Identifying Stalking: The Relevance of Intent in Commonsense ReasoningAuthor(s): Susan M. Dennison and Donald M. ThomsonSource: Law and Human Behavior, Vol. 26, No. 5 (Oct., 2002), pp. 543-561Published by: SpringerStable URL: http://www.jstor.org/stable/1394526Accessed: 11/05/2010 13:39

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=springer.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

Springer is collaborating with JSTOR to digitize, preserve and extend access to Law and Human Behavior.

http://www.jstor.org

Law and Human Behavior, Vol. 26, No. 5, October 2002 (? 2002)

Identifying Stalking: The Relevance of Intent in Commonsense Reasoning

Susan M. Dennison1,3 and Donald M. Thomson2

One method of distinguishing stalking from law-abiding behavior is to determine whether the accused intended to cause fear or harm to the target. However, this distinc- tion may not capture community concerns regarding intrusive or harassing behavior. The present research examines the effect of intent, persistence, relationship, and con- sequences on community perceptions of stalking. Responses of 1,080 members of the community to a series of scenarios indicated that the presence of explicit evidence of intent was not the only way stalking behavior was identified. Behavior was also iden- tified as stalking as a greater degree of persistence was depicted. Females more often than males perceived the behavior as stalking and inferred intent to cause fear or harm. Most participants who identified the behavior as stalking also indicated that it should be illegal. These results may assist in guiding ongoing debates over appropriate stalk- ing legislation and strategies to reduce the incidence of stalking, as well as indicating whether education regarding stalking laws is required.

KEY WORDS: stalking; legislation; intent; public perceptions; commonsense justice.

INTRODUCTION

A decade ago the crime of stalking came on to the world stage, and yet a clear

dividing line between stalking and acceptable behavior has never been established. This is mainly because in some contexts everyday behaviors such as telephoning and

sending gifts can be construed as stalking. Although stalking is increasingly acknowl- edged as a serious crime and public health issue (Violence Against Women Grants Office, 1998), research aimed at guiding the development of stalking policies has been lacking and the question is left begging as to whether the new laws are meeting community needs. As a way to advance the debate over an appropriate definition, this research examined the conditions under which the community perceives behavior

1 School of Criminology and Criminal Justice, Griffith University, Queensland, Australia. 2School of Social Sciences and Liberal Studies, Charles Sturt University, Bathurst 2075, Australia. 3To whom correspondence should be addressed at School of Criminology and Criminal Justice, Griffith University, Mt Gravatt Campus, Queensland 4111, Australia; e-mail: susan.dennison@mailbox.gu.edu.au.

543 0147-7307/02/1000-0543/1 ? 2002 American Psychology-Law Society/Division 41 of the American Psychology Association

Dennison and Thomson

as stalking. The paper discusses whether community sentiments match definitions of unacceptable behavior provided by stalking laws.

Criminal Law and Stalking Legislation

At the very foundation of most criminal law is the notion that two elements must be present for a crime to have been committed: actus reus, that certain behaviors took place; and mens rea, that the accused intended to bring about a particular outcome by engaging in those acts (Findlay, Odgers, & Yeo, 1999). A third element that often comes into play when determining a criminal charge is the consequences to a victim (e.g., a person cannot be charged with murder if the victim did not die). While actus reus is not overly difficult to establish if the acts in question took place, establishing mens rea is more difficult given that a person's intentions can be less obvious. Nevertheless, intent can often be inferred by the actions that have taken place, the circumstances surrounding the event, and by considering objectively, what a reasonable person could have foreseen as a likely outcome. While most stalking laws were initially enacted containing elements of actus reus and mens rea, there has been considerable debate regarding the necessary elements of stalking.

What typically sets stalking aside from acceptable behavior are aspects such as the repetitive nature of the behavior and content of communications (actus reus), the intent to instil fear or cause harm (mens rea), and in some cases, the outcome to the recipient. Although most commentators agree that the behavior should be repetitive, debates continue over whether laws should require that the stalker intended to cause fear or harm, or that the victim experienced fear or harm.

In the United States in 1998, 32 states included a requirement of intent to instil fear in their stalking provisions (Violence Against Women Grants Office, 1998). Of those that did not, 14 states required that the conduct constituting stalking be done purposefully. This left only four states that did not require some proof that the behavior was intentional. Twenty-six states required that the victim fear death or bodily injury, five states referred to fear of physical safety, and another 14 states protected against other fears, including emotional distress. Only six states did not require that the stalking arouse fear (Violence Against Women Grants Office, 1998).

In England and Wales, the Protection from Harassment Act (1997) created two criminal offences to address not only stalking, but also other forms of harassment, including bullying and harassment by neighbors. The term "stalking" has no legal status in England and Wales (Budd & Mattinson, 2000). The first offence within the Act is that of putting people in fear of violence (Sec. 4) and the second is a summary offence for causing harassment or distress (Sec. 2). Neither offence requires proof of intent to cause fear or distress (Budd & Mattinson, 2000).

Although the requirement of intent was initially present in all Australian stalk- ing statutes, some jurisdictions have since eliminated requirements of intent to cause fear or harm (i.e., Criminal Law Amendment Act [No. 1] 1998 [WA]; Criminal Code [Stalking] Amendment Act 1999 [QLD]). Instead, the fundamental element is whether fear or apprehension would reasonably arise in all the circumstances (determined both subjectively and objectively). In some Australian jurisdictions it is also immaterial whether fear or harm is actually caused. An example of the basic

544

Identifying Stalking: Commonsense Reasoning

criteria in stalking legislation is included in the Appendix, depicting a strict intent requirement, an intent requirement that allows for foreseeability, or no intent re- quirement.

In sum, there are three elements usually addressed in stalking legislation, these being the conduct itself, intent to instil fear or cause harm or the foreseeability of doing so, and the consequences to the target. However, there are clearly some ju- risdictions that are concerned that the legislation is not wide enough to meet public needs and have eliminated the burden of establishing intent. The dilemma facing law makers is that laws that require intent to cause fear or harm may not capture a wide range of behaviors that may be perceived as stalking (see, e.g., Ogilvie, 2000). How- ever, broad laws that rely almost solely on the victim's perception of the events have the potential to be misused (see, e.g., Harris, 2000), similar to apprehended violence orders where there are concerns that they are being sought over trivial matters (New South Wales Attorney General's Department, 1999). Legislation devoid of mens rea is also at odds with traditional measures of criminal responsibility (He Kaw Teh, 1985). If the problem of definition is to be addressed, an examination of community perceptions of stalking is one tool to enlighten potential law reform.

