Sexual Sadism and the Psychopathy Facets as Predictors of

General Violence and Sexual Crime Behaviors

Masters Thesis

Presented to

The Faculty of the Graduate School of Arts and Sciences

Brandeis University
Department of Psychology
Raymond A. Knight, Advisor

In Partial Fulfillment
of the Requirements for

Masters Degree

by
Carrie A. Robertson

August 2011
 Copyright by

Carrie A. Robertson

2011
The author gratefully acknowledges Raymond A. Knight, her advisor without his
extensive databases this research could not have been conducted

iii
 ABSTRACT

Sexual Sadism and the Psychopathy Facets as Predictors of

General Violence and Sexual Crime Behaviors

A thesis presented to the Department of Psychology

Graduate School of Arts of Sciences

Brandeis University
Waltham, Massachusetts

By Carrie A. Robertson

Sexual sadism and psychopathy have been linked theoretically, clinically, and empirically

to violence. Whereas evidence suggests that sadists derive sexual gratification from the

physical and emotional suffering of others, psychopaths are violent for varied reasons:

failing to attend to anothers distress cues, overreacting to perceived aggression, or as a

means to an external goal. Despite this overlap of predominantly predatory violence

displayed by both sadists and psychopaths, few studies have sought to explore the

interface of the two constructs, and even fewer have sought to conceptualize the

covariation of violence in each. Those studies that have explored the covariation have

suffered from limited assessments of violence, poorly operationalized definitions of

sadism, and failure to examine all four Psychopathy Checklist-Revised (PCL-R) facets of

psychopathy. The current research sought to remedy these problems by a) using detailed

analyses of crime behaviors to generate dependent measures of sexual and non-sexual

violence, b) using well-defined, validated measures of sadism, and c) considering all four

psychopathy facets. Two independent samples of 314 and 599 adult, male sex offenders

iv
were assessed using a multimethod approach with both ratings from archival records and

administration of a computerized self-report inventory, the Multidimensional Inventory

of Development, Sex, and Aggression (the MIDSA). Exploratory factor analyses of

detailed crime scene descriptions yielded three consistent sexual crime behavior factors:

Violence Severity, Sexual Control, and Sexual Behavior. Hierarchical multiple

regressions were performed with sadism and the PCL-R facets as predictors and with

general violence and the sexual crime behavior factors as outcomes. As hypothesized, the

PCL-R facets collectively predicted both the Violence Severity factor and general

violence, and sadism predicted general violence, the Violence Severity factor, and the

Sexual Control factor. Sadism and psychopathy overlapped, but the two constructs were

not co-extensive and appeared to capture different aspects of aggression.

v
 TABLE OF CONTENTS

I. List of Tables .... vi

II. List of Figures .. vii

III. Introduction 1

IV. Study 1

a. Methods .. 9

b. Results .. 13

V. Study 2

a. Methods 15

b. Results .. 18

VI. General Discussion .. 21

VII. Appendix A: Tables . 29

VIII. Appendix B: Figures 33

VIII. References 36

vi
 LIST OF TABLES

1. Sexual Crime Behavior Factor Scale Variables and Reliabilities by Rating Type

2. Relative Strengths of Sadism, Psychopathy, and their Interaction in the Prediction

of General Violence and Sexual Crime Behavior Factors

3. Relative Strengths of Psychopathy Facets and Cross-Factor Interactions (F1 x F3,

F1 x F4, F2 x F3, F2 x F4) in the Prediction of General Violence and Sexual

Crime Behavior Factors

4. Correlation Matrices for Sadism, Psychopathy, and the Psychopathy Facets

vii
 LIST OF FIGURES

1. Scatterplot of the interaction of low, moderate, and high self-reported sadism

levels with archivally-rated total PCL-R scores on the Violence Severity factor

during sexual crimes.

2. Scatterplot of the interaction of low, moderate, and high Facet 1 (Interpersonal)

scores with Facet 4 (Antisocial) scores on the Sexual Behavior factor during

sexual crimes; holding Facets 2 and 3 constant at zero.

3. Scatterplot of the interaction of low, moderate, and high Facet 1 (Interpersonal)

scores with Facet 4 (Antisocial) scores on the commission of Juvenile Assaults;

holding Facets 2 and 3 constant at zero.

viii
 INTRODUCTION

Both sexual sadism and psychopathy have been linked theoretically, clinically,

and empirically to sexual offending (Knight, 2010; Knight & Guay, 2006), as well as to

non-sexual generic violence (Porter & Woodworth, 2006). Whereas sadists have been

shown to derive sexual gratification from the physical and emotional suffering of others

(Breslow, 1989; Herron & Herron, 1982; Holt, Meloy, & Strack, 1999; Johnson &

Becker, 1997b; Kirsch & Becker, 2007; Knight, 2010; Marshall & Kennedy, 2003;

Marshall, Kennedy, & Yates, 2002), psychopaths appear to be violent for a variety of

reasons: failing to attend to anothers distress cues (Hare, Cooke, & Hart, 1999; Kirsch &

Becker, 2007), overreacting to perceived aggression (Porter & Woodworth, 2006), or as a

means to an external goal (Hare et al., 1999; Hare & Neumann, 2009; Porter &

Woodworth, 2006; Serin, 1991).

Despite the consistent covariation of these two constructs with violent behavior,

their interface has remained largely unexplored. The increased risk of violence associated

with high levels of each of these warrants increased research attention to both and to their

potential overlap because of the important implications for theory, clinical practice, and

social policy (Hare et al., 1999; Seto & Lalumire, 2000). Thus far, those studies that

have sought to explore the interface of sadism and psychopathy have suffered from

limited assessments of violence, poorly operationalized definitions of sadism, and failure

to consider all four facets of the Psychopathy Checklist-Revised (PCL-R; Hare, 2003).

The current research sought to remedy these problems by using detailed analyses of

1
crime behaviors to generate dependent measures of sexual and non-sexual violence, by

employing well-defined and validated measures of sadism, and by considering all four

psychopathy facets.

The term sadism emerged in the early 19th century in connection with the

writings of the Marquis de Sade (Hucker, 1997). Later, Krafft-Ebing (1886/1965)

examined the construct in his Psychopathia Sexualis a classic source of case material

on sexual sadism (Johnson, 1973; Kirsch & Becker, 2007). According to Krafft-Ebing,

sadistic violence required both sexual and personality pathology in the perpetrator. He

was the first to propose an association between cruelty, violence, and lust (Kirsch &

Becker, 2007; Krafft-Ebing, 1886/1965). Unfortunately, like Krafft-Ebings clinical

speculations most theories about sadism have relied on case histories and clinical

interviews of small, select samples using mainly exploratory and descriptive methods of

analyses (Breslow, 1989; Grubin, 1994a, 1994b; Kirsch & Becker, 2007).

Consistent with the DSM-IV-TR (American Psychiatric Association, 2000),

sadism is most frequently defined as the range of cognitions and behaviors associated

with the derivation of sexual excitement from the infliction of physical and/or emotional

pain (e.g. Abel & Osborn, 1992; Breslow, 1989; Kirsch & Becker, 2007; Porter &

Woodworth, 2006; Porter, Woodworth, Earle, Drugge, & Boer, 2003). Other researchers,

however, contend that other characteristics such as control/domination over another

individual are central to the definition (Grubin, 1994a; Johnson & Becker, 1997a; Knight,

2010; MacCulloch, Snowden, Wood, & Mills, 1983). Given this lack of consensus at the

conceptual level, attempts to operationalize sexual sadism have been inconsistent (Knight

& Prentky, 1990; Marshall & Hucker, 2006a; Marshall & Kennedy, 2003; Marshall et al.,

2
2002; Marshall, Kennedy, Yates, & Serran, 2002; McLawsen, Jackson, Vannoy,

Gagliardi, & Scalora, 2008; Mollinger, 1982; Prentky & Knight, 1991).

The current research circumvented these operational difficulties by employing

both a reliable, well-validated self-report scale of sadism and carefully anchored

classification criteria. Study 1 used the Multidimensional Inventory of Development, Sex,

and Aggressions (the MIDSA, 2011) Sexual Sadism scale. This self-report scale taps

both cognitive and behavioral aspects of sadism and has been shown to have reasonable

internal consistency and test-retest reliability, as well as concurrent and divergent validity

(Knight & Cerce, 1999; Knight, Prentky, & Cerce, 1994; MIDSA, 2011). Study 2 used

the sadism classifications from the Massachusetts Treatment Center Sex Offender

Typologies for rapists and child molesters created by Knight and colleagues (Knight,

1988, 1989, 1999, 2010; Knight, Carter & Prentky, 1989; Knight & Cerce, 1999; Knight

& King, in press; Knight & Prentky, 1990; Knight et al., 1994). Both typologies have

evinced concurrent and predictive validity, as well as adequate inter-rater reliability

(Barbaree, Seto, Serin, Amos, & Preston, 1994; Knight, 1988, 1989, 1999, 2010; Knight

et al., 1989).

