Approaching and Attacking Public Figures: A Contemporary Analysis of Communications and Behavior
J. Reid Meloy
There has been significant research during the past decade on abnormal or threatening communication and its relationship to escalation, approach, or attack behavior toward public figures (Meloy et al., 2008b). This paper is a review and critical integration of that research, which is pertinent to the operational needs of both public and private security, law enforcement, and intelligence agencies tasked with protecting public figures. Included are findings from new empirical studies (James et al., 2009b, 2010a, 2010b; Meloy et al., 2010; Unsgaard and Meloy, 2011) and theoretical advances not yet empirically tested.
The paper is divided into two sections: problematic approaches and attacks. The former refers to any behavior that entails physical movement toward a target that is potentially disruptive or threatening. The latter refers to any near-lethal approach, attack, or assassination of a targeted individual. This division is not arbitrary. It is necessary given the disparate research that has been conducted on samples of problematic approachers and samples of attackers and, in some cases, the divergence of results. It is the author’s hope that detailing these differences and similarities will broaden and deepen the understanding of such behaviors and also contribute to advances in operational research while ensuring the safety of public figures.
Predicting Movement from Communication to Approach
A detailed analysis of six studies, five of which were random samples, of problematic approaches to public figures, both politicians and celebrities, in the United States and Europe indicates a high degree of consistency across six headings that predict movement from communication to an approach (Meloy et al., 2010). The six studies in this analysis (Dietz et al., 1991a, 1991b; Scalora et al., 2002a, 2002b; James et al., 2009a; Meloy et al., 2008a) utilized variables that were similar enough to each other to warrant these six headings. They also provided quantitative data that show a statistically significant difference between approachers and nonapproachers toward the six samples of public figures. The following headings indicate the direction of a greater likelihood of an approach:
No threatening communications
Serious mental illness
Requests for help
Multiple means of communication
Multiple contacts and targets
No antagonistic communications
No threatening communications refers to the absence of an expressed desire to do harm to, or have physical harm occur to, a target. Serious mental illness refers to the presence of psychosis, indicated by evidence of hallucinations, delusions, or formal thought disorder, during the activity of concern. Requests for help refers to the subject asking for help from the target. Multiple means of communication refers to the subject using at least two methods of communication, such as writing letters, telephoning, e-mailing, sending gifts or enclosures, or faxing. Multiple contacts or targets is the most disparate heading and combines a subject’s repetitive contact of a target through any means of communication and the subject’s contact of other public figure targets—both have the characteristics of repetitiveness and dispersion. No antagonistic communications refers to the absence of any hostile, abusive, or degrading aspects to the communications.
Four of these six studies also conducted logistic regressions to see how accurately an approach could be predicted. Overall correct classification ranged from 76 to 83 percent, which is 25 to 30 percent better than chance, depending on the base rates for approach in each study. Although the predictor variables across the four studies differed, multiple communications and/or contacts with other targets emerged as a predictor variable in all four studies. It appears that a common thread across these predictor studies, as well as the other two studies, is a level of energy and fixation
on the part of the subject as a necessary prelude to approaching the target, operationally measured by multiple communications to the target and/or contacts with other public figures.
Although the consistency of these findings is promising, further research is necessary to cross-validate the results. Research designs could simply compare random samples of both approachers and nonapproachers to learn whether these six variables continued to both discriminate between the two groups and function in some circumstances as predictors of a problematic approach.
The operational application of these findings, heretofore individually known but not integrated prior to this study (Meloy et al., 2010), was somewhat misguided. A proportion of subjects whose communication characteristics are the opposite of these variables will move from communication to approach. This statistical reality was often overlooked in interpretations of the early research by readers who focused on significant differences between approachers and nonapproachers instead of actual frequencies. The assumption made by readers of this research was that significantly less difference meant that the lesser category had a zero frequency of the behavior. This is a logical error but appears to have become operationalized in some threat assessments—for example, the false belief that if there is no communicated threat, there is no risk, or that those who make a direct threat do not pose a threat. Consider the following data, which indicate the proportion of subjects who directly threatened a public figure and then did make a subsequent approach:
23 percent (Dietz et al., 1991a)
33 percent (Dietz et al., 1991b)
21 percent (Scalora et al., 2002a)
41 percent (Scalora et al., 2002b)
35 percent (Meloy et al., 2008a)
Even though these percentages are, in most cases, significantly lower than those for subjects who directly threatened but did not subsequently approach, they tended to be minimized, and often dismissed, when the results of the studies (particularly the Dietz studies) were informally discussed by threat assessment professionals. The within-study interpretations of these data were also problematic. For example, concerning threats and approaches to members of Congress, Dietz et al. (1991b) wrote, “Subjects who sent threats to a member of Congress were significantly less likely to pursue a face-to-face encounter with him or her” (p. 1466). This is statistically accurate but could be incorrectly interpreted as meaning that articulation of a direct threat would reduce risk in any one
subject who threatened—an interpretation that did not apply to one-third of their sample.
Future studies should emphasize this point and detail not only significant differences but also frequencies, effect size of the differences (preferably measured in odds ratios), and confidence intervals of the odds ratios. Also, Receiver Operating Characteristic analysis can be used to interpret predictive findings to ensure that base rates do not influence predictive outcome statements. The difficult problem of applying nomothetic (large-group) data to an individual case, wherein membership in a class does not necessarily imply individual predictive accuracy, also should be noted (Hart et al., 2007).
Behavioral Pathway, Motivation, and Mental Disorder
Odd, inappropriate, bizarre, or threatening communication addressed to a public figure cannot be fully understood by itself without other information about the sender, especially the behavioral pathway, motivation, and nature of the sender’s mental disorder. Behavioral pathway refers to the path along which an individual might progress in moving from communicating with a target to close physical proximity to the target. It was first mentioned by Dietz and Martell (1989) and then systematically studied by Fein and colleagues (1995) and Fein and Vossekuil (1998, 1999); later it was demarcated into stages by Calhoun and Weston (2003). Most recently it has been applied to studies of problematic approaches by various individuals to the British Royal Family.
James et al. (2009a) divided the stages into preapproach communications, communications and approach, approach without communications, unsuccessful breach of security, successful breach of security, and attack. Such a pathway analysis yields important behavioral findings, most notably the degree to which a perpetrator is influenced by both motivation and mental disorder. Motivation refers to the reason for the behavioral approach; it can be driven by “psychotic action” (Junginger, 1996)—behaviors driven by delusions or hallucinations. The nature of the mental disorder, if present, is most important when analyzed according to symptoms and behaviors—not diagnosis—and whether or not it causes, mediates, correlates with, or is unrelated to the motivation for the approach.
Three recent typologies attempt to address these aspects of public figure stalkers. Phillips (2006, 2008) identified five categories among an unknown number of subjects who approached, in a problematic way, protectees of the U.S. Secret Service: resentful, pathologically obsessed, infamy seeking, intimacy seeking, and nuisance or attention seeking. His typology focused on motive, positive symptoms of psychosis, and intent
to do harm but did not incorporate a behavioral pathway analysis. James et al. (2009a) identified eight motivational types among a random sample of 275 problematic approachers toward the British Royal Family: (1) delusions of royal identity, (2) amity seekers, (3) intimacy seekers, (4) sanctuary and help seekers, (5) royally persecuted, (6) counselors, (7) querulants (vexatious litigants), and (8) chaotics (those whose behaviors and motivations were highly disorganized). Their motivational typology was studied in relation to both the behavioral pathway and serious mental disorder.