Perceptions of Stalking: Commonsense Reasoning

Asking community members whether particular behaviors are examples of stalk- ing fits closely with what Finkel (1995) calls "commonsense justice." He describes it as "what ordinary people think is just and fair... it is what ordinary people think the law ought to be" (Finkel, 1995, p. 2). This is best evidenced in cases of jury nul- lification, when the jury fails to follow the law in reaching a verdict (Finkel, 2000a). Finkel (1997, 2000a, 2000b) posits that understanding commonsense justice provides a mechanism to reform law and public policy and curb prejudices and biases that taint commonsense justice and channel community sentiments to accord with the law. Many commonsense justice researchers suggest that there are times when the law should follow community sentiment, and times when the community might best be served by being guided by the law (Boekmann & Tyler, 1997; Finkel, 2000a, 2000b; Haney, 1997; Horowitz, 1997). A question that reoccurs throughout this paper is whether shaping stalking laws to reflect community sentiment would lead to poten- tial injustices? Regardless of the final answer, community sentiment about stalking can assist in understanding the extent to which stalking laws reflect commonsense justice and the potential problems in community members applying the law as ju- rors. An assessment of community sentiment will also help to identify the need for a review of stalking laws.

Jury nullification can occur in large part because people have prototypes of par- ticular crimes and they may draw on these prototypes when serving on a jury. In some instances, these prototypes may differ to the elements that define the crime (Finkel, 2000a) and may not always be commonsensical (Finkel & Sales, 1997). Stalking leg- islation was initially enacted in the United States amongst a media frenzy in which a popular young actress was killed by her stalker and other celebrities complained of harassment by overexuberant and sometimes dangerous fans (Anderson, 1993; Sohn, 1994). Similarly there was an intense media role in the introduction of stalking

545

Dennison and Thomson

in Australia, Canada, and the United Kingdom (Goode, 1995; Sheridan, Davies, & Boon, 2001b; Way, 1994). Given the exposure of the public to accounts of stalking, real or otherwise, it is likely that prototypes exist in the minds of community mem- bers as to what stalking is, what types of people engage in stalking, and what types of people are victims. But prototypes are not only shaped by the media, legislative enactments may also play a role (Finkel, Burke, & Chavez, 2000; Finkel & Sales, 1997). In the case of stalking, new criminal laws might have been interpreted as a sign that there was a real danger of becoming a stalking victim. Media reports of judi- cial rulings may also frustrate the community, when a person is acquitted of stalking for lack of intent when the behavior is seemingly harassing and detrimental to the victim (Burstin, 1997; McMahon, 1999).

It is possible that community members pay greater attention to the consequences to the victim than the intentions of the alleged stalker. In a study on community justice reasoning about deserved punishment for the crimes of robbery and rape, Hills and Thomson (1999) found that the consequences to the victim were a focus for participants in reasoning about their decisions, but the intentions of the offender were not mentioned. It is not known whether a similar process takes place when determining guilt for stalking.

While Mullen, Pathe, and Purcell (2000) suggest that what constitutes stalking should be dictated by common sense, they also acknowledge that common sense is derived from common values of privacy and freedom from harassment. These common values may not exist for stalking. Indeed, Haney (1997) argues that it is "in- creasingly problematic to assume that there is one popular commonsense worldview, and that there exists anything remotely resembling one 'people' who can be said to see legally relevant matters in even remotely similar ways" (p. 310). In grappling with the definition of stalking it is important to examine what is perceived as stalk- ing, and if perceptions are so varied that a widely accepted definition of stalking is impossible.

Stalking Surveys and Empirical Work

A national survey in the United States by the National Institute of Justice (NIJ; Tjaden & Thoennes, 1998), and the stalking questionnaire in the 1998 British Crime Survey (Budd & Mattinson, 2000) found that women (8-16%), more so than men (2-7%), reported being stalked at some stage in their lifetime. Whether the term "stalking" incorporated a presence of fear played an important role in reporting stalk- ing. The reported prevalence of stalking increased as the criteria for fear decreased in the definitions provided in the NIJ survey. Neither of these surveys included a requirement that the offender intended to cause fear or harm.

The British Crime Survey also revealed that men perpetrated 81% of reported incidents. In 29% of all cases the offender was an ex-intimate, 34% involved a stranger, and the remaining cases involved an acquaintance of some sort. Female victims were more likely to report that the offender wanted to start or continue a relationship with them, whereas male victims were more likely to say the offender intended to annoy or upset them (Budd & Mattinson, 2000). It may be that men and women are subjected to different forms of harassment or unwanted attention or, that

546

Identifying Stalking: Commonsense Reasoning

men do not perceive the unsolicited pursuit of a relationship as stalking. Potential differences in the way men and women perceive stalking has implications for the incidence and prosecution of stalking.

Despite the important insights that could be gained, very little research has been directed toward understanding "commonsense" notions of stalking. The research conducted thus far is confined to the United Kingdom and Australian populations. Sheridan, Davies, and Boon (2001a) examined perceptions of stalking in a female sample in the United Kingdom, in which participants were asked to rate 42 intrusive behaviors on the degree to which they were "stalking." Sheridan et al. concluded that the female respondents were able to distinguish between overbearing courtship and obsessive and potentially dangerous behavior, although participants were given a broad definition of stalking. Further research conducted by Sheridan and Davies (2001) showed that British people have a definition of stalking in line with harassment laws in England and Wales, which do not require that the offender intended to cause fear or harm to a person.

Hills and Taplin (1998) manipulated the presence or absence of a threat and the accused-target relationship in a heterosexual stalking scenario, conducted in Perth (Western Australia). Participants more often reported the likelihood of feeling frightened and of calling the police when placed in the scenario of being stalked by a stranger in comparison to an ex-intimate. However, upon close examination of the scenarios involving a male stalker, there were no differences in female responses across the relationship conditions. The likelihood of calling the police when a threat was either absent or present was equal for females, whereas males were more likely to call the police in the threat condition than the no-threat condition.