Like sadism, conceptualizations of psychopathy vary. Indeed, there is no

consensus about whether it comprises a multifaceted construct, involving the co-

occurrence of distinctive by synergistic components (Lilienfeld & Fowler, 2006) or a

unitary entity with a single underlying etiology (Neumann, Hare, & Newman, 2007).

Nonetheless, the descriptive characteristics proposed by Cleckley (1976) have achieved

wide acceptance, and the operationalization of these characteristics in the Psychopathy

Checklist-Revised (Hare, 1991, 2003) constitutes the most extensively validated and

3
widely used measure of psychopathy. The PCL-R has consistently yielded two

superordinate factors (e.g., Harpur, Hakstian, & Hare, 1988; Harpur, Hare, & Hakstian,

1989) that assess Affective-interpersonal features (Factor 1) and Impulsivity-antisocial

behavior (Factor 2). It remains controversial whether three-(Cooke & Michie, 2001) or

four-factor (Hare, 2003; Hare & Neumann, 2008) solutions better account for the

variance of the PCL-R. The three-factor model subdivides Factor 1 into Interpersonal and

Affective facets and focuses the third factor solely on the Impulsivity components of the

higher order Factor 2. The latter model similarly divides Factor 1, but subdivides Factor 2

into both Lifestyle Impulsivity and Antisocial facets.

In exploring the relation between violence and psychopathy only the two-factor

solution has been examined. Although some support for the covariation of Factor 1 and

various measures of violence has emerged (First & Halon, 2008; Porter et al., 2003),

Factor 2 has more consistently predicted violent behavior (Hare & McPherson, 1984;

Porter & Woodworth, 2006). Indeed, Kennealy, Skeem, Walters, and Camps (2010)

recent meta-analysis found that in roughly 80% of studies Factor 2 had more unique

predictive power than did Factor 1. Recent research using the four-facet solution of the

PCL-R has identified links between the specific psychopathy facets and various external

correlates (e.g., Hall, Benning, & Patrick, 2004; Hill, Neumann, & Rogers, 2004; Salekin,

Neumann, Leistico, & Zalot, 2004; Vitacco, Neumann, & Jackson, 2005), suggesting that

this model promises greater process specification. Indeed, some data indicate that the four

facets may be related to specific types of violent behavior and support further exploration

of these correlates (Walsh & Kosson, 2008; Walsh, Swogger, Walsh, & Kosson, 2007).

Consequently, the current study aims to provide a more refined understanding of the

4
relation between psychopathy and violence by studying the psychopathy construct at its

most basic level: the psychopathy facets.

Like the independent measures in the present study, the dependent construct,

violence, requires greater specification and delineation. It is important to refine our

understanding of what types of violence are related to psychopathys constituent parts

and to sadism (Walsh et al., 2007). Violence is a heterogeneous construct with multiple

etiologies, motivations, and manifestations, and it requires definitional and measurement

specification (Lion, 1991). Walsh and Kosson (2008) have suggested that examination of

more narrow-band indices of violence might add to our understanding of the relation

between violence and psychological constructs. The current study seeks to refine the

operationalization of violence and attempts to capture its complexities by using both

theoretically and empirically derived definitions. In addition to more traditional

measures, detailed analyses of sexual crime behaviors will generate dependent measures

of sexual violence. Moreover, a multimethod assessment will be implemented, using both

ratings from archival records and self-report inventories.

In the current study both sexual and non-sexual violence were assessed. Measures

of non-sexual violence included general, non-sexual aggression and both juvenile and

adult fighting and assaultive behavior. Because sadism has been linked to non-sexual

aggression (Ahlmeyer, Kleinsasser, Stoner, & Retzlaff, 2003; Hazelwood & Warren,

2000; Marshall et al., 2002), to childhood/juvenile behavior problems (DSM-IV-TR;

2000; Hare et al., 1999; Hill, Habermann, Berner, & Briken, 2006), and to adult

assaultive behavior (Briken, Habermann, Kafka, Berner, & Hill, 2006; Dietz, Hazelwood,

& Warren, 1990; Kirsch & Becker, 2007; Johnson & Becker, 1997b; Warren,

5
Hazelwood, & Dietz, 1996), it was hypothesized that it would predict all three non-sexual

violence or general aggression measures. Similarly, sadism has been found to covary with

behaviors analogous to all three of the empirically-derived sexual crime behavior factors:

Violence Severity (Beauregard & Proulx, 2007; Dietz et al., 1990; Fedora et al., 1992;

Herron & Herron, 1982; Hill et al., 2006; Knight, 2010), Sexual Control (Chan & Heide,

2009; Dietz et al., 1990; Gratzer & Bradford, 1995; Johnson & Becker, 1997b; Hare et

al., 1999; Langevin et al., 1988; Marshall & Kennedy, 2003), and Sexual Behavior

(Geberth & Turco, 1997; Gratzer & Bradford, 1995; Hazelwood & Warren, 2000;

Knight, 1999; Krafft-Ebing 1886/1965). Consequently, sadism was also hypothesized to

predict all three of these sexual crime behavior factors.

Psychopathys ties to non-sexual and sexual violence have also been well

documented. In fact, characteristically elevated levels of aggression are central to most

contemporary conceptualizations of the syndrome (Hare et al., 1999; Hare & McPherson,

1984; Porter et al., 2003; Walsh & Kosson, 2008; Walsh et al., 2007). Because

psychopathy has been associated with violence both in adolescence (Brown & Forth,

1997; Gretton, Hare, & Catchpole, 2004; Hare et al., 1999; Porter & Woodworth, 2006),

and in adulthood (Hare et al., 1999; Hare & McPherson, 1984; Hare, Hart, & Harpur,

1991; Porter & Woodworth, 2006; Serin, 1991), it was hypothesized that it would predict

all three non-sexual violence measures. As with general violence, psychopathy has been

linked to sexual violence (Hare et al., 1999; Porter et al., 2003) and was hypothesized to

predict the Violence Severity factor of sexual crimes in the current study. Although

studies have not specifically addressed elements of psychopaths use of sexual control

over others, their use of intimidation and violence as means of achieving control over

6
others (Hare & Neumann, 2009; Porter & Woodworth, 2006) informs the hypothesis that

psychopathy would be associated with the Sexual Control factor of sexual crimes.

Although psychopathys relation with coercive sexuality has been well-documented (First

& Halon, 2008; Harris, Rice, Hilton, Lalumire, & Quinsey, 2007; Knight & Guay, 2006;

Porter et al., 2003; Walters, Marcus, Edens, Knight, & Sanford, 2011), no studies have

focused on the presence of sexual behaviors extraneous to coitus (i.e., fondling,

cunnilingus) that are assessed in the Sexual Behavior factor. Nonetheless, because of

prior data suggesting covariation between scales capturing the interpersonal aspects of

psychopathy and sexualization (Knight & Sims-Knight, 2003; 2004), it was hypothesized

that the interpersonal facet would be related to the Sexual Behavior factor.

Due to the exploratory nature of the research on the relation between the various

forms of violence and the individual PCL-R facets, hypotheses about these individualized

relations must remain speculative. Although the facets have not been linked to violence

empirically, the psychopathic elements thought to comprise the facets have been

theoretically coupled with violence. The interpersonal detachment of psychopaths and

their cold, callous manipulation of others (characteristics of Facet 1) have been associated

with general violence (Benning, Patrick, Hicks, Bloningen, & Krueger, 2003; First &

Halon, 2008; Hare & McPherson, 1984). Psychopaths failure to respond to the distress

of victims (Facet 2 traits) has been implicated as a possible contributor to their increased

violence (Blair, Jones, Clark, & Smith, 1997; Blair et al., 1995). Facet 3s thrill-seeking

and poor behavioral inhibitions have both been deemed catalysts of violence (Benning et

al., 2003; Hare & McPherson, 1984; Porter & Woodworth, 2006; Porter et al., 2003).

Finally, proactive and reactive aggression, as well as antisocial behaviors (all symbolic of

7
Facet 4), have been associated with violence (Harris et al., 2007; Serin, 1991; Walsh &

Kosson, 2008). Although the psychopathic characteristics akin to Facets 1 and 2 have

been theoretically linked with violence, Factor 1 has inconsistent empirical associations

with violence (Kennealy et al., 2010). Consequently, it was hypothesized that Facets 1

and 2 would not predict any of the violence measures. In contrast, there is compelling

empirical evidence linking Factor 2 with violence (i.e., Kennealy et al., 2010). This, in

addition to the strong theoretical relationships between violence and psychopathic

components analogous to Facets 3 and 4, influenced the hypothesis that Facets 3 and 4

would predict both generic and sexual violence.

