Content area
We assessed the relationships between elements of the Macdonald triad (i.e., enuresis, cruelty to animals, and fire setting) and parental abuse (i.e., psychological, physical, and sexual abuse) within a population of serial killers. Data from 280 serial killers were retrieved from the Radford/Florida Gulf Coast University Serial Killer Database. This unique source of information was compiled from biographical books, newspaper articles, court documents, self-reports, and online sources. A series of chi-square tests for independence and binary logistic regression analyses were used to assess goodness of fit and regression relationships, respectively. The findings indicated significant statistical relationships among enuresis, fire setting, and animal cruelty to key dimensions of parental physical and psychological abuse, respectively. The use of binary logistic regression not only validated these joint associations but also elucidated the very potent relationships of fire setting and enuresis to that of psychological and physical parental abuse within this unique sample of serial killers. Associations between enuresis, fire setting, and animal cruelty to parental sexual abuse were not statistically significant. From a diagnostic perspective, enhanced awareness of these toxic relationships reinforces the importance of clinical interventions that may preempt adolescent delinquency.
We assessed the relationships between elements of the Macdonald triad (i.e., enuresis, cruelty to animals, and fire setting) and parental abuse (i.e., psychological, physical, and sexual abuse) within a population of serial killers. Data from 280 serial killers were retrieved from the Radford/Florida Gulf Coast University Serial Killer Database. This unique source of information was compiled from biographical books, newspaper articles, court documents, self-reports, and online sources. A series of chi-square tests for independence and binary logistic regression analyses were used to assess goodness of fit and regression relationships, respectively. The findings indicated significant statistical relationships among enuresis, fire setting, and animal cruelty to key dimensions of parental physical and psychological abuse, respectively. The use of binary logistic regression not only validated these joint associations but also elucidated the very potent relationships of fire setting and enuresis to that of psychological and physical parental abuse within this unique sample of serial killers. Associations between enuresis, fire setting, and animal cruelty to parental sexual abuse were not statistically significant. From a diagnostic perspective, enhanced awareness of these toxic relationships reinforces the importance of clinical interventions that may preempt adolescent delinquency.
From its inception, components of the Macdonald triad (1963) and parental abuse have been considered contributing factors to severe psychopathy (e.g., serial killing). To date, however, research regarding the unique relationship between each element of the triad (i.e., enuresis, cruelty to animals, and fire setting) and specific dimensions of parental abuse (i.e., psychological, physical, and sexual) among serial killers has been somewhat vague. The purpose of this study was to provide greater clarity through an analysis of these discrete associations within a select population of serial killers.
Serial Killers
The term "serial killer" was first used in the 1970s when FBI Special Agent Robert Ressler coined the term "serial murderer" during the notorious Son of Sam killings (Miller, 2014a). This definition established serial murder as distinctly different from that of traditional murders. The Federal Bureau of Investigation defines serial murder as "the unlawful killing of two or more victims by the same offender in separate events" (FBI, 2008). All serial killers in the database used for the present study met these criteria.
Historical prototypes of modern-day serial killers include the Roman Emperor Caligula (1st century), the Marshal of France, Gilles de Rais (15th century), the Hungarian countess Elizabeth Bathory (16th century), and Soviet Premier Josef Stalin (20th century). These individuals used powerful positions to exploit subordinates and satiate their twisted desires (Cawthorne, 2012; Hickey, 2010; Miller, 2014a; Miller, 2014b; Stone, 2001). Today, the term "serial killer" is used to describe those who commit serial murder, regardless of their title or position; modern serial killers include David Berkowitz (aka "Son of Sam"), Ted Bundy, Ed Gein, and Dennis Rader (aka "BTK").
The terms "psychopath" and "serial killer" are not synonymous. However, the term "criminal psychopath" is reserved for those psychopaths who demonstrate a wide range of persistent and antisocial behavior (Bartol & Bartol, 2014); antisocial personality disorder encompasses criminally psychopathic behaviors marked by a lifelong pattern of the manipulation and violation of others' rights, as defined in the Diagnostic and Statistical Manual of Mental Disorders V (APA, 2013).Serial killers are criminal psychopaths and represent an extreme variant on the spectrum of antisocial personality disorder.