Another Australian study examined the effect of explicit evidence of intent to cause fear or harm, consequences to the target, and the relationship between the ac- cused and target on community perceptions of stalking (Dennison, 2000; Dennison & Thomson, 2000). Even when there was no explicit evidence of intent to cause fear or harm to the target, and even when the target did not in fact experience any fear or harm, but rather an invasion of privacy, the behavior was still perceived as stalking. One explanation of these findings is that a high degree of persistence was depicted in the scenario. Therefore the repetitive nature of the unwanted attention may play an important role in identifying stalking.

From the limited research conducted to date, some factors have emerged that appear relevant in commonsense judgements of stalking. These factors include the presence of threats or intent to cause fear or harm, degree of persistence, and the level of fear or distress experienced by the recipient. These three factors accord with the core elements of most stalking laws. Additional factors that seem to affect commonsense notions of stalking are whether the recipient of the unwanted attention is male or female and possibly the relationship between the stalker and victim. In order to examine these factors more closely this study was conducted.

The experimental design is a 2 x 2 x 3 x 3 x 2 between-subjects factorial de- sign. The independent variables (IVs) are (1) alleged stalker's intentions-explicit evidence of intent, no explicit evidence of intent; (2) persistence-moderate, low; (3) consequences to the target-extreme fear, moderate fear, no fear; (4) relation- ship of alleged stalker to the target-stranger, ex-intimate, acquaintance; and (5) sex

547

Dennison and Thomson

of participant. The first four variables were controlled in this experiment; however, the fifth variable of sex was not, with more females participating in the study.

The research aims to develop a better understanding of when behavior is per- ceived as stalking. Four questions are addressed in this paper. First, which conditions are necessary and sufficient for behavior to be identified as stalking and by whom it is perceived as stalking? Based on previous research it is hypothesized that more females than males will perceive the behavior as stalking and that perceptions of stalking will be greater when there is explicit evidence of intent, when persistence is moderate rather than low, and when the consequences to the target are extreme fear. No hypotheses are made in relation to the offender target relationship, as the findings of Hills and Taplin (1998) are equivocal. It is not known how each of these factors interacts with each other to affect perceptions of stalking.

The second question is, in identifying behavior as stalking, do participants also believe the behavior should be illegal? Based on Finkel's and others work on com- monsense justice (Finkel, 1995, 2000a, 2000b), particularly that of jury nullification, it is hypothesized that where behavior is perceived as stalking, participants will also respond that it should be illegal, regardless of whether their perceptions correspond with "black letter" law.

The third question is, which conditions are necessary and sufficient for the per- ception of intent to arouse fear or apprehension? Finally, which conditions are nec- essary and sufficient for the perception of intent to cause mental or physical harm? It is hypothesized that in both these cases, perceptions of intent will be greater when there is explicit evidence of intent and when persistence is moderate rather than low. No hypotheses are made in relation to the effect of consequences and relationship. How perceptions of intent are related to perceptions of stalking will also be inves- tigated. Questions three and four were structured as two separate questions, given that in some jurisdictions, particularly Australian, intent is deemed to be present if the accused intended to arouse fear or apprehension, or cause mental or physical harm. As legislation does not define mental harm, it is not known whether fear or apprehension is perceived differently to the mental or physical harm criteria. As well as the main theme of identifying stalking, a secondary theme is to examine how community members might apply the intent requirement.

METHOD

Participants

There were 1,080 participants in the study, 30 participants in each of the 36 mani- pulated conditions. The 1,080 participants comprised 383 males and 685 females (a further 12 participants did not indicate their sex). The ages of participants ranged from 18 to 87 years. The mean age of males was 33.88 years (SD = 12.83), and the mean age for females was 30.68 years (SD = 12.16). All participants were members of the Melbourne (Victoria, Australia) community and were approached in public spaces such as shopping malls and pedestrian areas and invited to participate in the study. All participants were treated in accordance with the ethical guidelines of the National Health and Medical Research Council.

548

Identifying Stalking: Commonsense Reasoning

Materials

Participants were given a 4-page scenario, which depicted a male accused of stalking a female target over a period of 5 months. The scenario describes the ac- cusations made by the target as contained in a police report, a summary of the prosecution's case during the trial, and a summary of the defendant's case presented at the trial. This scenario had been used in Dennison and Thomson (2000), but in the present study, alongside the manipulations of intent, consequences, and relationship, an additional variable of persistence was manipulated. The level of persistence de- picted in the previous study was deemed to be high, with responses at ceiling level. Because of this, the present study reduced the amount of persistence depicted and defined these as moderate or low.

In the moderate persistence condition, the target believes she has been followed to work approximately 10 times during the 5-month period. The target complains that every other Friday the accused would appear behind her while she was walking, either a few steps behind her, from a distance of about 10 m, or parallel to her on the opposite side of the street.

The target also claimed that the accused had made four (silent) hang-up phone calls to her in a month. She received two hang-up calls in the second week of that month, a third in the third week, and one more in the fourth week of that month. A caller identification function on the target's phone indicated the phone number of the caller, which was later identified as being the accused's phone number. The target also reported that on three occasions the accused appeared at the same restaurant or cafe at which she was eating. On one occasion when the target attended a movie by herself, the accused also attended the same movie, sitting two rows behind her.

The low persistence condition differed from the moderate persistence condition in that the target did not receive any hang-up phone calls. The seemingly subtle difference between the moderate and low persistence conditions was deliberately chosen to determine if there was a notable difference in perceptions of stalking when the behavior was reduced to potentially coincidental acts (i.e., walking the same route to work once a fortnight and appearing in the same cafe and movie theatre). Hang-up phone calls that can be traced to the accused might be perceived as particularly invasive as it occurs in the targets home and cannot be construed as coincidental.

There were two versions of intent, one where there is explicit evidence of intent to create fear or apprehension, or mental harm or physical harm to the target, and one where this explicit evidence is absent. Explicit intent is depicted by a threatening message on the target's answering machine-"If you don't give me a chance then there will be trouble. You owe me at least that much"-and the discovery of the accused's diary that details the movement of the target, and the dates and times that he had watched her. It also contained an entry made by the accused that stated "I wonder how long I have to follow her until she becomes frightened of me. She's starting to look nervous already. That will teach her for rejecting me." All other behavior by the accused remained the same as that in the no-intent condition.

For consequences to the target, in the extreme fear condition, the target is ter- rified by the behavior, increases personal security, is unable to function properly at

549

Dennison and Thomson

work, and makes a number of life changes. In the moderate fear condition, the target becomes increasingly frightened and increases personal security. In the no fear con- dition, the target experiences annoyance and frustration by the alleged stalker's per- sistent behavior but indicates that she is not afraid of him and experiences feelings of resentment at the invasion of privacy.

