers were diagnosed with personality disorders, although the authors suggest that this may have stemmed from “a systematic bias among the diagnosticians” (p. 327).  Of the 17 non-psychotic stalkers, 11 were personality disordered.  Of these, eight received Cluster B diagnoses (or personality disorder NOS with Cluster B traits) and three were diagnosed with dependent personality disorder.  Of the Cluster B diagnoses, three were antisocial, two were borderline, two were narcissistic, and two were NOS with narcissistic, antisocial, and dependent traits.

In summary, although the evidence is not overwhelmingly clear, it appears that, at least within forensic psychiatric settings, approximately half of stalkers on average are personality disordered, and that Cluster B disorders and traits are the most common types of personality disorders.  Also present are dependent, avoidant, and schizoid disorders, and to a lesser extent paranoid and schizotypal disorders.  The predominance of Cluster B disorders will form an important part of the assessment of the overlap between stalkers and domestically abusive people, discussed later.

            Nature of stalking behaviours.  What has research identified to be common aspects of stalking behaviour?  There are commonalties across studies.  Stalking can persist for many months, and often years (Department of Justice Canada, 1996; Meloy & Gothard, 1995; Mullen & PathJ, 1994; Zona et al., 1993).  Common stalking behaviours include telephone calls, visiting work places or residences, besetting, letter writing, following, face-to-face confrontations, and unwanted gifts (for reviews, see Meloy, 1996, 1997, 1998).  In addition, as Table 1 indicates, a disturbingly high proportion of stalkers either make threats to their victims, or actually are physically violent toward them.  The range across studies of stalkers who threaten is 19% to 78%, with an approximate average of 50%. Approximately 30% to 40% of stalkers are actually physically violent to their victims.

            Relationship between the stalker and the stalked.  As shown in Table 1, many studies find that stalkers pursue former or, to a much lesser extent, current intimate partners.  This finding is consistent across different methodologies of studies, and hence may be considered quite reliable and robust.  On average, across studies, a reasonable estimate would be that in 50% to 60% of cases, (ex-)intimate partners are being stalked.  A few of the studies in Table 1 warrant some additional description given their large-scale scope. 

            The U.S. Department of Justice, in a telephone survey of 8000 men and 8000 women,[3] found that 59% of women who were stalked, and 32% of men who were stalked at one point had an intimate relationship with the stalker.  This discrepancy between men and women (that women are more likely to be stalked by ex-intimates than are men) has been found in other research as well.  In a large-scale study of Canadian crimes of criminal harassment, Kong (1996) found that across 7462 cases of criminal harassment, and 49% involved ex-intimates. The Department of Justice (1996) in Canada, based on 601 cases, also found that 57% involved ex-intimates.  The British Columbia Attorney General (1995), in a small though random study of 101 cases of criminal harassment, determined that 71% involved ex-intimate partners.  These four studies provide a reliable and steady anchor for the estimate of the percent of stalkers who were once intimately involved with their victims.  The other studies in Table 1 are subject to greater selection bias, although, on average, are consistent with this estimate of 50% to 60%.  This group of victims is the largest or most frequently occurring.  Other groups of victims of stalking include friends and acquaintances, strangers, family members, and colleagues (for reviews, see Meloy, 1996, 1997, 1998).

           

Typologies of Stalkers

            There have been a number of “typologies” of stalking put forward.  Unfortunately, many of these are based not on rigorous data analytic techniques, but on intuition and clinically informed speculation. Clinical typologies of stalkers that have been replicated across studies do not yet exist.  Some of the more common typologies will be summarized below.

Typologies can be divided into those that focus on the perpetrator, and those that focus on the relationship between the perpetrator and the victim (Meloy, 1998).  An example of a perpetrator-based typology is provided by Zona et al. (1993), who classified stalkers into three main categories – erotomanic (n = 7), love obsessional (n = 32), and simple obsessional (n = 35).  The first group was purely and solely delusional.  The love obsessional group also had erotomanic symptoms but these were secondary to other disorders.  The simple obsessional group was defined by the presence of some sort of prior relationship that had ended. 

This typology has gained favour in the stalking literature (Meloy, 1997, 1998).  Unfortunately, little description of how the groups were selected or clear definitional criteria for group membership is given.  The typology was not subjected to formal testing or data-analytic techniques.  In fact, the stalkers were not directly assessed.  However, the typology has some utility on a conceptual basis in terms of at least distinguishing between stalkers with and without prior relationships (although the nature of the relationship is not specified for this group).  Future validation research may help to develop this typology.

Relational typologies seem more conceptually sensible at this early stage of stalking research.  Meloy (1998) suggests that three groups be defined – intimates, strangers, and acquaintances.  Meloy (1997, 1998) has focused upon developing one of these sub-groups – the intimate stalker (or obsessional follower, as he calls them).  Other researchers add a number of different categories.  For instance, Harmon et al. (1995) classify stalkers into groups who are “affectionate/amorous” and those who are “persecutory/angry,” and then cross these with seven relational categories, resulting in 14 groups of stalkers.  Little validity or supporting data is provided for how these groups systematically differ in terms of behavioral and psychological correlates.  Nonetheless, some differences are provided between the two main groups (affectionate vs. angry), and the relational breakdown may be pragmatically useful to groups such as law enforcement.

Although there are other “typologies,” (i.e., psychotic versus not psychotic – Kienlen et al., 1997), the main point is that the current research on stalkers appears not to be able to offer any firm typologies. There has been little or no sophisticated data-driven creation of typologies.  A sensible approach for the current time is to view stalkers along lines of their relationship to their victims, as Meloy (1998) has done.  In so doing, one can draw from more developed areas of research as the stalking research begins to grow.

Stalking Summary

            Stalking is a fairly common phenomenon, experienced by perhaps five percent of the general population, and is perpetrated more often against women than men.  Although the state of the research corpus cannot support very firm conclusions, it appears that Axis I and II disorders are prevalent.  Concerning Axis II disorders, Cluster B personality disorders and traits predominate.  Typologies of stalking have been forwarded, but tend not to have been empirically or systematically derived.  However, a useful distinction to draw is based on the stalker-victim relationship.  In particular, perhaps the most common relationship is between ex-intimate partners.  This latter point introduces the bridge between stalking and domestic violence.  A framework of domestic violence will be forwarded, and then juxtaposed on the empirical and conceptual literature on stalking.

 

Prevalence and Nature of Domestic Violence

            Although this paper deals both with stalking and domestic violence, its main purpose is not to thoroughly review the domestic violence literature.  Rather, the following sections will describe definitions, incidence data, and common typologies of domestically violent persons that may stem from review sources.  The aim is to apply these typologies of domestic violence to the stalking literature.  Following the summary of domestic violence concepts, the link between stalking and domestic violence is described in some detail. 

Definition and Incidence of Domestic Violence

            Although asserting that a definition of a construct is clear can often be dangerous, “domestic violence,” for the purposes of this paper at least, can be defined as violence or aggression perpetrated against one person in a domestic, marital, conjugal, or dating relationship, whether heterosexual or homosexual.  Dutton (1995), more specifically, has defined wife assault as “any physical act of aggression by a man against a woman with whom he is in an intimate (i.e., sexual-emotional) relationship” (p. 3).  The U.S. Department of Justice (1996) has initiated a data-driven approach to developing a “uniform definition of domestic violence” (p. 2).  At the current time, however, the definition given above will be used.

            The U.S. Department of Justice (1996) reports that the National Crime Victimization Survey of the early 1990s estimated approximately 1.15 million violent crimes are committed each year by intimate partners against intimate partners.  As with most criminal victimization reporting, this likely is an underestimate.  Dutton (1995) summarized various sources of incidence data for domestic violence within North America and estimated that severe violence occurs in roughly 9% to 12% of marriages.  More liberal definitions of domestic violence that include “any violence ever” return incidence rates between 19% and 28% (Dutton, 1995).  Domestic violence clearly affects a large proportion of the population and an enormous number of people in absolute terms.