In The Stalking Risk Profile, MacKenzie et al. (2009) identified eight motivational categories for stalkers of public figures: (1) resentful, (2) intimacy seekers, (3) incompetent suitors, (4) predatory (sexual motivation), (5) help seekers, (6) attention seekers, (7) the chaotic, and (8) unclassified. The profile was designed for risk management of such cases and is a structured professional judgment instrument (Monahan, 2000). Although typologies may seem irrelevant to operational tasks, they are not. A typology developed from a random sample of subjects of concern can bring more efficiency to the assignment and utilization of protective intelligence resources. Such work, along with research on mental disorders and behavioral pathways, could eventuate in an iterative decision-tree model for estimating the risk of problematic approaches toward or stalking of a protectee, much like the Classification of Violence Risk, developed to help predict the risk of short-term violence among persons discharged from acute care psychiatric facilities (Monahan et al., 2001; Monahan, 2010).
To determine operational validity, typologies need to be empirically tested for both inter-rater reliability and various kinds of validity across a number of variables important to protection, such as the prediction of a successful breach of security. The Phillips’ (2006, 2008) typology has not yet been empirically tested, but holds promise given its derivation from actual threateners and approachers identified by the U.S. Secret Service. Moreover, across all of the typologies there appears to be a supraordinate variable called fixation (from the Latin figo, meaning to be bound fast) that has both clinical and behavioral significance.
Emerging research indicates the importance of fixation, an intense preoccupation with an individual, activity, or idea (Meloy et al., 2008b). Normal fixations are a part of everyday life and include such states as romantic love, parental devotion, intense loyalty, and adulation. Pathological fixations are obsessive preoccupations that typically result in deterioration of the subject’s intimate, social, and occupational lives (Leets et al., 1995; Mullen et al., 2009a; Schlesinger, 2006). Such pathological fixations focus on a person or cause, the latter an intensely personal grievance
or quest for justice that inhibits effective social functioning and alienates others. Research in Europe indicates that fixation on a cause is related to risk of attack. In a study of nonterrorist attacks on Western European politicians between 1990 and 2003, 50 percent of attackers were found to be fixated on a cause (James et al., 2007; N = 24). In a study of attacks on the British Royal Family between 1778 and 1994 (James et al., 2008; N = 23), 63 percent of subjects whose motivation could be discerned (n = 19) were fixated on a cause.
Although it is difficult to make this distinction in the U.S. Secret Service Exceptional Case Study Project (ECSP) (Fein and Vossekuil, 1999), 67 percent of near-lethal approachers, attackers, and assassins had a grievance, as well as motivations that suggested focus on a cause, such as avenging a perceived wrong, bringing national attention to a perceived problem, saving the country/world, and bringing about political change (Fein and Vossekuil, 1998, 1999). Fixation on a cause may be a moderating variable between problematic approach and intent to attack, but it has not been empirically studied. Such fixations are distinguished from political extremism, which usually emerges in interactions of an actual or virtual group on the fringes of the traditional political process and is not as intensely personalized.
The nature of the fixations evident in abnormal communications to public figures has been studied in the context of British and Western European attackers (James et al., 2007, 2008). Although they may predict certain subsequent behavioral pathways or escalations, the empirical question is whether fixation on a cause incrementally contributes to risk of an attack on a public figure, especially politicians and government officials. The supraordinate, and perhaps clinically obscure concept of fixation, moreover, is often evident in warning behaviors.
Emerging research supports the belief that warning behaviors are important and should be construed as much broader than a specific threat (Meloy et al., 2004b; Scalora et al., 2002a, 2002b, 2003). Warning behaviors are dynamic and acute behaviors that precede an act of targeted violence, are related to it, and are therefore a risk factor for it. Warning behaviors show an intense and accelerating effort to further a particular quest, usually some highly personal cause. They often predict an approach (Meloy et al., 2010), but with some exceptions (Scalora et al., 2003). Intensity is usually measured by frequency of contact, duration of contact, multiple means of contact, and multiple contacts with other figures (target dispersion) and is associated in the research with the presence of serious mental disorder (James et al., 2009a; Scalora et al., 2002b).
Warning behaviors are also present in research on attacks. In contemporary Western European attacks (James et al., 2007), 46 percent of subjects evidenced warning behaviors before attacking1 and were more likely at the time of the attack to have a mental disorder (phi = 0.77 effect size), to be psychotic (0.65), and to show clear evidence of delusional beliefs (0.65). In the ECSP study (Fein and Vossekuil, 1998, 1999)—despite the very low frequency of direct threats toward the target or law enforcement (7 percent)—most subjects had a history of verbal or written communication about the target (77 percent); one out of four communicated to the target (23 percent); and 63 percent had a history of indirect, conditional, or direct threats about the target.
Specific warning behaviors may be another moderating variable between the research on problematic approaches and attacks. As yet, there are no studies of specific warning behaviors as predictors of a targeted attack. There are many case studies, though, that have retrospectively identified certain warning behaviors after an attack as predictors of that attack, but such circular reasoning does not advance predictive science. It would be most useful to determine both the specificity (accuracy of not predicting) and the sensitivity (accuracy of predicting) of certain warning behaviors in relationship to an attack—a task easier said than done. Moreover, the fundamental difficulty with warning behaviors is a lack of clarity in definition.
Meloy et al. (unpublished) have recently proposed that warning behaviors can be divided into seven categories:
Pathway warning behavior—acts that indicate research, planning, preparation, or implementation of an attack (Calhoun and Weston, 2003).
Fixation warning behavior—increasingly pathological preoccupations with a public figure or a highly personalized cause (Mullen et al., 2009a).
Identification warning behavior—a psychological desire to be a “pseudocommando” (Dietz, 1986; Knoll, 2010); development of a “warrior mentality” (Hempel et al., 1999); interest in and study of previous assassins or public figure attackers; or fascination with weapons, as indicated by collection, approach, skill development, or fantasy-based associations (Meloy, 1992a).
Novel aggression warning behavior—acts of violence unrelated to the planned and targeted attack that are committed for the first time.
Energy-burst warning behavior—increase in the frequency, duration, or variety of any warning behavior prior to an attack.
Leakage warning behavior—communication to a third party of intent to do harm to a target through an attack (O’Toole, 2000; Meloy and O’Toole, in press).
Direct-threat warning behavior—communication of a direct threat to the target or law enforcement before an attack on a public figure.
These seven categories have face validity and are commonly encountered in threat assessment cases, but they have not been subjected to empirical testing to determine their inter-rater reliability or their validity in predicting an attack.
Grandiosity and Entitled Reciprocity
Grandiosity and entitled reciprocity have emerged as two important psychological characteristics of subjects who approach public figures. They suggest both psychopathology in general and pathological narcissism—a sense of specialness that diminishes empathy for others. Grandiosity, an exaggerated sense of self-importance evident in communications, was tested in a logistic regression model in a study of those who approached or did not approach members of the British Royal Family (James et al., 2010a).
A regression for a model comprising the single factor of grandiosity produced an Area Under the Curve (AUC) of 0.74 (95 percent confidence interval 0.65 to 0.82) and correctly predicted almost 74 percent of the cases—nearly 74 percent of the approachers and over 73 percent of the nonapproachers. The effect size was moderate (phi = 0.47). Dietz and Martell (1989) found in their study more than 20 years ago that those who approached celebrities were significantly more likely (X2 = 4.85, p < .03) to evidence an excessive sense of self-importance or uniqueness (52 percent) than those who did not approach (36 percent). If subjects who problematically approached members of Congress took a “special constituent role,” it significantly increased the risk of an approach (46 versus 16 percent, X2 = 7.77, p = .0053). Grandiosity can be somewhat grounded in reality (e.g., “I can dramatically influence the votes in my district!”) or delusional (e.g., “I am the president!”). In many cases, grandiosity among such subjects compensates in fantasy for real-life failures in both work and love.