The accused is depicted as being either (1) a stranger who bumped into the target in the street and subsequently contacted her and invited her out, to which she declined; (2) an ex-partner who the woman had dated for 12 months; the woman felt that he was possessive and broke up with him, there was no violence involved in the relationship; (3) an acquaintance who worked in the same office as the target, the accused had invited her out and she had declined. The harassing behavior begins after the relationship break-up in the ex-intimate scenario, or in the case of the stranger and acquaintance, after the offer for a date was declined.

Participants completed a questionnaire on their perceptions of the events in the scenario. As well as demographic information the items in the questionnaire included (1) whether the behavior described in the scenario was stalking (No/Yes); (2) whether it should be illegal for someone to behave the way the accused did (No/Unsure/Yes); (3) whether the accused intended to create fear or apprehension in the target (No/Unsure/Yes); and (4) whether the accused intended to cause mental harm or physical harm to the target (No/Unsure/Yes).

Procedure

Participants were supplied with an information sheet that explained the aim of the study and the tasks involved, a scenario describing the alleged stalking event, and a questionnaire. The task required approximately 20-30 min to complete. Partic- ipants were randomly assigned to read one of the 36 vignettes. Participants were also supplied with a reply-paid envelope to return the completed questionnaire. Two thou- sand, seven hundred and fifty questionnaires were distributed in the Melbourne com- munity, 1,145 were returned, providing a response rate of 42%. As 1,080 completed questionnaires were required for this study, once each cell had 30 participants in it, any remaining questionnaires were not included in the analysis. There were no notable differences between those questionnaires that were discarded and those in- cluded in the analysis. Because of the random allocation of questionnaires, it was sometimes the case that scenarios were handed out despite the cell for that scenario being filled the same day. An effort was made to minimize missing data by replac- ing any questionnaires containing three or more unanswered questions with spare, returned questionnaires.

RESULTS

Results were analyzed using SPSS for Windows. Frequencies in some analyses do not add up to 1,080 as some participants failed to respond to certain questions. To examine the effects of each of the IVs on each of the dependent variables logit log-linear analysis was conducted. Because of the capacity of the program, the logit

550

Identifying Stalking: Commonsense Reasoning

analysis failed to run with five IVs. Therefore, to screen for differences between males and females, the consequences conditions were combined (reducing the analysis to four variables) and a saturated model was run. The analysis was then run again combining the relationship conditions and including the consequence variable in the analysis. Once differences according to sex had been examined, male and female responses were then combined and logit analyses were then conducted on the four manipulated variables. (Note, z = 1.96 or greater is significant at p < .05).

The analysis sought to first address which conditions in the scenarios were suf- ficient to return a judgement that the behavior was stalking. Main effects of sex, intent, and persistence and an interaction between persistence and relationship were found. There also appeared to be a nonsignificant interaction between intent and per- sistence. Second, whether participants believed the behavior should be illegal was examined. There was a high level of correspondence between perceptions of stalking and judgements of what the legality of the behavior ought to be. Third, the analysis sought to determine under which conditions the accused was perceived to intend to create fear or apprehension. There were main effects of sex, intent, and relationship and a two-way interaction between intent and relationship. There were also three- way interactions between persistence, consequences, and intent, and persistence, consequences, and relationship. Participants were less likely to see the behavior as stalking if they believed the accused did not intend to create fear or apprehension.

Finally, an analysis was conducted to determine the conditions that affected perceptions of intent to cause mental or physical harm. There were main effects of sex, intent, persistence, and relationship. There were also significant interactions between consequences and relationship and intent and relationship, as well as a three- way interaction between persistence, intent, and relationship. Participants were less likely to see the behavior as stalking when they believed the accused did not intend mental or physical harm. The direction of these findings is explained in further detail in the subsequent section.

Which Conditions are Sufficient for Identifying Stalking? In total, 83% of the participants perceived the behavior of the accused to be

stalking. There was a main effect of sex (z = 3.21; z = 3.61), with more females re- sponding that the behavior was stalking (86.2%) compared with males (78.0%). A nonhierarchical logit analysis was conducted, using the four manipulated variables. The model produced a good fit between observed and expected frequencies, like- lihood ratio G2(24) = 20.39, p = .675. Accuracy in prediction from the model was reasonable, with 26% of the variation of responses predicted from the model.

There were main effects of intent (z = 8.81) and persistence (z = -3.18), al- though there also appeared to be an interaction between these two variables. Al- though the logit analysis did not reveal a significant interaction (z = -1.59), further analysis revealed some noteworthy differences. When explicit evidence of intent was present there was no difference in perceptions that the behavior was stalking be- tween the moderate (97.4%) and low persistence (96.3%) conditions, with almost all participants identifying the behavior as stalking, X2(1, N = 539) = 0.56, p = .46. When there was an absence of explicit evidence of intent there was a difference across

551

Dennison and Thomson

the persistence conditions, X2(1, N = 525) = 50.12, p < .01. Perceptions of stalking were reduced in the moderate persistence condition (83.3%) from when intent was present and in the low persistence condition (54.8%), such that only about half of the participants in this latter condition identified the behavior as stalking. Therefore an effect of persistence occurs only in the absence of explicit evidence of intent; when intent is present this is sufficient to perceive the behavior as stalking, regardless of persistence.

There was an interaction between persistence and relationship (z = 2.22). With moderate persistence, it made very little difference what the relationship was, whereas when there was low persistence, the pattern is a reduction in the percentage of persons who classified the behavior as stalking for the stranger, 89.4 vs. 79.5; X2(1, N = 355) = 6.56, p = .01, and the acquaintance, 92.4 vs. 64.8; X2(1,N = 353) = 41.00, p < .01. For the ex-intimate there was no significant reduction in classifying the low persistent behavior as stalking, 89.3 vs. 83.1; X2(1, N = 356) = 2.86, p = .09. The reduction in classifying low persistent behavior as stalking was much greater for the acquaintance condition.

