study notes crjs 493.docx

Unit 1 Criminal behaviour is a legitimate social problem, and the courts have come to rely on various professional groups to assist with the adjudication and sentencing of criminals. Chiefly, this assistance comes in the form of assessments that gauge an individual's risk of re-offence. Risk assess...

Unit 1 Criminal behaviour is a legitimate social problem, and the courts have come to rely on various professional groups to assist with the adjudication and sentencing of criminals. Chiefly, this assistance comes in the form of assessments that gauge an individual's risk of re-offence. Risk assessments have thus become a routine component of sentencing. Formal risk assessments are also relied upon during bail applications and for parole-board decisions. As I noted earlier, Canadian researchers have been at the forefront of scientific efforts to develop sound risk assessment practices. These efforts have led to the well-articulated set of principles that are discussed in the Bloom et al. (2005) article. In a nutshell, current risk assessment practices require that information about an individual be drawn from multiple sources and that multiple aspects of his or her functioning (e.g., mental health, community supports) be directly evaluated. It is also widely agreed that assessments should be specific and rely upon appropriate risk measures. Thus, it would be wrong to assess risk for violence against strangers with an instrument calibrated on domestic violence perpetrators, for example. The entire process should rely on clearly defined procedures and be amenable to review by external parties. Before the articulation of these principles, risk assessment procedures were highly subjective, difficult to understand, and occasionally bizarre. Although the Bloom et al. (2005) article refers to a dizzying array of risk assessment instruments, I would like to focus your attention on the difference between actuarial and structured checklist approaches to risk assessment. The former are fixed statistical formulas that provide a numerical probability that a given offender who fits a certain profile will reoffend over a specific period of time. For example, provided that a given offender has the appropriate demographic profile, application of the VRAG (an actuarial instrument) may show a risk to reoffend of 40% over the next five years. As you consider this statement, you should note that the figure of 40% is an average re-offence rate for offenders with similar profiles. In other words, the VRAG will provide an average re-offence rate for similar offenders but *will not* provide information about the specific offender that is being evaluated. In spite of the great promise of actuarial instruments, they have been the subject of considerable criticism. The most salient of these criticisms involves their inability to consider case-specific information. In other words, idiosyncratic features of the individual and his or her crime are inadequately captured or completely missed by actuarial prediction formulas. A related problem is that the majority of available actuarial instruments focus heavily on static variables (i.e., unchangeable historical variables) and, therefore, say very little about how to manage risk. Some authorities have also questioned the morality of decisions that punish an individual for membership in a statistical category (i.e., a category associated with a 40% re-offence rate). The other common approach to risk assessment relies on checklists of empirically identified risk markers. These checklists have been developed by identifying variables that are specifically linked to various forms of violence. We have, for instance, checklists for violence against strangers and checklists for violence against romantic partners. Although these checklists do not provide statistical estimates of recidivism---the final outcome is typically framed as low, moderate, or high risk---they are quiet flexible, and direct comparisons with actuarial formulas has typically shown equal performance. Currently, structured checklists are the most widely employed risk assessment instruments. The risk assessment process that relies on these empirically derived checklists is most commonly referred to as *Structured Professional Judgment* (SPJ). Parenthetically, you will notice frequent reference to a statistical index---the AUC (Area Under the Curve)---in your readings. Full explanation of this statistical index and its derivation is beyond the scope of this course. For our purposes, you should simply interpret the AUC figures as an approximate index of a risk assessment instrument's accuracy. For example, an AUC of.70 indicates that the instrument can distinguish a recidivist from a non-recidivist with 70% accuracy. An AUC of.50 equals chance levels---about the same accuracy as flipping a coin. **Study Questions** 1. Outline and discuss the basic principles of contemporary risk assessment. 2. What are the two complementary methods for measuring crime in Canada, and how do they differ? 3. Discuss overall crime rates in Canada, including regional differences. The idea that socioeconomic disadvantage contributes to crime and violence has a long history in both sociology and psychology. In his 2003 paper, Markowitz outlines sociological factors related to *instrumental violence*. Instrumental violence is violence that is perpetrated to obtain a specific goal. Instrumental violence is not emotionally driven, and it is not idealistic. A robber who assaults an elderly man to force him to give up his wallet is using instrumental violence. He is not angry (at least not initially) at the victim, and he is not stealing money for some noble cause. The robber assaults the victim simply to achieve a specific goal: personal gain. In other words, he is using violence as a means-to-an-end. Markowitz (2003) links the tendency to employ instrumental violence to social forces that loosen community ties and engender attitudes conducive to violent behaviour. He specifically identifies a "code of the streets" associated with low SES neighbourhoods as an attitudinal variable that promotes violent confrontation. The need for status among one's peers and a general "disputatious" social orientation are also identified as variables that contribute to violence. The latter is assumed to develop as a result of negative living conditions (e.g., crowding) that cause residents of low SES neighbourhoods to feel mistreated and resentful. Economic disadvantage also tends to undermine *neighbourhood cohesion* and *collective efficacy*. The former is a sociological term essentially synonymous with the lay notion of "sense of community." In other words, *neighbourhood cohesion* refers to how much residents feel connected to their community. Do they feel like they belong to a group called a "neighbourhood" or do they feel like a collection of individuals with little connection and responsibility to other neighbourhood residents? Sociological theorists propose that lack of neighbourhood cohesion produces a set of social circumstances that promote or, at least, fail to inhibit violent behaviour. *Collective efficacy* refers to the degree of influence that a neighbourhood can exert over its individual residents. Collective efficacy occurs as a result of the connection and sense of responsibility that community members feel toward each other. In a cohesive neighbourhood, collective efficacy will manifest with members of higher status assuming mentorship and informal policing responsibilities for the community. Collective efficacy is undermined as neighbourhood cohesion is eroded. These arguments are nicely summarized on page 152 of the Markowitz (2003) article. Note also figure 1 on the same page. The links with a "+" indicate that the presence of one variable will lead to the other. Links with "-" indicate that the presence of one variable will inhibit the other. For example, economic disadvantage is positively related (increases) to disputatiousness which, in turn, increases the likelihood of violence. Neighbourhood cohesion, on the other hand, is negatively related to (decreases) violence. The *predatory violence* discussed by Meloy (2000) is a specific type of instrumental violence. Predatory violence involves the unemotional pursuit of "prey." The ultimate goal of predatory violence is the satisfaction of some personal goal or desire. Meloy (2000) identifies the following as the most common goals of predatory violence: "money, power, dominance and control, revenge, territoriality, and sexual gratification" (p. 93). Although revenge can be a motive for predatory violence, the actual act of aggression is unemotional (cold) and calculated. I urge you to carefully review the features of predatory violence discussed in your readings (pages 91--97). Table 8-1, on page 88, also provides a very useful comparison of predatory and affective violence. *Affective violence* is, essentially, the opposite of predatory violence: it is extremely emotional, its only goal is threat reduction, and it is primarily defensive. Affective violence is discussed on pages 89--91 of Dr. Meloy's book. From a criminal justice perspective, affective violence raises an interesting question: if it is primarily defensive, how can affective violence be of much interest to the criminal justice system? The answer to this question lies in the definition of "threat." Specifically, threats can be objective (plain for all to see) or subjective (apparent only to the person being threatened). Having someone point a gun at you is an example of an objective threat. Anyone viewing this event would agree that you are being threatened. Having the person that you love more than anyone else tell you he or she is leaving is an example of a subjective threat. Casual observers are unlikely to view you as being threatened. Yet, the imminent loss of a loved one is a powerful threat that is capable of triggering affective violence. As you have learned (or will learn) in this unit, affective violence is also easily displaced so that victims can be innocent bystanders or even good Samaritans. Although *intended violence* is a type of instrumental violence, it is distinguished from all other forms of criminal violence by the absence of common criminal motives. Thus, while acts of intended violence can reflect a wide range of specific motives they are never, for example, committed purely for profit or sexual gratification. Individuals who commit acts of intended violence are invariably found to have acted out of some very personal desire to settle a score. In the words of Calhoun and Weston (2003), intended violence inevitably reflects a personal "search for justice." The seemingly impersonal killing of a politician or an abortion clinic physician, for instance, can reflect a personal desire for some sort of social justice. Just like threats, however, grievances can be both subjective and objective. An individual who commits an act of intended violence does not necessarily have to have an objective score to settle. The score, so to speak, may be ideological or even delusional. The simple point is that a perceived grievance will invariably underpin intended violence. The general category of intended violence includes *targeted* and *opportunistic violence*. The cardinal difference between these types of violence is that the former focuses on individuals, while the latter is focused on places of gathering. The victims of opportunistic violence, therefore, are random. They are simply and tragically at the wrong place at the wrong time. Another way of looking at this distinction is that the goal of intended violence is to harm an individual while the goal of opportunistic violence is to make a statement of some sort. School shootings are examples of opportunistic violence. The assassination of a public figure, on the other hand, is an example of targeted violence. Opportunistic violence tends to occur at predetermined locations (e.g., schools, hospitals, churches), while the location for targeted violence can vary widely because individuals are mobile. **Study Questions** 1. Discuss and critically evaluate the Social Interactionist theory of violence. 