Grandiosity is a facet of pathological narcissism, an aspect of personality that is quite apparent in stalkers (Meloy, 1998; Mullen et al.,
2009b) and is an abnormal variant of narcissism, most clearly defined by Rothstein (1980) as “a felt quality of perfection” (p. 4). Grandiosity is also apparent in the attack research, specifically the ECSP study. Fein and Vossekuil (1998, 1999) reported that in 38 percent of principal incidents of near-lethal approach, attack, or assassination (N = 74) there was evidence that attention/notoriety was a goal. Of the eight motives they cited for attacks, grandiosity, or the wish to achieve such importance, can be inferred in five: (1) to achieve notoriety/fame, (2) to bring national attention to a perceived problem, (3) to save the country or the world, (4) to achieve a special relationship with the target, or (5) to bring about political change.
Years after the assassination of President John F. Kennedy in 1963 (Bugliosi, 2007), a close female friend of Lee Harvey Oswald reflected on Oswald’s personality in Minsk during the years 1959 to 1962: “I could paint a portrait of him as someone who thinks too much of himself but doesn’t work to become the person he wants to be…. The most important thing for Lee was that he wanted to become famous. Idea number one. He was fanatic about it, I think. Goal number one. Show that he was different from others, and you know, he achieved this goal” (Mailer, 1995, p. 321). A psychiatric social worker at the Youth House in Manhattan where Oswald was briefly placed as an adolescent for chronic truancy recorded similar findings: “He acknowledged fantasies about being all-powerful and being able to do anything he wanted. When asked if this ever involved hurting or killing people, he said that it did sometimes but [he] refused to elucidate on it” (Mailer, 1995, p. 365). She later wrote: “There is a rather pleasant, appealing quality about this emotionally starved, affectionless youngster which grows as one speaks to him” (p. 365).
Entitled reciprocity is the belief that a particular public figure owes the subject time and attention because of the time and attention the subject has paid to the public figure (Meloy et al., 2008b). It is also an aspect of pathological narcissism and is related to grandiosity: The subject’s importance demands that he receive the attention he deserves. In the British Royal Family study (James et al., 2010a), three motivations together accounted for nearly 72 percent of cases in which the communicators went on to approach—those with delusions of royalty, amity seekers, and intimacy seekers. Among all these motivations is the subject’s often delusional belief that he or she is owed a debt of gratitude through blood ties, friendship, or love.
Entitled reciprocity, however, has not been measured but may be an important predictor of resentment and perhaps aggression in certain subjects who are shunned by their public figure target. This could develop over time when communications are not responded to, or it could be an acutely negative emotional reaction when a highly anticipated personal
encounter with the public figure results in disappointment or the humiliation of being ignored. An example of grandiosity and entitled reciprocity from a letter writer to Prince Charles appears below (from author’s files, courtesy of the Fixated Threat Assessment Centre, London):
Dear Charles—God dam it. God dam you! Charles Prince of Wales! You know that the Catholic Church is a cult, right? You do know that, don’t you? Well, it is. And you shouldn’t be worshipping the Virgin Mary. She’s not the Queen of Heaven. I AM! I’m God’s wife, and you better make room for me there now! How dare you make me grovel in the dirt. Charles, I’m your Heavenly Mother! And you best start respecting me as such with a whole lot of hugs and kisses (on the cheek), well wishes, and tender loving care, or you are going to die a very long death starting right now! Stick to the Word of God, Charles.
Electronic Communications to Public Figures
Although it might seem that the written letter to Prince Charles is a relic of the past given the various social media platforms available today for communications, there is virtually no research contrasting the use of electronic communications (e.g., e-mails) toward public figures—with one exception, described by Schoeneman-Morris et al. (2007). This random study of e-mails and letters to members of the U.S. Congress found that letter writers were more problematic in that they were significantly more likely to exhibit symptoms of severe mental illness, engage in multiple target contacts, use multiple methods of contact, and approach. In fact, e-mail senders focused on government concerns, used obscene language, and displayed disorganization significantly more often. Threatening language was found in about half of all communications, with no significant differences between the two types of communication.
The research possibilities concerning electronic communications are endless. Any attempts to contrast samples of written letters and e-mails to public figures, with a focus on variables predicting a problematic approach, would contribute to this nascent area of investigation. Historically, written communications to public figures held a central place in threat assessment investigation, until challenged by the work of Fein and Vossekuil (1998, 1999), and further research could prove operationally useful.
Making or Posing a Threat
The distinction between making and posing a threat, first enunciated by Fein et al. (1995) and utilized in the ECSP study, has permeated the threat assessment community over the past 15 years as an important
theoretical construct and operational focus (Calhoun, 1998; Calhoun and Weston, 2003, 2008). Calhoun and Weston have challenged the assumptions that those who make a direct threat pose the greatest risk and that articulated threats are central to threat assessment. The challenge derived from the fact that none of the subjects who attacked or assassinated a U.S. public figure in the second half of the twentieth century communicated a direct threat to law enforcement or the target beforehand (Fein and Vossekuil, 1999). Subsequent research with other data on attacks of public figures makes clear that suspicious behavior (“warning behav-iors”) should be considered more important than a directly communicated threat when assessing the risk of an approach (Meloy et al., 2004b) to any public figure. For instance, James et al. (2007) also found that none of the subjects who attacked a Western European politician between 1990 and 2004 had directly communicated a threat beforehand. Such findings have moved threats from principal actor to supporting role in the theater of public figure threat assessment. However, warning behaviors—the somewhat obscure elements of a decision called “posing a threat”—are not clearly enunciated in the research, as noted earlier, and characteristics that lead to the decision that a subject “poses a threat” are also unknown. To further complicate matters, in certain cases, those who make threats also pose threats (Scalora et al., 2002a, 2002b).
To yield predictive data, the elements of such behaviors must be consistently defined and further studied, which could include a standardized definition of “posing a threat” and identifying the decision-making tree that leads to the perception of a “posed threat” by threat assessment professionals. One approach is to empirically study the various levels of concern and threat currently utilized by public and private security agencies to determine if they are reliably applied to various cases and the degree to which they predict certain approach behaviors or necessary interventions to curtail such approaches (e.g., arrest, hospitalization, surveillance). An important group that also merits study consists of those who problematically approach a public figure without communicating beforehand and without intending to attack. This group has been mostly neglected in the research to date. In the British Royal Family study (James et al., 2009a), this group was found to be more likely to behave in an intimidating manner, more likely to attempt to breach security, and much less likely to be fixated on the target than were approachers who communicated beforehand. This group (approach/no communication/no intent to attack) was not utilized as a comparison group in the ECSP research, which might have provided useful information.
ATTACKS AND ASSASSINATIONS
Mental Disorders, Attacks, and Assassinations
Research indicates the importance of mental disorders in a large proportion of subjects who attack public figures (Fein and Vossekuil, 1998, 1999; James et al., 2007, 2008; Meloy et al., 2004b). In the ECSP study, the finding of a substantial presence of mental disorder contrasts with the recommendations that diminished the importance of mental disorder (Fein and Vossekuil, 1999): 61 percent of subjects had been evaluated by a mental health professional, 43 percent had a history of delusions and were delusional during their attack or near-lethal approach, and 21 percent had a history of auditory hallucinations. However, Fein and Vossekuil made two important points: (1) within the delusion, the behavior toward the target may be quite rational and (2) focusing on the “thinking that leads a person to see assassination as an acceptable, or necessary action” (p. 332) is operationally much more useful than labeling or diagnosing the person with a particular mental disorder. Their position is supported by a large meta-analysis of the relationship between psychosis and violence (Douglas et al., 2009), which found that studies coded at the level of the symptom had significantly higher effect sizes, particularly active positive symptoms (delusions, hallucinations, etc.), when studying the relationship between violence and psychosis.