8
 STUDY 1

Methods

Participants

From a sample of 529 sex offenders who were administered the Multidimensional

Assessment of Sex and Aggression (the MIDSA, 2011) while incarcerated in prisons and

special commitment facilities in Massachusetts and Minnesota, a subsample of 314 was

selected because their archival records were sufficiently complete to rate both the PCL-R

and crime-scene behavior. The sample comprised predominantly repeat offenders, some

of whom had been deemed to have been sexually dangerous persons and civilly

committed. The sample was predominantly Caucasian (n = 211, 67%), with 60 African-

Americans (19%), 15 Native Americans (5%), 15 Hispanics (5%), two Asians (<1%) and

11 undisclosed ethnicities (3%). At the time of assessment, the offender ages ranged from

20 to 68 years of age, with a mean of 39.06 years (SD = 9.66). Fifty-six percent of

offenders had assaulted a child under the age of 16 (n = 175), 40 percent had committed

crimes against exclusively adult 16 years of age or older (n = 124), and the victim history

of five percent were not included in their files. As juveniles under 17 years of age, these

offenders committed 74 (M = .25, n = 281) serious sexual offenses and committed

another 724 (M = 2.41, n = 298) after turning seventeen.

Measures

Predictor variables. Independent variables included self-report measures of

sexual sadism and archival ratings of the PCL-R facets.

9
 Sexual sadism ratings. The MIDSA (MIDSA, 2011) is a contingency-based,

computerized inventory (formerly the Multidimensional Assessment of Sex and

Aggression (the MASA; Knight et al., 1994) that assesses multiple domains relevant to

sexual aggression. The overall strategy for creating and validating the MIDSA has been

described in detail elsewhere (Knight & Cerce, 1999; Knight et al., 1994; MIDSA, 2011).

The MIDSAs Sexual Sadism scale, used in the present study, was a composite of two

subscales: a seven-item Sadistic Fantasies scale and an eight-item Sadistic Behaviors

scale. The internal consistencies for the fantasy scale and the behavior scale are .85 and

.84, respectively, for adult male sex offenders. The correlation between the two scales is

.79 in the same population, and the test-retest reliability on comparable samples has been

reported as .81 (MIDSA, 2011).

Psychopathy ratings. Archival clinical and criminal files were rated using the

PCL-R (Hare, 2003) by a research assistant trained by an approved PCL-R expert. The

PCL-R consists of 20 items each rated on a 3-point scale based on the degree to which

the item described the offender. The rater was blind to the hypotheses of the current

study. Total scores reflect the extent to which the offender matches the description of a

prototypical psychopath; the scores do not convey diagnoses. Total PCL-R scores and

individual facet scores were calculated.

General violence outcome variables. The study included both an archivally-

rated scale that measures general, non-sexual aggression, and two self-report scales from

the MIDSA that assess juvenile and adult fighting and assaultive behavior.

Unsocialized general aggression (USAG). This archivally-rated scale measures

the amount and frequency of general aggression displayed by the offender throughout his

10
life prior to imprisonment; the USAG excludes sexual offense related aggression. It is a

7-point Gutman scale, ranging from 0 (no evidence of unsocialized aggression) to 6

(evidence of occasional or frequent extreme unsocialized aggression, e.g., extreme

mutilation or brutal murders). The USAG was rated by a single rater who was blind to the

expectations of the current study. An interrater reliability estimate for this scale is

presented in the description of Study 2.

Juvenile assault. The MIDSAs factor-generated Juvenile Fighting and

Assaultive Behavior scale includes five subscales measuring instances of fighting,

bullying, aggressive, and impulsive behavior. This scale excludes sexual aggression. The

aggressive scale spans childhood and adolescence, and the impulsivity scale focuses on

acting out in grammar school, middle school, and high school. Higher scores indicate

high incidence rates of fighting and impulsivity. The Cronbach alpha reported for adults

was .84 (MIDSA, 2011).

Adult assault. The MIDSAs factor-generated, self-report scale, Adult Fighting

and Assaultive Behavior, was used to measure non-sexual adult assaultive behavior. This

scale includes five subscales that measure instances of carrying weapons, weapons

charges and convictions, assaultive crime charges and convictions, robbery charges and

convictions, as well as fighting and assaultive behavior. Higher scores indicated high

incidence rates of fighting, violent crimes, and weapons possession. The Cronbach alpha

for adult sexual offenders was .71 (MIDSA, 2011).

Sexual crime outcome measures. Exploratory factor analyses of ratings of

crime-scene behavior during sexual offenses from archival records yielded comparable

factors for both the Study 1 and Study 2 samples. Up to nine sexual offenses were coded

11
for the Study 1 sample. Three methods of concatenating behavioral information (the

mean level of severity across all crimes, the maximum level of severity for any behavior,

and the severity rating of the index crime behavior) yielded similar results. Only the

mean severity ratings were used in the data analyses because these ratings captured the

greatest amount of data and yielded the highest internal consistencies. A principal

components analysis with iterations yielded three factors with eigenvalues greater that

were rotated to VARIMAX criteria. Three factors consistently emerged: Violence

Severity, Sexual Control, and Sexual Behavior. These factors accounted for 38.48% of

the variance in Study 1 and 34.04% of the variance in Study 2. The sexual crime behavior

factors scale variables and reliabilities by rating type are presented in Table 1.

Violence severity. The Violence Severity factor scale accounted for 20.12% of the

variance and measured the mean severity of the items: broken bones; cuts, bruises, and

abrasions; expressive aggression before, during and after; injury requiring a doctor;

instrumental aggression; instrumental aggression; presence of a weapon; sadistic assault

of the genitals and/or breasts; stabbing; and the use of a weapon. The standardized

internal consistency of this factor scale was .80.

Sexual control. The Sexual Control factor scale, accounting for 9.63% of the

variance, measured the mean instances of blindfolding and/or gagging the victim and

tying the victim up during sexual crimes. The standardized Cronbach alpha for the Sexual

Control factor scale was .69.

Sexual behavior. The Sexual Behavior factor scale accounted for 8.72% of the

variance and measured the mean instances of anal penetration with the penis; performing

cunnilingus on the victim; fondling the victims genitals and/or breasts; kissing; vaginal

12
penetration using fingers/hands; the victim performing fellatio on the offender; and the

victim masturbating the offender. The standardized internal consistency of the Sexual

Behavior factor scale was .45.

Results

Reverse hierarchical multiple regressions were calculated to determine the unique

contributions of sadism and the psychopathy facets to the prediction of all six outcome

variables. Table 2 presents the respective R2s that occurred when the predictor variable

was entered in Block 2 after the alternative was entered in Block 1. Sadism significantly

contributed to the prediction of Juvenile and Adult assault, but not to the prediction of

USAG. Sadism also contributed to the prediction of violence and control during the

course of sexual crimes, but did not contribute to sexual behavior during a crime. The

psychopathy facets uniquely contributed to the prediction of all three non-sexual violence

measures. Psychopathy was a significant predictor of violence and sexual behavior during

a crime, but not of sexual control. The only significant interaction between sadism scores

and Total PCL-R Scores was in predicting the Violence Severity factor during the sexual

crimes, F (1, 279) = 5.82, p = .017; R2 = .094). Figure 1 illustrates that psychopathy

scores increased violence most for those with the highest sadism scores.

Hierarchical multiple regressions with the four psychopathy facets entered in the

first block and cross-factor facet interactions (F1 x F3, F1 x F4, F2 x F3, F2 x F4) in the

second block were performed to determine the unique contributions of each interaction to

the prediction of all six outcome measures. The relative predictive potencies of the PCL-

R facets and their cross-factor interactions are shown in Table 3. Facet 1 (Interpersonal)

significantly contributed to the prediction of Juvenile and Adult Assault, but negatively

13
predicted violence during the commission of a sexual crime. Facet 2s (Affective) only

contribution was to predicting violence during a sexual crime. Impulsivity (Facet 3) and

Antisociality (Facet 4) predicted all three non-sexual violence measures. Antisociality

also contributed to the prediction of Violence Severity during the sexual crimes.

Interestingly, Antisociality covaried negatively with the Sexual Behavior during the

course of a crime. The inclusion of the cross-factor interactions was largely

inconsequential producing no significant R2s. Facet 1 did, however, interact with Facet

4 to suppress the level of the Sexual Behavior factor during sexual crimes. Figure 2

illustrates that Facet 4 suppressed the Sexual Behavior factor most for those with the

lowest Facet 1 scores.