The Macdonald Triad
According to Macdonald (1963), components of the triad, first identified by Yarnell (1940) as a unified entity, include persistent enuresis (e.g., bedwetting after age 5), fire setting (e.g., arson without clear motives), and acts of animal cruelty (e.g., children who abuse animals). Sigmund Freud (1905) referred to evidence of psychopathology related to these triadic elements when he hypothesized the relationships between and among fire setting, enuresis, and sexual problems (Heath, Hardesty,& Goldfine, 1984). More recently, Margaret Mead (1964) suggested that acts of animal cruelty were serious diagnostic signs and advocated for immediate psychological intervention in order to preempt adolescent delinquency.
Researchers have assessed the triad for associations with antisocial behaviors such as homicide and aggression. Initially, Macdonald's clinical case studies of homicidal patients (1963), Hellman and Blackman's (1966) retrospective study on aggressive adult prisoners, and Wax and Haddox's (1974)findings regarding incarcerated male adolescents have provided substantial evidence for the association of Macdonald's triad to highly aggressive and/or homicidal behavior. However, the use of inadequate methodologies, such as case studies and retrospective assessments (Heath, Hardesty, & Goldfine, 1984), and inconsistent findings on the predictive validity of the triad (Khunley et al., 1982; Sendi & Blomgren, 1975), reinforced concerns that the triad, as a whole, had not been adequately tested (Wright & Hensley, 2003). More recently, Slavkin (2001) has found significant associations among juvenile delinquents between animal cruelty and fire setting. His data, however, revealed no significant relationship between enuresis and fire setting. Since research on the triad has generally been limited and somewhat inconclusive, a focus on studies relative to individual characteristics of the triad has become a priority (Singer & Hensley, 2004). This reorientation has provided greater clarity on the relative contributions to criminal psychopathology of each element compared to the triad as a unified entity.
An FBI study that explored the specific association between the triadic element of animal cruelty and serial killers (Humane Society of the United States, 2014), and work by Ressler and colleagues (1988) found that a substantial number of convicted serial murderers have engaged in animal cruelty, heightening awareness of this potentially venomous predictor of behavior. Using quasi-experimental techniques, Kellert and Felthous (1985) found a statistically significant presence of animal cruelty among incarcerated adults characterized as "highly aggressive," compared to their criminally "non-aggressive" counterparts. More recently, Wright and Hensley(2003) found that, among a population of 354 serial killers,75(21%) had engaged in acts of animal cruelty; out of five randomly selected perpetrators for clinical follow-up, all demonstrated a direct connection between childhood animal cruelty and serial homicide (Tallichet & Hensley, 2004). However, not all homicidal and/or extremely violent individuals have a history of animal cruelty, even though a majority do (Ascione, 2001).
Comorbidity between fire setting and serial murder has not been investigated to any great extent (Singer & Hensley, 2004). However, Ressler's classic (1988) study linking fire setting to serial murder and Singer and Hensley's (2004) clinical assessment of three serial killers demonstrated a clear association between arson and, prospectively, adult serial homicide. Other key findings of interest among inpatient psychiatric juvenile fire setters indicated a significant, dyadic relationship between animal cruelty and fire setting; juveniles who engaged in acts of animal cruelty were more likely to engage in recidivistic fire setting, in contrast to their counterparts who did not engage in such acts (Kuhnley et al., 1982; Slavkin, 2001).
Enuresis is an unconscious, involuntary, and non-violent act; linking it directly to violent crime may be problematic (Hickey, 2010). Coupled with the scant research relative to enuresis and serial homicide, the nature of this relationship is somewhat unclear; however, in previous analyses of case studies among serial killers, the presence of enuretic behavior was found to be the most prominent common trait when compared to: exposure to porn prior to the age of 12, isolation, fire setting, petty stealing, arrests, sexual deviance, and cruelty to animals (McKenzie, 1995; Ressler, 1988).Ressler found a 68% rate of enuresis among 22 cases, while McKenzie found a 50% rate among 16 subjects. Findings by Heath, Hardesty, and Goldfine (1984), and corroborated by Slavkin (2001), among juvenile delinquents found no significant relationship between enuresis and fire setting; however, relative to a typical sample of juveniles, the level of enuresis among fire setters was statistically significant. Enuresis may not be causal relative to subsequent homicidal behaviors, but its presence may predict the likelihood of future predatory behavior patterns (Ressler et al., 1988).