Typologies of Batterers

            There have been numerous typologies developed in the domestic violence literature (for reviews, see Dutton, 1998; Holtzworth-Munroe & Stuart, 1994; see also Table 2).  Commonalties have emerged across these efforts, and the purpose of the present section is to summarize some of these typologies. We draw on Dutton’s (1998) recent summary and integration of both primary empirical research and reviews.  The general picture that emerges from the batterer research is that there appear to be three basic groups that appear across studies.

Dutton has in several places (Dutton, 1995, 1997, 1998) summarized much of the this literature and stated that there are essentially three definable types of domestic abusers who fall along two dimensions — overcontrolled versus undercontrolled, and instrumental versus impulsive. These include the psychopathic, avoidant, and borderline subtypes (see Table 2).  Dutton (1998) argued that, despite different labels, these three types are found in the work of other researchers and commentators (see Hamberger & Hastings, 1986; Holtzworth-Munroe & Stuart, 1994; Saunders, 1998). Psychopathic batterers are those who are generally antisocial and violent, both within and without intimate relationships.  As Dutton (1997) points out, these batterers are instrumental in their violence use, rather than emotional and impulsive.[4] They are generally antisocial and hence often may be incarcerated for other crimes. They tend to have Dismissing attachment styles on the Relationship Style Questionnaire (RSQ; Bartholomew & Horowitz 1991). This group has been identified by others as well – the “generally violent/antisocial” group of Holtzworth-Munroe and Stuart (1994), “Cluster 2” or the “Generally Violent/Antisocial” group of Hamberger, Lohr, Bonge, and Tolin (1996), the “Type 2” or “generally violent” group of Saunders (1992), the “Type 1” generally antisocial batterers of Gottman et al. (1995), and the “instrumental/undercontrolled” group of Tweed and Dutton (1998).

                                               

Insert Table 2 about here

                                               

Avoidant batterers, according to Dutton, tend to suppress their anger as a function of being unassertive.  Periodically, their “bottled-up” anger and negative affect is expressed as violence toward an intimate partner.  These over-controlled batterers are abusive less often than the other types. They tend to have dependent “Preoccupied” attachment styles. Researchers have identified this general group as “dependent/compulsive” (Hamberger & Hastings, 1986), “family only/nonpathological” (Hamberger et al., 1996), “passive-aggressive (family only)” (Holtzworth-Munroe & Stuart, 1994), “emotionally suppressed” (Saunders, 1992), and “impulsive/overcontrolled” (Tweed & Dutton, 1998).

Finally, Borderline batterers are impulsive, unstable in affect and interpersonal relations, and, as the label implies, most like borderline personality disordered people.  This group of batterers tends to be at highest risk for repeat domestic violence.  Dutton et al (1994) found this group to have Fearful attachment styles. Dutton (1995, 1998) proposes that this group is the only one to display the cyclical nature of abuse characterized by tension-building, release (abuse), and contrition.  The group is dysphoric, emotionally volatile, angry, and often devalues others.  It is this group of batterers that appears to be most alike stalkers who pursue intimate partners. Others have described groups with these characteristics as “schizoid/borderline” (Hamberger & Hastings, 1986), “negativistic-dependent” (Hamberger et al., 1996), “dysphoric/borderline” (Holtzworth-Munroe & Stuart, 1994), “dependent/unstable” (Gottman et al., 1995), “emotionally volatile” (Saunders, 1992), and “impulsive/undercontolled” (Tweed & Dutton, 1998).

Comorbidity of Stalking and Domestic Violence

Empirical Evidence

The logical starting point for this section is to estimate the proportion of stalkers who pursue former intimate partners.  This information was summarized above, and will be mentioned only briefly here.  A reasonable estimate of the percent of stalking victims who were once intimate partners of their pursuers is 50% to 60%.  Focusing somewhat, the next question becomes – what proportion of this group involved domestic violence situations?  That is, a person may stalk a former intimate partner to whom he or she was or was not abusive while the relationship existed.  In a broad sense, one may conceptualize any stalking of a previous partner to be domestic abuse, regardless of whether the previous relationship was marred by abuse.  However, a narrower conceptualization would seek to establish stalking that followed or stemmed from a relationship that was itself violent.  There seem to be two routes to answering this latter question:  (1) the proportion of domestically violent people who stalk their current or former partners; and (2) the proportion of stalkers who were once abusive to their victims within a context of an intimate relationship. 

Stalking Among Intimate Abusers

            There appears to be only one direct study that falls within this category.  Burgess et al. (1997) studied a group of 120 persons who were charged with felony domestic violence and were attending a treatment program.  Of these 120 batterers, 36 (30%) admitted to stalking their partners.  More support, albeit indirect, is found in correlational studies of relational intrusion.  Dutton, van Ginkel, and Landolt (1996) found that a measure of relationship intrusiveness, created from what appears to be the criminal harassment (stalking) provisions in the Canadian Criminal Code, was correlated with partners’ ratings of batterers’ jealousy, which was itself correlated with abusiveness.  Spitzberg and Rhea (1999) found that a measure of “obsessive relational intrusion,” a stalking-like construct that is stalking when combined with threatening behaviour, was correlated with measures of sexual coercion and forcefulness.  These indirect studies support the notion that abusive partners may be likely to display stalking-like behaviours.

Intimate Abuse Among Stalkers

            These studies include samples of stalkers, some of whom have been violent within past intimate relationships.  The study by the Canadian Department of Justice (1996) determined that of the 57% of stalking victims who were stalked by former intimate partners, half of these intimate partners were also violent within the previous relationship.  Kienlen et al. (1997) found that, within their sample of 25 stalkers, 15 (60%) stalked former intimate partners.  Of these, 7 (47%) had been violent within that previous intimate relationship.  In a sample of 128 stalkers, Kileen and Dunn (1998) found that 65% of cases had a history of domestic violence.

These few studies that do provide direct data on the specific issue of stalking and previous domestic violence suggest that somewhere between 30% and 65% of stalking cases that involve former intimates also involved a previous violent relationship.  Broadly construed, then, approximately 50% to 60% of all stalking cases may be considered “domestic” in the sense that cases involve former intimates.  In a narrow sense, roughly half of this domestic stalking group (30% to 65%) involve previous violent relationships.

Unfortunately, little research has been carried out that specifically evaluates the characteristics of this sub-sample of stalkers who pursue former intimate partners against whom they were previously violent within a relationship.  Less has been done concerning how this sub-sample may differ from other sub-samples of stalkers.  What has been done is presented below.

One finding that seems to emerge from the research is that persons who stalk ex-intimate partners tend to display more violence towards their victims than do persons who stalk others. Meloy and Gothard (1995) determined that threats made by stalkers were more common where the victim was a former intimate partner.  In a study of 101 randomly selected cases of criminal harassment in Canada, the only group of offenders who threatened their victims with a weapon were those whose victims were former intimate partners (British Columbia Attorney General, 1996).  Similarly, Schwartz-Watts and Morgan (1998) found that violent stalkers were more likely to have a prior attachment to their victims (80%) than were non-violent stalkers (55%).  The Canadian Department of Justice (1996) reported that in criminal harassment cases involving former partners, 50% of the previous relationships involved violence.  In the U.S. Department of Justice (1997) study, violence directed at stalking victims by the stalker was very common in situations of former intimate abuse.  Approximately 80% of women who were stalked by former intimate partners had physical violence directed toward them during the stalking episodes (no direct comparison data were provided for non-intimate stalking victims).  Burgess et al. (1997) also found that batterers who later stalked their victims were more likely than other batterers to have inflicted serious violence upon their partners.  That is, they were more likely to have beaten the faces and limbs of their victims, to have sexually assaulted them, and to have strangled them with their hands.  Kienlen et al. (1997) noted in her sample of 25 persons charged with stalking that, with one exception, all physical violence that occurred in the context of the stalking behaviour was directed at former intimate partners.