The operational implications of these findings and opinions are significant. Threat assessment professionals will find the description of behaviors and symptoms related to a mental disorder more useful than the particular diagnostic label. For example, discovering through investigative efforts that a particular subject has paranoid schizophrenia is much less relevant to threat assessment than noting that the subject believes the public target is an alien from another planet and needs to be killed so that he does not propagate and threaten other humans. On the other hand, although diagnostic labeling can obscure symptoms and behaviors relevant to threat assessment professionals, it can also function as an efficient communicator of probable symptom clusters for mental health professionals and signify the likely prognosis, or clinical outcome, if psychiatric or psychological treatment can be applied to the subject of concern.
Psychosis and delusions have also been found to be positively correlated with lethality risk (death or serious injury) in contemporary attacks on Western European politicians (James et al., 2007; phi = 0.49). Delusional content has strongly influenced the motivation, and thus the behavioral pathway, toward historical attacks on members of the British Royal Family (James et al., 2008). Hoffmann et al. (in press) have found that the majority of the small universe of potentially lethal attackers of public figures in Germany (1968 to 2004) were psychotic at the time (N = 9). All
but one of these attacks occurred since 1990 and were mostly directed at politicians. A related study concerning mass murderers also found that psychosis is significantly and positively correlated with a higher casualty rate (Meloy et al., 2004a). However, in the ECSP study, Fein and Vossekuil (1999) found a significantly lower frequency of delusion (p = .004) between the attackers (25 percent) and the near-lethal approachers (60 percent) in their analysis (n = 73) of mental state at the time of the principal incident.
Moreover, the studies of attacks on public figures in the United States, the United Kingdom, and Western Europe discussed here underscore the fact that serious mental disorders do not mitigate the risk of a planned attack on a public figure. All of these studies indicate that despite the presence of mental illness, subjects can carefully plan an attack over the course of days, weeks, or months. What has not been studied is whether the nature of the mental disorder (e.g., a delusional belief in one’s mission) may bring a resolve and commitment to the planning that would otherwise be absent, or at least marked by ambivalence, in the subject who was not delusional while planning an attack.
Predatory and Affective Violence
Most acts of violence toward public figures are predatory (instrumental)2 and involve a weapon, most likely a firearm (de Becker et al., 2008; Meloy et al., 2004b). This was documented in the ECSP study and confirmed by the study of Western European attacks. In emerging research there is also a suggestion that most individuals who embarked on a pathway toward violence did not use mind-altering substances at the time of the attack. This is in stark contrast to affective violence,3 in which substance abuse at the time of an attack is common (e.g., spousal violence, which is usually affective, commonly involves alcohol intoxication by the perpetrator, victim, or both; Miller, 1990).
It is a reasonable hypothesis, although untested, that subjects who engage in attack behavior toward a public figure will not use substances to increase the probability of tactical success, just as they typically will not explicitly threaten before an attack. There are a few cases, moreover, where subjects who engaged in predatory violence used psychotropic medications (barbiturates or sedatives/hypnotics) to deliberately main-
tain a state of calmness during the violence. However, these cases did not involve attacks on public figures (Meloy and Mohandie, 2001). There is also the anecdotal finding in certain cases that specific loss (e.g., job, family, reputation, income) precedes an attack and may actually be the point at which the date and time of the approach or attack is set—even though an attack had been contemplated for weeks or months. This loss is either cumulative or sudden, and there is likely to be a predisposition to attack a public figure that precedes it but without specificity of time, target, or location. These patterns of loss have yet to be studied among attackers and assassins of public figures, particularly in relation to location, timing, and target selection.
Although the attack research indicates that most acts of violence toward public figures are predatory (planned, purposeful, emotionless) rather than affective (reactive, impulsive, emotional), the latter do occur. In one celebrity study (Meloy et al., 2008a), a majority of the small number of attacks (n = 5) against a sample of 159 celebrity stalking victims were affective and did not involve a weapon. They usually involved attempts to grab the celebrity or assault security personnel during a public appearance. This celebrity sample was embedded in the largest study of stalkers to date (Mohandie et al., 2006; N = 1,005). When all acts of violence (n = 337) were compared in this latter study, those stalkers who had an actual relationship (prior intimate or acquaintance groups) with the target were more likely to be affectively violent, and those without a relationship (public figure and private stranger groups) were more likely to be predatorily violent (p = .001). Affective violence toward a public figure appears most likely to occur when there is a perceived rejection by the public figure, which could happen in a moment, such as the public figure not shaking hands or making eye contact with the subject in a rope line or security personnel interfering with attempted contact between the subject and the public figure (James et al., 2010b). Clinically, this may be more likely in an individual who has a strong sense of entitled reciprocity and grandiosity.
Conflation of the Politics of Hatred
One of the most important emerging trends in threats toward public figures in the United States is conflation of the various politics of hatred, which then becomes a pathological fixation. This contemporary conflation usually includes hatred of African Americans, Jews, the federal government, abortion rights supporters, or gun control advocates. Pathological fixation strongly suggests the existence of a major mental disorder—or at least a paranoia-tinged rigid and intolerant belief system, which draws its content from the various politics of hatred. This conflation has a number of real-world stimuli: It is likely accelerated by the election of an African
American as president of the United States in November 2008, his initial appointment of a Jewish chief of staff characterized by the press as aggressive, a perceived expansion of the federal government through actions taken in response to the recession of 2008 to 2009, and President Obama’s support of abortion rights.4
This conflation raises the question of whether there should be a strict operational demarcation between a terrorist threat and a fixated threat, especially if a fixated subject’s secondary motivation is to instill public fear or foment revolution.5 It also emphasizes the risk of an ideologically driven “lone terrorist” who acts outside of a terrorist cell or extremist group, often despite—or because of—failed attempts to associate with the latter (Puckitt, 2001; Biesterfeld and Meloy, 2008). Puckitt’s finding that lone terrorists often unsuccessfully try to affiliate with an extremist group and are rejected (thus intensifying their bond to a radical ideology) has direct operational implications for surveillance of such groups. Timothy McVeigh and Terry Nichols attempted to associate with the Michigan Militia in the years prior to the Oklahoma City bombing in April 1995, but they were rejected by the group for advocating direct violence against the government (from author’s files on U.S. v. Timothy James McVeigh).
What may at first appear to be a purely political motivation might actually mask a diagnosable psychiatric condition, wherein political, religious, or racial hatred provides the rationale for homicidal aggression.6 Lance Corporal Kody Brittingham was arrested, along with two other Marines, in December 2008 for attempted armed robbery of a motel. In his barracks at Camp Lejeune, North Carolina, investigators found maps, photos, and personal vital statistics on then president-elect Barack Obama and white supremacist materials. There was also a letter titled “Operation Patriot”:
I, Kody Brittingham, write this as a letter of intent. I’m in full mental health and clear judgment, having consciously made a decision, and in turn do so choose to carry out the actions entailed. I have sworn to defend my country, my Constitution, and the values and virtues of the aforementioned. My vow was to protect against all enemies, both foreign and domestic. I have found, through much research, evidence to support my current state of mind. Having found said domestic enemy (BHO), it is my duty and honor to carry out by all means necessary to protect my nation and her people from this threat. (Zeleny and Rutenberg, 2009)
Lance Corporal Brittingham pled guilty in August 2009 to threatening to kill the president and to attempted armed robbery. Although there is no publicly available evidence that he has a diagnosable mental disorder—again, the signs, symptoms, and behavior caused by any mental disorder are more important to threat assessment than the label—a conflation of racism and patriotism is a reasonable inference. Increased threat associated with the politics of hatred, at this time coming from the extreme right, has been documented in reports by the U.S. Department of Homeland Security (2009) and the Southern Poverty Law Center (2009).