Parental Abuse
Key components of parental abuse include psychological abuse; intentional humiliation causing emotional conflict and/or psychological damage; physical abuse, the causing or allowing any non-accidental physical injury; and sexual abuse, any illegal sexual act upon a child considered unhealthy for a child that meets the criminal definition as such (Mitchell & Aamodt, 2005). Previous research linked parental abuse to psychopathy and/or one's propensity to commit serial murder. For example, Ressler and Schactman (2002) reported that over 40% of serial murderers reported childhood psychological abuse; Mitchell and Aamodt (2005) found that 70% of serial killers were abused as children, either psychologically (50%), physically (36%), or sexually (26%); Borja and Ostrosky (2013) extrapolated a strong alignment between parental victimization/abuse and severe psychopathy (e.g., serial killing). Such findings suggest potentially lethal outcomes associated with highly dysfunctional, abusive parental relationships. Stein (2009) found that 80% of 65 violent offenders were subject to extreme levels of abuse as children and attributed the offenders' outcomes to the underlying deviance of early attachments, making their interpersonal landscapes highly toxic. According to Stein (2009), "abused children grown up may be driven to recreate this toxic attachment, to share the dread and torment they once experienced-by finding victims of their own" (pp. 322-323).
The Macdonald Triad and Parental Abuse
Prior research on the triadic elements and parental abuse suggested that animal cruelty, fire setting, and enuresis are products of parental abuse, and parental abuse is suggested to possess greater predictive validity, in regards to future homicidal tendencies, than the triad alone (Ryan, 2009). However, Macdonald (1963), Justice, Justice, and Kraft (1974), and Slavkin (2001) conceded that triadic behaviors can occur simultaneously with factors such as highly dysfunctional parenting, which may also predict violent, antisocial behavior. Much of the past empirical research linking specific types of parental abuse (psychological, physical, and sexual) to criminal psychopathy has been lacking, both quantitatively and methodologically (Marshall & Cooke, 1999). Other than more recent findings by Mitchell and Aamodt (2005) relative to the prevalence of parental abuse among serial killers (50% psychological ,36% physical, 26% sexual), the nature of specific relationships between each dimension of parental abuse to serial homicide was, at best, vague.
In order to provide greater clarity regarding each triadic element, separately, to that of each dimension of parental abuse among serial killers, or as Macdonald (1963) stated, "his most sadistic patients," we took a closer look at the significance of these specific, discrete associations in relation to one's propensity to commit serial murder. Access to the Radford/Florida Gulf Coast University Serial Killer Database (Aamodt, 2013) made it possible to extrapolate and elucidate the nature of these associations.
We hypothesize that there is a statistically significant relationship between each triadic component (animal cruelty, fire setting, and enuresis) and each dimension of parental abuse (psychological, physical, and sexual) within the sample of serial killers, and that examining them makes it possible to elucidate specific relationships among these variables.
METHOD
Participants
Two hundred eighty participants were selected for this study from the 3,191 serial killers included in the Radford/Florida Gulf Coast University Serial Killer Database (Aamodt, 2013).These participants were the only serial killers in the database with recorded data relevant to the present study; in other words, these serial killers had provided "yes" or "no" data pertaining to one or more elements of both the Macdonald triad (enuresis, cruelty to animals, and fire setting) and parental abuse (psychological, physical, and sexual abuse). All participants committed crimes between 1950 and the present. Sources of information for each profile included books, court documents, newspapers, self-reports, and websites (Aamodt & Surrette, 2010). Demographically, 239 participants were male, and 41 were female; 228 were Caucasian, 37 were Black, 9 were Hispanic, 5 were Asian, and 1 was Native American.
Procedure
Research was conducted using the database to examine potential relationships between elements of the Macdonald triad and parental abuse for 280 serial killers who met the inclusion criteria described above. To increase the integrity of the data, a hierarchal system was used to rank the various source materials. For example, data obtained from the State of Kansas vs. Dennis L. Rader court documents are given higher credence than information collected from M. H. Stone's book Anatomy of Evil. In addition, three separate independent sources containing the same data were required to validate each piece of information. The following queries were activated to determine the prevalence of each element pertinent to the study:
Cruelty to animals - Did the subject abuse animals?
Enuresis- Did the subject have a habit of wetting the bed after age five?
Fire setting - Did the subject commit arson or have an obsession with fire?
Psychological abuse - Was the subject psychologically abused during childhood?
Sexual abuse - Was the subject sexually abused during childhood?
Physical abuse - Was the subject physically abused during childhood?