Should the Behavior be Illegal? The results show that what was considered illegal behavior corresponded well

with classifications of stalking. Of the 883 participants who said that the behavior was stalking, 82.7% said it should be illegal, 14.2% were unsure, and 3.2% said that it should not be illegal. Of the 176 participants who did not perceive the be- havior as stalking, 5.1% said it should be illegal, 22.2% said they were unsure, and 72.7% said that it should not be illegal. These differences were significant, X2(2, N = 1059) = 614.54, p < .001. The results for the group not seeing the be- havior as stalking are intriguing. While a small percentage said that the behavior should be illegal, almost one quarter was unsure, even though they did not iden- tify the behavior as stalking. Perhaps participants were unclear as to what behavior constituted stalking, but nevertheless perceived the behavior as unacceptable and possibly warranting legal intervention.

Did the Accused Intend to Create Fear or Apprehension? Almost half (46.5%) of the participants perceived that the accused intended

to create fear or apprehension to the target, with 31.5% responding that they were unsure and 22% indicating that the accused did not intend to create fear or ap- prehension. There was a main effect of sex (z = 2.13, z = 1.88), with more females responding that the accused intended to arouse fear or apprehension in the target (48.6%) compared with males (42.8%). A nonhierarchical logit analysis was then con- ducted. The model produced a good fit between observed and expected frequencies, likelihood ratio G2(40) = 51.29, p = .109. Accuracy in prediction from the model is reasonable, with 18% of the variation in perceptions of intent to arouse fear or apprehension predicted by the model.

There was a main effect of intent (z = 9.10; z = 4.13) and offender-target rela- tionship (z = -2.13). However, there was also a significant interaction between these

552

Identifying Stalking: Commonsense Reasoning

variables (z = -2.20). When explicit evidence of intent was present, there was no significant difference between the relationship conditions, X2(4, N = 538) = 4.73, p = .32. Responses for the ex-intimate, stranger, and acquaintance conditions, re- spectively, were 76.4, 67.2, and 71.1%. When explicit evidence of intent was absent, more participants in the ex-intimate condition responded that the accused intended to create fear or harm (31.3%) than those in the stranger (16.8%) and acquaintance conditions (16.1%), X2(4, N = 538) = 24.04, p < .01. Not surprisingly, perceptions of intent to arouse fear or apprehension were significantly reduced when explicit evidence of intent was absent compared with when it was present.

There was also a complex interaction between persistence, consequences, and intent (z = 2.16; z = -2.03) and also between persistence, consequences, and re- lationship (z = -2.00; z = 2.27). These interactions are difficult to interpret as no consistent patterns emerged. Given that the interactions do not contribute in a mean- ingful way to an understanding of how the behavior was perceived, the interactions will not be discussed here.

In terms of the correspondence between perceptions of stalking and intent, sig- nificantly fewer participants responded that the behavior was stalking when they believed that the accused did not intend to create fear or apprehension, X2(2, N =

1060)= 256.22, p < .001. Of the 496 participants who responded that the accused did intend to create fear or apprehension in the target, 98.6% classified the behavior as stalking. Of the 331 participants who were unsure whether the accused intended to create fear or apprehension, 82.8% classified the behavior as stalking. Only 51.1% of the 233 participants who responded that the accused did not intend to arouse fear or apprehension in the target classified the behavior as stalking. Therefore when par- ticipants do not perceive that the accused intended to create fear or apprehension in the target, this leads to a reduction in the number of number of participants who identify the behavior as stalking. However, given the number of participants who were unsure whether intent was present and yet still identified the behavior as stalk- ing, and that approximately 50% still identified stalking when they did not perceive intent, it seems that for many people, the intention to create fear or apprehension is not a necessary feature of stalking.

Did the Accused Intend to Cause Mental or Physical Harm?

Only 29.8% of participants responded that the accused intended to cause mental or physical harm to the target, with 37.9% indicating that they were unsure whether this intent was present and 32.3% indicating that this intent was not present. These responses show a marked difference from perceptions of intent to create fear or apprehension, with fewer participants perceiving intent to cause mental or physical harm. Of those who perceived intent to create fear or apprehension, only 59.2% also perceived intent to cause mental or physical harm. There were very few participants (7.5%) who perceived intent to cause harm but do not perceive intent to arouse fear or apprehension.

There was a main effect of sex (z = 2.84; z = 2.82), with more females respond- ing that the accused intended to cause mental or physical harm to the target (32.3%) compared with males (25.1%). A nonhierarchical logit analysis was conducted and

553

Dennison and Thomson

the model produced a good fit between observed and expected frequencies, like- lihood ratio G2(42) = 36.79, p = .699. Although the model produced reasonable frequencies, accuracy in prediction from the model is relatively low, with only 10% of the variance in responses predicted by the model.

There was a main effect of persistence (z = -3.78). More participants in the low persistence condition (37.3%) responded that the accused did not intend to cause harm to the target compared with those in the moderate persistence condition (27.3%). In the low persistence condition 35.3% were unsure and 27.5% indicated that this intent was present. Similarly, when the conduct consisted of moderate persis- tence 40.6% were unsure and 32.1% indicated that intent to cause harm was present. The degree of persistence appears to contribute some evidence that the accused intended to cause mental or physical harm.

There was also a main effect of intent (z = 9.85; z = 3.45) and relationship (z = -2.20) and a significant Intent x Relationship interaction (z = -2.39). When explicit evidence of intent was present there was no significant difference in re- sponses that the accused intended to cause mental or physical harm to the target across the stranger (46.4%), ex-intimate (48.6%), and acquaintance (46.1%) condi- tions, X2(4, N = 536) = 2.60, p = 6.63. In contrast, when explicit evidence of intent was absent more participants in the ex-intimate condition responded that the accused intended to cause mental or physical harm to the target (19.0%) compared with those in the stranger (10.6%) and acquaintance conditions (8.3%), X2(4, N = 538) = 15.66, p < .01. However, this interaction is moderated by a three-way interaction between persistence, intent, and relationship (z = 2.66).

There was a significant difference between the relationship conditions when persistence was low and there was no explicit evidence of intent, X2(4, N = 270) = 19.27, p < .01. When the accused was an ex-intimate, more participants responded that the accused intended to cause mental or physical harm (18.9%) than when the accused was a stranger (8.9%) or acquaintance (3.3%). When persistence was moderate, the same pattern occurred, although the differences were not significant. Community members seem to be less likely to give the benefit of the doubt to the ex-intimate than to either the stranger or acquaintance in the absence of strong confirming evidence.