2. Discuss the notion of intended violence, giving examples of both targeted and opportunistic violence. 3. Compare and contrast predatory and affective violence. Chapter 1 of Webster et al. (2014) provides a brief overview of how legal and mental health systems are typically linked. Figure 1.1, on page 3, provides a general layout of how these systems interact and identifies a number of decision points that are often influenced by risk assessment findings. Note, for example, that a police officer has the option of routing a suspect into either the health-care system or the legal system. Note also that the systems interact so that a suspect initially placed in the court system can be transferred into the mental health system, and vice versa. The point that you should take away from this discussion, as well as the Chapter 2 review of theories, is that risk assessments do not occur in a vacuum. Rather, they occur in a socio-legal and theoretical context that shapes their specific parameters. In Chapter 3, Webster et al. (2014) explain that assessments conducted for the purpose of evaluating risk include four activities: a) *risk attribution*, b) *risk prediction*, c) *risk assessment*, and d) *risk management*. Risk attribution is an indefensible process that involves labelling an individual dangerous or not dangerous based on largely irrelevant observation (e.g., size). Risk attribution statements generally reflect incompetence or indifference. Risk prediction is the result of risk assessment. As we have seen (e.g., Bloom et al., 2005), risk assessments can be clinical or actuarial, or they can be based on structured checklists. Although many evaluators believe that risk prediction is the ultimate goal of risk assessment, this belief is misguided for two reasons: a) it fails to consider that many high-risk individuals are successfully released into the community on a regular basis, and b) it fails to consider that both the courts and correctional gatekeepers ultimately seek information about how to contain risk. In other words, it is not sufficient to simply label an individual as low/medium/high risk or to indicate that he or she poses a 40% chance of recidivism over five years. What everyone really wants to know is what do we do to contain the risk of re-offence? Does the individual require some sort of treatment? Should he or she be monitored daily, or is biweekly reporting to a parole officer sufficient? These are the sort of questions that ultimately interest a judge or parole-board member. More will be said about risk management later in this course. As I noted earlier, it is critical that you thoroughly understand the technical problems discussed by Dr. Meloy. Full appreciation of these problems necessarily begins with acceptance of two basic premises of risk assessment: in all cases there will be things that you do not know, and not all abnormal behaviour can be controlled. Too often people overestimate their own abilities. This arrogance and naïveté, along with ignorance of basic issues like the *base-rate problem*, contribute significantly to *false negatives* in risk assessment. *False positives*, on the other hand, tend to reflect cynicism about human nature and the ability to change, or simply fear of being wrong (remember you are less likely to be wrong when your default response is high risk). In either case, prediction errors carry serious, potentially life-altering consequences. Errors in prediction are discussed on pages 8--11 of the Meloy (2000) readings. Pay particular attention to figure 1 on page 8, and be sure you understand its 2x2 nature before moving on. For example, assume that you have assessed a woman charged with assault as posing a high risk for violence, and as a result, you recommend against conditional release (bail). In spite of your recommendation to the Court, her lawyer is successful in securing bail for his client. She is consequently released and behaves as a model citizen until she is acquitted at trial. What type of error have you committed? I hope you recognized the upper left-hand quadrant of figure 1 as the box where you would have landed because you have committed a false-positive error in this case. The negative consequences of a false-positive error tend to fall primarily on the individual who was assessed. Chiefly, these consequences will involve some restriction of liberty (e.g., incarceration or restrictive release conditions). Conversely, the consequences of a false-negative error can include serious injury or even death to a victim. Many would argue that false negatives are much more dangerous, and therefore, the *over-prediction bias* (i.e., tendency to err in the direction of false positives) that mars unstructured risk assessment is justified from a public safety perspective. The counter-argument would involve appeal to civil liberties and individual rights. Ultimately, the decision to err in one direction or another is a matter of social and legal policy that can only be addressed by judges and politicians. This brings us full circle to the legal and social issues discussed by Webster et al. (2014). The only responsibility that a professional evaluator of risk carries is to minimize all types of risk. Contemporary risk assessment practices require that all evaluators understand the so-called base-rate problem. The base-rate problem in risk assessment is really quite simple: Violence in society is statistically very rare, and rare events are very difficult to predict. Consider the following example: your co-worker, Joe, has the peculiar habit of altering his route to work on one day of the month. This change in routine never occurs on the same day in consecutive months, and there appears to be no discernible pattern to the selection of this alternate route. Now, would it be easier to predict which day Joe will alter his route to work or which day Joe will travel along the usual route? Of course, it is much easier to predict selection of the usual route. Why? Simply because selection of the usual route occurs much more frequently. In fact, assuming a twenty-day-per-month work schedule, your chances of successfully predicting that Joe will travel along the usual route are 95% (19/20), while your chances of successfully predicting the alternate route day are 5% (1/20). Now work through this problem with the base rates for homicide in Canada, and you will have a real-life example of the base-rate problem in risk assessment. Full appreciation of this problem should help you understand why risk assessments that rely exclusively on clinical judgment can be quite dangerous. **Study Questions** 1. What are the six facets of risk assessment? Are these facets a feature of actuarial (statistical) risk assessment, or are they more closely related to contextualized, individual case assessments? 2. Discuss the base-rate problem in risk assessment. Link your discussion to the 2x2 prediction matrix by indicating how base rates for violent behaviour contribute to each quadrant of the matrix. The already complex relationship between mental illness and violence is further complicated by very high comorbidity (co-occurrence) rates between mental illness and substance abuse. In their attempts to disentangle these important predictor variables, researchers have discovered that some forms of mental illness (e.g., anxiety disorders) tend to inhibit violence while others (e.g., paranoia) are powerful risk markers for violent behaviour. In most cases, the presence of a substance abuse disorder increases the risk for violence. Some research, for example, indicates that the presence of a substance abuse disorder increases the risk of violence by a factor of 12-16 compared to individuals who do not have a substance abuse disorder. In other words, the presence of a substance abuse disorder can increase the risk of violence by up to 16 times. The relationship between substance abuse and violence has received a great deal of attention. Generally, this research shows that substance abuse increases the risk of violence among individuals who are predisposed to violence or who find themselves in environments that promote aggressiveness. Stimulants (e.g., cocaine, methamphetamine), in particular, are a class of drug invariably associated with an increased risk for violence and, in the long term, can cause psychotic episodes with pronounced paranoia. By way of brief explanation, paranoid psychotic reactions involve a loss of contact with reality and a cluster of symptoms (paranoia) that leave one feeling persecuted and vulnerable to attack. Paranoia---especially drug-induced paranoia---is a powerful risk factor for violence. Stimulants like cocaine contribute to violence by stimulating the central nervous system. Although it affects all neurotransmitters (biochemical substances that transmit information between neurones), cocaine exerts its main influence by blocking the re-uptake of dopamine. This stimulation of brain dopamine is believed to produce the euphoric and powerfully addictive effects of cocaine. Cocaine's addictive properties are such that animals allowed free access to this drug will select it over food to the point of starvation. Although the initial effects of cocaine use are subjectively positive, with continued dosing, cocaine abusers become increasingly suspicious, irritable, anxious, and angry. Their judgment (which is impaired at all stages of use) is also progressively undermined, and they become increasingly aggressive. Chronic users are also vulnerable to episodes of frank psychosis that are often indistinguishable from paranoid schizophrenia. During the acute phases, cocaine-induced psychotic episodes include delusions of persecution and hallucinations (auditory, visual, and tactile). Intoxicant-related violence tends to fall into three categories: a) *psychopharmacological violence* committed during acute intoxication (e.g., when one is drunk), b) *systemic violence* committed by those involved in the distribution of illicit drugs (e.g., a turf war between rival gangs), and c) *economic-compulsive violence* committed by addicts who turn to crime to subsidize their addictions. Although the relationship between mental illness and crime has a long and controversial history, a growing body of evidence now indicates a modest relationship between violence and some types of mental illness. *Persecutory delusions*, for example, seem to increase the risk of violence, particularly when they are accompanied by *delusional distress* (e.g., anger, fear, anxiety). Note the obvious association between persecutory delusions and paranoia. Note also that while delusional distress may be a reasonable