RESULTS
A series of chi-square tests of independence were performed on each triadic element, separately, in relation to the discrete dimensions of parental abuse (see Table 1). Bar charts were created to represent the nature of these relationships visually (Figures 1-8, see end note). Phi coefficients were calculated to demonstrate the effect size of these dyadic relationships (see Table 1), and binary logistic regression analyses were performed to analyze relationships among the type of abuse and the triadic elements (see Table 2). Additionally, to gain a better understanding of the somewhat enigmatic triadic element of enuresis, a single sample chi-square test was performed to compare the percentage of serial killers exhibiting enuresis in the present study with that of the general population.
Chi-Square Analysis
A series of chi-square tests of independence were performed to analyze relationships between discrete elements of the MacDonald triad, separately (fire setting, enuresis, and animal cruelty), relative to each dimension of parental abuse (psychological, physical, sexual) (see Table 1). An alpha level of .05 was chosen to indicate whether relationships were statistically significant.
The first chi-square test (using Yates' continuity coefficient)was between enuresis and psychological abuse; x2 (1, и=211) =7.396 (p=0.007) was significant at the p< .05 level. A phi coefficient of 0.20 indicated a small but noteworthy effect size (Cohen, 1988). The second chi-square test (using Yates' continuity coefficient) showed a significant correlation between enuresis and physical abuse, x2 (1, n=216) =11.737 (p=.007). A phi coefficient of 0.245 indicated a small to medium effect size (Cohen, 1988). The third chi-square test of independence, between enuresis and sexual abuse, revealed a x2 (1, n=211) =2.446 (p=.118), indicating the relationship was not statistically significant.
To better understand the enigmatic nature of enuresis, a single sample chi-square test was performed to compare its presence among our participants versus the general adult population. According to Mitchell and Aamodt (2005), between 0.5% and 1.0% of the adult population experiences enuresis, whereas 18.5% of the serial killer sample exhibits this behavior. The resulting x2 (n=221) = 687.72 was statistically significant at the p < .05 level, indicating that the proportion of enuretic serial killers in our sample was found to be significantly higher than that of the general population (18.5% vs. 1%).
The next set of chi-square tests evaluated the relationship between the triadic element of fire setting and each discrete variable of parental abuse. The first test, between fire setting and psychological abuse, revealed a chi-square (1, n=270) =25.967 (p=.000), which is significant at the p < .05 level. A phi coefficient of 0.319 indicated a medium effect size (Cohen, 1988). The second chi-square test, between fire setting and physical abuse, revealed a significant association at the p < .05 level, (1, n=280) =17.357 (p=.000). A phi coefficient of 0.258 indicated a small to medium effect size. The third test, between fire setting and sexual abuse, showed no statistically significant relationship, x2 (1, n=280) =1.407 (p=.206).
The final set of chi-square tests of independence evaluated the relationship between animal cruelty and each discrete variable of parental abuse. The first test examined animal cruelty and psychological abuse and found a chi-square of (1, n=267) =4.154 (p=.042), which was significant at the p < .05 level. A phi coefficient of 0.113 indicated a small effect size (Cohen, 1988). A second test, comparing animal cruelty and physical abuse, showed a chi-square of(1, n=276)= 4.706 (p=.03), which is significant at the p < .05 level. A phi coefficient of 0.139 indicated a small effect size. The third test, focused on cruelty to animals and sexual abuse, revealed a chi-square (1, n=273) =3.053 (p=.081) that is non-significant.
Binary Regression Analyses
Binary regression analyses were used to test the relationships between three predictor variables (enuresis, fire setting, and cruelty to animals) and psychological abuse, the outcome variable; the model comprised all three predictors. Chi-square (3, n=208) =16.224 (p= .001) was statistically significant at the p < .05 level (see Table 1). Furthermore, the model explained between 7.5% (Cox and Snell R2) and10% (Nagelkerke R2) of the variance in psychological abuse; 63.5% of the cases were correctly classified. Two of the three predictive variables, enuresis and fire setting, made unique, statistically significant contributions to the model. The strongest predictor of the reporting of psychological abuse was fire setting, recording an odds ratio of 8.68; respondents who reported fire setting were more than eight times more likely to have experienced psychological abuse, controlling for all other factors in this model.