Pearsons correlation coefficients were calculated to determine the quantitative

relationships between sadism, psychopathy, and each of the PCL-R facets. Table 4

presents the coefficients for these relationships. Sadism has low-order positive

relationships, at the p < .001 significance level, with total PCL-R score, Facet 1, Facet 3,

and Facet 4. The relationship between sadism and Facet 2 (Affective) was statistically

insignificant, r (294) = .09, p = .125.

14
 STUDY 2

Methods

Participants

From a sample of 902 sex offenders who had been evaluated for commitment at a

Massachusetts treatment facility between 1959 and 1991, a subsample of 599 was

selected as part of a follow-up study (Knight & Thornton, 2007). The selection process,

which required record completeness similar to that described in Study 1, is detailed

elsewhere (Knight & Thornton, 2007). The subsample comprised predominantly repeat

offenders, some of whom were deemed to have been sexually dangerous persons and

civilly committed (n = 266; 44%). The sample was predominantly Caucasian (n = 545,

91%), with 54 non-Caucasian offenders (9%), and three offenders whose ethnicities were

undisclosed (<1%). At assessment, offenders ranged in age from 17 to 73 years of age,

with a mean of 36.26 years (SD = 11.54). Forty-six percent of the offenders had assaulted

children under the age of 16 (n = 277), 37 percent had assaulted adults aged 16 years or

more (n = 222), and 17 percent of the offenders had both child and adult victims. Prior to

the age of 17, offenders in this sample amassed a total 139 (M = .23, n = 585) serious

sexual offenses; after the age of 17 onward, they had committed 1,479 (M = 2.50, n =

585) additional sexual offenses.

Measures

Predictor variables. Independent variables included archivally-rated measures of

sexual sadism and of the PCL-R facets.

15
 Sexual sadism ratings. Archival clinical and criminal files were rated using the

two-axis MTC:CM3 Child Molester Typology (Knight et al., 1989; Knight & King, in

press) and the MTC:R3 Rapist Typology (Knight & Prentky, 1990; Knight, 1999, 2010).

The overall strategies for creating and validating the CM3 and R3 have been described in

detail elsewhere (Knight, 2010; Knight & King, in press). In previous studies of

committed sex offenders, the interrater reliability of CM3 Axis I was = .75 and that of

CM3 Axis II was = .79 (Knight, 1988). The overall reliability for CM3s sadism

classification was = .60 (Knight et al., 1989). The interrater reliability of R3 Axis I was

= .68 (Knight, 1999). Child molesters who met the criteria for inclusion in the CM3

Axis II sadistic categories and rapists who met the criteria for inclusion in the R3 sadistic

categories were classified as sexual sadists.

Psychopathy ratings. Similar to Study 1, the archival clinical and criminal files

for the offenders in this study were rated using the PCL-R (Hare, 2003). Two trained

raters who were blind to the hypotheses of the current study completed the archival

ratings. Both raters were trained by the same expert who trained the raters in Study 1,

who was also available to monitor the ratings and answer questions during the coding

process (cf. Knight & Thornton, 2007). Total PCL-R scores and individual facet scores of

dual-rated offenders were the average of the two ratings. Random subsets of offenders in

this study were independently dual-coded and the inter-rater reliability was .79 (n = 177)

for total PCL-R scores; .72 (n = 148) for Facet 1; .59 (n = 169) for Facet 2; .57 (n = 177)

for Facet 3; and .78 (n = 179) for Facet 4.

General violence outcome variables. Study 2 included only archivally-rated

variables.

16
 Unsocialized general aggression (USAG). This same scale used in Study 1 to

measure the amount and frequency of general aggression displayed by the offender

throughout his life prior to imprisonment excluding sexual offense related aggression

was used in Study 2. Here, however, two independent raters blind to the expectations of

the study coded the clinical and institutional files; the interrater reliability of 493

offenders in this sample was .67. When two raters coded an offender, the consensus

rating of the two raters was used.

Juvenile assault. An archivally-rated, rational scale was created, measuring

similar variables to those in its self-report companion MIDSA scale used in Study 1. The

Juvenile Assault scale assessed instances of verbal and physical aggression toward

schoolmates and teachers, instigation and involvement in fights, and truancy and conduct

problems in grammar school and junior high school. This scale does not include instances

of sexual violence. The standardized Cronbach alpha for this scale was .94 on the sample

of 902 offenders from which this subsample was drawn. The interrater reliability of 370

offenders in this sample was .85.

Adult assault. An archivally-rated, rational scale was created as an analogue of

the MIDSA scale used in Study 1. The Adult Assault scale measured instances of verbal

and physical aggression toward coworkers and supervisors, instigation and involvement

in fights, repeated aggressive and destructive behavior, physical aggression, verbal

aggression, and assault arrests and charges. This scale excludes instances of sexual

aggression. The standardized Cronbach alpha for this scale was .91 on the sample of 902

offenders from which this subsample was drawn. The interrater reliability of 397

offenders in this sample was .78.

17
 Sexual crime outcome measures. The exploratory factor analysis of the Study 2

crime scene measures closely paralleled the factors in Study 1.

Violence severity. The same factor scale used in Study 1 was used in Study 2. On

the originating sample of 902 offenders, the scale accounted for 18.09% of the variance,

and the standardized Cronbach alpha was .78.

Sexual control. The same factor scale used in Study 1 was used here. The factor

accounted for 9.77% of the variance in the originating sample, and the standardized

Cronbach alpha for this factor scale was .74.

Sexual behavior. The same factor scale used in Study 1 was used in Study 2, with

the addition of mean instances of the offender masturbating the victim. This variable was

not coded in a mutually exclusive manner with the variable of performing cunnilingus on

the victim as it had been in Study 1. In the sample with 902 offenders, the standardized

internal consistency of this factor scale was .64, and it accounted for 6.18% of the

variance.

Results

A strategy similar to the one used in Study 1 for analyzing the relative

contributions of sadism and psychopathy to aggression was employed in Study 2.

Reverse hierarchical multiple regressions were calculated to determine the unique

contributions of sadism and the psychopathy facets to the prediction of all six outcome

variables. For Study 2 the predictor variables respective R2s when entered in Block 2

after the alternative was entered in Block 1 are also presented in Table 2. The archivally-

rated sadism of Study 2 replicated the predictive patterns of self-report sadism in Study 1.

The psychopathy facets in Study 2 similarly mimicked the predictive patterns of Study 2,

18
except that here the facets did significantly contribute to the prediction of the Sexual

Control factor during sexual crimes, F (4, 437) = 2.62, p = .035. The interaction between

sadism ratings and Total PCL-R scores did not contribute significantly to any of the

models.

As in Study 1, hierarchical multiple regressions with the four psychopathy facets

entered in the first block and cross-factor facet interactions (F1 x F3, F1 x F4, F2 x F3,

F2 x F4) in the second block were calculated to determine the unique contributions of

each to the prediction of all six outcome measures. The relative predictive potencies of

the PCL-R facets and interactions for Study 2 are presented in Table 3. As in Study 1,

Facet 1 (Interpersonal) significantly contributed to the prediction of Juvenile and Adult

Assault and was ineffective in predicting the Sexual Control and Sexual Behavior sexual

crime factors. In contrast, Facet 1 showed no relation to the Violence Severity sexual

crime factor. Facet 2 (Affective) played a more versatile role in prediction for Study 2,

mimicking the prediction of the Violence Severity sexual crime factor in Study 1, but

also contributing to the prediction of all three non-sexual violence measures. Again,

Impulsivity did not predict sexual crime behaviors. Unlike Study 1, however, Facet 3

only predicted Juvenile Assault and not the other non-sexual violence measures.

Antisociality (Facet 4) predicted all six outcomes and consistent with Study 1 it covaried

negatively with sexual behavior during the course of a crime. Cross-factor interactions

did not produce significant R2s in the models. Nevertheless, Facet 1 and Facet 4 did

interact to increase the instances of Juvenile Assault. As can be seen in Figure 3, Facet 4

increased the instances of Juvenile Assault most in those with high Facet 1 scores.

19
 Point biserial correlation coefficients were calculated to determine the

quantitative relationships between sadism, psychopathy, and each of the PCL-R facets.

The coefficients for these relationships are presented in Table 4. Sadism has low-order

positive relationships, at the p < .01 significance level, with total PCL-R score, Facet 1,

and Facet 4. The relationship between sadism and Facet 2 was significant at the p < .05

level; the relationship between sadism and Facet 3 (Impulsivity) was statistically

insignificant, r (489) = .04, p = .329.