research concept, in practice persecutory delusions are invariably associated with distress (few people who feel persecuted are happy, relaxed, and friendly), and therefore, as a general rule you should treat any evidence of paranoia as a risk marker for violence. *Threat and Internal Control Override (TCO)* delusions also appear to have a relationship to violence. These delusions leave one feeling threatened (again, note the link to paranoia) and controlled by external forces. Although one recent study (see Bjørkly, 2002) failed to find evidence of a link between TCOs and violence, in practice these delusions are always treated as a risk marker for violence. Together, *Antisocial Personality Disorder (APD)* and *psychopathy* account for a disproportionate amount of crime and violence in society. The former is the strongest predictor of non-violent offending, and the latter is a powerful predictor of violent offending. Meloy (2000) provides a solid review of the shared heritage of these two disorders, how they diverged, and their current status. Of the two, psychopathy has the strongest empirical base and is of more immediate relevance to applied forensic practice. Although the precise origins of psychopathy remain unknown, suggestive data exist indicating that psychopaths are genetically predisposed predators. For those of you familiar with the theory of Natural Selection, *serial polygyny*---a mating strategy that leads to the impregnation of multiple partners without establishing long-term emotional bonds (see Meloy, 2000; p. 117) ---is believed to account for the survival of the psychopathic gene pool. In any event, the prevalence of psychopathy in the general population is about 1% for males. APD, on the other hand, has a prevalence rate of approximately 3% for males. The general population prevalence estimates for psychopathy and APD among women are.3% and 1%, respectively. Psychopaths differ in important ways from non-psychopaths: they tend to be more aggressive, they are very grandiose, and they seem to have no capacity for emotional bonding with others. The inability to bond with others is especially problematic because emotional bonds provide the seedbed for a spectrum of emotions (e.g., empathy, remorse, love) that promote caregiving activities and act to inhibit victimizing behaviours. Some data also indicate the presence of neurocognitive peculiarities that leave psychopaths inclined to abuse intoxicants (especially stimulants) and to use predatory modes of violence. The original Cleckley criteria listed on pages 102--103 of Dr. Meloy's book provide a good qualitative description of a psychopath. The Psychopathy Checklist-Revised (2nd edition) is the benchmark diagnostic test for psychopathy. It includes 20 variables that capture various aspects of the disorder. Each variable is scored on a three-point scale (0--2), producing a total score that ranges from 0 to 40. Scoring of each item is based on explicit criteria, and the resulting total score is used for diagnostic purposes. Conventional use of this instrument identifies 30 as a diagnostic cut-off score for the condition of psychopathy. For clinical use, this diagnostic cut-off is raised to 33 to account for measurement error and to reduce the likelihood of a false-positive diagnosis. Meloy (2000) provides an excellent description of each PCL-R variable (pages 104--128). I urge you to carefully review this information; not to memorize the variables but rather to gain an appreciation of the psychopath as a different type of human being. Research with the PCL has shed considerable light on the psychological makeup of a psychopath. We now know, for instance, that psychopathy is composed of a set of personality features that are unique to the psychopath (factor 1 traits) and a set of personality features that are shared with APD criminals (factor 2 traits). Research has also demonstrated that psychopaths reoffend more quickly, more often, and more violently than non-psychopathic offenders, and that their violent behaviour does not usually diminish with age. Finally, and perhaps most troublesome, we have yet to develop effective treatment for psychopathy and criminal psychopaths remain largely unmanageable in a community setting. **Study Questions** 1. Discuss the relationship between substance abuse and violence. 2. Which intoxicants have been most strongly linked to violence, and what forms of violence are associated with intoxicants? 3. The relationship between mental illness and violence is complex. Which disorders are most commonly encountered in the criminal justice system? 4. What are the specific research findings related to delusions and TCOs? Professional judgment approaches include three distinct procedures: 1) *unstructured clinical judgment*, 2) *anamnestic risk assessment*, and 3) *SPJ*. The first category represents the original and historically dominant approach to violence risk assessment, but has fallen out of favour as evidence accumulates that it is extremely unreliable. Anamnestic risk assessment procedures add some structure to the clinical judgment approach but still allow a great deal of latitude regarding the evaluation process. Anamnestic procedures are essentially untested. SPJ is the most rigorous of the professional judgment approaches. SPJ decisions are based on an explicit set of criteria, clearly specify what variables need to be reviewed, and rely on well-researched instruments. SPJ instruments also provide information about risk-management decisions. This is a critical advantage because risk management is the ultimate focus of all risk assessments. Judges and parole boards may be interested in the level of risk posed by an offender, but they will always be more interested in how that risk can be managed (i.e., how the public can be protected). The Meloy (2000) readings offer a bit of a departure from conventional SPJ procedures. As I noted earlier, his biopsychosocial model includes a list of empirical correlates of violence but has never been offered as a unified measure of violence risk. In a sense, this model can be seen as falling somewhere between anamnestic and SPJ procedures. The advantage to this approach is that the list of variables can change quickly in response to new research findings. SPJ instruments, on the other hand, are set and thus require a much slower revision process. The HCR-20, for instance, has been under revision for a number of years. The chief disadvantage of Dr. Meloy's model is that it has never been tested as a whole and, therefore, has unknown validity (even though all the variables are empirically supported). Consider, for example, that research has established that the HCR-20 can identify recidivists with an accuracy in the range of 70--75%. This is a robust finding that has been replicated around the world. By using the HCR-20, an expert can respond directly to questions regarding the accuracy of his or her procedures. Questions of this sort are very common in courtroom settings. Questions regarding the overall accuracy of the biopsychosocial model, in contrast, can only be answered with reference to empirical findings for each individual variable. Lengthy technical answers of this sort tend to have less impact on a judge or jury. They can also be construed as evasive. As a general rule, whenever possible, you should work with an established, well-researched instrument. This is not to say that the biopsychosocial model is without merit. On the contrary, it is an exceptional model, and you should review it very carefully because it includes a wealth of practical information that can be used to supplement an established SPJ instrument. My only point is that the HCR-20 and instruments like it are easier to explain and defend in a courtroom. **Study Questions** 1. Some of the variables in Dr. Meloy's model are not included in the HCR-20. Do you think it would be useful to review both instruments during a risk assessment? If yes, give an example of how you would report your aggregate findings. 2. Compare and contrast the SPJ and actuarial approaches to risk assessment. Your answer should include a critical analysis of the strengths and weaknesses of each approach. Which model do you prefer and why? Which model do you think provides the most useful information to the courts or other consumer groups, and why? My commentary for this section will be brief because the WHO report is purely descriptive and because the process of risk assessment that was discussed in the previous section also applies to sexual offenders. In other words, the general process of risk assessment remains the same, but the specific variables and instruments change. You should, therefore, simply view Dr. Hanson's chapter as providing information that is specific to sexual offenders. By reading this information against the backdrop of the general principles that you learned earlier, you will acquire a solid foundation for assessing risk of re-offence among sexual offenders. When working through Dr. Hanson's chapter, pay particular attention to the base-rate data that he provides. However you may feel about these figures (e.g., do they underestimate recidivism?), they are the most comprehensive data available to us, and they provide the foundation for properly conducted sexual offender risk assessments. If, for instance, you conclude that an offender poses a "high" risk for re-offence, you are necessarily implying that his risk is greater than 14% over five years (see table 17.1, page 150). Absent these base rates, terms such as "high," "medium," or "low" become meaningless and indefensible. Base-rate information can also be used to counter popularly held misconceptions. Note, for example, that incest offenders have the lowest re-offence rates of any group. This finding runs counter to the popularly held belief that child molesters universally pose a high risk. Note also that these data clearly indicate that most sexual offenders do not re-offend. I encourage you to carefully consider how you feel about this finding as it also runs counter to popularly held beliefs about the dangerousness of sexual offenders. Established predictors of sexual recidivism are listed in table 17.2 of Dr. Hanson's chapter (page 152). It is important that you carefully review this table. It is equally important that you carefully review table 17.3 (page 153), which includes variables that are *unrelated* to sexual recidivism. Note that this table refutes the commonly held notion that sexual offenders who deny guilt pose a high risk for re-offence. Unfortunately, this belief continues to be held by many non-forensic practitioners and the courts. As I noted earlier, the WHO report is a descriptive document that requires little explanation. Nonetheless, it is filled with valuable information, and I urge you to study it carefully. My recommendation is that you organize this material into a number of summaries to make it more manageable for study purposes. It would be useful, for example, to create brief summaries of the variables that increase a woman's risk of being victimized or that promote sexual violence by men. Similar summaries could be constructed for social and cultural variables linked to sexual violence. You should also achieve some appreciation of statistical trends in sexual violence, particularly those related to North America. **Study Questions** 1. Do the available base-rate data underestimate re-offence rates among sexual offenders, or do you think they are relatively accurate? Explain and defend your answer. 