There was also an interaction between consequence and relationship (z = 2.38), which was only significant when the consequences consisted of moderate fear, X2(4, N = 356) = 10.52, p < .05. Frequencies are provided in Table 1. When the consequences to the victim consist of moderate fear, more participants respond that the accused intended to cause mental or physical harm when the accused is an ex- intimate, compared to the stranger condition and acquaintance condition. In the stranger condition more participants were unsure of the intentions, whereas in the acquaintance condition they were either more unsure of the intentions or did not believe they were present.

Significantly fewer participants responded that the behavior was stalking when they believed the accused did not intend to cause mental or physical harm, X2(2, N = 1058) = 207.98, p < .001. Of the 318 participants who responded that the accused did intend to cause harm to the target, 99.4% classified the behavior as stalking. Similarly, of the 398 participants who were unsure whether the accused intended to cause harm,

554

Identifying Stalking: Commonsense Reasoning

Table 1. Frequencies for Perceptions of Intention to Cause Mental Harm or Physical Harm as a Function of Consequences and Relationship

Intention to cause harm

Consequences Relationship No Unsure Yes

Extreme fear Stranger 42 (35.0) 41 (34.2) 37 (30.8) Ex-intimate 31 (25.8) 47 (39.2) 42 (35.0) Acquaintance 43 (35.8) 44 (36.7) 33 (27.5) Total 116 (32.2) 132 (36.7) 112 (31.1)

Moderate fear Stranger 29 (24.6) 58 (49.2) 31 (26.3) Ex-intimate 34 (28.3) 41 (34.2) 45 (37.5) Acquaintance 44 (37.3) 42 (35.6) 32 (27.1) Total 107 (30.1) 141 (39.6) 108 (30.3)

No fear Stranger 41 (34.2) 45 (37.5) 34 (28.3) Ex-intimate 36 (30.5) 48 (40.7) 34 (28.8) Acquaintance 47 (39.2) 41 (34.2) 32 (26.7) Total 124 (34.6) 134 (37.4) 100 (27.9)

Total 347 (32.3) 407 (37.9) 320 (29.8)

Note. Values in parentheses represent percentage.

90.2% classified the behavior as stalking. Only 59.6% of the 342 participants who believed that the accused did not intend to cause mental or physical harm to the target classified the behavior as stalking. Once again, given the numbers that still identified the behavior as stalking when they were unsure of whether the accused intended to cause mental or physical harm or indicated that he did not, it appears that many members in the community do not see this form of intent as a necessary element of stalking.

DISCUSSION

The results of this study show that identifying stalking appears to be a more complex process than was evident in earlier research (Dennison & Thomson, 2000). Although the results of this study cannot be directly compared to that of Sheridan and Davies (2001), they do tend to indicate that members of the Australian commu- nity do not hold relatively simple, broad, or homogenous views on what constitutes stalking. For example, the hypothesis that females would perceive the behavior of the accused in a more negative light than males was supported. More females than males identified the behavior as stalking, perceived that the accused intended to cause fear or apprehension, and intended mental or physical harm. These findings indicate that there are differences in how males and females define or understand harassing and intrusive behavior. However, as the scenarios depicted a female target, female par- ticipants may have identified more closely with the target than did males. Although males and females differed in their views by less than 10%, suggesting that percep- tions of males and females are not overly discrepant, this finding still points toward differences in the way men behave toward women and the potential for victimiza- tion. Another consequence of the differences in males and females perceptions of stalking is that when laws are made by male-dominated parliaments, these laws may not reflect concerns of females.

555

Dennison and Thomson

Another hypotheses receiving support was that the behavior would be perceived as stalking more often when explicit evidence of intent was present and persistence was moderate. However, the hypothesis that more extreme consequences would lead to greater perceptions of stalking was not supported. When explicit evidence of intent was present, this was sufficient to identify the behavior as stalking, regardless of persistence, relationship, and consequences. When persistence was moderate, this was generally also sufficient to identify stalking. However, when persistence was low, a number of other factors became important, particularly whether there was explicit evidence of intent to cause fear or harm and the relationship between the accused and target. Support was also given to the hypothesis that behavior identified as stalking should be illegal.

Despite support for the various hypotheses, the analysis of explained variance suggests that there are other elements that people consider when identifying stalk- ing. These elements may involve personal experiences with stalking, concerns over personal safety and invasion of privacy, or the nature of the behavior itself along- side perceived intentions and degree of persistency. Nevertheless, these findings have important implications for legal definitions and amendments to stalking legislation.

The results of this study suggest that intent to cause fear or harm will be recog- nized when there is some explicit evidence of its presence. Furthermore, persistence contributed to the perception of intent, so that intent is more likely to be constructed or inferred when persistence is moderate or high, possibly allowing the intent require- ment to be satisfied. However, it is important that legislation clarify what constitutes mental harm. If a distinction is to be made between fear and mental harm then this distinction needs to be consistent and identifiable.

In terms of legislation, there are two problems that need to be reconciled. First, should the less intrusive and possibly ambiguous behaviors be called stalking? Second, should the behavior be deemed criminal? Commonsense justice notions of stalking suggest that in the absence of explicit evidence of intent and when persis- tence is low, intent is less likely to be perceived or constructed, and there is some ambiguity regarding whether such behavior is "stalking," with only 50% of partic- ipants identifying stalking. Given the uncertainty over the less intrusive scenarios, concerns that an intent requirement in stalking legislation prevents convictions of a wide range of behavior as stalking may be premature. The element of intent ensures that the behaviors that many people engage in innocently each day-such as walk- ing the same route to work as another-are not also captured within stalking laws. On the other hand, individuals who are subjected to unwanted attention that is not at the extreme end of the scale may still have legitimate causes for concern.

Although there was no main effect of consequences for perception of stalking, it did interact with the other variables for perceptions of intent in a complex way. This finding could be interpreted three ways. First, the manipulation of consequences may not have been successful. Participants may have perceived the consequences to the target similarly across the conditions. Future research should conduct a manipulation check. Second, the consequences to the target may not be relevant to members of the community when deciding whether stalking has occurred or not. Third, that the target at the very least experienced an invasion of her privacy may have been enough to satisfy community members, in conjunction with the actual behavior, that stalking

556

Identifying Stalking: Commonsense Reasoning

had occurred. Regardless, it appears that jurisdictions that require that the target did in fact suffer fear or harm as a consequence of the stalking may not actually match the concerns of the community. In fact, in Australia, the Model Criminal Code Officers Committee (1998) recognized that there were those in the community who did not believe that consequences were relevant in deciding whether behavior constituted stalking and did not include such a requirement in their model stalking law. The results of this research indicate that such a move is indeed reflective of the concerns of the community with respect to the event of stalking in this study. Further research is required on how the nature of the consequences to the recipient affect perceptions of stalking, particularly as many stalking laws in the United States and Australia, and harassment laws in England and Wales, require that the victim experience some level of fear or distress.