20
 GENERAL DISCUSSION

This study addressed several critical issues about sadism and psychopathythe

overall relation between the two, the similarities and differences in the types of

aggressive and sexual crime behaviors to which they relate, and the differential

correlations of the psychopathy facets with aggressive and sexual crime behaviors.

Results were consistent across multimethod assessmentself-report and ratings from

archival filesand different methods of scalingrationally-derived Gutman scales,

factor scales, and categorical judgments. Psychopathy and sadism, although related, are

not co-extensive and capture different aspects of aggression. Psychopathy predicted

USAG, but sadism did not; whereas, both constructs related to the early manifestation of

and the more extreme levels of violence (Juvenile and Adult Assault, respectively).

Within psychopathy, the individual facets covary with different aspects of aggression.

Whereas general aggression relates primarily with antisociality and secondarily with

impulsivity, sexual aggression correlates with affective and antisocial components not

impulsivity. The predominant correlation between antisociality and the non-sexual

aggression variables, the inconsistent covariance of the interpersonal and affective facets

with such violence, and the low incidence of interactions between facets in predicting

non-sexual violence support Kennealy et al.s (2010) assertions that Factor 2 is more

involved in non-sexual violence than Factor 1, and that interactions between Factors 1

and 2 do not seem critical. Similarly, the interactions between sadism and psychopathy

were not instrumental in predicting aggression. The current findings highlight the need to

21
differentiate between types of violence when determining the role that psychopathy facets

and sadism play in prediction.

Relation of Sadism and Psychopathy

Despite the theoretical and clinical overlap of sexual sadism and psychopathy,

little empirical work has specifically studied the relation between the two constructs

(Geberth & Turco, 1997; Holt et al., 1999; Kirsch & Becker, 2007). The current study

corroborated that the two are significantly, positively correlated (Holt et al., 1999; Porter

& Woodworth, 2006). Sadism correlated with total PCL-R, Facet 1, and Facet 4. The

psychopath and the sadists shared desire to control and dominate others, often as a

means of obtaining a goal (Chan & Heide, 2009; Gratzer & Bradford, 1995; Hare et al.,

1999; Hare & Neumann, 2008; Meloy, 1997; Walsh et al., 2007), may account for the

relationship between sadism and Facet 1. Both sadism and psychopathy have been found

to covary with early behavior problemsnamely verbal and physical aggressionthat

continue well into their adult lives (Briken et al., 2006; Brown & Forth, 1997; Grubin,

1994b; Hare & Neumann, 2008; Harpur et al., 1989; Hill et al., 2006; Johnson & Becker,

1997b), which may be reflected in the relation between sadism and Facet 4. Although the

correlations between sadism and the psychopathy dimensions were significant, they

represented low effect sizes, indicating substantial divergences in the two constructs. Not

surprisingly, they each yielded different patterns of covariation with varying types of

violence.

Correlations of Sadism with Violence

Given the association between violence and heightened sexual arousal in sadists

(Breslow, 1989; Herron & Herron, 1982; Holt et al., 1999; Johnson & Becker, 1997b;

22
Kirsch & Becker, 2007; Knight, 2010; Marshall & Kennedy, 2003; Marshall et al., 2002;

Porter & Woodworth, 2006), it was hypothesized that sadism would predict both

increased instances of and elevated severity of aggression. Sadism predicted both

Juvenile and Adult Assault, consistent with the hypothesis that sadism covaries with

increased levels of non-sexual violence throughout the lifespan. Interestingly, sadism did

not correlate with increased levels of non-sexual violence severity, as measured by

USAG. The increased number of assaults in the absence of severity elevations suggests a

stronger relation to behavioral disinhibition or greater reactive aggression to minor

frustrations, possibly a result of the sadists exposure to chronically unstable, abusive

environments (Chan & Heide, 2009; Dietz et al., 1990; Gratzer & Bradford, 1995; Hare

et al., 1999; Hill et al., 2006; MacCulloch, Gray, & Watt, 2000; Warren et al., 1996).

In contrast to its weak prediction of non-sexual violence severity, sadism

significantly predicted the Violence Severity factor measured in sexual crimes. This

correlation, in conjunction with sadisms prediction of the Sexual Control factor,

corroborates that control of a victim, as well as the victims pain and humiliation,

synthesize with sexual arousal in sadists (Grubin, 1994a; MacCulloch et al., 1983;

Marshall, Kennedy, Yates, et al., 2002; Siomopoulus & Goldsmith, 1976; Weinberg,

1987). On the contrary, sexual behaviors that are extraneous to achieving climax, as

measured by the Sexual Behavior factor, were seemingly irrelevant and did not covary

with sadism.

Given the disparities between the predominantly theory-driven hypotheses

presented here and the empirically-derived associations, future researchers should

incorporate more evidence-based operationalizations of sadism into their research. The

23
empirically-derived, behavioral correlates to sadism identified herein would serve as

enhanced criteria for research operationalization as compared to the current practice of

using either DSM-IV-TR (2000) based criteria that require inferences to be made or

idiosyncratic definitions based on clinical experience (Knight & Thornton, 2007;

Marshall & Hucker 2006a, 2006b; Marshall & Kennedy, 2003; Marshall et al., 2002;

Marshall, Kennedy, Yates, et al., 2002; McLawsen et al., 2008; Mollinger, 1982). The

use of frameworks empirically proven to be valid and reliable, such as the MIDSA and

the MTC:R3/CM3 typologies, will allow future research to be analyzed in a collective

manner without fear of obfuscations of sadisms operationalization or measurement.

Correlations of Psychopathy with Violence

As in the case of sadism, there has been debate over the most appropriate means

of measuring psychopathy (Cooke & Michie, 2001; Hare, 2003; Walsh & Kosson, 2008).

The current research used the four facet approach to gain a more refined understanding of

the relationships between the four psychopathic dimensions and violence. Collectively,

the four facets predicted all three measures of general violence, the Violence Severity

factor, and the Sexual Behavior factor, corroborating that psychopathy is associated with

increased instances of violence and violence severity (Gretton et al., 2004; Hare et al.,

1991; Hare & McPherson, 1984; Porter & Woodworth, 2006; Porter et al., 2003; Serin,

1991; Walsh & Kosson, 2008), as well as, a propensity to sexually exploit others (Hare &

Neumann, 2009; Harris et al., 2007; Porter & Woodworth, 2006). The four facets did not

aggregately predict the Sexual Control factor suggesting that the Sexual Violence factor

may be the primary means of obtaining both a victims cooperation and coitus with the

24
Sexual Control factor playing a lesser role (Hare & Neumann, 2009; Hare et al., 1999;

Porter & Woodworth, 2006; Serin, 1991).

The motivational underpinnings of psychopathic behavior are central to the study

of the psychopathic personality. Here, the interpersonal dimension of psychopathy (Facet

1) significantly predicted both Juvenile and Adult Assault, suggesting that psychopaths

successfully used verbal and physical aggression to intimidate and dominate as children

behavior that was solidified by adulthood (Hare & Neumann, 2009; Meloy, 2002; Porter

& Woodworth, 2006). Impulsivity appears to play a crucial role in the young

psychopaths violent behavior, with Facet 3 predicting Juvenile Assault. It does not

appear to play as integral a role as the psychopath ages. The affective component of

psychopathy (Facet 2) relates to the Violence Severity factor of sexual assaults. In such

situations, the attentional biases and difficulties processing affective cues might render

psychopaths less likely to attend to/understand victim distress or to feel guilt/empathy in

relation to victim plight (Blair et al., 1995; Hare et al., 1999). Finally, the Antisocial

dimension (Facet 4) covaried with all measures of violence and violence severity,

illustrating that antisocial/criminal activity begins at a relatively early age for

psychopaths and continues unabated throughout their adult lives (Brown & Forth, 1997;

Hare et al., 1999; Harris et al., 2007; Porter & Woodworth, 2006).

Limitations of the Current Research

Advances in assessment should facilitate much needed research on the etiology,

function, manifestation, and treatment of psychopathy, as well as sadism. The current

research attempted to bring us closer to understanding these two constructs, however, the

research was not without limitations. Even though Study 1 participants were ensured

25
anonymity in their responses on the MIDSA, the accuracy of their responses must be

treated with caution given the host of distortions associated with retrospective data such

as memory failures and biased reporting of stressful experiences (Grabell & Knight,

2009). Similarly, we cannot attest to the accuracy of the information contained within the

archival records used in both studies. Although it is recognized that official records are

sometimes unreliable, this limitation should apply equally to both studies and to whatever

extent unreliability was encountered, it should have contributed unsystematically. Finally,

it is preferable to infer personality traits on the basis of both interview and file data rather

than file data alone, but scoring of the PCL-R exclusively from file review is defensible

for research purposes (Gretton et al., 2004; Hare, 2003; Marcus, Sanford, Edens, Knight,

& Walters, 2011). It is possible that the PCL-R scores in the current study are

underestimates, as is often the case when interviews are not performed, due to the

inability to accurately assess the personality traits associated with psychopathy (Factor 1;

Facets 1 and 2), even though some can be inferred from the archival data (Hare, 2003;

Williamson, Hare, & Wong, 1987). Future studies using file and interview information

are needed to address questions about the relative contribution of interpersonal, affective,

impulsive, and antisocial components of psychopathy to the prediction of violence.