Binary regression analyses were also used to test the relationship between the three predictor variables (enuresis, fire setting, and cruelty to animals) and that of physical abuse, the outcome variable. The model comprised all three predictors, and chi-square (3, n=212) =13.413 (p= .004) was statistically significant at the p < .05 level. Furthermore, the model explained between 6.1% (Cox and Snell R2) and8.3% (Nagelkerke R2) of the variance in psychological abuse; 65.1% of the cases were classified correctly. Two of the three predictive variables, enuresis and fire setting, made unique, statistically significant, contributions to the model. The strongest predictor of the reporting of physical abuse was enuresis, recording an odds ratio of 2.892; respondents who reported enuresis were almost three times more likely to have experienced physical abuse, controlling for all other factors in this model.
DISCUSSION
A number of statistically significant relationships were found in support of the hypotheses. The ability to assess these associations quantitatively using the Radford/FGCU Serial Killer Database (Aamodt, 2013) provided a unique opportunity not found in prior research. Using both chi-square and binary regression analyses, the most powerful statistically significant relationships were evident between fire setting and psychological abuse, followed by enuresis and physical abuse. Specifically, the strongest predictor of psychological abuse was fire setting, with a recorded odds ratio of 8.68; respondents who reported fire setting were more than eight times more likely than the general population to have experienced psychological abuse (an outcome variable). Similarly, the robust relationship between enuresis (a predictive variable) and physical abuse (an outcome variable) recorded an odds ratio of 2.892; respondents who reported enuresis were almost three times more likely to have experienced physical abuse. Even though associations between the third triadic element, animal cruelty, to parental physical (p=.03) and psychological (p= .042) abuse were statistically significant (p<0.05) its impact on the binary regression model, unlike that of enuresis and fire setting, was not statistically significant. Finally, associations between each triadic element to the outcome variable of sexual abuse were not statistically significant; this implies a negligible relationship between these variables for predictive purposes. The use of binary regression analyses to elucidate the relative contribution of these robust, comorbid relationships (e.g., fire setting and psychological abuse; enuresis and physical abuse) enhanced awareness of the strength of these toxic relationships and the critical nature of clinical interventions that may preempt adolescent delinquency and/or the propensity to commit serial murder. Although the present findings support associations between elements of the triad and components of parental abuse, the directionality of this relationship cannot be ascertained.
Research Implications
Elucidating relationships between discrete components of the triad (enuresis, fire setting, and animal cruelty) to dimensions of parental abuse (psychological, physical, sexual) provided greater clarity on the relative contributions of each element and a person's propensity to become a serial killer. The use of chi-square tests for independence and binary logistic regression analyses unraveled the robust relationships between fire setting and psychological abuse and between enuresis and physical abuse. In other words, from a practitioner's perspective, an adolescent's proclivity to wet the bed and/or engage in arson may indicate the presence, simultaneously, of parental psychological or physical abuse; these warning signs should escalate the need for immediate clinical intervention to preempt more serious ramifications (e.g., criminally psychopathic behaviors). Even though linking enuresis directly to violent crime may be problematic (Hickey, 2010), our study demonstrated that there is a robust, statistically significant relationship between enuresis and physical abuse. Furthermore, enuresis and fire setting demonstrated unique, statistically significant contributions to the binary regression model that validated the strong associations between both triadic elements and parental psychological and physical abuse.
The less powerful, yet statistically significant, association between animal cruelty and psychological and physical abuse does not minimize the relevance of this relationship relative to a person's propensity to commit serial homicide. The strength of this destructive relationship has been substantiated previously (Kellert & Felthous,1985; Ressler et al., 1988); however, engaging in secretive acts that are perceived as "wrongdoing," such as animal cruelty, may not always be reported by their perpetrators for obvious reasons. As a result, participation in these acts may be underrepresented; Wright and Hensley (2003) reported that, out of a population of 354 serial killers, only 75 (21%) reported previous engagement in animal cruelty. The absence of statistically significant relationships among serial killers' parental-inflicted sexual abuse and discrete triadic elements may indicate an insidious, diabolical collusion with the abuser (parents); in other words, according to Stein (2009), victims many not report a crime as a means of protecting themselves from sharing their perceived vulnerability and allaying fears of losing the only attachment they have ever had.
In summary, these associations suggest that key relationships between certain triadic elements, in conjunction with dimensions of parental abuse, may serve as potent indicators of one's propensity to commit serial murder. This does not exclusively predict an act of serial homicide; rather, these are merely risk factors for such behaviors. As Kiehl (2014) points out, "many people who commit homicides as adults had these symptoms (triadic elements) as children, but the majority of children who have these symptoms do not go on to commit homicide as adults," (p. 135). Regardless, the results of this study provide a clearer understanding of the potency of associations between discrete triadic elements and dimensions of parental abuse and how they might relate to the identification of serious "at-risk" behaviors in adolescents.