2. Identify and discuss the social and cultural variables that are believed to contribute to sexual violence against women. 3. How do the variables identified in question 2 differ from the risk factors for sexual re-offence? Why are the lists not identical? The WHO report provides a sobering overview of intimate partner violence. For the most part, this form of violence conforms to one of two patterns: a) an escalating pattern of domestic terrorism that includes severe violence and near complete domination of the victim, and b) a less severe pattern that involves periodic eruptions of aggressive behaviour that are linked to underlying frustration and anger. These underlying feelings need not necessarily be linked to the relationship. Frustration and stress related to work, for instance, can trigger aggressive behaviour between intimates. The first pattern of violence is perpetuated almost exclusively by men and is responsible for most of the admissions to hospitals and women's shelters. The second pattern is much more common and includes mutual violence between the partners in a relationship. The list of potential triggers for violence between intimates is virtually limitless. Some of the most common triggers of male violence against female intimates are listed on page 95 of the WHO report. Perusal of these variables quickly reveals that they reflect the power and control issues that drive male efforts to dominate and oppress women. Table 4.5 (page 98) includes a very good list of risk factors for male violence against women. Most of these variables are included in the risk instruments described by Dr. Kropp in the Webster et al. book. As a rule, the level of risk posed by an individual will increase as a function of the number of risk factors that are present. In other words, more variables typically equal higher risk. It is also possible, however, for an individual to pose a very high risk of violence in spite of having only one or two risk factors in his profile. Consider the example of an individual who expresses a clear intention to kill his or her partner. This individual will necessarily pose a high risk for violence irrespective of the presence or absence of other risk factors. The principles that underlie properly conducted risk assessments for intimate partner violence are outlined Chapter 16 of your textbook readings for this unit. These principles are self-explanatory and by now should sound somewhat familiar to you. One word of caution though: namely, the recommendation that victims of violence should be interviewed as part of the risk assessment process is not universally accepted. Objections against this recommendation centre on the concern that subjecting victims of violence to multiple interviews (remember victims are often interviewed by police, health-care providers, social workers, well-intentioned relatives, etc.) may re-victimize them and is, therefore, potentially unethical. It is also unclear how a defence expert can ethically contact the complainant in a case without first obtaining explicit consent. Generally, if victim statements to the police are available and sufficiently detailed, it is best to rely on these documents rather than attempting to interview the victim. None of this is intended to suggest that third-party information should not be pursued as part of the risk assessment process. On the contrary, third-party information (e.g., relatives, witnesses, etc.) is critically important and risk assessments should never be based exclusively on information provided by the accused. **Study Questions** 1. Discuss the factors (social, cultural, financial, psychological) that keep women in abusive relationships. Include in your discussion a description of the process that ultimately concludes with a woman leaving an abusive partner. 2. Identify and discuss the principles that underlie risk assessments for intimate partner violence. Your discussion should include identification of the empirically identified risk markers for this type of violence. Research indicates that successful intervention programs for violent offenders are extremely complex. Successful intervention programs, for example, need to be tailored to specific offender groups and, in some cases, to individual offenders. There is very clear evidence, for instance, that individual offender characteristics will often determine whether a program is successful or not. In forensic intervention, one size very definitely does not fit all. Historically, the Risk-Needs-Responsivity (RNR) Model had been the dominant framework for offender assessment and rehabilitation. Recently, the Good Lives Model (GLM) has been offered as an alternative. Together, the RNR and GLM either in combination or individually or in some modified fashion provide the underpinnings for most offender treatment programs in Western countries. Although there is considerable overlap between the two models, there is ongoing debate regarding the relative merits of each approach and, ultimately, which model is best suited to reduce criminal behaviour. RNR proponents argue that their approach is supported by a mature literature showing its effectiveness in reducing recidivism. GLM proponents argue that the RNR model is too narrowly focused on individual deficits and does little to assist offenders to live a satisfying, prosocial lifestyle. In the end, GLM proponents argue that their model subsumes and enhances the RNR model. In other words, they argue that the GLM includes all RNR principles and more, resulting in a broader more humane treatment approach that assists offenders in their inherent goal of leading a better life without offending. The research articles that are required reading in this unit compare and contrast the RNR and GLM. As you work through these articles, you will see that the two models differ in their fundamental assumptions about the causes of criminal behaviour: the GLM assumes that criminal behaviour occurs as a result of frustrated needs, while the RNR model assumes that criminal behaviour occurs as a result of personal deficits and/or negative social influences. The courts, and by extension society, depend on our ability to develop programs that will curb criminal behaviour. Unfortunately, the research evidence indicates that our efforts have not been entirely successful. To be sure, several studies exist that support the view that forensic intervention is helpful. However, a broader review of this literature does not support any conclusion beyond that *some* treatment is effective *some* of the time with *some *offenders. Optimists argue that any treatment effect is better than none. Pessimists counter that current treatment programs are largely unsupported by empirical evidence and that touting them as effective is misleading to the public. Your readings also include discussion of transition points during risk management. The main point that you should take away from this discussion is that changes (i.e., periods of transition) are often accompanied by fluctuations in risk. Changes that are particularly stressful (e.g., death of a loved one) are especially likely to increase the risk of violence among previously violent individuals. The implication here is that risk assessments need to be conducted periodically because risk is rarely static. Report writing is an important component of the risk assessment process because even the most brilliant conclusions will be lost if they cannot be conveyed with clarity. As a rule, reports should be free of jargon, transparent (the reader should understand how you reached your conclusion), and as brief as possible. At best, entire sections of lengthy reports will be ignored. At worst, lengthy reports are vexing to consumers who do not have the time to be impressed by your writing skills. The best advice I can give with respect to report writing is get to the point as soon as possible but get there in a manner that is understandable to readers. **Study Question** 1. Compare and contrast RNR and GLM intervention approaches. Part 2 The readings for this unit introduce you to contemporary threat management, a relatively new area of practice in the criminal justice system. Chapter 1 of Calhoun and Weston (2003) provides an overview of the threat management process and four simple questions that can be used to determine whether a formal threat management process is required in any given setting (e.g., corporate, government, etc.). In addition to evaluating the need for a formal threat management process, these questions are intended to increase vigilance to out-of-context or inappropriate events. They also highlight the importance of not discounting emotional reactions. Although the need to pay particular attention to these unusual events and reactions may seem obvious to you as you study this course, a hectic and demanding work environment often has the effect of narrowing a person's attention on his or her immediate sphere of responsibility. An effective threat management attitude, however, requires constant vigilance to both external (i.e., environmental) and internal (i.e., emotional) events. In the second chapter of their book, Calhoun and Weston (2003) introduce *targeted violence* and *opportunistic violence* as subcategories of *intended violence*. Targeted violence has victim selection as its most distinguishing feature; opportunistic violence, on the other hand, tends to be more symbolic, with specific victims typically selected simply because of their location or because they represent something larger than their individual identity (e.g., a specific race or gender). You should also note that Calhoun and Weston are quite pessimistic about actuarial tables and the process of prediction. This position stands in sharp contrast to the process of risk assessment that was described in the first half of this course. Chapter 3 of Calhoun and Weston (2003) and the chapter from Meloy (2000) both offer reviews of similar bodies of research. The tone of these reviews is, however, very different. Calhoun and Weston seem very focused on the individual and reiterate the belief that, stripped of common criminal motives, all intended violence is motivated by the pursuit of personal justice. Conversely, Meloy's review of the literature is much more accepting of group statistics and large-scale research. Meloy also places much less emphasis on the notion of violence as a means to justice. The important point for you to take from this unit is that threat management is not synonymous with risk assessment---at least not the risk assessment process that was described in the first half of this course. As you already know, the latter is focused on individuals whose behaviour is driven, at least in part, by common criminal motives. Threat management, on the other hand, is concerned with individuals who act violently out of a sense of personal conviction and who intend to harm a specific victim, group of victims, or institution. In other words, contemporary threat management is focused on individuals who intend to harm someone or something for very personal reasons. Fortunately, these individuals are quite rare. Unfortunately, these low base rates for intended violence preclude the development of the actuarial prediction tables that underpin much of modern risk assessment practice. It is precisely for this reason (i.e., very low base rates) that Calhoun and Weston (2003) are correct in rejecting the actuarial approach for threat management. Simply stated, the argument is that the frequency of occurrence for intended violence is insufficient for the development of accurate actuarial tables. Conversely, although still relatively rare, more common criminal behaviour is, well, more common. The greater frequency of occurrence of common assault, for instance, allows researchers to develop statistical prediction models of the sort employed by weather forecasters. For the reasons stated in this paragraph, it is doubtful that anyone will ever develop similar prediction models for intended violence. **Study Questions** 1. How do modern risk assessment practices differ from the contemporary threat management process? How are the two processes similar? 