The warning behavior that is arguably the most important from an operational perspective is “leakage” of intent to harm a target, whether vague or specific, to third parties (O’Toole, 2000; Meloy and O’Toole, in press). Leakage is one of the seven types of warning behaviors noted earlier and is characteristic of both assassins of public figures and mass murderers (Hempel et al., 1999; Meloy et al., 2004a). Individuals in both groups want to carry out a very low frequency but highly catastrophic act of violence against an intended target (either identified beforehand or opportunistic). These are “black swan” events.7
In the ECSP study (Fein and Vossekuil, 1998, 1999), 63 percent of subjects (N = 83) had a history of indirect, conditional, or direct threats about their target, usually to family, friends, co-workers, or others known to the target. There were no direct threats to the target or law enforcement officials beforehand by those who attacked or assassinated their target. If near-lethal approachers are included, this direct threat frequency
increases from 0 to 7 percent.8 In the Western European study, 46 percent of attacks (N = 24) were preceded “by obvious and often flamboyant warning behaviors in the form of threatening or bizarre communications to politicians, public figures, or police forces” (James et al., 2007, p. 342). There were no cases in which the attack was preceded by a direct threat to kill the individual who was eventually attacked. Among adult mass murderers, the majority leak their intent to attack to third parties, but only a minority communicate a direct threat to their targets beforehand (Meloy et al., 2004a). Although the reasons for this dynamic likely vary from case to case, it is most plausible that the absence of a direct threat is motivated by a desire for tactical success. The prevalence of leakage is the inability of the subject to contain his or her excitement, satisfaction, or anxiety while researching, planning, and implementing an attack.
Leakage is also evident in journals, diaries, and electronically via the Internet. Online leaks have not yet been formally studied. A recent example of leakage on the Internet is the nine-month blogging carried out by George Sodini before committing mass murder near Pittsburgh, Pennsylvania, in August 2009. In his blog he discussed his intent, timing, preparations, and one “false start” but not an exact location. The irony of this case is that the entire world could have paid attention to him, but no one did.
New Threat Research
Although leakage is typically much more prevalent than a direct threat when investigating a problematic approacher or potential attacker of a public figure, new homicidal threat research, mostly related to stalking of nonpublic figures and in a mental health context, empirically supports the conventional belief that all threats should be taken seriously. Warren et al. (2007) found substantially higher rates of assault and even homicide in Australia following threats in 1993 and 1994 to kill among a large sample (N = 613) of subjects. The offense required that the threat produced fear in the victim. The individuals at highest risk for subsequent violence were young, had mental disorders, abused substances, and did not have prior criminal convictions. Among homicidal threateners, the rates of homicide and suicide almost exactly mirrored the results found in a classic study done more than 40 years ago (MacDonald, 1963) and were orders of magnitude higher than expected by chance (Warren et al., 2007).
Within 10 years, 44 percent of the threateners were convicted of further violent offenses, including 3 percent (n = 19) for homicides. Twenty-six percent (n = 5) of the homicide victims were those originally threatened by the subject. Sixteen threateners (2.6 percent) committed suicide, and three were murdered. Substance abuse, prior violence, limited education, and untreated mental disorder contributed to any risk of violence by those who threatened to kill.
In another study (Smith, 2008), a sample of FBI threateners (N = 96) were more likely to act harmfully if their communications showed lower ambivalent hostility and higher conceptual complexity. Lower ambivalent hostility was related to a lack of paranoia; higher conceptual complexity was related to deliberative thinking. This finding of a lack of paranoia among those who harmed is consistent with the British Royal Family problematic approach studies discussed earlier, which found paranoia negatively associated with breach activity. Likewise, the ECSP and European attacks studies documented the minor role of paranoia among assassins, attackers, and near-lethal approachers. In the FBI study the author believed that the results could be generalized to all written threat cases of the FBI’s National Center for the Analysis of Violent Crimes, although only 10 percent of the cases involved public figures as targets (Smith, 2008). Data continue to emerge to indicate that threatening communicators, if they are subsequently violent, might not attack the original target of the communicated threat.
Depression and Suicidality
Emerging research suggests the importance of depression and suicidality in the clinical motivation for an approach toward or attack of a public figure. Meloy et al. (2004b) found that many subjects evidenced a downward spiral in their lives in the months or year preceding the approach or attack, usually a combination of social failure and personal vulnerability to chronic anger, depression, or psychosis. Fein and Vossekuil (1998, 1999) found that 44 percent of subjects had a history of serious depression or despair and 24 percent had a history of suicidal attempts. James et al. (2007) found that 12 percent of Western European attackers clearly intended to die during the assault. Mohandie et al. (2006) found in their large study of stalkers that 25 percent evidenced suicidality (e.g., threats, gestures, attempts) in their histories. Perpetration of violence by persons with major mental disorder is correlated with adverse outcomes such as suicide and self-harm (Nicholls et al., 2006). As Douglas et al. (2009) wrote, “Negative symptoms that result in depression or suicidality may increase violence risk, as morbid thoughts of self-harm may change or expand in focus to include others” (p. 4).
Attacks on a public figure, depression, and suicidality appear to be linked for several reasons:
the wish to “suicide by cop” (Mohandie et al., 2009) while attempting to attack or assassinate is a more public forum for ending one’s life and may satisfy other narcissistic needs for attention;
the “suicidal” communication beforehand may be one aspect of “final act” behavior;
suicidal intent as one of several motivations for an attack on a public figure may be positively correlated with the amount of “lethal force” security surrounding the target; and
suicidal desires or intent can be given a positive valence by redefining them as motivations for martyrdom and linking them to a religious or political cause (Menninger, 1938; Reik, 1941).
These motivations are, in turn, usually fueled by hatred of a particular race, ethnic group, religion, or political position, often combined with a fear of conspiracy or persecution by the targeted individual or group. Such fear may be paranoid, without any basis in reality, or it can be historically factual and reasonable given personal or group suffering at the hands of another.
At the other end of the clinical spectrum, and typically devoid of depressive symptoms, is the psychopathic attacker or assassin. The construct of psychopathy has received virtually no attention in the research on stalking, threatening, and attacking public figures. Psychopathy, or psychopathic personality, is characterized by affective deficiency (i.e., absence of empathy, bonding, guilt, or remorse) and chronic antisocial behavior (criminal and noncriminal exploitation of others; Hare, 2003). Psychopathy has never been measured in either problematic approachers or attackers of public figures, although it has been theoretically proposed as an important construct (O’Toole et al., 2008). Most importantly, psychopathy accounts for the largest proportion of explainable variance in research on the risk of both criminal and civil violence (Monahan et al., 2001). It is a reliable and valid scientific construct that is relatively easily measured by trained professionals (Hare, 2003); it correlates with the risk of predatory (i.e., planned, purposeful) violence, which is the most likely mode of violence when a public figure is attacked (Meloy et al., 2008b); and there now exists a security and law enforcement assessment tool for measuring psychopathy (P-SCAN, available from http://www.mhs.com).
The relationship between psychopathy and psychosis is also notable.
When they coexist in a violent subject, the former will typically play a much larger role than the latter in accounting for the violence. The effect of psychosis on violence risk indicates a small, though reliable, effect size of r = .12 to .16 (Douglas et al., 2009).9 Psychopathy and its impact on violence show effect sizes between .25 and .30 (Douglas et al., 2009). In general, psychosis shows a significantly lower odds ratio for the prediction of violence than personality disorder.
The relevance to attacks on public figures is the operational importance, though not yet measured, of psychopathy in particular and personality disorder in general in motivating a near-lethal approach, attack, or assassination. In the ECSP study (Fein and Vossekuil, 1998, 1999), 39 percent of the subjects were never evaluated by a mental health professional, and 57 percent had no history of delusional ideas. Seventy-five percent of attackers were not delusional during the principal incident, and 40 percent of near-lethal approachers were not delusional. In the European attacks study (James et al., 2007), 46 percent were determined to have no mental disorder, highlighting the reliable absence of mental disorder in a proportion of public figure attackers and the likely presence, though unmeasured, of character pathology (such as psychopathy) as a motivation for the assault.