More indirectly, in a study by Coleman (1997), 141 female participants were asked to identify whether they had been stalked by former partners.  Thirteen people (9.2%) indicated that they had been.  Participants were asked to complete the Conflict Tactics Scale (Strauss, 1979, 1990) concerning their former partners’ behaviours during the relationship.  Results indicated that former partners who stalked their ex-intimates had higher scores on the Verbal Violence and Physical Violence scales of the CTS.  These findings suggest that men who stalked their former intimate partners after a break-up were more likely than other men to have been abusive in the relationship.

            Some instructive, although, again, somewhat indirect, research has been carried out on a concept called “obsessive relational intrusion” (ORI), mentioned briefly above (Spitzberg et al., 1998; Spitzberg & Rhea, 1999).  Spitzberg and Rhea (1999) defined ORI as the unwanted pursuit of a relationship.  ORI is conceptualized to be broader than stalking.  It includes such behaviour as constant calling or requesting a date, break and enter, or besetting.  When it becomes threatening, it is stalking.  As such, ORI and stalking are related though not synonymous.

            Spitzberg and Rhea (1999) carried out a study on 360 college students (185 female; 178 male) who were asked to recall a situation of ORI that they had experienced.  The purpose of the study was to assess the link between ORI and sexual coercion.  Approximately one third (34%) of the sample identified ex-partners; the remainder identified friends (38.4%) and acquaintances (24.6%).  The gender of the pursuer was female 46% of the time, and male 54% of the time.  Spitzberg and Rhea used an instrument called the “victim short form” of the ORI (ORI-VSF) to assess ORI.  Items on the ORI-VSF differ in severity of intrusiveness and are rated according to frequency of occurrence.  The scale has two factors — “experience of pursuit” and “experience of aggression” — each with high internal consistency (alphas of .90 and .86, respectively).  Sexual coercion was rated on a similarly constructed scale.  This scale produced four components: (1) psychological coercion (alpha = .93); (2) severe force (alpha = .90); (3) mild force (alpha = .90); and (4) deception (alpha = .87).  ORI factors were moderately to strongly related[5] to sexual coercion experiences (mild force, severe force, deception coercion, and psychological coercion). 

In another study on ORI, Spitzberg et al. (1998) found that the ORI factors of aggression and pursuit were related to whether participants self-defined themselves as victims of stalking.  The researchers also found that stalking was related to angst, fear, stress, and hopelessness.  Taken together, the interpretation of these findings are that ORI and stalking are strongly related, and that people who engage in the stalking-like behaviour of ORI are also more likely to be sexually coercive and forceful than those who do not engage in such behaviours.

Limited research has investigated how stalkers of former partners, whether violent or otherwise in that previous relationship, may differ in other respects.  Burgess et al. (1997) attempted to differentiate between batterers who also stalked their victims and those who did not.  They found that “stalking batterers” were more likely than “non-stalking batterers” to have a history of stalking, a history of assault, to have abused alcohol, to have lived alone, and to have been unmarried; battering stalkers were less likely to have perceived provocation in their victims.  Kienlen et al. (1997) pointed out a trend for stalkers of ex-intimate partners to be more likely to be non-psychotic, whereas those who stalked others were more likely to be psychotic.[6]

Again, indirect evidence suggests that stalkers of non-intimates have different motives, and likely different psychological characters underlying this motivation.  For instance, in the study by Dietz et al. (1991a) of threatening letters written to Congress, apparent motives were often government-related and persecutory.  In the other Dietz et al. (1991b) study of letters to Hollywood celebrities, the motives for writing were more often sexual and romantic than in the Congress letters, with approximately half having such a theme.  Not knowing or ever having met or had a relationship with the victim, a reasonable inference would be that the letter-writers here were more likely primarily mentally ill rather than characterologically dysfunctional.  Thus, although the Hollywood letter writers may be more similar to domestic stalkers than are the Congress letter writers, they are nonetheless generally dissimilar in terms of psychopathology. 

 

A Profile of Intimate Stalkers

 

The empirical evidence to this point seems to suggest that stalkers tend to possess certain characteristics: threatening and violent behavior, prior intimate partners as victims, substance abuse, other Axis I disorders such as depression or schizophrenia, and Axis II pathology, including especially Cluster B personality traits, as well as dependent, schizoid, anxious, and avoidant traits.  Little research has actually addressed whether “domestic stalkers” differ from other stalkers.  However, given that it does appear that the largest group of stalkers target intimate partners, it may be a reasonable to hypothesize that the characteristics of stalkers, as described above, generally apply to domestic stalkers. Further, some indirect research suggests that non-domestic stalkers tend to be politically motivated and persecutory.  These underlying motives appear to differ from domestic stalkers.

 

Conceptual Basis for an Overlap Between Stalking and Domestic Violence

           

The main feature of domestic stalkers,[7] other than they appear to be more violent than other stalkers, is that they often possess Cluster B personality disorders or characteristics.[8]  Cluster B personality disorders include narcissistic, antisocial, borderline, and histrionic.  The traits of these disorders form the basis of Meloy’s (1997, 1998; Meloy & Gothard, 1995) characterization of stalkers.  As will be described shortly, they also form the basis of one of the typologies of domestic violence.

            Meloy and Gothard (1995) described stalkers as possessing narcissistic character pathology.  The authors adopt an object relations or self psychology perspective, stating that “abandonment rage arising out of a narcissistic sensitivity appears to defend against the grief of object loss, which then drives the obsessional pursuit” (p. 262).  Meloy (1997) describes that personality characteristics and abandonment rage may co-incide, such that rejection (real or perceived) by another person is humiliating and shameful, and is defended against with rage.  The abandonment rage leads to pursuit and the attempt to devalue the other person in real life, which reinforces the “narcissistic linking fantasy to the idealized object.” (p. 183).  Narcissistic wounding occurs for the stalker every time his approaches or contacts are rebuked.  Given the difficulty tolerating shame, humiliation, and loss, stalkers may react with rage, fantasies of entitlement, and possibly retaliation.  This may explain why violence seems more common in stalkers of ex-intimates than in other stalkers. 

            Meloy (1997) describes that narcissistic characteristics are most apparent in the sense of entitlement and gross disregard for the suffering of the victim.  Narcissistic self-absorption increases and empathy for others systematically decreases.  Hysterical traits include emotional instability, dependency, hypersexuality, and overinvolvement, in which the person may seek to be in close proximity to the object, despite attempts to discourage or even punish this.

            In an earlier publication, Meloy (1989) reviewed the literature on erotomania and proposed a division — the classical delusional erotomanic, and the nondelusional “borderline” erotomanic, who is typified by narcissistic, hysterical, paranoid, borderline, and antisocial traits.  Borderline erotomania ought to be considered a differential diagnosis for delusional erotomania, and characterized as a “gross disturbance of attachment or bonding” (p. 480).  The term “borderline” was chosen to refer to the unstable nature of this condition, and refers to the level of personality organization.  Not all borderline erotomanics would necessarily fulfill the criteria for borderline personality disorder in the DSM system (III-R at the time of publication).  As opposed to delusional erotomania, borderline erotomania usually involves a history of emotional attachment to the object.  Separation from the object is seen as abandonment, and, as reported above, is defended against with narcissistic rage.