A potentially important finding was that there was a trend for participants to more often perceive that the ex-intimate intended to cause fear or harm than the stranger or acquaintance. These findings differ with that of Hills and Taplin (1998), where there was no difference in females' responses to a male stalker according to relationship. In this study participants may have focussed on the scenario informa- tion that indicated that the ex-intimate had been possessive in the relationship and interpreted the target's account of the stalking as being continued possessiveness of the ex-intimate in a refusal to let go of the relationship. Stereotypes of these types of domestic situations may have led participants to perceive some form of malicious intent more often, whereas this prior history was not present for the stranger or acquaintance.

In drawing conclusions from the present study some circumspection is required. It is possible that participants may have been biased toward perceiving the behavior as stalking, given that this was the topic of study, potentially inflating the number of participants who perceived the behavior as stalking. Furthermore, that participants were forced to make a choice as to whether or not the behavior was stalking (i.e., there was no "unsure" option) may have also affected the pattern of responses. How- ever, regardless of the absolute levels, there is nevertheless a clear impact of intent, persistence, and relationship on perceptions of stalking relative to other factors. Fur- thermore, the context surrounding the behavior of the accused may be important in identifying stalking, particularly as females in this study, as well as in Hills and Taplin (1998) and the British Crime Survey (Budd & Mattison, 2000), were more likely to report the unsolicited pursuit of a relationship as stalking. The type of behavior may also be important, for example walking behind someone may be perceived differently to sitting on the same train or bus, or writing letters or sending gifts.

Further research is required to explore community perceptions of stalking in a variety of other contexts and using other behaviors that might be construed as harassing, intrusive, or a nuisance to the recipient. The sex of the accused and target may also be important in identifying stalking. The debate over appropriate stalking legislation and how to best convey the type of behavior that is unacceptable may benefit from obtaining a greater understanding of the features that distinguish ac- ceptable social interaction from behavior that is perceived as stalking. Yet even when we identify these features, perhaps the most important question still remains. What should be done about stalking laws if they are discrepant with public opinion? Should

557

Dennison and Thomson

the law follow commonsense justice or should the law guide community sentiment? The requirement of intent in stalking laws is not overly discrepant with community responses in this study. Until additional research suggests otherwise, perhaps commu- nity education would be useful to advise the public on the general type of behavior that is now illegal, particularly when there are differences in males' and females' perceptions of what is harassing and intrusive behavior. This type of education may prevent some individuals coming into contact with the law. Alternatives to criminal prosecution in less serious cases of harassment or unwanted attention also need to be explored.

APPENDIX: EXAMPLES OF STALKING LEGISLATION

Example 1-Strict Intent Requirement (South Australia: Criminal Law Consolidation Act 1935) Illegal Stalking, 19AA (1) A person stalks another if-

(a) on at least two separate occasions, the person- (i) follows the other person; or

(ii) loiters outside the place of residence of the other person or some other place frequented by the other person; or

(iii) enters or interferes with property in the possession of the other person; or

(iv) gives offensive material to the other person, or leaves offensive material where it will be found by, given to or brought to the attention of the other person; or

(v) keeps the other person under surveillance; or (vi) acts in any other way that could reasonable be expected to arouse the

other person's apprehension or fear; and (b) the person-

(i) intends to cause serious physical or mental harm to the other person or a third person; or

(ii) intends to cause serious apprehension of fear.

Example 2-Intent Requirement Fulfilled if Foreseeability was Likely (Victoria: Crimes Act 1958) 21A. Stalking

(1) A person must not stalk another person. Penalty: Level 5 imprisonment (2) A person (the offender) stalks another person (the victim) if the offender

engages in a code of conduct which includes any of the following: (a) following the victim or any other person; (b) telephoning, sending electronic messages or otherwise contacting, the

victim or other person; (c) entering or loitering outside or near the victim's or any other person's

place of residence or of business or any other place frequented by the victim or the other person;

558

Identifying Stalking: Commonsense Reasoning

(d) interfering with property in the victim's or any other person's possession (whether or not the offended has an interest in the property);

(e) giving offensive material to the victim or other person or leaving it where it will be found by, given to or brought to the attention of, the victim or the other person;

(f) keeping the victim or any other person under surveillance; (g) acting in any other way that could reasonably be expected to arouse fear

or apprehension of fear in the victim for his or her own safety or that of any other person-with the intention of causing physical or mental harm to the victim or of arousing apprehension of fear in the victim for his or her own safety or that of any other person and the course of conduct engaged in actually did have that result.

(3) For the purposes of this section an offender also has the intention to cause physical or mental harm to the victim or to arouse apprehension or fear in the victim for his or her own safety of that of any other person if the offender knows, or in all the particular circumstances that offender ought to have understood, that engaging in a course of conduct of that kind would be likely to cause such harm or arouse such apprehension or fear and it actually did have that result.

Example 3-No Requirement of Intent to Cause Fear or Harm (Queensland: Criminal Code [Stalking] Amendment Act 1999) 359B. "Unlawful stalking" is conduct-

(a) intentionally directed at a person (the "stalked person"); and (b) engaged in on any 1 occasion if the conduct is protracted or on more than

1 occasion; and (c) consisting of 1 or more acts of the following, or a similar, type-

(i) following, loitering near, watching or approaching a person; (ii) contacting a person in any way, including, for example, by telephone,

mail, fax, e-mail or through use of any technology; (iii) loitering near, watching, approaching or entering a place where a per-

son lives, works or visits; (iv) leaving offensive material where it will be found by, given to or brought

to the attention of, a person; (v) giving offensive material to a person, directly or indirectly; (vi) an intimidating, harassing or threatening act against a person, whether

or not involving violence or a threat of violence; (vii) an act of violence, or threat of violence, against property of, anyone,

including the defendant; and (d) that-

(i) would cause the stalked person apprehension or fear, reasonable arising in all the circumstances, of violence to, or against property of, the stalked person or another person; or

(ii) causes detriment, reasonably arising in all the circumstances, to the stalked person or another person.