Clarke (1982) identified Type III subjects in his typology of U.S. assassins as psychopaths, who experience life as meaningless, and the motivation to assassinate is the nonpolitical expression of rage in someone devoid of human attachments who does not experience the more socialized emotions of guilt, shame, or remorse. As he wrote, “They are belligerently contemptuous of morality and social convention” (1982, p. 15). He identified three American assassins who fit this third type: (1) Guiseppe Zangara, an Italian immigrant who attempted to kill President Franklin Roosevelt on February 13, 1933; (2) Arthur Bremer,10 who shot and crippled Alabama Governor George Wallace on May 15, 1972; and (3) John Hinckley, Jr., who shot and wounded President Ronald Reagan on March 30, 1981. Again, empirical measurement of psychopathy in these individuals has not been done, but given the extensive published materials on these subjects, it could be accomplished without a clinical interview by using a standardized observational instrument, the Psychopathy Checklist–Revised (Hare, 2003).
Pathway to Violence
Emerging research confirms the existence of a pathway to violence (Calhoun and Weston, 2003)—consisting of the stages of grievance, ideation, research/planning, preparation, breach, and attack—but it is more complex than first formulated. Most approaches to a public figure are not intended to be or are predictive of violence (Meloy et al., 2008b). A pathway to violence depends on the motivation for communication and approach and the perceived reaction of the public figure, which will virtually always be personalized by the subject. For example, a subject whose initial approach is motivated by a desire for help might subsequently become aggressive and hostile if the expected response is not forthcoming. Likewise, there may be no pathway at all, other than an initial approach resulting in a successful breach of security and an attack. An example is the assassination of Swedish Foreign Minister Anna Lindh on September 10, 2003. Her attacker, Mijailo Mijailovic, had a fantasy of killing someone famous and actually reported to a psychiatric clinic that he had murdered someone six days before the Lindh assassination. He was diagnosed with “a personality disorder intermittently bordering on psychosis” and prescribed medications. He then subsequently and accidentally encountered Ms. Lindh, who was without a security detail, in a Stockholm department store. Minutes later he stabbed her to death in front of her friend. She was a target of opportunity, and a thorough investigation indicated no evidence of prior planning (Unsgaard and Meloy, 2011).
Besides a pathway to violence, there are other domains of risk. A subject might disrupt the public figure’s schedule, there may be recidivism or persistence of pursuit (James et al., 2009b), or a problematic approacher might embarrass or inconvenience a public figure target through behaviors that pose no physical threat.
Communicated Threats Schematic
A schematic has been proposed to improve the analytic clarity of communicated threats; it includes motivation, means, manner, and material content (Meloy et al., 2008b). Motivation refers to whether the threat is expressive (to regulate affect of the threatener) or instrumental (to control the behavior of the target). Means refers to the method of communication, such as letters, e-mails, telephone calls, text messages, and faxes. Manner refers to whether the threat is communicated directly or indirectly to the target. Material content refers to all material aspects of the threat itself, usually analyzable through the use of forensic technology, such as linguistics, DNA transfer evidence, fingerprint evidence, or graphic presentation. This face-valid schematic has not yet been tested for any predictive or concurrent validity but is an attempt to clarify terms used to
study communications that are not mutually exclusive (e.g., conditional threat, veiled threat, direct threat) and have been inconsistently defined in previous studies.
Children of Public Figures
Another concern—inordinate interest in the children of a public figure—also deserves attention from a problematic approach or threat perspective. Inappropriate communication (frequent, long-duration, bizarre, or odd) to the minor children of famous people usually arises from three psychological sources: (1) nonpsychotic transference, or the shifting of emotions from one’s own children, or oneself as a child, or the absence of children, to the offspring of a public figure; (2) psychotic transference, or a delusional belief that the subject is related to or has an important role in the children’s actual lives; and (3) pedophilic interest—an almost exclusively male subject’s interest in minors as sexual objects. There is no published research on this topic concerning the children of public figures, although there are many safety programs in place in schools and elsewhere for all children who may encounter a relative or stranger with malevolent intent. There is at least one private study that has been completed related to crimes against children of public figures, but the findings of the study are unavailable.
The absence until 2009 of two prepubescent children in the White House since the presidency of John F. Kennedy warrants careful and immediate study of these potential concerns and threats. Perhaps the most onerous threat toward the children of public figures is kidnapping. Although research in this area is dated and no published research has focused exclusively on the children of public figures, there were 115 stereotypical kidnappings in 1999, defined as abductions perpetrated by a stranger or slight acquaintance and involving a child who was transported 50 or more miles, detained overnight, held for ransom or with the intent to keep the child permanently, or killed. In 40 percent of these cases the child was killed, and in another 4 percent the child was not recovered. Two-thirds of these stereotypical kidnappings involved female victims between the ages of 6 and 14 (Finkelhor et al., 2002). Other studies involving large national samples have found that offender and offense characteristics in child abductions vary significantly according to the victim’s age, gender, and race (Boudreaux et al., 1999). For example, sexual gratification is the most likely motivation for stranger abduction of a girl 5 to 10 years old. Time and distance intervals are also critical to case solvability in child abduction murders (Brown and Keppel, 2007). Most child abductions, though, are perpetrated by family members or close relatives (Boudreaux et al., 1999).
FUTURE OPERATIONAL RESEARCH
The map is not the territory. In other words, despite a theory’s elegance or the comprehensiveness of data collection, research results will not exactly reflect reality. There will always be known unknowns, unknown unknowns, and individual differences that are not captured by large-group research, which is the cornerstone of the social and behavioral sciences. On rare occasions, “black swans”—events that are completely unpredictable yet catastrophic—will appear (Taleb, 2007), challenging historical beliefs and assumptions that have guided operational decisions, even when based on a robust research program.
Research studies of individuals who problematically approach, escalate, and in a few cases attack public figures should not only utilize nomothetic (large-group) data randomly drawn from recent case management files but also focus on select cases and the individual differences defining them. Subjects of particular interest to law enforcement, security, and intelligence agencies because of their unusual or outlier behaviors could yield important data by being forensically evaluated with standardized tests and measures if possible.11 Sensitivity to all forms of methodological challenges in research (including study design, measurement, and confounding factors) should be rigorously maintained to minimize their impact on findings and, when unavoidable, should be set forth as limitations.
The study of those who approach or attack public figures is a nascent science, but it can bring an operational efficiency to those tasked with protecting public figures. Research continues to refine our understanding of the interplay of protective intelligence gathering and personal protection and contributes to minimizing the vulnerability of public figures to an attack. The danger in many cases is quite real. As Hoffmann and Meloy (2008, p. 191) have written, “Disappointment or humiliation is the very predictable outcome when a public figure is pursued. The idolized figure is now beneath contempt. Yearning becomes disgust. Love may even become hatred. Rationalizations are put into place. Delusion may bring a resolve that is immutable. Aggression intensifies. Revenge is in the air.”
Biesterfeld, J., and J.R. Meloy. 2008. The public figure assassin as terrorist. In J.R., Meloy, L. Sheridan, and J. Hoffmann, eds., Stalking, Threatening, and Attacking Public Figures: A Psychological and Behavioral Analysis (pp. 143-162). New York: Oxford University Press.
Boudreaux, M., W. Lord, and R. Dutra. 1999. Child abduction: Age-based analysis of offender, victim, and offense characteristics in 550 cases of alleged child disappearance. Journal of Forensic Sciences, 44:539-550.
Brown, K., and R. Keppel. 2007. Child abduction murder: An analysis of the effect of time and distance separation between murder incident sites and solvability. Journal of Forensic Sciences, 52(1):137-145.