            Meloy (1997) has drawn on a broader ego or self psychology model to explicate his model of stalking.  Stalkers are described as being prone to fits of rage in response to abandonment.  Rage has a connection to shame.  Lewis (1992) described shame as a state of self-devaluation.  It is an exposed self. Rage is a response to prolonged shaming; it is an attempt to ward off shame which is expected from another object, and also to switch the attribution from an internal one made of the self to an external one made of another. Rage, stemming from shame, sets up a “feeling trap” – the shame-rage-shame spiral.  Narcissism, often attributed to stalkers, is a shame disorder (Kohut, 1972). Given the sensitivity to shame, and the characterologic lengths which are gone to in order to avoid it, when shame is experienced, it is more likely to lead to rage in narcissistic people than in people who are not shame-prone (Lewis, 1992).

            Shame may become internalized after repeated experience and shape a person’s identity, affecting sense of worth, adequacy, and value (Kaufman, 1992).  Shame is an alienating affect.  A shamed person feels exposed to others and to oneself.  With shame there is a painful examination of the self. Shame is a terribly powerful and intense experience – attention is focused solely upon the inadequacy of the self.  If this is done publicly, by, say, a significant other, the feelings may be magnified.  Kaufman (1992) also writes that rage is the natural response to shame, in that it protects the self from prolonged exposure to continued pain. Rage insulates the vulnerable self from further exposure to unpleasant emotions.  Entrenched rage borders on generalized hatred, which in turn promotes revenge-seeking behaviour and potential violence.  The behaviour of stalking is consistent with this vengeance.

            The major difficulty comes when the child identifies with the parent who induces shame (Kaufman, 1992).  The child has a need to be loved; this need will persist whether love-based or shame-based.  The child who is shamed may come to internalize beliefs about the self which have been consistently communicated to them by their parents, and which may be reinforced by peers.  The shamed child may internalize beliefs about the self, including worthlessness, stupidity, and the like.  If parents continually blame the child for events gone wrong, the child may learn to blame him- or herself later in life for anything that goes awry.  Certain patterns of relating to others will develop.  The child may learn to be deferent and submissive in relationships, or may be aggressive to guard against being vulnerable to humiliation and shame.

            Associated with this characterological style are certain defenses such as denial, projection, minimization, and splitting (Meloy & Gothard, 1995).  The shamed self may begin to disown aspects of itself – splitting.  This is the defense against internalized shame.  There may be repression of disavowed aspects of the self, which reflect an effort to disown the parental image.  The result may be internalized insecurity.  This parent image is then projected onto others who may at times remind the person of the parent.  Other defenses to rage may include a pervasive tendency to hold others in contempt or disdain.  Or, the individual may have a near-pathological drive for power and perfection.

            Shame-laden interpersonal styles will repeat across the lifespan, in dealings with others. Meloy (1997) asserts that “stalking is a pathology of attachment” (p. 178).  Bartholomew (1990) has proposed a model of adult attachment that involves two dimensions – avoidance (model of other) and dependence (model of self), each of which can be high or low.  Four attachment styles result.  Securely attached people are low on both avoidance and dependence.  Dismissing attachment styles are characterized by high avoidance (negative model of the partner) and low dependence (positive model of self).  This involves denial of attachment and emotional distance.  The two styles that are typified by anxious attachment possess negative models of the self.  The preoccupied style, in addition, possesses a positive model of others, and is characterized by high dependence and attempts to seek approval from others so as to compensate for low self-esteem.  The fearful attachment style is high on both dependence and avoidance (negative models of both self and other), and has the paradoxical trait of desiring intimacy though distrusting others.  The latter two styles, and particularly the fearful style, has been found to be prevalent in batterers.

Kienlen (1998) has described how stalking and attachment styles may relate.  Although not assessing this empirically, she postulated that stalking may stem from disrupted childhood attachments that have persisted into adulthood.  She postulated that the Preoccupied and Fearful styles may be most strongly associated with stalking. As indirect support for this conceptualization, she reported finding that the minority of stalkers were raised by both parents (Kienlen et al., 1997).  Most stalkers experienced the loss of a primary caregiver early in life.  Similarly, more than half reported childhood abuse, and were described as withdrawn children with behaviour problems.  Many stalkers experienced a relationship loss just prior to the onset of stalking.  The authors reported that anger, hostility, projection of blame, obsession, dependency, minimization, denial, and jealousy were common “psychological factors” that motivated stalking. 

The data of Kienlen et al. (1997), though far from providing strong support, in that attachment patterns actually were not assessed in the study, are consistent with attachment theory.  However, attachment theory explanations for stalking are, empirically undeveloped.  Other facets of the character of stalking include obsessionality, jealousy, anger, and blaming the intimate partner for one’s sense of suffering (projection).  Kienlen et al. (1997) found that among non-psychotic stalkers, anger and jealousy were prominent motives for stalking.  Further, stalkers often blamed the victim for their misfortune.

Meloy (1997, 1998; Meloy & Gothard, 1995) has written that stalkers are obsessional.  In fact, he often labels stalkers “obsessional followers.”  Along these lines, Spitzberg and colleagues (Spitzberg et al., 1998; Spitzberg & Rhea, 1999) have proposed the concept of obsessional relational intrusion as a stalking-like phenomenon that, when accompanied by threatening behaviour, is isomorphic with stalking. 

The Conceptual Bridge between Stalking and Domestic Violence

            What has been proposed so far about stalking, drawn from empirical and theoretical bases, is that stalkers often target their former intimate partners, and possess Axis II characteristics of Cluster B personality disorders, as well as other personality disorder traits such as avoidant, dependent, and schizoid.  They may have mood disorders such as depression and dysthymia.  Stalking may derive from an attachment-based borderline personality that is tinged by jealousy, anger, and depression.  Without a stable sense of self, stalkers are proposed to draw on intimate others for ego strength.  Reacting with primitive defenses to perceived or actual insults or abandonments, stalkers may devalue their former intimate partners, and react with rage to defend against these insults to the ego.

            How does this conceptualization line up with the typologies of batterers presented earlier?  The overlap is found, quite clearly, in the “borderline/cyclical” group proposed by Dutton (1995, 1997, 1998).  More specifically, the concepts of “borderline personality organization” and “abusive personality” are conceptually similar to domestic stalkers.  These concepts may elucidate and inform the understanding of stalkers.  The “generally antisocial” group of batterers, though likely to engage in repeat domestic (and non-domestic) violence, are unlikely to engage in stalking because of a tendency toward detachment rather than dysfunctional attachment.  That is, antisocial individuals, likely typified by dismissing attachment styles, do not have the interest to sustain a “relationship” through stalking.  Rather, once a relationship ends, they are more likely to dismiss the partner and “move on,” rather than to obsess over and stalk the partner.  Table 2 represents this low risk for stalking among this group of batterers.

Dutton (1995, 1997, 1998) explains that the “borderline personality organization” is a cyclical personality, typified by instability in relationships and affective states, devaluing and undermining the intimate partner, intolerance of and rage in response to rejection or abandonment, and substance abuse. The borderline personality is often angry, depressed, and anxious.  According to Dutton (1998), the borderline personality relies on the intimate other to sustain his or her sense of self.  When the partner leaves or threatens to leave, this may be received as devastating to the self in that one’s own ego will disintegrate along with the loss of the partner.  This manifests in increased demands, anger, and abusiveness.  There may be an elevated attempt to control and devalue the partner.  The abuser may be depressed and anxious.  If the partner has left, vestigial control may be experienced through stalking. Such stalking may temporarily suppress anxiety-dysphoric states.