Future Directions

The current study used data from two different samples and incorporated both

self-report and archival-rating methods of measurement to maximize the generalizability

of the findings; however, the data from the current study might have implications only for

convicted male offenders. The author is not in a position to say anything about the

violence of female psychopaths or sadiststwo groups whose existence is not well

26
understood (Kirsch & Becker, 2007; Seto, Khattar, Lalumire, & Quinsey, 1997;

Weinberg, 1987). Similarly, although there are sadists and psychopaths who do not

offend (Dietz et al., 1990; Drukteinis, 1992; Krueger, 2010; Marshall & Kennedy, 2003;

Meloy, 1997; Serin, 1991; Serin, Malcolm, Khanna, & Barbaree, 1994), only those who

committed sexual crimes were the focus of this paper. There is a need to explore whether

offenders versus non-offenders are qualitatively or quantitatively different (Hare et al.,

1999; Hucker, 1997; Krueger, 2010). Finally, research differentiating among subgroups

of offenders (i.e., child molesters versus rapists) might provide insight into whether the

groups have distinctive psychopathology, which could have implications for assessment

and treatment (Ahlmeyer et al., 2003; Beauregard, Stone, Proulx, & Michaud, 2008).

Clearly, a better understanding of psychopathy and sadism are needed so that

earlier and more appropriate intervention and management strategies can be developed.

While there are many logistical and financial reasons to have fixed treatment

programming, the highly varied nature of the psychopathology found here indicates that

the best treatment success will probably be found through relatively narrow treatment

protocols focusing on specific dimensions of pathology (Ahlmeyer et al., 2003; Vitacco

et al., 2005). The existence of a relationship between psychopathy and sadism has

potentially important implications for the treatment and supervision of offenders. For

instance, treatment selection criteria, intervention strategies, and treatment goals may

vary according to the extent to which psychopathy and sadism are present in a particular

offender (Serin et al., 1994). Although the current results provide valuable clues for

treatment strategies, more must be done to assess the risks for pathology development so

27
that someday we may be able to focus our attention on prevention measures for at-risk

youths.

28
 APPE
ENDIX A: T
TABLES

Table
T 1

Sexual
S Crimee Behavior Factor
F Scale Variables a nd Reliabilitties by Ratinng Type

a
Note.
N loaded onnly on mean raating; b loaded only on maxim mum rating; c looaded only on index rating; abb did
ac
no
ot load on indeex rating; didd not load on maximum
m ratingg; *did not loadd in Study 1 duue to mutually
ex
xclusive coding g with Cunnilin
ngus on Victim m.

29
Table
T 2

Relative
R Stren
ngths of Sad
dism, Psychoppathy, and ttheir Interacction in the P
Prediction off
General
G Violeence and Sexxual Crime Behavior
B Faactors

Note.
N The R2 reflects
r the chaange that occurrred when this ccomponent waas entered into Block 2 of the
model
m with the alternative
a 1 the interactioon R2 reflectss the change thhat occurred whhen it
enteered in Block 1;
was
w entered in Block
B 3. Interacctions were callculated using the total PCL-R score, not thhe individual faacets;
in
nsignificant inteeractions were not included ini the table. Saadism ratings inn Study 1were self-report; saddism
raatings in Study 2 were based on data in arch hival records. **p < .05. **p < .01. ***p < .0001.

30
Table
T 3

Relative
R Stren
ngths of Psycchopathy Fa
acets and Crross-Factor IInteractions (F1 x F3, F
F1 x
F4,
F F2 x F3, F2F x F4) in the Predictio
on of Generaal Violence aand Sexual CCrime Behavvior
Factors
F

Note.
N The reflects the standaardized coefficiient for this commponent in thee full model wiith all four faceets
enntered in Blockk 1 and all fourr interactions entered in Blockk 2. Insignificaant interactionss were not inclluded
in
n the table. Juvenile and Adullt Assault ratin ngs in Study 1 wwere self-reporrt; Juvenile andd Adult Assaullt
raatings in Study 2 were based on data in arch hival records. **p < .05. **p < .01. ***p < .0001.

31
Table
T 4

Correlation
C Matrices
M for Sadism, Psyychopathy, aand the Psychhopathy Faccets

Note.
N Number ofo scores correlated provided in parentheses. Study 1 sadissm was measurred via self-repport;
Sttudy 2 sadism was determineed via archival ratings. *p < .005. **p < .01. ***p < .001.

32
 APPE
ENDIX B: F
FIGURES

Figure
F 1. Study 1; n = 29
94. Scatterploot of the inteeraction of loow, moderatte, and high self-
reeported sadissm levels wiith archivally
y-rated total PCL-R scorres on the V Violence Seveerity
faactor during sexual crimes.

33
Figure
F 2. Study 1; n = 233. Scatterploot of the inteeraction of loow, moderatte, and high
Facet 1 (Interrpersonal) sccores with Faacet 4 (Antissocial) scorees on the Sexxual Behavioor
faactor during sexual crimes; holding Facets
F 2 andd 3 constant at zero.

34
Figure
F 3. Study 2; n = 40 09. Scatterploot of the inteeraction of loow, moderatte, and high
Facet 1 (Interrpersonal) sccores with Faacet 4 (Antissocial) scorees on the com
mmission of
Ju
uvenile Assaaults; holding Facets 2 an nd 3 constannt at zero.

35
 REFERENCES
Abel, G. G., & Osborn, C. (1992). The paraphilias: The extent and nature of sexually deviant and criminal
behavior. Psychiatric Clinics of North America, 15(3), 675-687.

Ahlmeyer, S., Kleinsasser, D., Stoner, J., & Retzlaff, P. (2003). Psychopathology of incarcerated sex
offenders. Journal of Personality Disorders, 17(4), 306-318. doi: 10.1521/pedi.17.4.306.23969

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders: DSM-IV-
TR (Fourth ed.). Washington, DC: Author.

Barbaree, H. E., Seto, M. C., Serin, R., C., Amos, N. L., & Preston, D. L. (1994). Comparisons between
sexual and nonsexual rapist subtypes: Sexual arousal to rape, offense precursors, and offense
characteristics. Criminal Justice and Behavior, 21(1), 95-114.

Beauregard, E., & Proulx, J. (2007). A classification of sexual homicide against men. International Journal
of Offender Therapy and Comparative Criminology, 51(4), 420-432.

Beauregard, E., Stone, M.R., Proulx, J., & Michaud, P. (2008). Sexual murderers of children:
Developmental, precrime, crime, and postcrime factors. International Journal of Offender
Therapy and Comparative Criminology, 52(3), 253-269.

Benning, S. D., Patrick, C. J., Hicks, B. M., Blonigen, D. M., & Krueger, R. F. (2003). Factor structure of
the psychopathic personality inventory: Validity and implications for clinical assessment.
Psychological Assessment, 15(3), 340-350. doi: 10.1037/1040-3590.15.3.340

Blair, R. J. R., Jones, L., Clark, F., & Smith, M. (1997). The psychopathic individual: A lack of
responsiveness to distress cues? Psychophysiology, 34(2), 192-198. doi: 10.1111/j.1469-
8986.1997.tb02131.x

Blair, R. J. R., Sellars, C., Strickland, I., Clark, F., Williams, A. O., Smith, M., & Jones, L. (1995). Emotion
attributions in the psychopath. Personality and Individual Differences, 19(4), 431-437. doi:
10.1016/0191-8869(95)00080-p

Breslow, N. (1989). Sources of confusion in the study and treatment of sadomasochism. Journal of Social
Behavior and Personality, 4(3), 263-274.

Briken, P., Habermann, N., Kafka, M. P., Berner, W., & Hill, A. (2006). The paraphilia-related disorders:
An investigation of the relevance of the concept in sexual murderers. Journal of Forensic
Sciences, 51(3), 683-688. doi: 10.1111/j.1556-4029.2006.00105.x

Brown, S. L., & Forth, A. E. (1997). Psychopathy and sexual assault: Static risk factors, emotional
precursors, and rapist subtypes. Journal of Consulting and Clinical Psychology, 65(5), 848-857.