559

Dennison and Thomson

What is immaterial for unlawful stalking, 359C

(1) For section 359B(a), it is immaterial whether the person doing the unlawful stalking- (a) intends that the stalked person be aware the conduct is directed at the

stalked person; or (b) has a mistaken belief about the identity of the person at whom the

conduct is intentionally directed.

(2) For section 359B(a) and (c), it is immaterial whether the conduct directed at the stalked person consists of conduct carried out in relation to another person or property of another person.

(3) For section 359B(d), it is immaterial whether the conduct throughout the occasion on which the conduct is protracted, or the conduct on each of a number of occasions, consists of the same or different acts.

(4) For section 359B(d), it is immaterial whether the person doing the unlaw- ful stalking intended to cause the apprehension or fear, or the detriment, mentioned in the section.

(5) For section 359B(d)(i), it is immaterial whether the apprehension or fear, or the violence, mentioned in the section is actually caused.

REFERENCES

Anderson, S. C. (1993). Anti-stalking laws: Will they curb the erotomanic's obsessive pursuit? Law and Psychology Review, 17, 171-191.

Boekmann, R. J., & Tyler, T. R. (1997). Commonsense justice and inclusion within the moral community: When do people receive procedural protections from others? Psychology, Public Policy, and Law, 3(2/3), 362-380.

Budd, T., & Mattinson, J. (2000). The extent and nature of stalking: Findings from the 1998 British Crime Survey. Home Office Research Study No. 210. London: Home Office.

Burstin, F (1997, July 9). Court rejects stalk charges. Herald Sun, 9. Criminal Code [Stalking] Amendment Act 1999 [QLD]. Criminal Law Amendment Act [No. 1] 1998 [WA]. Dennison, S. (2000). Congruence between stalking legislation and public perceptions: How accurately do

stalking laws reflect public needs? Doctoral thesis, Charles Sturt University, NSW, Australia. Dennison, S., & Thomson, D. M. (2000). Community perceptions of stalking: What are the fundamental

concerns? Psychiatry, Psychology and Law, 7(2), 159-169. Findlay, M., Odgers, S., & Yeo, S. (1999). Australian criminal justice (2nd ed.). South Melbourne, VIC:

Oxford University Press. Finkel, N. J. (1995). Commonsense justice: Jurors' notions of the law. Cambridge, MA: Harvard University

Press. Finkel, N. J. (1997). Commonsense justice, psychology, and the law: Prototypes that are common, senseful,

and not. Psychology, Public Policy, and Law, 3(2/3), 461-489. Finkel, N. J. (2000a). Commonsense justice and jury instructions: Instructive and reciprocating connections.

Psychology, Public Policy, and Law, 6(3), 591-628. Finkel, N. J. (2000b). But it's not fair! Commonsense notions of unfairness. Psychology, Public Policy, and

Law, 6(4), 898-952. Finkel, N. J., Burke, J. E., & Chavez, L. J. (2000). Commonsense judgements of infanticide: Murder,

manslaughter, madness, or miscellaneous? Psychology, Public Policy, and Law, 6(4), 1113-1137. Finkel, N. J., & Sales, B. D. (1997). Commonsense justice: Old roots, germinant ground, and new shoots.

Psychology, Public Policy, and Law, 3(2/3), 227-241. Goode, M. (1995). Stalking: Crime of the nineties? Criminal Law Journal, 19, 21-31. Haney, C. (1997). Commonsense justice and capital punishment: Problematising the "Will of the People".

Psychology, Public Policy, and Law, 3(2/3), 303-337.

560

Identifying Stalking: Commonsense Reasoning

Harris, J. (2000). An evaluation of the use and effectiveness of the Protection from Harassment Act 1997. Home Office Research Study No. 203. London: Home Office.

He Kaw Teh (1985). 15 Australian Criminal Reports 203. Hills, A. M., & Taplin, J. L. (1998). Anticipated responses to stalking: Effect of threat and target-stalker

relationship. Psychiatry, Psychology and Law, 5(1), 139-146. Hills, A. M., & Thomson, D. M. (1999). Should victim impact influence sentences? Understanding the

community's justice reasoning. Behavioral Sciences and the Law, 17, 661-671. Horowitz, I. A. (1997). Reasonable doubt instructions: Commonsense justice and standard of proof.

Psychology, Public Policy, and Law, 3(2/3), 285-302. McMahon, G. (1999, May 29). Stalk charges dismissed. Herald Sun, 19. Model Criminal Code Officers Committee. (1998). Model criminal code. Model Criminal Code Officers

Committee of the Standing Committee of Attorneys-General. Mullen, P. E., Pathe, M., & Purcell, R. (2000). Stalkers and their victims. UK: Cambridge, Cambridge

University Press. New South Wales Attorney General's Department. (1999). Discussion paper: Apprehended violence

orders-A review of the law. Sydney, NSW: Criminal Law Review Division. Ogilvie, E. (2000). Stalking: Legislative, policing and prosecution patterns in Australia. Canberra, ACT:

Australian Institute of Criminology Research and Public Policy Series, No. 34. Australian Institute of Criminology.

Protection from Harassment Act (1997). England and Wales. Sheridan, L., & Davies, G. M. (2001). What is stalking: The match between legislation and public percep-

tion, Legal and Criminological Psychology, 6, 3-17. Sheridan, L., Davies, G. M., & Boon, J. C. W. (2001a). Stalking: Perceptions and prevalence, Journal of

Interpersonal Violence, 16, 151-167. Sheridan, L., Davies, G. M., & Boon, J. (2001b). The course and nature of stalking: A victim perspective,

The Howard Journal of Criminal Justice, 40(3), 215-234. Sohn, E. F (1994). Antistalking statute: Do they actually protect victims? Criminal Law Bulletin, 30(3),

203-241. Tjaden, P., & Thoennes, N. (1998). Stalking in America: Findings from the National Violence Against

Women Survey. Washington, DC: National Institute of Justice and Centers for Disease Control and Prevention (NCJ 169592).

Violence Against Women Grants Office. (1998). Stalking and domestic violence: The Third annual report to Congress under the Violence Against Women Act. Washington, DC: Office of Justice Programs, U.S. Department of Justice.

Way, R. C. (1994). The criminalization of stalking: An exercise in media manipulation and political op- portunism. McGill Law Journal, 39, 379-400.

561