Dutton (1998) also has elaborated upon abandonment rage, but within the context of domestic abuse rather than stalking.  Rage may be expressed as a defense to the injury caused by abandonment.  This conceptualization is clearly congruent with that of the stalker proposed above. Drawing on many of the same constructs as does Meloy (1997, 1998), Dutton explains that borderline personalities employ primitive defenses such as splitting and projection, and defend against insults to the ego with rage.  This tendency toward rage and a weak ego stems from harmful childhood climates.  In early development, negative interactions with the caregiver, usually the mother, result in the splitting of the mother “object,” into the “good” and “bad” object.  Rage and hatred toward the bad object can then occur without the possibility of destroying the entire mother object.  Developmentally, these rage reactions to rejection and perceived negativism are preserved, and hence when persons mature, they may respond similarly when in relationships that are emotionally threatening.  Dutton (1998) states, with respect to batterers, that “rage is the magic elixir that restores an inner sense of power” (p. 103).  Abuse that may accompany rage devalues the partner and instills feelings of low self-worth within her, making her less attractive to other men.  As such, rage serves to defeat or suppress feelings of jealousy and powerlessness.

            These constructs have received empirical support as well.  The Borderline Personality Organization (BPO) scale (Oldham et al., 1985) includes three subscales:  (1) identity diffusion; (2) primitive defenses; (3) reality testing.  Dutton (1995, 1998) has summarized research on batterers using the BPO, and reports that the BPO total score correlates with moderate strength to the CTS (both verbal and physical aggression), and strongly to anger and trauma symptoms.  In addition, persons scoring high on BPO displayed more jealousy, and scored higher on a measure of psychological maltreatment of women in terms of dominance and isolation and emotional abuse.  Based on these studies and findings, Dutton (1995) proposed that a “borderline personality organization” is a key element of domestic assault.  He describes that such persons within relationships depend on their partners to maintain ego identity.  That is, such persons fundamentally have weak or poorly developed ego strength, and hence rely upon others to fulfill this function.  Associated with this organization, as Meloy (1997, 1998) has described of stalkers, is the tendency to employ primitive defenses such as splitting and projection. 

            The BPO is central to Dutton’s conception of the “abusive personality.”  Added to the notion of the borderline personality is the construct of child trauma.  Essentially, scores on the BPO scale were found to correlate with the experience of abuse in childhood, including rejection and physical abuse.  The reader is reminded of findings within the stalking literature that stalkers tend to have had abusive relationships (Kienlen et al., 1997).  This abusiveness was proposed to lead to dysfunctional attachment styles in adulthood (Kienlen, 1998).

Dutton (1995) has relied on Bartholomew’s conceptualization of attachment, which is the same model used by Kienlen (1998) to describe stalkers. Fearful attachment was most highly related to measures of abusiveness, BPO, anger, and jealousy, although the preoccupied style also had moderate correlations to these constructs (Dutton, Saunders, Starzomski, & Bartholomew, 1994).  Early patterns of abuse and rejection can lead to later abusiveness and to measures of the abusive personality.  Tweed and Dutton (1998) found that while groups of both antisocial/ aggressive/ narcissistic batterers and impulsive/ borderline batterers scored high on preoccupied attachment styles, only the impulsive group also demonstrated a fearful attachment style.  Generally, because the fearful and preoccupied attachment styles include abandonment anxiety, Dutton expects that they would be related to anger, jealousy, and affective instability, all characteristics of the BPO.  He posits that the fearful pattern may most aptly fit domestic abusers.  These individuals, while seeking intimacy, are distrustful and fear rejection. Table 2 describes how Bartholomew’s attachment styles likely map onto types of batterers.

In somewhat more detail, Dutton described the role of shame to the development of the abusive personality in batterers.  In terms of shame, Dutton (1998) reported that the correlations between shame and BPO, anger, trauma symptoms, maltreatment of women, and abusiveness were in the moderate to large range.  Correlations between recollections of shame and guilt experiences as children, on the one hand, and parental abusiveness on the other were in the large range (Dutton, van Ginkel, & Starzomski, 1995).  Correlations between shame and guilt and later domestic abuse, once parental abuse was partialled out of the correlations, on average were small to moderate in size with respect to BPO.  These data provide correlational support for the connection between shaming as a particularly insidious parental characteristic on later development and abusiveness.  Of course, in theory, these principles should apply to stalkers.  However, the research simply has not been done. 

Generally, jealousy has been found to predict abusiveness in couples (Dutton, van Ginkel, & Landolt, 1996).  It seems to play some role in stalking as well (Kienlen et al., 1997).  Concerning separated couples, jealousy predicted “intrusiveness,” as constructed from descriptors within the Code’s definition of criminal harassment, or stalking (Dutton et al., 1996).  Jealousy was strongly correlated with intrusiveness.  What is interesting is the similarity between this study and the research on ORI, which has been defined as a concept that is related to stalking.  A similar pattern of results has been found with ORI correlating with violence and pursuit (Spitzberg et al., 1998; Spitzberg & Rhea, 1999).

Dutton et al. (1996) explained that intrusiveness, which is “typical of men who stalk former partners” (p. 420), may stem from the emotional reaction to loss of the relationship.  Dutton et al. (1996) claim that the interactive process of early shaming, attachment insecurity, and the abusive personality may manifest as stalking behaviour during estrangement.  Stalking may be more likely among persons with fragile egos. This conceptualization is consistent with the work on ORI (Spitzberg et al., 1998; Spitzberg & Rhea, 1999), as well as Meloy’s (1997, 1998) position on stalking.

            Generally, then, there are many overlaps between conceptualizations of domestic stalkers and the “borderline/cyclical” type of batterer.  Table 3 presents these in summary form.  Most of the listed characteristics have received empirical support, and all have theoretical support, in both the stalking and domestic violence literatures.  The similarities between these groups is high.  Both are construed as consisting of borderline personalities, as well as some other psychopathological traits such as mood disorder, avoidance, and dependence.  Attachment dysfunction is present in both groups, with the anxious attachment style of Fearful (and, to a lesser extent, Preoccupied) dominating.  This attachment dysfunction stems from putative abuse and shaming experiences during childhood.  Shame produces primitive defenses such as splitting, resulting in the tendency for both groups to oscillate between devaluation and idealization of the intimate partner.  Both groups are described as rageful in response to perceived or actual abandonment or rejection.  Anger, jealousy, a sense of entitlement, and general emotional volatility characterize domestic stalkers and borderline/cyclical batterers alike.  Both groups are likely to abuse substances, and to use violence and threats against their partners.

                                               

Insert Table 3 about here

                                               

            A question that arises is how the cyclical nature of the abusive personality manifests within stalkers.  Dutton et al. (1996) propose that stalking may be a facet of estrangement.  There is unfortunately little research on how stalking may fit into the cycle of violence that typifies this group of domestic batterers.  The underlying pathology may be the same between these groups of stalkers and batterers.  However, the situational context differs in that there typically is not an ongoing relationship between stalkers and their victims, where the same cannot be said of batterers and their victims.  As such, the phases of tension building, battering, and contrition may be expressed differently.  That is, the qualitative nature of stalking may vary.  Stalkers may still oscillate between devaluation and idealization of their victims, but this may be manifest in different types of stalking behaviour.  For instance, during the tension-building phase, a stalker may display many of the threatening and intrusive stalking behaviours such as menacing phone calls, besetting, threats of harm, and face-to-face contacts.  After continuing rejections by the object of their pursuit, the tension could culminate in an attempt to be physically assaultive.  After this, a stalker may enter the contrition phase and display qualitatively different stalking behaviour, such as unwanted gifts, non-menacing phone calls, and the like.  The cycle may perpetuate itself, as it does when this pathology is situated in the context of an on-going relationship.

            In this regard, stalking is the manifestation of abuse once an intimate partner leaves a relationship, either as a continuation of past abuse or as a behaviour that has been triggered by the ending of the relationship.  We know that there is a high prevalence of violence directed at ex-intimate partners by stalkers.  Due to reporting biases, the prevalence may actually be higher.  Threats occur even more commonly than physical violence, but can be viewed as verbally abusive behaviour.  Furthermore, it may be that the opportunities to be physically abusive are simply less available for stalkers as opposed to batterers who live with their intimate partners.  Figure 1 displays a proposed relationship between stalking and domestic violence.