2. Is prediction really "the province of angels and fools" (Calhoun and Weston, 2003; p. 25)? 3. What area of modern risk assessment is most similar to threat management? If you are able to identify an area of overlap, why is prediction possible for risk evaluators in this case? The readings in this unit provide you with a conceptual foundation for effective threat management. The Path to Intended Violence that is depicted on page 58 of Calhoun and Weston (2003) outlines an invariable process---a journey toward violence that always follows the same sequence of steps. This model provides a framework for evaluating immediate danger, for conducting a thorough and well-grounded threat investigation, and for managing the threat to a target once it has been identified. The journey toward violence always begins with a personal grievance and, unless interrupted, always ends violently. Grievances do not have to be objectively legitimate. An individual simply has to feel wronged to begin moving along the Pathway. Grievance, like every stage in the Pathway, has a set of telltale signs that can be used to identify potentially violent individuals. You should carefully study the telltale signs for each step of the Pathway. As you study this information, you should keep in mind that your goal is to understand the individual's subjective world. This will require that you seek to understand the reason that he or she feels aggrieved (as opposed to simply determining that he or she feels resentful). Has he or she been rejected by the target? Does the target represent something the subject finds intolerable? Does the target represent an opportunity for the fame the subject feels he or she justly deserves? These are the sorts of questions that provide the foundation for effective threat management. Obviously, the answers to these questions will require comprehensive analysis of the target's history and current circumstances. In an applied setting, you would also use this framework to determine the level of risk that is posed by the subject. An individual who demonstrates evidence of planning, for example, clearly poses a more imminent threat to the target than a person who is only thinking about violence (the Ideation stage of the pathway). Although the movement toward a violent outcome can occur very rapidly (or very slowly), as a general rule, the closer one is to the Attack stage, the greater the danger to the target. The *Ideation* stage of the Pathway simply involves consideration of violence as a reasonable response to a grievance. *The willingness to consider violence as a justifiable response is the cardinal distinction between individuals who pose a risk for violence and the rest of us who refuse to justify the use of violence.* Again, pay special attention to the telltale signs listed in Figure 3 (p. 65). Inappropriate Communications or Contacts (IC&Cs) tend to occur at this stage. IC&Cs are clearly defined, and examples are offered in Exhibit 1 (p. 68). The *Research/Planning* and *Preparation* stages involve actively preparing for an attack on the target. Planning at this point may be very simple or very complex. Typically, the complexity of the planning will vary as a function of the subject's intelligence: more intelligent subjects will generate more complex plans than less intelligent ones. *Final act behaviours* (p. 80) also tend to occur at this point. These behaviours often reflect suicidal ideation and may, for example, include making funeral arrangements. Final act behaviours can also be lethal to others. Killing loved ones is an example of homicidal final act behaviours. The *Breach* and *Attack *stages involve actual violent behaviours. There is very little to say about these stages beyond indicating that they involve physical movement toward the target. Each stage of the Pathway to Intended Violence is accompanied by illustrative vignettes that end with a summary statement of the issue that is illustrated. I urge you to carefully study these vignettes as they provide a wealth of very practical information. **Study Question** 1. Outline and critically discuss each step and the associated telltale signs of the Pathway to Intended Violence. Your readings for this unit offer practical steps for identifying individuals of violent intent and for evaluating the risk they pose. The first step in this process is to develop a reliable network of communication. This will require the training of front-line staff or other individuals of relevance. The creation of this communication network---as well as the appropriate training of its individual components---is absolutely critical for effective threat management. Once a network of trained individuals is in place to collect information, the next step involves a careful evaluation of the reports they provide. This process of information gathering and evaluation is repeated throughout the threat management process, with each evaluative step informing the next stage of the process. At each stage of this process---but especially during the early stages---the most salient question is "How close is the danger to the target right now?" (p. 92). The best way to answer this question is through a careful analysis of the *Inappropriate Communications or Contact* (IC&C) that is reported to the threat manager. Figure 1 on page 94 of your textbook provides a flow chart of the threat management process. I recommend you review this chart carefully and commit the sequence of steps to memory. Contemporary threat management is a process that is anchored by objective, verifiable observations. Speculation can distort this process and, at worst, increase the danger to the target. Calhoun and Weston (2003) conceptualize the fact-finding process as a four-step procedure: 1. *Establish a reporting process*: I made this point earlier. Simply stated, a group of individuals must be identified and trained to report IC&Cs to the threat manager. Ideally, this training process should be repeated at regular intervals to guard against complacency and to ensure that reporting guidelines remain salient for everyone. 2. *Locate the target*: This step requires that the potential target be located and interviewed immediately once an IC&C is received. Locating the target helps ensure his or her safety. Interviewing the target also assists the threat manager in his or her analysis of the IC&C. 3. *Data collection*: Your text refers to this step of the process as "getting the facts." I prefer data collection because I believe the term "data" emphasizes the dispassionate, objective search for information that is critical for effective threat management. Regardless of what you call it, this stage of the process requires: a) an analysis of the IC&C, b) an interview of the person who reported it, and c) an interview of the target. Throughout this process, the threat manager must remain vigilant to the biases and distortions that may unwittingly accompany the reporting of an IC&C. The process of analyzing an IC&C is detailed on pages 101--103 of your text. You should pay particular attention to the four questions that are listed on page 101. 4. *Data analysis ("Analyzing the facts")*: The final step of this process requires a systematic, standardized, and well-documented analysis of the available data. Those of you with science backgrounds will recognize the words "systematic, standardized, and well-documented" as central to the scientific method. The goal of this final step is threefold: a) to determine the level of protection that is required for the target, b) to outline a process of investigation, and c) to identify threat management strategies. Your readings for this unit conclude with three cardinal rules for a threat manager: a) avoid speculation, b) seek both positive and negative information because the case may include variables that reduce risk, and c) assess risk on an ongoing basis. The latter is absolutely necessary for both risk and threat management. Things change very quickly in a real-world setting and previously accurate information can become misleading as a result of new developments. **Study Questions** 1. What is the first goal of the threat assessment process, and how is it achieved? 2. Re-analyze the case study "Underestimating the Danger" (pp. 93--94), correcting the errors that were made. Your answer to this question should include a detailed process for dealing with the threat appropriately. An organizing framework is indispensable in any assessment of human behaviour because it imposes structure on a complex process and ensures that all critical variables are considered. In this unit, you have been introduced to a simple yet effective framework for conducting threat assessments. The four questions that provide the foundation for this framework are explicitly identified on page 113 of your readings. Briefly, they require that a threat manager determine the circumstances surrounding the receipt of the IC&C and that he or she seek to understand the subject's point of view. The latter requires an evaluation of what is at stake for the subject. For example, are the subject's actions motivated by a desire for revenge or financial gain, or is he or she simply seeking to terrorize? The third and fourth questions are related and, in practice, can be collapsed into a single question that asks whether the subject is behaving like a hunter or a howler. An unwavering focus on these questions increases investigative rigour and reduces the danger of bias during a threat assessment. For ease of presentation, the first step---a Cir-Con evaluation---can be broken down into two parts. The first part involves a detailed analysis of the circumstances under which the IC&C was received. Detailed and practical examples of how this task can be performed are included in Table 6-1 on page 115 of your reading. I urge you to pay close attention to this table. By way of summary, effective analysis of the circumstances of arrival for an IC&C requires consideration of the following issues: 1. *Method of delivery*: As a rule, the closer the subject gets to the target the higher the risk he or she poses. For example, a threat that is delivered verbally to the target usually signals a level of risk that is higher than that associated with a threatening telephone call. 2. *The intended message of the IC&C*: In more practical terms, what is the subject seeking to accomplish with his or her threat? Is the apparent motive financial gain, terrorism, or revenge? Is there indication of mental illness or is the subject simply using the IC&C as a method for collecting more information about the target (e.g., level of security)? These questions and the others posed in Table 6-1 focus the threat manager's attention on the content of the IC&C and the actions that accompanied its delivery. 