Chan, H. C., & Heide, K. M. (2009). Sexual homicide: A synthesis of the literature. Trauma Violence &
Abuse, 10(1), 31-54. doi: 10.1177/1524838008326478

Cleckley, H. (1976). The mask of sanity (6th ed.). St. Louis: Mosby.

36
Cooke, D. J., & Michie, C. (2001). Refining the construct of psychopathy: Towards a hierarchical model.
Psychological Assessment, 13(2), 171-188. doi: 10.1037//1040-3590.13.2.171

Dietz, P. E., Hazelwood, R. R., & Warren, J. (1990). The sexually sadistic criminal and his offenses.
Bulletin of the American Academy of Psychiatry & the Law, 18(2), 163-178.

Drukteinis, A. M. (1992). Serial murder: The heart of darkness. Psychiatric Annals, 22(10), 532-538.

Fedora, O., Reddon, J. R., Morrison, J. W., Fedora, S. K., Pascoe, H., & Yeudall, L. T. (1992). Sadism and
other paraphilias in normal controls and aggressive and nonaggressive sex offenders. Archives of
Sexual Behavior, 21(1), 1-15.

First, M. B., & Halon, R. L. (2008). Use of DSM paraphilia diagnoses in sexually violent predator
commitment cases. Journal of the American Academy of Psychiatry and the Law, 36(4), 443-454.

Geberth, V. J., & Turco, R. N. (1997). Antisocial personality disorder, sexual sadism, malignant narcissism,
and serial murder. Journal of Forensic Sciences, 42(1), 49-60.

Grabell, A. S., & Knight, R. A. (2009). Examining childhood abuse patterns and sensitive periods in
juvenile sexual offenders. Sexual Abuse-a Journal of Research and Treatment, 21(2), 208-222.
doi: 10.1177/1079063209333133

Gratzer, T., & Bradford, J. M. W. (1995). Offender and offense characteristics of sexual sadists: A
comparative study. Journal of Forensic Sciences, 40(3), 450-455.

Gretton, H. M., Hare, R. D., & Catchpole, R. E. H. (2004). Psychopathy and offending from adolescence to
adulthood: A 10-year follow-up. Journal of Consulting and Clinical Psychology, 72(4), 636-645.
doi: 10.1037/0022-006x.72.4.636

Grubin, D. (1994a). Sexual murder. British Journal of Psychiatry, 165(5), 624-629. doi:
10.1192/bjp.165.5.624

Grubin, D. (1994b). Sexual sadism. Criminal Behaviour and Mental Health, 4(1), 3-9.

Hall, J. R., Benning, S. D., & Patrick, C. J. (2004). Criterion-related validity of the three-factor model of
psychopathy: Personality, behavior, and adaptive functioning. Assessment, 11(4), 4-16.

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

Hare, R. D. (2003). Manual for the Revised Psychopathy Checklist (2nd ed.). Toronto, ON: Multi-Health
Systems.

Hare, R. D., Cooke, D. J., & Hart, S. D. (1999). Psychopathy and sadistic personality disorder. In T.
Millon, P. H. Blaney & R. D. Davis (Eds.), Oxford textbook of psychopathology. (pp. 555-584).
New York, NY US: Oxford University Press.

Hare, R. D., Hart, S. D., & Harpur, T. J. (1991). Psychopathy and the DSM-IV criteria for antisocial
personality disorder. Journal of Abnormal Psychology, 100(3), 391-398.

Hare, R. D., & McPherson, L. M. (1984). Violent and aggressive-behavior by criminal psychopaths.
International Journal of Law and Psychiatry, 7(1), 35-50.

Hare, R. D., & Neumann, C. S. (2008). Psychopathy as a clinical and empirical construct. Annual Review of
Clinical Psychology, 4, 217-246. doi: 10.1146/annurev.clinpsy.3.022806.091452

37
Hare, R. D., & Neumann, C. S. (2009). Psychopathy. In P. H. Blaney & T. Millon (Eds.), Oxford textbook
of psychopathology (2nd ed.). (pp. 622-650). New York, NY US: Oxford University Press.

Harpur, T. J., Hakstian, A. R., & Hare, R. D. (1988). Factor structure of The Psychopathy Checklist.
Journal of Consulting and Clinical Psychology, 56(5), 741-747.

Harpur, T. J., Hare, R. D., & Hakstian, A. R. (1989). Two-factor conceptualization of psychopathy:
Construct validity and assessment implications. Psychological Assessment: A Journal of
Consulting and Clinical Psychology, 1(1), 6-17.

Harris, G. T., Rice, M. E., Hilton, N. Z., Lalumire, M. L., & Quinsey, V. L. (2007). Coercive and
precocious sexuality as a fundamental aspect of psychopathy. Journal of Personality Disorders,
21(1), 1-27. doi: 10.1521/pedi.2007.21.1.1

Hazelwood, R. R., & Warren, J. I. (2000). The sexually violent offender: Impulsive or ritualistic?
Aggression and Violent Behavior, 5(3), 267-279.

Herron, M. J., & Herron, W. G. (1982). Meanings of sadism and masochism. Psychological Reports, 50(1),
199-202.

Hill, A., Habermann, N., Berner, W., & Briken, P. (2006). Sexual sadism and sadistic personality disorder
in sexual homicide. Journal of Personality Disorders, 20(6), 671-684.

Hill, C. D., Neumann, C. S., & Rogers, R. (2004). Confirmatory factor analysis of the Psychopathy
Checklist: Screening Version in offenders with Axis I disorders. Psychological Assessment, 12(1),
90-95.

Holt, S. E., Meloy, J. R., & Strack, S. (1999). Sadism and psychopathy in violent and sexually violent
offenders. Journal of the American Academy of Psychiatry and the Law, 27(1), 23-32.

Hucker, S. J. (1997). Sexual sadism: Psychopathology and theory. In D. R. Laws & W. T. O'Donohue
(Eds.), Sexual deviance: Theory, assessment, and treatment (pp. 194-209). New York, NY:
Guilford Press.

Johnson, B. R., & Becker, J. V. (1997a). Natural born killers? The development of the sexually sadistic
serial killer. Journal of the American Academy of Psychiatry and the Law, 25(3), 335-348.

Johnson, B. R., & Becker, J. V. (1997b). Sexual wickedness. Psychiatric Annals, 27(9), 634-637.

Johnson, J. (1973). 'Psychopathia sexualis.'. British Journal of Psychiatry, 122(567), 211-218. doi:
10.1192/bjp.122.2.211

Kennealy, P. J., Skeem, J. L., Walters, G. D., & Camp, J. (2010). Do core interpersonal and affective traits
of PCL-R psychopathy interact with antisocial behavior and disinhibition to predict violence?
Psychological Assessment, 22(3), 569-580. doi: 10.1037/a0019618

Kirsch, L. G., & Becker, J. V. (2007). Emotional deficits in psychopathy and sexual sadism: Implications
for violent and sadistic behavior. Clinical Psychology Review, 27(8), 904-922. doi:
10.1016/j.cpr.2007.01.011

Knight, R. A. (1988). A taxonomic analysis of child molesters. Annals of the New York Academy of
Sciences, 528, 2-20.

Knight, R. A. (1989). An assessment of the concurrent validity of a child molester typology. Journal of
Interpersonal Violence, 4(2), 131-150. doi: 10.1177/088626089004002001

38
Knight, R. A. (1999). Validation of a typology for rapists. Journal of Interpersonal Violence, 14(3), 303-
330.

Knight, R. A. (2010). Is a diagnostic category for paraphilic coercive disorder defensible? Archives of
Sexual Behavior, 39(2), 419-426. doi: 10.1007/s10508-009-9571-x

Knight, R. A., Carter, D. L., & Prentky, R. A. (1989). A system for the classification of child molesters:
Reliability and application. Journal of Interpersonal Violence, 4(1), 3-23. doi:
10.1177/088626089004001001

Knight, R. A., & Cerce, D. D. (1999). Validation and revision of the multidimensional assessment of sex
and aggression. Psychologica Belgica, 39(2-3), 135-161.

Knight, R. A., & Guay, J. P. (2006). The role of psychopathy in sexual offenders against women. In C. J.
Patrick (Ed.), Handbook of psychopathy. (pp. 512-532). New York, NY: Guilford Press.

Knight, R. A., & King, M. W. (in press). Typologies for child molesters: The generation of a new structural
model. In B. K. Schwartz (Ed.), The Sexual Offender, Vol. 7, Kingston, NJ: Civic Research
Institute, Inc.

Knight, R. A., & Prentky, R. A. (1990). Classifying sexual offenders: The development and corroboration
of taxonomic models. In W. L. Marshall, D. R. Laws, & H. E. Barbaree (Eds.), Handbook of
sexual assault: Issues, theories, and treatment of the offender. (pp. 23-52). New York: Plenum
Press.