                                     

Insert Figure 1 about here

                                               

           

            However, there do seem to be some differences between the borderline/cyclical batterer and stalkers who pursue intimate partners.  For instance, stalkers may possess a higher prevalence of major mental (psychotic) disorders, and they may be more socially maladjusted (in terms of unemployment, lack of previous relationships, and criminal histories).  This may stem from methodological artifact, however, in that much of what is known about the characteristics of stalkers is based on highly selective groups that have been court-referred for psychiatric assessment.  In less selected samples (i.e., Lyon, 1998), psychopathology and unemployment have not been as prevalent.  Unfortunately, such samples have been unable to provide much clinical information, and no diagnostic information.

            An alternative is that there is truly some different or additional dynamic that is present in stalkers.  Perhaps stalkers represent the severe end of the continuum of the psychopathology that underlies borderline/cyclical batterers.  Or, perhaps stalkers are more obsessional that batterers.  They may be less phasic than batterers. It may be that domestic stalkers and borderline/cyclical batterers are actually one and the same.  Or, at least, a large proportion of the groups overlap.

            There may be some stalkers who simply do not fit the typology of the borderline/cyclical batterer, but are more akin to the Generally Antisocial or the Avoidant types.  Some of the characteristics of these groups do match up with a minority of stalkers.  For instance, antisocial personality disorder occurs in a consistent minority of stalking samples that have been able to measures personality disorders.  Generally, however, one would not expect persons with antisocial personality disorder to possess the obsessional quality of stalkers, or the commitment to maintain an attachment, despite its dysfunctional nature. Dependent and avoidant traits occurred in the stalking samples as well.  As such, some proportion of stalkers could fit into this category.  Of course, there may be some blurring of these category distinctions, such that a person may not always fit cleanly into one or the other.  Finally, the universe of stalkers, even those who target ex-intimates, may not divide in the same way as the universe of domestic batterers, and hence, while there may be overlap, there will be some systematic error as well.

            These are all hypotheses that can be tested.  What seems clear is that the ultimate stalking study has yet to be done, and what is known about stalking has to be patched together from small scale, highly selective samples with fairly decent clinical measures, and larger scale, more representative samples with less thorough clinical data.  Various research strategies could be used in conjunction to address the issues raised in this manuscript. Research that investigates the link between stalking and domestic violence ought to include all the measures common to both areas of research, including specially designed measures of stalking (i.e., the Stalking Checklist) and stalking-related behaviours such as obsessive relational intrusion.  Large-scale prospective studies of randomly chosen couples could be followed longitudinally and assessed periodically for domestic violence, break-ups, and stalking.  Ideally, a host of clinical and demographic data could be collected at the outset as well as at the follow-up periods.

            More focused research could be carried out on selected groups of batterers and stalkers, similar to that which has been done, but for the specific purpose of evaluating the bridge between the two constructs.  Within stalking samples, it would be important to administer the commonly used measures.  This may entail contacting the victim of the stalkers to complete measures such as the CTS.  In groups of domestically violent persons, the incidence of stalking could be assessed with the aforementioned measures of stalking and related constructs.

Conclusion

            These research suggestions may begin to offer more concrete support for the observations and hypotheses made throughout this paper.  It appears that the most typical stalking scenario involves ex-intimate partners.  Violence is common in the past relationship, and is common during the stalking episode.  Domestic stalkers and certain batterers share a host of common characterological similarities, such as borderline personality organizations, jealousy, anger, abandonment rage, poorly integrated ego and primitive defenses, dysfunctional attachment styles, substance abuse, and emotional volatility.  Both groups also tend to confine their aggression to relationships, although stalkers may more commonly have criminal histories (although this may stem from methodological artifact).

            It is proposed that a high proportion of domestic stalkers would fit into the borderline/ cyclical batterer subgroup – they are one and the same.  There is certainly a theoretical and conceptual basis for this statement.  There also is some amount of empirical support, both direct and indirect.  It is of course only through systematic research that the exact overlap between these groups can be evaluated.  The model of the borderline/cyclical batterer may be a rich source of theory and research from which to draw in order to inform research on stalking.

References

            American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

            Attorney General of British Columbia.  (1995).  The report of the criminal harassment unit.  Part II:  The nature and extent of criminal harassment in British Columbia.  Vancouver:  Attorney General.

            Bartholomew, K.  (1990).  Avoidance of intimacy: An attachment perspective.  Journal of Social and Personal Relationships, 7, 147-178.

            Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61, 226-244.

            Bland, R., & Orn, H.  (1986).  Family violence and psychiatric disorder.  Canadian Journal of Psychiatry, 31, 127-137.

            Burgess, A. W., Baker, T., Greening, D., Hartman, C. R., Burgess, A. G., Douglas, J. E., & Halloran, R.  (1997).  Stalking behaviors within domestic violence.  Journal of family Violence, 12, 389-403.

            Cohen, J.  (1992).  A power primer.  Psychological Bulletin, 112, 155-159.

            Coleman, F. L. (1997). Stalking behavior and the cycle of domestic violence. Journal of Interpersonal Violence, 12(3), 420-433.

            Dempster, R. J., Hart, S. D., Boer, D. P. (1997, August). Psychopathy and violent criminal histories in male offenders. Poster session presented at the annual meeting of the American Psychological Society, Chicago, Ill.

            Department of Justice Canada. (1996). A review of Section 264 (criminal harassment) of the Criminal Code of Canada. Ottawa: Author.

            Dietz, P. E., Matthews, D., Martell, D., Stewart, T., Hrouda, D., & Warren, J. (1991). Threatening and otherwise inappropriate letters to Members of the United States Congress. Journal of Forensic Sciences, 36, 1445-1468.

            Dietz, P. E., Matthews, D., Van Duyne, C., Martell, D., Parry, C., Stewart, T., Warren, J., & Crowder, J. (1991). Threatening and otherwise inappropriate letters to Hollywood celebrities. Journal of Forensic Sciences, 36, 185-209.

            Dutton, D. G.  (1995).  The domestic assault of women:  Psychological and criminal justice perspectives.  Vancouver:  University of British Columbia Press.

            Dutton, D. G.  (1997).  A social psychological perspective on impulsivity/intimate violence.  In C. D. Webster & M. A. Jackson (Eds.), Impulsivity:  Theory, assessment, and treatment (pp. 32-41).  New York: Guilford.

            Dutton, D. G.  (1998).  The abusive personality:  Violence and control in intimate relationships.  New York:  Guilford.

            Dutton, D. G., van Ginkel, C., & Landolt, M. A.  (1996).  Jealousy, intimate abusiveness, and intrusiveness.  Journal of Family Violence, 11, 411-423.

Dutton, D. G., van Ginkel, C., & Starzomski, A.  (1995).  The role of shame and guilt in the intergenerational transmission of violence.  Violence and Victims, 10, 121-131.

            Dutton, D. G., Saunders, K., Starzomski, A., & Bartholomew, K.  (1994).  Intimacy-anger and insecure attachment as precursors of abuse in intimate relationships.  Journal of Applied Social Psychology, 24, 1367-1386.

            Fremouw, W. J., Westrup, D., & Pennypacker, J.  (1997).  Stalking on campus:  The prevalence and strategies for coping with stalking.  Journal of Forensic Sciences, 42, 666-669.

            Gottman, J. M., Jacobson, N. S., Rushe, R. H., Shortt, J. W., Babcock, J., La Taillade, J. J., & Waltz, J.  (1995).  The relationship between heart rate activity, emotionally aggressive behavior, and general violence in batterers.  Journal of Family Psychology, 9, 227-248.

            Hall, D. M.  (1998).  The victims of stalking.  In J. R. Meloy (Ed.), The psychology of stalking:  Clinical and forensic perspectives (pp. 113-137).  Toronto: Academic Press.