3. *The amount of target research indicated by the IC&C*: The goal here is to determine how much target research preceded the arrival of the IC&C. Does the content of the IC&C indicate personal knowledge of the target (e.g., mentions publicly unknown details of the target's family)? Does the method of delivery depend on generally available information (e.g., delivered to a public office) or does it indicate knowledge of the subject's private life (e.g., delivered to a personal, unpublished address)? Generally, the greater the evidence of target research the higher the level of threat. 4. *Target selection*: It is critically important for a threat assessor to seek to understand the link between the subject and the target. This link need not be objective: mental illness can create delusional links that are as powerful as more objective ones. In any event---real or imagined---a link between the subject and the target is always present and must be understood from the perspective of the subject. 5. *Intimacy Effect*: As we have previously seen in Unit 9, the presence of intense emotional feelings toward the target increases the threat level considerably. The second step of the Cir-Con evaluation is focused on contextual issues. Again, I recommend that you carefully review Table 6-1 in your readings for this unit. A practical illustration of a Cir-Con evaluation is presented in the case study on pages 122--123 of your text. In summary, the second part of a Cir-Con evaluation requires consideration of the following issues: 1. *Change in the subject's life*: Change is stressful for everyone. It is important for a threat manager to seek an understanding of the stressors that are affecting a subject at the time he or she delivers the IC&C. Common stressors that carry the potential to elevate the level of threat in a case include divorce, job loss, and separation from loved ones. Your book indicates, and I strongly agree, that change should be measured both in terms of amount and subjective importance to the subject. A divorce that brings welcome relief to a subject, for example, would not be expected to elevate risk in the same way that the unwanted loss of spouse would. In general, it is undesired change that produces the sort of stress that is most relevant in a threat assessment situation. 2. *Change in the subject's behaviour*: This issue needs to be considered within the context of the Path to Intended Violence that we discussed in Unit 10. In essence, the question at this stage of analysis is whether the current IC&C suggests a movement toward or away from violence. Of course, the IC&C may also reflect a static situation---i.e., no change in threat level. Usually, however, the arrival of a new communication signals change of some sort. 3. *Review of past IC&Cs*: In addition to providing an overall context for evaluating change, past IC&Cs can serve as a predictor variable because past behaviour is the best predictor of future behaviour. Thus, for example, a subject who has a history of sending IC&Cs without follow-up tends to pose a level of threat that is lower than that posed by a subject with a history of aggressive behaviour following IC&Cs. 4. *The target's current situation*: Note that the focus of the threat assessment has now switched to the target. Interviewing the target to evaluate his or her current activities, especially involvement in controversial issues, can provide clues about the identity of the subject. Any involvement by the target in highly controversial or confrontational activities should alert the threat manager to a potentially high-risk situation. 5. *Previous IC&Cs to the target*: As I noted under point \#3, the existence of previous IC&Cs allow a threat manager to assess whether a subject is moving toward or away from violence. A target who receives numerous IC&Cs from different subjects, however, is usually involved in activities that are offensive to a large number of people. In general, the larger the pool of potential subjects (i.e., the greater the number of people making threats), the more difficult it will be to manage the threat and, as a result, the greater the risk of harm to the target. This flows logically from the simple observation that predicting the behaviour of many people will be more difficult that predicting the behaviour of a single person. Following a thorough Cir-Con evaluation, the threat manager must seek to determine what motivates the subject, as well as his or her level of personal involvement in the case. In short, a threat manager must seek to understand what is at stake for the subject. It is very important that you understand that this evaluation focuses on the subject's perception and that this perception may not necessarily correspond with reality. As your textbook correctly points out, many people have grievances but only those who believe their grievances justify violence pose an active threat. Your text also indicates that an evaluation of what is at stake for a subject involves two procedures: 1) an unstructured examination of the subject's grievances to determine how strongly they are held, and 2) the application of a semi-structured assessment tool referred to by the acronym of JACA. Table 6-2 on pages 125--126 outlines the possible outcomes of an unstructured review of the subject's grievances. This range of outcomes is anchored at one end by "no stakes" and, at the other end, by "everything to lose." The latter obviously signals a very high level of risk, particularly when there are indications that the subject has lost interest in life or is preparing for death. The absence of *inhibitors*---variables that reduce the risk of violence---also signals a high level of threat. Your textbook uses the analogy of inhibitors as dominoes that stand in row. As each domino falls, the move toward violence gains momentum. A particularly dangerous situation occurs when the subject blames the target for the loss of inhibitors or when the target inadvertently removes the final inhibitor. The case study on page 128 of your readings clearly illustrates the effects of tumbling inhibitors. The brief description of "crazy Eddie" on the next page is also informative. I recommend that you take some time to review both of these case studies. Application of the JACA scale is described in Table 6-3 on pages 131--132 of your readings. Briefly, JACA stands for *justification*, *alternatives*, *consequences*, and *ability*. Thus, application of the JACA focuses the attention of the threat manager on whether the subject feels justified in acting violent, sees any alternatives to violence, is willing to accept the consequences of violence, and is able to commit violence. The answers to these questions often flow logically from the previous steps in a threat assessment. For example, a "nothing to lose" mentality would suggest the subject is willing to assume the consequences of violence. Similarly, evidence of extensive target research and a military background, coupled with evidence the subject blames the target for the loss of inhibitors, would suggest both that the subject feels justified in using violence and that he or she has the ability to be violent. The final step of the threat assessment process requires a determination of whether the subject is a howler or a hunter. The Path to Intended Violence described in Unit 10 provides the context for making this determination. Table 6-4 on page 135 of your reading provides examples of behaviours that indicate the subject is a hunter and is moving toward violence. Table 6-5 on pages 137--138 discusses the Dietz-10, a list of ten variables that have been identified as common to American assassins. Again, information for rating each of the Dietz-10 variables tends to emerge as the assessment process unfolds. Generally, the level of threat can be seen to increase as the number of Dietz-10 variables increases. Note, however, that certain variables can combine to produce a high-risk rating even though the overall number of variables remains low. Your book provides the example of a mentally ill subject who conducts extensive target research and purchases a handgun. The presence of these three variables alone is sufficient to indicate a high threat level. At this point you should recognize the Dietz-10 as a sort of SPJ instrument. **Study Questions** 1. Compare and contrast Hunters and Howlers. Be sure to include discussion of inhibitors for each group. 2. What JACA factors, if any, are present in the case study "Eight Out of the Dietz Ten" (pp. 139--140)? It has become a cliché to describe stalking as an old behaviour but a new crime. Nonetheless, the statement remains valid: centuries-old descriptions of what we would now call stalking exist, but the world's first anti-stalking law is less than 20 years old. Although there is some variation between jurisdictions, stalking laws tend to identify three components: a) a pattern of following or harassing, b) a threat, and c) a fearful victim. Note that the threat need not be explicit. Repeatedly following a victim (i.e., a pattern of behaviour) can be interpreted as threatening if the conduct would cause fear in a "reasonable" person. Legal definitions of "reasonable" and "fear" overlap with common sense understanding of the terms but remain subject to individual interpretation by judges and prosecutors. A more useful definition of stalking is provided on page 169 of Dr. Meloy's book. Briefly, he offers the term "obsessional follower" as a clinical definition of stalking. Note that this simple yet elegant definition encompasses both the obsessed thinking of a stalker and the behaviour (following) that is most likely to draw the attention of law enforcement officials. Stalking is surprisingly widespread. Depending on how the behaviour is defined, the available data indicate a lifetime prevalence rate for being stalked of between 8--16% for women. The lifetime prevalence rates for men is considerably lower. As a more concrete example of the magnitude of this problem, consider that the Tjaden and Thoennes (1998) data cited in the article by Kropp et al. (2002) translates into 1,000,000 women and 400,000 men that are stalked each year in the United States. The length of stalking varies widely, but it is not uncommon for a stalker to persist for several months or years. The impact on the victim is wide ranging and can include disruption of their social lives, emotional distress, and in the extreme, a diagnosable mental illness. Many victims also become suicidal. These findings justify the intense research and legal scrutiny that stalking attracts. Your readings for this unit provide demographic data indicating that the "average" stalker is a man in his mid-30s with above-average education but a history of underemployment. He is also likely to have a history of failed relationships and a prior criminal record. The available data also indicate a relatively high prevalence rate for mental illness, especially substance abuse disorders and psychosis. Among stalkers, psychosis is most likely to take the form of a *delusional disorder* or *schizophrenia*. Very briefly, delusional disorders involve an unshakeable false belief that is not