Bugliosi, V. 2007. Reclaiming History: The Assassination of President John F. Kennedy. New York: W.W. Norton.
Calhoun, F. 1998. Hunters and Howlers: Threats Against Federal Judicial Officials in the United States, 1789-1993. Arlington, VA: U.S. Department of Justice, U.S. Marshals Service.
Calhoun, F., and S. Weston. 2003. Contemporary Threat Management: A Practical Guide for Identifying, Assessing, and Managing Individuals of Violent Intent. San Diego, CA: Specialized Training Services.
Calhoun, F., and S. Weston. 2008. On public figure howlers. In J.R. Meloy, L. Sheridan, and J. Hoffmann, eds., Stalking, Threatening, and Attacking Public Figures: A Psychological and Behavioral Analysis (pp. 105-122). New York: Oxford University Press.
Clarke, J. 1982. American Assassins. Princeton, NJ: Princeton University Press.
De Becker, G., T. Taylor, and J. Marquart. 2008. Just 2 Seconds: Using Time and Space to Defeat Assassins. Studio City, CA: Gavin de Becker Center for the Study and Reduction of Violence.
Dietz, P. 1986. Mass, serial, and sensational homicides. Bulletin of the New York Academy of Medicine, 62(5):477-491.
Dietz, P., and D. Martell. 1989. Mentally Disordered Offenders in Pursuit of Celebrities and Politicians. Washington, DC: National Institute of Justice.
Dietz, P., D. Matthews, C. Van Duyne, D. Martell, C. Perry, T. Stewart, and J. Warren. 1991a. Threatening and otherwise inappropriate letters to Hollywood celebrities. Journal of Forensic Sciences, 36:185-209.
Dietz, P., D. Matthews, D. Martell, T. Stewart, D. Hrouda, and J. Warren. 1991b. Threatening and otherwise inappropriate letters to members of the United States Congress. Journal of Forensic Sciences, 36:1445-1468.
Douglas, K., L. Guy, and S. Hart. 2009. Psychosis as a risk factor for violence to others: A meta-analysis. Psychological Bulletin, 135(5):679-706.
Federal Bureau of Investigation. 1996. Terrorism in the United States. Washington, DC: Counterterrorism Threat Assessment and Warning Unit, National Security Division, Federal Bureau of Investigation.
Fein, R., and B. Vossekuil. 1998. Preventing attacks on public officials and public figures: A Secret Service perspective. In J.R. Meloy, ed., The Psychology of Stalking: Clinical and Forensic Perspectives (pp.176-194). San Diego, CA: Academic Press.
Fein, R., and B. Vossekuil. 1999. Assassination in the United States: An operational study of recent assassins, attackers, and near-lethal approachers. Journal of Forensic Sciences, 44(2):321-333.
Fein, R., B. Vossekuil, and G. Holden. 1995. Threat Assessment: An Approach to Prevent Targeted Violence. Washington, DC: Office of Justice Programs, National Institute of Justice, U.S. Department of Justice.
Finkelhor, D., H. Hammer, and A. Sedlak. 2002. Nonfamily Abducted Children: National Estimates and Characteristics. Washington, DC: Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, U.S. Department of Justice.
Hare, R.D. 2003. Manual for the Psychopathy—Checklist Revised, 2nd edition. Toronto, Canada: Multihealth Systems.
Hart, S., C. Michie, and D. Cooke. 2007. Precision of actuarial risk assessment instruments: Evaluating the “margin of error” of group v. individual predictions of violence. British Journal of Psychiatry, 190(Suppl. 49):S60-S65.
Hempel, A., J.R. Meloy, and T. Richards. 1999. Offender and offense characteristics of a nonrandom sample of mass murderers. Journal of the American Academy of Psychiatry and the Law, 27(2):213-225.
Hoffmann, J., and J.R. Meloy. 2008. Contributions from attachment theory and psychoanalysis to advance understanding of public figure stalking and attacking. In J.R. Meloy, L. Sheridan, and J. Hoffmann, eds., Stalking, Threatening, and Attacking Public Figures (pp. 165-194). New York: Oxford University Press.
Hoffmann, J., J.R. Meloy, A. Guldimann, and A. Ermer. In press. Public figure attacks in Germany, 1968-2004. Behavioral Sciences and the Law.
James, D.V., P. Mullen, J.R. Meloy, M. Pathe, F. Farnham, L. Preston, and B. Darnley. 2007. The role of mental disorder in attacks on European politicians, 1990-2004. Acta Psychiatrica Scandinavica, 116(5):334-344.
James, D.V., P. Mullen, M. Pathé, J.R. Meloy, F. Farnham, L. Preston, and B. Darnley. 2008. Attacks on the British Royal Family: The role of psychotic illness. Journal of the American Academy of Psychiatry and the Law, 36:59-67.
James, D.V., J.R. Meloy, P. Mullen, M. Pathé, F. Farnham, L. Preston, and B. Darnley. 2010a. Abnormal attentions toward the British Royal Family: Factors associated with approach and escalation. Journal of the American Academy of Psychiatry and the Law, 38(3):329-340.
James, D.V., P. Mullen, J.R. Meloy, M. Pathé, L. Preston, B. Darnley, F. Farnham, and M. Scalora. 2010b. Stalkers and harassers of British Royalty: An exploration of proxy behaviors for violence. Behavioral Sciences and the Law, doi: 10.1002/bsl.922.
James, D.V., P.E. Mullen, M. Pathé, J.R. Meloy, L.F. Preston, B. Darnley, and F. Farnham. 2009a. Stalkers and harassers of royalty: The role of mental illness and motivation. Psychological Medicine, 39(9):1479-1490.
James, D.V., T. McEwan, R. MacKenzie, J.R. Meloy, P. Mullen, M. Pathé, F. Farnham, L. Preston, and B. Darnley. 2009b. Persistence in stalking: A comparison of general and public figure stalking samples. Journal of Forensic Psychiatry and Psychology, 21(2):283-305.
Junginger, J. 1996. Psychosis and violence: The case for a content analysis of psychotic experience. Schizophrenia Bulletin, 22(1):91-103.
Knoll, J. 2010. The “pseudocommando” mass murderer: Part I, The psychology of revenge and obliteration. Journal of the American Academy of Psychiatry and the Law, 38(1):87-94.
Leets, L., G. De Becker, and H. Giles. 1995. Fans: Exploring expressed motivations for contacting celebrities. Journal of Language and Social Psychology, 14(1-2):102-123.
MacDonald, J.M. 1963. The threat to kill. American Journal of Psychiatry, 120:125-130.
MacKenzie, R., T. McEwan, M. Pathe, D. James, J. Ogloff, and P. Mullen. 2009. The Stalking Risk Profile: Guidelines for the Assessment and Management of Stalkers. Melbourne, Australia: Centre for Forensic Behavioural Science, Monash University.
Mailer, N. 1995. Oswald’s Tale: An American Mystery. New York: Random House.
Meloy, J.R. 1992a. Violent Attachments. Northvale, NJ: Aronson.
Meloy, J.R. 1992b. Revisiting the Rorschach of Sirhan Sirhan. Journal of Personality Assessment, 58(3):548-570.
Meloy, J.R., ed. 1998. The Psychology of Stalking: Clinical and Forensic Perspectives. San Diego, CA: Academic Press.
Meloy, J.R. 2006. The empirical basis and forensic application of affective and predatory violence. Australian and New Zealand Journal of Psychiatry, 40(6-7):539-547.
Meloy, J.R., J. Hoffmann, A. Guldimann and D. James. Unpublished. The concept of warning behaviors in threat assessment. Available: http://www.forensis.org [accessed June 2010].