            Hamberger, L. K., & Hastings, J. E.  (1986).  Personality correlates of men who abuse their partners:  A cross-validation study.  Journal of Family Violence, 1, 323-341.

            Hamberger, L. K., Lohr, J. M., Bonge, D., & Tolin, D. F.  (1996).  A large sample empirical typology of male spouse abusers and its relationship to dimensions of abuse.  Violence and Victims, 11, 277-292.

            Hare, R. D. (1991). The Hare Psychopathy Checklist-Revised. Toronto, Ontario: Multi-Health Systems.

            Hare, R. D.  (1996).  Psychopathy:  A clinical construct whose time has come.  Criminal Justice and Behavior, 23, 25-54.

            Harmon, R. B., Rosner, R., & Owens, H. (1995). Journal of Forensic Reports, 40(2), 188-196.

            Holtzworth-Munroe, A., & Stuart, G. L.  (1994).  Typologies of male batterers:  Three subtypes and the differences among them.  Psychological Bulletin, 116, 476-497.

            Kaufman, G.  (1992).  Shame:  The power of caring.  Rochester, Vermont: Schenkman Books, Inc.

            Kienlen, K. K.  (1998).  Developmental and social antecedents of stalking. In J. R. Meloy (Ed.), The psychology of stalking:  Clinical and forensic perspectives (pp. 52-67).  Toronto: Academic Press.

            Kienlen, K. K., Birmingham, D. L., Solberg, K. B., O’Regan, J. T., & Meloy, J. R.  (1997).  A comparative study of psychotic and nonpsychotic stalking.  Journal of the American Academy of Psychiatry and the Law, 25, 317-334.

            Kileen, K., & Dunn, J.  (August, 1998).  Victim and law enforcement strategies for managing forceful interaction and coercive pursuit.  Paper presented at the annual Threat Management Conference, Los Angeles, California.

Kohut, H.  (1972/1978).  Thoughts on narcissism and narcissistic rage.  In P. H. Ornstein (Ed.), The search for the self:  selected writings of Heinz Kohut (vol. 2) (pp. 615-658).  New York:  International Universities Press.  (Originally published in 1972)

            Kong, R.  (1996).  Criminal harassment.  Juristat, 16 (12), 1-13.

            Lewis, M.  (1992).  Shame:  The exposed self.  Toronto: Maxwell Macmillan Canada.

            Lyon, D. R.  (1998). The characteristics of stalkers in British Columbia:  A statistical comparison of persons charged with criminal harassment and persons charged with other Criminal Code offences.  Master’s thesis, Department of Psychology, Simon Fraser University, Burnaby, British Columbia.

            Lyon, D. R., & Douglas, K. S.  (in press).  Violence against educators: The British Columbia Teacher’s Federation Violence Survey.  Vancouver:  BCTF.

            Manitoba Law Reform Commission.  (1997).  Stalking.  Winnipeg, Manitoba: Manitoba Law Reform Commission).

            Meloy, J. R.  (1989).  Unrequited love and the wish to kill:  Diagnosis and treatment of borderline erotomania.  Bulletin of the Menninger Clinic, 53, 477-492. 

            Meloy, J. R. (1996).  A clinical investigation of the obsessional follower: “She loves me, she loves me not . . .”. In L. Schlesinger (Ed.), Explorations in Criminal Psychopathology: Clinical syndromes with forensic implications. (pp. 9 - 32). Springfield, Ill: Charles C. Thomas.

            Meloy, J. R.  (1997).  The clinical risk management of stalking:  “Someone is watching over me …”.  American Journal of Psychotherapy, 51, 174-184.

            Meloy, J. R. (1998).  The psychology of stalking.  In J. R. Meloy (Ed.), The psychology of stalking:  Clinical and forensic perspectives (pp. 1-23).  Toronto: Academic Press.

            Meloy, J. R., & Gothard, S. (1995). Demographic and clinical comparison of obsessional followers and offenders with mental disorders. American Journal of Psychiatry, 152(2), 258-263.

            Mental Health Law and Policy Institute. (1997). Surrey Pretrial Services Centre (SPSC) Mental Health Intake Interview Questions and Coding Criteria. Burnaby, BC: Simon Fraser University, Mental Health Law and Policy Institute.

            Mullen, P. E., & PathJ, M. (1994). Stalking and the pathologies of love. Australian and New Zealand Journal of Psychiatry, 28, 469-477.

            Oldham, J., Clarkin, J., Appelbaum, A., Carr, A., Kernberg, P., Lotterman, A., & Haas, G.  (1985).  A self-report instrument for Borderline Personality Organization.  In T. H. McGlashan (Ed.), The borderline:  Current empirical research (pp. 1-18).  Washington, D.C.: American Psychiatric Press.

            Overall, J. E., & Gorham, D. R. (1962). The Brief Psychiatric Rating Scale. Psychological Reports, 10, 799-812.

            Robins, L. N., & Reiger, D.  (1991).  Psychiatric disorders in America.  New York:  Free Press.

            Romans, J. S. C., Hays, J. R., & White, T. K.  (1996).  Stalking and related behaviors experienced by counseling center staff members from current or former clients.  Professional Psychology:  Research and Practice, 27, 595-599.

            Sandberg, D. A., McNiel, D. E., & Binder, R. L.  (1998, August).  Characteristics of psychiatric inpatients who stalk, threaten, and harass hospital staff after discharge.  Paper presented at the annual convention of the American Psychological Association, San Francisco, CA.

            Saunders, D. G.  (1992).  A typology of men who batter:  Three types derived from cluster analysis.  American Journal of Orthopsychiatry, 62, 264-275.

            Saunders, R.  (1998).  The legal perspective on stalking. In J. R. Meloy (Ed.), The psychology of stalking:  Clinical and forensic perspectives (pp. 25-49).  Toronto: Academic Press.

            Schwartz-Watts, D., & Morgan, D. W.  (1998).  Violent versus nonviolent stalkers.  Journal of the American Academy of Psychiatry and the Law, 26, 241-245.

            Schwartz-Watts, D., Morgan, D. W., & Barnes, C. J.  (1997).  Stalkers:  The South Carolina experience.  Journal of the American Academy of Psychiatry and the Law, 25, 541-545.

            Spitzberg, B. H., Nicastro, A. M., et al.  (1998).  Exploring the interactional phenomenon of stalking and obsessive relational intrusion.  Communication Reports, 11, 33-48.

            Spitzberg, B. H., & Rhea, J.  (1999).  Obsessive relational intrusion and sexual coercion victimization.  Journal of Interpersonal Violence, 14, 3-20.

            Strauss, M. A.  (1979).  Measuring intrafamily conflict and violence:  The Conflict Tactics Scale.  Journal of Marriage in the Family, 41, 75-88.

            Strauss, M. A.  (1990).  Measuring intrafamily conflict and violence:  The Conflict Tactics Scale.  In M. A. Strauss & R. J. Gelles (Eds.), Physical violence in American families: Risk factors and adaptation to violence in 8,145 families (pp. 29-48). New Brunswick, NJ: Transaction Publications.

            Tweed, R. G., & Dutton, D. G.  (1998).  A comparison of impulsive and instrumental subgroups of batterers.  Violence and Victims, 13, 217-230.

            United States Department of Justice. (1997).  Domestic violence and stalking:  The second annual report to congress under the Violence Against Women Act.  Washington, D.C.: U.S. Department of Justice Office of Justice Programs.

            Wright, J. A., Burgess, A. G., Burgess, A. W., Laszlo, A. T., McCrary, G. O., & Douglas, J. E. (1996). A typology of interpersonal stalking. Journal of Interpersonal Violence, 11(4), 487-502.

            Zona, M. A., Sharma, K. K., & Lane, J. (1993). A comparative study of erotomanic and obsessional subjects in a forensic sample. Journal of Forensic Sciences, 38(4), 894-903.