entirely impossible. For example, the belief that one is being followed or poisoned. Provided that these beliefs are demonstrably false, they would be examples of persecutory delusions. Note, however, that neither belief is absolutely impossible. Schizophrenia, on the other hand, includes active hallucinations and delusions that are bizarre. In the language of mental illness, the term "bizarre" is synonymous with impossible. Thus, for example, an individual who believes that his internal organs have been removed and replaced with someone else's internal organs is experiencing a bizarre delusion---i.e., something that is not possible. This information is provided to you only to assist you in understanding the data on stalker demographics. For the purpose of this course, you need only understand that both conditions (indeed all psychosis) involve a loss of contact with reality. The crime of stalking is typically unrelated to common criminal motives such a greed. Stripped of the usual criminal motives, stalking becomes a very personal crime that reflects a link (real or imagined) between the stalker and the victim. The existence of this perceived link between perpetrator and victim has understandably drawn some attention to the offender-victim relationship. This attention does not imply that the victims share responsibility for the stalking. Rather, the offender-victim relationship is simply another source of information that helps us understand the complex behaviours of stalkers. Analysis of the offender-victim relationship in stalking has revealed that most female victims are stalked by a past or present romantic partner. Men, on the other hand, tend to be stalked by male acquaintances. On page 175 of his book, Dr. Meloy suggests that victims of stalking tend to fall into one of three categories: a) prior sexual intimates, b) acquaintances, or c) strangers. Later, he allows for a possible fourth category made up of family members. Kropp et al. (2002) complement Dr. Meloy's analysis of the offender-victim relation by discussing the offender's behaviour toward the victim. In their analyses, stalking behaviour can be characterized according to how close it brings the stalker to the victim. Specifically, they argue that stalking behaviour can be characterized as a) remote, b) approach oriented, or c) direct contact. Remote behaviours are the most common form of stalking and may, for example, include unwanted phone calls or gifts. Approach-oriented behaviours bring a stalker closer to the victim but stop short of direct contact. These behaviours include following. Approach-oriented behaviours are the most common form of stalking. The final category, direct contact, is self-explanatory. Examples for this type of stalking behaviour include face-to-face communication or violence. Kropp et al. (2002) identify these behaviours as posing the greatest risk to the victim. By drawing on our past discussions about the Intimacy Effect, we can add that the victim at greatest risk for violence would be a family member or past sexual partner that is the recipient of direct contact behaviours. An analysis of the offender-victim relationship should occur as early as possible in a stalking assessments because it can provide a very quick estimate of risk. Even when the precise link between victim and stalker is unknown, simply categorizing the stalker's behaviours according to the Kropp et al. (2002) guidelines can provide a rough estimate of risk to the victim. As more information becomes available, you can turn your attention to the risk factors identified on pages 180--181 of Dr. Meloy's book. You should review these risk factors with his earlier (pages 176--177) comments about threats and violence in mind. In particular, you should carefully consider the apparently contradictory finding that threats are not a good predictor of violence in stalking situations but stalking victims who receive threats are at an elevated risk of being harmed. Although a satisfactory explanation for these discrepant data has yet to emerge, the lesson for applied stalking assessments is clear: one should never ignore a direct threat to the victim. Another very interesting finding that you should consider carefully is that the presence of a prior sexual intimacy between the stalker and the victim appears to predict violence with 90% accuracy. Again, this empirical finding supports the notion of an Intimacy Effect. Kropp et al. (2002) offer a more elaborate process for assessing risk in a stalking situation. Their model draws from the mainstream risk assessment literature to suggest a structured approach that employs a varied menu of risk factors depending on the individual stalker profile. The profiles for this risk assessment process are drawn from the typology they present on pages 607--609. This typology has considerable practical utility, and I recommend that you study it carefully. Once a risk assessment has been performed, the task becomes one of generating a viable risk management plan. A sound risk management framework is presented on pages 183--189 of Dr. Meloy's book. "Dramatic moments" are especially important and should always be seen as signalling the possibility of imminent violence. Although not specifically mentioned in your readings, it is important to remember that dramatic moments can occur as a result of events in *either* the lives of the offender or the victim. Your readings offer the appearance of a Protection Order as an example of an event that can trigger a dramatic moment. This event clearly occurs in the offender's life. That is, the offender is served with a Protection Order, and he or she reacts with anger which, in turn, increases the risk of violence against the victim. Consider, on the other hand, the case of a politician who is scheduled to deliver a controversial speech. Although this speech is an event that occurs in the life of the politician (i.e., it is not directed at the offender), it may trigger a dramatic moment for a stalker that has previously restricted him or herself to remote stalking behaviours. You should therefore think of dramatic moments as events that can occur anywhere in the offender-victim relationship. Even events that appear to be purely personal for the victim can have a dramatic effect on stalker, particularly if he or she is mentally ill. **Study Questions** 1. Compare and contrast the stalker typologies that are discussed in your readings. 2. Which typology do you find most helpful and why? 3. Is there a single typology that you prefer, or do you think case-specific variables should impact the selection of typology in any given case? On a continuum of severity, bullying is a relatively mild form of school violence. Nonetheless, bullying has demonstrably negative effects on victims. Victims of bullying, for example, tend to become highly sensitive to violence cues, and they become more likely to carry weapons. Although still a highly unlikely event, some victims become aggressors in an extremely violent and spectacular manner. Victims of bullying may also suffer mental health consequences. At the risk of oversimplification, these consequences can be categorized as either internalizing or externalizing disorders. The former primarily involve emotional disorders that, among other things, render a child shy, withdrawn, and irritable. These children are often diagnosed with anxiety or depressive disorders. Externalizing disorders, on the other hand, tend to manifest as problem behaviours such as aggression and substance abuse. Children with these disorders are often diagnosed with learning disabilities and attention deficit/hyperactivity disorders. The tragic case of Daniel Scruggs (Burgess et al., 2006, pp. 5--7) is an extreme example of an internalizing disorder linked to bullying. The externalizing consequences of bullying are clearly illustrated in Case 3 of the same article (pp. 7--9). Note that in Case 3, bullying led to an explosion of aggression against others. Daniel Scruggs, on the other hand, turned his aggression inward and committed suicide. This then is the hallmark distinction between internalizing and externalizing conditions: in the former, aggression and anger are turned inward, and in the latter, they are expressed against others. Case 3 in the Burgess et al. article also includes discussion of five factors that have been linked to school shootings. These factors converge to create the portrait of a school shooter as a psychologically vulnerable, socially isolated individual who harbours violent fantasies, endorses violence, and has access to weapons. In essence, these factors combine to produce the profile of a school shooter. Although at first blush this is not an unreasonable profile, you should exercise extreme caution whenever you encounter a profile in an applied forensic setting. Consider, for example, that research findings reported on page 169 of the Reddy et al. article reveal that juveniles who commit homicide are *less likely* to have a history of mental illness, criminal behaviour (including violence), or school problems. The data serve to counterbalance the intuitive appeal of profiles like the one offered by Burgess et al. Arguably the most useful finding to emerge from research on profiling is that profiles tend to limit the scope of inquiry to a degree that other important information is ignored. The point being made here is very simple: in a real-world setting, reliance on psychological profiles can produce dangerous consequences. A more detailed criticism of profiling can be found on pages 161--164 of the Reddy et al. article. In reviewing this section of the article, you should pay close attention to the distinction between *retrospective* and *prospective* profiling. Note also the research evidence indicating that profiling is based on a weak database, tends to produce biased decision making, and unfairly stigmatizes students because of high false-positive rates (i.e., individuals are mistakenly labelled as high risk). Having established lethal school violence as a rare occurrence does not remove the need to develop assessment and prevention strategies. Current assessment strategies draw heavily from the threat management literature that you have read about in this course. The term "targeted violence" used by Reddy et al., for example, is essentially a synonym for "intended violence"---the term preferred by Calhoun and Weston. Before the emergence of threat assessment as the preferred framework for evaluating the risk of intended (or targeted violence), school jurisdictions favoured established risk assessment procedures of the sort described earlier in this course. As you can see from your readings (Reddy et al., 2001), however, critical differences between targeted violence and general aggression conspire to undermine the utility of general risk assessment procedures for predicting the risk of targeted violence. Recall that risk assessment procedures have been derived from research among criminal populations and that they are specifically intended to predict general aggression. Logically then, it makes little sense to apply these procedures to a different form of violence (targeted) and to a different population (non-criminal youths). For these basic reasons, a threat assessment approach is preferred when evaluating the risk of targeted violence in a school setting. The guiding principles of threat assessment in a school setting