Knight, R. A., Prentky, R. A., & Cerce, D. D. (1994). The development, reliability, and validity of an
inventory for the Multidimensional Assessment of Sex and Aggression. Criminal Justice and
Behavior, 21(1), 72-94.

Knight, R. A., & Sims-Knight, J. E. (2003). Developmental antecedents of sexual coercion against women:
Testing of alternative hypotheses with structural equation modeling. In R. A. Prentky, E. Janus, &
M. Seto (Eds.), Sexual coercion: Understanding and management, (pp. 72-85). New York: New
York Academy of Sciences.

Knight, R. A., & Sims-Knight, J. E. (2004). Testing an etiological model for male juvenile sexual offending
against females. Journal of Child Sexual Abuse, 13, 33-55.

Knight, R. A., & Thornton, D. (2007). Evaluating and improving risk assessment schemes for sexual
recidivism: A long-term follow-up of convicted sexual offenders (Final Report No. NCJ 217618).
Retrieved from http://nij.ncjrs.gov/publications

Krafft-Ebing, R. v. (1886/1965). Psychopathia sexualis [Psychopathy of sex; translation]. London, UK:

Greenleaf Publishing.

Krueger, R. B. (2010). The DSM diagnostic criteria for sexual sadism. Archives of Sexual Behavior, 39(2),
325-345. doi: 10.1007/s10508-009-9586-3

Langevin, R., Bain, J., Wortzman, G., Hucker, S., Dickey, R., & Wright, P. (1988). Sexual sadism: Brain,
blood, and behavior. In R. A. Prentky & V. L. Quinsey (Eds.), Human sexual aggression: Current
perspectives. (pp. 163-171). New York, NY US: New York Academy of Sciences.

Lilienfeld, S. O., & Fowler, K. A. (2006) The self-report assessment of psychopathy: problems, pitfalls,
and promises. In C. J. Patrick (Ed.), Handbook of psychopathy. (pp. 107-132). New York, NY:
Guilford Press.

39
Lion, J. R. (1991). Pitfalls in the assessment and measurement of violence: A clinical view. The Journal of
Neuropsychiatry and Clinical Neurosciences, 3(2), S40-S43.

MacCulloch, M., Gray, N., & Watt, A. (2000). Brittain's sadistic murderer syndrome reconsidered: An
associative account of the aetiology of sadistic sexual fantasy. Journal of Forensic Psychiatry,
11(2), 401-418. doi: 10.1080/09585180050142606

MacCulloch, M. J., Snowden, P. R., Wood, P. J. W., & Mills, H. E. (1983). Sadistic fantasy, sadistic
behaviour and offending. British Journal of Psychiatry, 143, 20-29.

Marshall, W. L., & Hucker, S. J. (2006a). Issues in the diagnosis of sexual sadism. Sexual Offender
Treatment, 1(2). Retrieved from http://www.sexual-offender-treatment-org

Marshall, W. L., & Hucker, S. J. (2006b). Severe sexual sadism: Its features and treatment. In R. D.
McAnulty & M. M. Burnette (Eds.), Sex and sexuality, Vol 3: Sexual deviation and sexual
offenses (pp. 227-250). Westport, CT: Praeger Publishers/Greenwood Publishing Group.

Marshall, W. L., & Kennedy, P. (2003). Sexual sadism in sexual offenders: An elusive diagnosis.
Aggression and Violent Behavior, 8(1), 1-22.

Marshall, W. L., Kennedy, P., & Yates, P. (2002). Issues concerning the reliability and validity of the
diagnosis of sexual sadism applied in prison settings. Sexual Abuse: Journal of Research and
Treatment, 14(4), 301-311. doi: 10.1023/a:1019917519457

Marshall, W. L., Kennedy, P., Yates, P., & Serran, G. (2002). Diagnosing sexual sadism in sexual
offenders: Reliability across diagnosticians. International Journal of Offender Therapy and
Comparative Criminology, 46(6), 668-677. doi: 10.1177/0306624x02238161

McLawsen, J. E., Jackson, R. L., Vannoy, S. D., Gagliardi, G. J., & Scalora, M. J. (2008). Professional
perspectives on sexual sadism. Sexual Abuse-a Journal of Research and Treatment, 20(3), 272-
304. doi: 10.1177/1079063208320250

Meloy, J. R. (1997). The psychology of wickedness: Psychopathy and sadism. Psychiatric Annals, 27(9),
630-633.

Meloy, J. R. (2002). The 'polymorphously perverse' psychopath: Understanding a strong empirical

relationship. Bulletin of the Menninger Clinic, 66(3), 273-289. doi: 10.1521/bumc.66.3.273.23368

Mollinger, R. N. (1982). Sadomasochism and developmental stages. Psychoanalytic Review, 69(3), 379-
389.

Multidimensional Inventory of Development, Sex, and Aggression (MIDSA) clinical manual. (2011). (3
ed.). Bend, OR: Augur Enterprises.

Neumann, C. S., Hare, R. D., & Newman, J. P. (2007). The super-ordinate nature of the Psychopathy
Checklist-Revised. Journal of Personality Disorders, 21(2), 102-117. doi:
10.1521/pedi.2007.21.2.102

Porter, S., & Woodworth, M. (2006). Psychopathy and aggression. In C. J. Patrick (Ed.), Handbook of
psychopathy. (pp. 481-494). New York, NY: Guilford Press.

Porter, S., Woodworth, M., Earle, J., Drugge, J., & Boer, D. (2003). Characteristics of sexual homicides
committed by psychopathic and nonpsychopathic offenders. Law and Human Behavior, 27(5),
459-470.

40
Prentky, R. A., & Knight, R. A. Identifying critical dimensions for discriminating among rapists. Journal of
Consulting and Clinical Psychology, 59(5), 643-661.

Salekin, R. T., Neumann, C. S., Leistico, A. M. R., & Zalot, A. A. (2004). Psychopathy in youth and
intelligence: An investigation of Cleckleys hypothesis. Journal of Clinical Child and Adolescent
Psychology, 33(4), 731-742.

Serin, R. C. (1991). Psychopathy and violence in criminals. Journal of Interpersonal Violence, 6(4), 423-
431.

Serin, R. C., Malcolm, P. B., Khanna, A., & Barbaree, H. E. (1994). Psychopathy and deviant sexual
arousal in incarcerated sexual offenders. Journal of Interpersonal Violence, 9(1), 3-11.

Seto, M. C., Khattar, N. A., Lalumire, M. L., & Quinsey, V. L. (1997). Deception and sexual strategy in
psychopathy. Personality and Individual Differences, 22(3), 301-307. doi: 10.1016/s0191-
8869(96)00212-7

Seto, M. C., & Lalumire, M. L. (2000). Psychopathy and sexual aggression. In C. B. Gacono (Ed.), The
clinical and forensic assessment of psychopathy: A practitioner's guide. (pp. 333-350). Mahwah,
NJ US: Lawrence Erlbaum Associates Publishers.

Siomopoulos, V., & Goldsmith, J. (1976). Sadism revisited. American Journal of Psychotherapy, 30(4),
631-640.

Vitacco, M. J., Neumann, C. S., & Jackson, R. L. (2005). Testing a four-factor model of psychopathy and
its association with ethnicity, gender, intelligence, and violence. Journal of Consulting and
Clinical Psychology, 73(3), 466-476. doi: 10.1037/0022-006x.73.3.466

Walsh, Z., & Kosson, D. S. (2008). Psychopathy and violence: The importance of factor level interactions.
Psychological Assessment, 20(2), 114-120. doi: 10.1037/1040-3590.20.2.114

Walsh, Z., Swogger, M. T., Walsh, T., & Kosson, D. S. (2007). Psychopathy and violence: Increasing
specificity. Netherlands Journal of Psychology, 63(4), 136-143. doi: 10.1007/bf03061075

Walters, G. D., Marcus, D. K., Edens, J. F., Knight, R. A., & Sanford, G. M. (2011). In search of the
psychopathic sexuality taxon: Indicator size does matter. Behavioral Sciences & the Law, 29(1),
23-39. doi: 10.1002/bsl.964

Warren, J. I., Hazelwood, R. R., & Dietz, P. E. (1996). The sexually sadistic serial killer. Journal of
Forensic Sciences, 41(6), 970-974.

Weinberg, T. S. (1987). Sadomasochism in the United States: A review of recent sociological literature.
Journal of Sex Research, 23(1), 50-69. doi: 10.1080/00224498709551341

Williamson, S., Hare, R. D., & Wong, S. (1987). Violence: Criminal psychopaths and their victims.
Canadian Journal of Behavioural Science-Revue Canadienne Des Sciences Du Comportement,
19(4), 454-462.

41