Clinical and Policy Implications
From a practitioner's perspective, the simultaneous comorbidity between key triadic elements and specific acts of parental abuse indicate that enuresis and fire setting may be potent predictors of parental psychological and physical abuse at home. Results from this study strongly reinforce the need to probe for such toxic relationships when such behaviors are present. Furthermore, the less robust, yet statistically significant, relationship between cruelty to animals and both parental physical and psychological abuse may be under- or unreported; perpetrators' proclivity to not disclose such information, and incidences of parental sexual abuse, underscore the need for clinicians to probe more deeply for such toxicity.
Enhanced awareness that leads to appropriate interventions based on these associations may create a greater sense of urgency in dealing with such issues; this aligns with a critical need for intervention to preempt adolescent delinquency (Macdonald, 1963; Mead, 1964; Slavkin, 2001; Stein, 2009; Wax & Maddox, 1974). Furthermore, the critical need for intervention at an early age may expose and prevent continual parental victimization and abuse (Borja & Ostrosky, 2013), which leads to the cycle of toxic attachment so prolifically described by Stein (2009).
Limitations
One limitation of the present study concerns the low availability of data pertinent to our topic; however, it is rare to have access to such data. The identification of a population of 280 serial killers from the Radford/Florida Gulf Coast University Serial Killer Database (Aamodt, 2013) and its relevant information far exceeds any other source of meaningful data available today. Another limitation concerns the fact that this information was collected from serial killers who have been apprehended. No information exists in the database regarding serial killers who have not been apprehended. Concerning the collection of information, the validity of data collected by third parties, and/or selfreported, may be questionable. As Mitchell and Aamodt (2005) noted, misinformation concerning criminals may emerge because of (1) misreported wrongdoings to explain why a crime was committed or to seek forgiveness, and/or (2) a wrongdoing may not be reported to protect the victim from exposing his or her perceived vulnerability. Furthermore, according to Stein (2009), to allay fears of losing the only attachment they have ever had, some abuse victims perceive heinous acts by their parents as acceptable and will not report them.
Due to the associational focus of this study, cause and effect relationships cannot be determined. Nevertheless, although it is highly unlikely, it could be argued that participation in triadic behaviors prompts abuse from parents, or even an unknown variable could be mediating these factors. As we continue to organize elements of the Serial Killer Database, boundless future opportunities exist to examine other associations among criminally psychopathic variables (e.g., gender differences, differences between lust and anger, and differences between spree and serial killers).
REFERENCES
Aamodt, M. (2013). Serial Killer Database [Database]. Virginia: Radford University, Florida: Florida Gulf Coast University
Aamodt, M. G., & Surrette, M. A. (2010). Using the Radford Serial Killer Database to monitor trends in serial murder. Paper presented at the annual meeting of the Society for Police and Criminal Psychology, Honolulu, Hawaii.
American Psychiatric Association [APA]. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: APA.
Ascione, F. R. (2001). Animal abuse and youth violence. Washington, D.C.: U.S. Dept. of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.
Bartol, C., & Bartol, A. (2014). Criminal behavior: A psychological approach (10th ed.). Upper Saddle River, NJ: Pearson.
Borja, K., &Ostrosky, F. (2013). Early traumatic events in psychopaths. Journal of Forensic Sciences, 38(4), 927-931. doi: 10.1111/1556-4029.12104
Cawthorne, N. (2012). Tyrants: History's 100 most evil despots &dictators (1st ed.). New York: Metro Books.
Cohen, M. A. (1988). Some new evidence on the seriousness of crime. Criminology, 26(2), 343-353.
Federal Bureau of Investigation [FBI]. (2008). Serial murder: Multi-disciplinary perspectives for investigators. Behavioral Analysis Unit. National Center for the Analysis of Violent Crime. U.S. Department of Justice, Washington D.C.
Freud, S. (1905). On psychotherapy. Standard edition, 7(267), 64-145.
Heath, G.A., Hardesty, V.A., &Goldfine, P.E. (1984). Fire setting, enuresis, and animal cruelty. Journal of Child and Adolescent Psychotherapy, 1, 97-100.