are: a) no single profile for targeted violence exists, b) threats have limited value as a risk factor in threat assessment, and c) targeted violence does not occur randomly or spontaneously. At this point, you should recognize these principles as similar to those that underpin the threat assessment process proposed by Calhoun and Weston. When Reddy et al. speak of a "behavioral process" that precedes targeted violence, for instance, they are referring to a process that is very similar to the Path to Intended Violence that you learned about earlier in this course. The specific assessment process they describe on page 169 is also very similar to the process that you learned in Units 11 and 12. By comparing and contrasting these two procedures, you should see a unified framework emerge for conducting threat assessments and managing risk. Existing prevention programs for school violence place considerable emphasis on creating a school environment conducive to the reporting of bullying behaviours. Most programs also include additional adult supervision and instruction on conflict resolution strategies. An interesting and intuitively appealing development for these intervention programs is the establishment of peer support groups for vulnerable students. In theory, peer support groups should make students feel more connected and protected by their peers. These feelings of greater interpersonal connection and belonging should act as a barrier against aggressive acting out. Preliminary evaluations of school violence intervention programs have produced some encouraging results. In particular, children who participate in these programs appear to leave feeling better about themselves and less anxious about their peers. **Study Questions** 1. Discuss and critically evaluate data regarding the prevalence of school violence. Do these data suggest that concerns about school violence are exaggerated, or do other concerns exist that supersede these data and justify the widespread and very strong public response to school violence? 2. Should researchers work to develop actuarial prediction formulas for school violence? Why or why not? For classification purposes, workplace violence incidents can be placed into one of four categories. Type 1 incidents are committed by criminals who *do not* have a direct connection to the workplace. Bank robberies are an obvious example of this type of workplace violence. Type 2 incidents involve violence perpetrated by consumers. An irate customer who assaults an employee or a patient who assaults a doctor, for example, is committing type 2 workplace violence. Type 3 violence is committed by past or present employees against co-workers or management. Type 3 workplace violence is associated with the stereotype of the disgruntled former employee who returns to seek vengeance for perceived mistreatment. Finally, type 4 violence is perpetrated by an individual who has no direct link to the workplace but who has a personal (usually intimate) relationship with an employee of the company. Type 3 and type 4 violence are the focus of your readings for this unit. Note that both types of violence are very likely to involve the desire for justice emphasized by Calhoun and Weston (2003) in their analysis of intended violence, and that type 4 violence will always include the Intimacy Effect (see Unit 9). The case study described on page 41 of your reading is a chilling example of workplace violence that involves the Intimacy Effect. The goals of workplace violence prevention programs are threefold: 1) create an environment conducive to the early identification of risk markers for violence through the training of staff, 2) create an atmosphere that supports the reporting of violence or threatening behaviour by supporting staff who are victimized or feel threatened, and 3) establish community links with threat assessment/management experts who can provide external support to in-house threat managers. Typically, these external consultants will include psychologists or psychiatrists who specialize in threat assessment and management. Pages 19--20 of your reading include specific guidelines for developing a workplace violence prevention program. Pre-employment screening is identified as an important preventive measure. Although often impractical to perform on a large scale, the screening of prospective employees can identify some obvious risk markers for violence. Chief among these risk markers are a history of substance abuse and violence, particularly violent convictions and violent conflict with co-workers. You should recognize these two variables as powerful predictors of violence from the risk assessment portion of this course (see Unit 3). For existing employees, staff can be trained to identify other risk markers. Disciplinary action against an employee with a history of volatile outbursts or psychological instability, for example, should be recognized as a potentially dangerous situation. The danger associated with this scenario is heightened if the employee feels publicly shamed. An excellent list of risk markers can be found on pages 21--22 of your reading. Though this list should not be viewed as a one-size-fits-all checklist that can be used mechanically (i.e., without much thought), the combined presence of the following variables should *always *be interpreted as signalling a high risk situation, particularly if assessment reveals a lack of inhibitors: - Recent acquisition/fascination with weapons - Preoccupation with violence - Homicidal/suicidal threats The risk of workplace violence is also increased by several *contextual* factors. Contextual factors are variables that originate in the environment rather than in the individual. Paranoia, for example, is an individual risk factor for violence because it is a variable that exists in the individual. Absent effective treatment, a paranoid individual will remain mentally ill regardless of where he or she finds himself. Heat and crowding, on the other hand, are environmental risk factors for violence. Note that these variables---heat and crowding---exist outside of the individual. All things being equal, removing an individual from these unpleasant (i.e., hot and crowded) surroundings will lower his or her risk for aggressive behaviour. Management styles that leave employees feeling exploited and under-appreciated are contextual factors related to workplace violence. A comprehensive workplace violence prevention program must address both individual and contextual risk factors. **Study Questions** 1. Identify and describe the four types of workplace violence. What prevention strategies would you suggest for each type of violence? 2. Describe a proactive approach to workplace violence prevention. Your answer should include mention of the responsibilities of employers and roles for potential outside consultants. The core principles of applied threat management are: 1) threat assessments have a short shelf life and must be updated continuously, 2) any intervention carries the potential to aggravate a situation, and 3) threat management is a fluid process that requires flexible and innovative problem solving. Principle 1 should be familiar to you by now. In essence, threat assessments have a very short shelf life because they only provide a snapshot of the situation. Future developments may alter the situation in the direction of increasing or declining risk to the target. Effective threat management depends on continuous threat assessment updates, and an effective threat manager must be open to both good (i.e., threat is declining) and bad (i.e., threat is increasing) news. As a rule then, you should assume that the value of a threat assessment will decline rapidly as time passes. There are, of course, exceptions to this general rule. A threat assessment performed shortly before a subject is incarcerated, for example, can be expected to change minimally until he or she is released from custody. Release from custody may, however, become a "dramatic moment" (see Unit 13) and should therefore trigger immediate reassessment of threat posed by the subject. The case study on pages 187--188 nicely illustrates how a situation can change from low to high risk based on new developments. As you review this case study, you should note that the interventions described in the last two paragraphs cannot be performed under Canadian privacy legislation. In a Canadian context, the threat manager would be limited to interviewing the subject unless he or she obtained consent to access personal records. Although your book indicates that threat management interventions will produce one of three outcomes---no effect, positive effect (risk declines), or negative effect (risk increases)---it is imperative that you appreciate that any intervention carries the potential to aggravate a situation. Appreciating this potentially negative outcome brings one back to principle 1 and the need for constant re-evaluation of risk to the target. In practice, I recommend that all interventions be followed immediately by a re-evaluation of risk. Principle 2 also requires that each intervention be evaluated for its potential outcome *before *it is implemented. Only interventions likely to reduce risk should be implemented. There is no room for emotionally driven knee-jerk reactions in threat management. Principle 3 is the most difficult to follow because it is relatively ambiguous. In a nutshell, principle 3 requires that each situation be met with a uniquely tailored and flexible response. Threat managers do not have a menu of available responses for every potential situation. They must respond flexibly and creatively based on a full appreciation that each presenting situation reflects the individual circumstances and psychological make-up of the subject. Although they have their place, cookie-cutter approaches (i.e., inflexible, predetermined procedures) are very dangerous in applied threat management settings. Instead, a threat manager needs to be innovative and maintain an open mind to the endless possibilities that accompany a specific situation. He or she should also consult a team of experts on a regular basis. Regular attendance at workshops and conferences is also essential to maintain an up-to-date knowledge base. Application of principle 3 is illustrated on pages 192--193. The remainder of your readings discuss specific strategies for defusing risk. Each strategy is clearly outlined and accompanied by specific discussion of its advantages and disadvantages. Each strategy is also accompanied by an illustrative case study and a discussion of when to consider applying it and when it is best to avoid using it. I urge you to review these chapters carefully because they provide the foundation for a solid threat management process. **Study Questions** 1. Which intervention strategy paradoxically carries the greatest potential danger? Discuss the advantages and disadvantages of applying this intervention strategy, and when it is best used. 2. Subject interviews are difficult to conduct, and the decision to initiate direct contact with a subject should be preceded by a careful consideration of his or her potential reactions. Describe when one should consider interviewing a subject and when it is best to avoid contacting him or her. 3. Compare the "refocus" and the "assist" interview strategies, and discuss the advantages and disadvantages of each approach. 11495866