Author Notes

 

1 Kevin S. Douglas. Department of Psychology and Mental Health, Law, and Policy Institute, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia, Canada, V5A 1S6. Phone: (604) 291-5945. Electronic Mail: kdouglas@arts.sfu.ca.

 

2 Donald G. Dutton. Department of Psychology, University of British Columbia.

 

 

Table 1

Select Characteristics of Stalkers

 

 

 

 

 

 

 

 

 

 

 

 

Study

N

Age

(M)

Male stalker

Female victim

Intimate partner

Use of threats

Use of violence

Substance use

Other Axis I

PD

Cluster B Dx or traits

 

 

 

%

%

%

%

%

%

%

%

%

BC AG (1995)

100g

na

89

85

71

78

0 – 42h

na

na

na

na

Burgess et al. (1997)

36d

na

83

na

100i

na

100i

50

na

na

na

Coleman (1997)

13e

26

na

na

100j

xk

xk

na

na

na

na

Department of Justice (1996)

601

37

91

88

57

19

15

24l

14l

na

na

Fremouw et al. (1997)

593

na

na

62

60

na

na

na

na

na

na

Harmon et al. (1995)

48

40

67

64

13

46

21

2

75

19

na

Kienlen et al. (1997)

25

38

84

88

58

76

32

na

78

44

36o

Kileen & Dunn (1998)

128

30-39

99

na

65

75

52

22-33

na

na

na

Kong (1996)

7462

34

88

80

49

24 n

25n

na

na

na

na

Lyon (1998)

54

32

100m

na

na

48n

49n

55

na

na

na

Manitoba L.R.C. (1997)

38

na

na

na

61

na

na

na

na

na

na

Meloy & Gothard (1995)

50c

35

90

na

55

70

25

70

65

85

60

Mullen & PathJ (1994)

14

40

79

79

na

36

79

na

100

100

na

Romans et al. (1996)

10f

na

< 50

40

na

na

na

na

na

60

na

Sandberg et al. (1998)

17

38

82

na

na

na

na

na

71

47

na

Spitzberg & Rhea (1999)

na

na

na

na

na

na

na

na

na

na

na

Spitzberg et al. (1998)

44

na

na

na

na

na

na

na

na

na

na

Schwartz-Watts et al.  (1998)

42

35

95

na

67

na

48

60

36

na

na

Schwartz-Watts et al. (1997)

18

34

100

na

na

na

na

61

44

na

na

U.S. Dept. of Justice (1997)

16000

na

60/94

79

59a

na

na

na

na

na

na

Wright et al.  (1996)

30

na

na

na

53

na

na

na

na

na

na

Zona et al. (1993)

74

na

74

74

47b

45

3

9

45

na

na

 

 

 

 

 

 

 

 

 

 

 

 

 

Notes.

a This is the figure for female victims.  The figure for male victims was 32%.

b This figure indicates that 47% of stalkers knew their victims, and that this was comprise “mostly” of former intimates.

c This was the size of the total sample.  Of these, 20 were stalkers.

d The total sample size was 120, 36 of whom were stalkers.

e The total sample size was 141, 13 of whom were stalked.

f The total sample size was 178, 10 of whom were stalked.

g There were 100 cases involving 102 victims and 101 offenders.

h Zero percent was for the professional group, and 42% for the family member group.  For the intimate partner group, 41% experienced physical violence.

i The prevalence is 100% because the sample consists of persons charged with felony domestic violence.

j The prevalence is 100% because this study focussed on stalking by former partners.

k Percent of victims experiencing violence was not reported.  However, victims of stalking filled out the CTS for their partners, and reported mean scores of 23.5 and 63.4 for the Physical Violence, and Verbal Violence subscales, respectively.  These mean scores were significantly greater than for the control group and a group of harassed but not stalked persons (the Cohen’s d averaged 1.59 for differences between groups on verbal violence, and 1.44 for physical violence, both very large effect sizes).

l These figures may be underestimates, as all information was collected from police files.  Further, the figures apply only generally to “mental” or “psychological” problems that were noted in police files, rather than diagnoses.

m This stems from the remand centre handling only males.

n Based on associated criminal charges.

o 9 of 11 (82%) people with personality disorder had Cluster B diagnoses.

Table 2

 

Typologies of Domestic Abusers

 

 

 

 

 

Group

 

 

 

Psychopathic (generally violent)

 

 

Borderline

(cyclical)

 

Avoidant (overcontrolled)

 
 
Source

 

 

 

 

Hamberger and Hastings (1986)

 

 

Antisocial/ Narcissistic

 

Schizoid/ Borderline

 

Dependent/ Compulsive

Hamberger , Lohr, Bonge, & Tolin (1996)

 

Generally violent/ Antisocial

Negativistic-Dependent

[Family only/ nonpathological]

Holtzworth-Munroe and Stuart (1994)

Generally violent/ antisocial

Dysphoric/ Borderline

Passive-aggressive (family only)

 

Gottman et al. (1995)

 

Type 1 (generally antisocial and violent)

 

Type 2 (dependent/ unstable)

N/A

Saunders (1992)

Type 2 (generally violent

Type 3 (emotionally volatile)

Type 1 (emotionally suppressed)

 

 

 

 

 

Attachment Style

 

 

 

 

Dismissing

Fearful

Preoccupied

 

 

 

 

Risk for Stalking

 

 

 

 

Low

High

Moderate

 

 

 

 

 

Note.  The typology labels and several of the rows are adapted from Dutton (1998; Table 1.1)

 

 

 

 

 

Table 3

 

Points of Overlap between Conceptualizations of Domestic Stalking and the Borderline/Cyclical Type of Batterer

 

 

*      Borderline personality organization

 

*      Violence and threats

*      Cluster B personality traits

*      Jealousy

*      Low prevalence of antisocial personality

*      Substance abuse

*      Mood disorders (dysthymia, depression)

*      Abandonment rage

*      Attachment dysfunction (fearful, preoccupied)

*      Avoidant and dependent traits

*      Early childhood trauma (shaming experiences)

*      Anger

*      Emotional instability and volatility

*      Dependence

*      Fantasies and sense of entitlement

*      Intrusiveness

*      Primitive defenses (splitting; projection)

 

 

 

 

 

Figure 1

Proposed Relationship between Stalking and Domestic Violence

 

 

 

 

Note.  This figure is adapted and modified from Dutton (1995, 1998).
 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 


Endnotes

 


[1] These figures describe people who “self-defined” as stalking victims, after reading a definition that comported with the stalking law in the particular jurisdiction.

[2] The subtype was not specified.

[3] These figures were not the response rate.  Approximately 71% of the 16000 people actually responded to questions.

[4] It should be pointed out that, while psychopathic batterers may use instrumental aggression, psychopaths typically are regarded as impulsive (see Hare, 1991, 1996).  Their impulsivity, however, is not marked by a heightened affective experience, but rather by a deficient “behavior inhibition system.”  That is, psychopaths are unable to learn from past errors to the same extent as non-psychopaths, and are quick to act as a result.

[5] Correlational indices ranged from .29 to .62.  According to Cohen (1992), a moderate effect size, in correlational terms, is ± .30, and a large effect size is ± .50.

[6] Nonpsychotic subjects pursued ex-intimates 71% of the time (n = 12 of 17), whereas psychotic stalkers pursued ex-intimates 38% of the time (n = 3 of 8).  Due to low power, this difference is not significant, and caution should be exercised in generalizing it.

[7] This is again assuming that the main characteristics of domestic stalkers can be inferred from the characteristics of stalkers more generally, based upon the reason that the largest group of stalkers is domestic stalkers.

[8] The reader is reminded that research seems to suggest that antisocial personality disorder may be less prevalent among stalkers than other Cluster B disorders, or as among other offenders.