Meloy, J.R., A. Hempel, T. Gray, K. Mohandie, A. Shiva, and T. Richards. 2004a. A comparative analysis of North American adolescent and adult mass murderers. Behavioral Sciences and the Law, 22(3):291-309.
Meloy, J.R., D.V. James, P.E. Mullen, M. Pathé, F. Farnham, L. Preston, and B. Darnley. 2004b. A research review of public figure threats, approaches, attacks, and assassinations in the United States. Journal of Forensic Sciences, 49(5):1086-1093.
Meloy, J.R., D.V. James, P.E. Mullen, M. Pathé, F. Farnham, L. Preston, and B. Darnley. 2010. Factors associated with escalation and problematic approaches toward public figures. Journal of Forensic Sciences, doi: 10.1111/j.1556-4029.2010.01574.x.
Meloy, J.R., and K. Mohandie. 2001. Investigating the role of screen violence in specific homicide cases. Journal of Forensic Sciences, 46(5):1113-1118.
Meloy, J.R., K. Mohandie, and M. Green. 2008a. A forensic investigation of those who stalk celebrities. In J.R. Meloy, L. Sheridan , and J. Hoffmann, eds., Stalking, Threatening, and Attacking Public Figures: A Psychological and Behavioral Analysis (pp. 37-54). New York: Oxford University Press.
Meloy, J.R., and M.E. O’Toole. In press. The concept of leakage in threat assessment. Behavioral Sciences and the Law.
Meloy, J.R., L. Sheridan, and J. Hoffmann, eds. 2008b. Stalking, Threatening, and Attacking Public Figures: A Psychological and Behavioral Analysis. New York: Oxford University Press.
Menninger, K. 1938. Man Against Himself. New York: Harcourt Brace.
Miller, B. 1990. The interrelationships between alcohol, drugs and family violence. In M. De la Rosa, E. Lambert, and B. Gropper, eds., Drugs and Violence: Causes, Correlates, and Consequences (pp. 186-216). Washington, DC: National Institutes of Health, U.S. Department of Health and Human Services.
Mohandie, K., J.R. Meloy, M. McGowan, and J. Williams. 2006. The RECON typology of stalking: Reliability and validity based upon a large sample of North American stalkers. Journal of Forensic Sciences, 51(1):147-155.
Mohandie, K., J.R. Meloy, and P. Collins. 2009. Suicide by cop among officer-involved shooting cases. Journal of Forensic Sciences, 54(2):1-7.
Monahan, J. 2000. Clinical and actuarial predictions of violence. In D. Faigman, D. Kaye, M. Saks, and J. Sanders, eds., Modern Scientific Evidence: The Law and Science of Expert Testimony (pp. 300-318). St. Paul, MN: West Publishing.
Monahan, J. 2010. Classification of violence risk. In R. Otto and K. Douglas, eds., Handbook of Violence Risk Assessment (pp. 187-198). New York: Routledge.
Monahan, J., H. Steadman, E. Silver, P. Appelbaum, P. Robbins, E. Mulvey, L. Roth, T. Grisso, and S. Banks. 2001. Rethinking Risk Assessment: The MacArthur Study of Mental Disorder and Violence. New York: Oxford University Press.
Mullen, P., D. James, J.R. Meloy, M. Pathé, F. Farnham, L. Preston, B. Darnley, and J. Berman. 2009a. The fixated and the pursuit of public figures. Journal of Forensic Psychiatry and Psychology, 20(1):33-47.
Mullen, P., M. Pathé, and R. Purcell. 2009b. Stalkers and Their Victims, 2nd edition. London, UK: Cambridge University Press.
Nicholls, T., J. Brink, S. Desmarais, C. Webster, and M. Martin. 2006. The Short-Term Assessment of Risk and Treatability (START): A prospective validation study in a forensic psychiatric sample. Assessment, 13(3):313-327.
O’Toole, M.E. 2000. The School Shooter: A Threat Assessment Perspective. Quantico, VA: Critical Incident Response Group, FBI Academy, National Center for the Analysis of Violent Crime.
O’Toole, M.E., S. Smith, and R.D. Hare. 2008. Psychopathy and predatory stalking of public figures. In J.R. Meloy, L. Sheridan, and J. Hoffmann, eds., Stalking, Threatening, and Attacking Public Figures: A Psychological and Behavioral Analysis (pp. 215-243). New York: Oxford University Press.
Phillips, R.T. 2006. Assessing presidential stalkers and assassins. Journal of the American Academy of Psychiatry and the Law, 34(2):154-164.
Phillips, R.T. 2008. Psychiatric consultation to the United States Secret Service. In J.R. Meloy, L. Sheridan, and J. Hoffmann, eds., Stalking, Threatening, and Attacking Public Figures: A Psychological and Behavioral Analysis (pp. 363-385). New York: Oxford University Press.
Puckitt, K.M. 2001. The Lone Terrorist: The Search for Connection and Its Relationship to SocietalLevel Violence. Washington, DC: Counterterrorism Division, Federal Bureau of Investigation, U.S. Department of Justice.
Reik, T. 1941. Masochism in Modern Man. New York: Farrar Straus.
Rothstein, A. 1980. The Narcissistic Pursuit of Perfection. New York: International Universities Press.
Scalora, M.J., J.V. Baumgartner, D. Callaway, W. Zimmerman, M.A. Hatch-Maillette, C.N. Covell, R.E. Palarea, J.A. Krebs, and D.O. Washington. 2002a. An epidemiological assessment of problematic contacts to members of Congress. Journal of Forensic Sciences, 47(6):1360-1364.
Scalora, M.J., J.V. Baumgartner, D. Callaway, W. Zimmerman, M.A. Hatch-Maillette, C.N. Covell, R.E. Palarea, J.A. Krebs, and D.O. Washington. 2002b. Risk factors for approach behavior toward the U.S. Congress. Journal of Threat Assessment, 2(2):35-55.
Scalora, M.J., J. Baumgartner, and G. Plank. 2003. The relationship of mental illness to targeted contact behavior toward state government agencies and officials. Behavioral Sciences and the Law, 21(2):239-249.
Schlesinger, L. 2006. Celebrity stalking, homicide, and suicide: A psychological autopsy. International Journal of Offender Therapy and Comparative Criminology, 50(1):39-46.
Schoeneman-Morris, K., M.J. Scalora, G. Chang, W. Zimmerman, and Y. Garner. 2007. A comparison of email versus letter threat contacts toward members of the United States Congress. Journal of Forensic Sciences, 52(5):1142-1147.
Smith, S. 2008. From violent words to violent deeds: Assessing risk from FBI threatening communication cases. In J.R. Meloy, L. Sheridan, and J. Hoffmann, eds., Stalking, Threatening, and Attacking Public Figures: A Psychological and Behavioral Analysis (pp. 435-455). New York: Oxford University Press.
Southern Poverty Law Center. 2009. The Second Wave: Return of the Militias. Montgomery, AL: Southern Poverty Law Center.
Taleb, N. 2007. The Black Swan. New York: Random House.
Unsgaard, E., and J.R. Meloy. 2011. The assassination of the Swedish Minister for Foreign Affairs. Journal of Forensic Sciences. Available: http://onlinelibrary.Wiley.com/ doi/10.1111/j.1556-4029.2010.01653.x/full [accessed January 2011].
U.S. Department of Homeland Security. 2009. Rightwing Extremism: Current Economic and Political Climate Fueling Resurgence in Radicalization and Recruitment. Washington, DC: Office of Intelligence and Analysis, U.S. Department of Homeland Security.
Warren, L., P. Mullen, S. Thomas, J. Ogloff, and P. Burgess. 2007. Threats to kill: A follow-up study. Psychological Medicine, 38(4):599-605.
Zeleny, J., and J. Rutenberg. 2009. Threats against Obama spiked early. The New York Times, May 19, p. A1.