Hellman, D., & Blackman, N. (1966). Enuresis, firesetting and cruelty to animals: A triad predictive of adult crime. American Journal of Psychiatry, 122(12), 1431-1435.
Hickey, E.W. (2010). Serial murderers and their victims(5th ed.). Belmont, CA: Wadsworth, Cengage Learning.
Humane Societyof the United States. (2014). Animal cruelty facts and statistics. Retrieved from http://www.humanesociety.org/issues/abuse_neglect/facts/ animal_cruelty_facts_statistics.html
Justice, B., Justice, R., & Kraft, I. (1974). Early-warning signs of violence: Is a triad enough? American Journal of Psychiatry, 131(4), 457-459.
Kellert, S. R., & Felthous, A. R. (1985). Childhood cruelty toward animals between criminals and noncriminals. Human Relations, 38, 1113-1129.
Kuhnley, E. J., Hendren, R. L., & Quinlan, D. M. (1982). Fire-setting by children. Journal of the American Academy of Child Psychiatry, 21(6), 560-563.
Kiehl, K. (2014). The psychopath whisperer: The science of those without conscience (1st ed.). New York: Crown.
Macdonald, J.M. (1963). The threat to kill. American Journal of Psychiatry, 120(2), 130-135.
Marshall, L. A., & Cooke, D. J. (1999). The childhood experiences of psychopaths: A retrospective study of familial and societal factors. Journal of Personality Disorders, 13(3), 211-225.
McKenzie, C. (1995). A study of serial murder. International Journal of Offender Therapy and Comparative Criminology, 39(1), 3-10.
Mead, M. (1964). Cultural factors in the cause and prevention of pathological homicide. Bulletin of the Menninger Clinic, 28(1), 11-22.
Miller, L. (2014a). Serial killers: 1. Subtypes, patterns, and motives. Aggression and Violent Behavior, 19(1), 1-11.
Miller, L. (2014b). Serial killers: II. Development, dynamics, and forensics. Aggression and Violent Behavior, 19(1), 12-22.
Mitchell, H., & Aamodt, M. (2005). The incidence of child abuse in serial killers. Journal of Police and Criminal Psychology,20(1), 40-47.
Ressler, R. K., Burgess, A. W., & Douglas, J. E. (1988). Sexual homicide: patterns and motives. New York: Lexington Books.
Ressler, R. K., & Shachtman, T. (2002). Whoever fights monsters. New York: St. Martins.
Ryan, K. (2009). The Macdonald triad: Predictor of violence or urban myth? (Unpublished Master's thesis). California State University, Fresno.
Sendi, I. B., & Blomgren, P. G. (1975). A comparative study of predictive criteria in the predisposition of homicidal adolescents. The American Journal of Psychiatry, 132(4), 423-427.
Singer, S. D., & Hensley, C. (2004). Applying social learning theory to childhood and adolescent firesetting: can it lead to serial murder?. International Journal of Offender Therapy and Comparative Criminology, 48(4), 461-476.
Slavkin, M. L. (2001). Enuresis, firesetting, and cruelty to animals: Does the ego triad show predictive validity?Adolescence, 36(143), 461-466.
Stein, A. (2009). From their cradle to your grave: How child abuse and dissociation drive violent crime. The Journal of Psychohistory, 36(4), 320-327.
Stone, M. H. (2009). The anatomy of evil. Prometheus Books.
Stone, M. H. (2001). Serial sexual homicide: Biological, psychological, and sociological aspects. Journal of Personality Disorders, 15(1), 1-18.
Tallichet, S. E., & Hensley, C. (2004). Exploring the link between recurrent acts of childhood and adolescent animal cruelty and subsequent violent crime. Criminal Justice Review, 29(2), 304-316.
Wax, D. E., & Haddox, V. G. (1974). Enuresis, fire setting, and animal cruelty: A useful danger signal in predicting vulnerability of adolescent males to assaultive behavior. Child Psychiatry and Human Development, 4(3), 151-156.
Wright, J., & Hensley, C. (2003). From animal cruelty to serial murder: Applying the graduation hypothesis. International Journal of Offender Therapy and Comparative Criminology, 47(1), 71-88.
Yarnell, H. (1940). Firesetting in children. American Journal of Orthopsychiatry, 10(2), 272-286.
Note: All figures are available from the first author upon request.
Copyright North American Journal of Psychology Dec 2017